12183 World Development Report 1993 Investing in Health Published for the World Bank Oxford University Press Oxford University Press OXFORD NEW YORK TORONTO DELHI BOMBAY CALCUTTA MADRAS KARACHI KUALA LUMPUR SINGAPORE HONG KONG TOKYO NAIROBI DARES SALAAM CAPE TOWN MELBOURNE AUCKLAND and associated companies in BERLIN IBADAN © 1993 The International Bank for Reconstruction and Development I THE WORLD BANK 1818 H Street, N. W, Washington, D.C. 20433 U.S.A. Published by Oxford University Press, Inc. 200 Madison Avenue, New York, N.Y. 10016 Oxford is a registered trademark of Oxford University Press. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Manufactured in the United States of America First printing June 1993 The maps that accompany the text have been prepared solely for the convenience of the reader; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area, or of the authorities thereof, or concerning the delimitation of its boundaries or its national affiliation. The map on the cover, which shows the eight demographic regions used in the analysis in this Report, seeks to convey an impression of the general improvement in health experienced worldwide during the past forty years. ISBN 0-19-520889-7 clothbound ISBN 0-19-620890-0 paperback ISSN 0163-5085 Text printed on recycled paper that conforms to the American National Standard for Permanence of Paper for Printed Library Materials, 239.48-1984 Foreword World Development Report 1993, the sixteenth in this Second, government spending on health should annual series, examines the interplay between hu- , be redirected to more cost-effective programs that man health, health policy, and economic develop- do more to help the poor. Government spending ment. The three most recent reports-on the envi- accounts for half of the $168 billion annual expen- ronment, on development strategies, and on diture on health in developing countries. Too poverty-have furnished an overview of the goals much of this sum goes to specialized care in ter- and means of development. This year's report on tiary facilities that provides little gain for the health, like next year's on infrastructure, examines money spent. Too little goes to low-cost, highly in depth a single sector in which the impact of effective programs such as control and treatment public finance and public policy is of particular of infectious diseases and of malnutrition. Devel- importance. oping countries as a group could reduce their bur- Countries at all levels of income have achieved den of disease by 25 percent-the equivalent of great advances in health. Although an unaccepta- averting more than 9 million infant deaths-by re- bly high proportion of children in the developing directing to public health programs and essential world-one in ten-die before reaching age 5, this clinical services about half, on average, of the gov- number is less than half that of 1960. Declines in ernment spending that now goes to services of low poverty have allowed households to increase con- cost-effectiveness. sumption of the food, clean water, and shelter nec- Third, governments need to promote greater di- essary for good health. Rising educational levels versity and competition in the financing and deliv- have meant that people are better able to apply ery of health services. Government financing of new scientific knowledge to promote their own public health and essential clinical services would and their families' health. Health systems have leave the coverage of remaining clinical services to met the demand for better health through an ex- private finance, usually mediated through insur- panded supply of services that offer increasingly ance, or to social insurance. Government regula- potent interventions. tion can strengthen private insurance markets by Yet developing countries, and especially their improving incentives for wide coverage and for poor, continue to suffer a heavy burden of disease, cost control. Even for publicly financed clinical ser- much of which can be inexpensively prevented or vices, governments can encourage competition cured. (If the child mortality rate in developing and private sector involvement in service supply countries were reduced to the level that prevails in and can help improve the efficiency of the private high-income countries, 11 million fewer children sector by generating and disseminating key infor- would die each year.) Furthermore, increasing mation. The combination of these measures will numbers of developing countries are beginning to improve health outcomes and contain costs while face the problems of rising health system costs enhancing consumer satisfaction. now experienced by high-income countries. Significant reforms in health policy are feasible, This Report advocates a three-pronged ap- as experience in several developing countries has proach to government policies for improving shown. The donor community can assist by fi- health in developing countries. First, governments nancing the transitional costs of change, especially need to foster an economic environment that en= in low-income countries. The reforms outlined in abies households to improve their own health. this Report will translate into longer, healthier, and Growth policies (including, where necessary, eco- more productive lives for people around the nomic adjustment policies) that ensure income world, and especially for the more than 1 billion gains for the poor are essential. So, too, is ex- poor. panded investment in schooling, particularly for The World Health Organization (WHO) has girls. been a full partner with the World Bank at every iii step of the preparation of the Report. I would like World Bank is grateful to them as well . Specific to record my appreciation to WHO and to its many acknowledgments are provided elsewhere in the staff members at global and regional levels who Report. facilitated this partnership . The Report has bene- Like its predecessors, World Development Report fited greatly from WHO's extensive technical ex- 1993 includes the World Development Indicators, pertise. Starting from the Report's conception, which offer selected social and economic statistics WHO participated actively by providing data on on 127 countries. The Report is a study by the various aspects of health development and sys- Bank's staff, and the judgments made herein do tematic input for many technical consultations. not necessarily reflect the views of the Board of Perhaps WHO's most significant contribution was Directors or of the governments they represent . in a jointly sponsored assessment of the global burden of disease, which is a key element of the Report. I look forward to continued collaboration between the World Bank and WHO in the discus- Lewis T. Preston sion and implementation of the messages in this President Report. The United Nations Children's Fund The World Bank (UNICEF), bilateral agencies, and other institu- tions also contributed their expertise, and the May 31, 1993 This Report has been prepared by a team led by Dean T. jamison and comprising jose-Luis Bobadilla, Robert Hecht, Kenneth Hill, Philip Musgrove, Helen Saxenian, jee-Peng Tan, and, part-time, Seth Berkley and Christopher J. L. Murray. Anthony R. Measham drafted and coordinated contributions from the Bank's Population, Health, and Nutrition Department. Valuable contributions and advice were provided by Susan Cochrane, Thomas W. Merrick, W. Henry Mosley, Alexander Preker, Lant Pritchett, and Michael Walton. Extensive input to the Report from the World Health Organization was coordinated through a Steering Committee chaired by jean-Paul jardel. An Advisory Committee chaired by Richard G. A. Feachem provided valuable guidance at all stages of the Report's prepara- tion. Members of these committees are listed in the Acknowledgments . Peter Cowley, Anna E. Maripuu, Barbara J. McKinney, Karima Saleh, and Abdo S . Yazbeck served as research associates, and interns Lecia A. Brown, Caroline) . Cook, Anna Godal, and Vito Luigi Tanzi assisted the team . The work was carried out under the general direction of Lawrence H. Summers and Nancy Birdsall. Many others inside and outside the Bank provided helpful comments and contributions (see the Bibliographical note). The Bank's International Economics Department contributed to the data appen- dix and was responsible for the World Development Indicators . The production staff of the Report included Ann Beasley, Stephanie Gerard, jane Gould, Kenneth Hale, jeffrey N . Lecksell, Nancy Levine, Hugh Nees, Kathy Rosen, and Walton Rosenquist. The support staff was headed by Rhoda Blade-Charest and included Laitan Alii and Nyambura Kimani. Trinidad S. Angeles served as admin- istrative assistant. John Browning was the principal editor, and Rupert Pennant-Rea edited two chapters . Preparation of this Report was immensely, aided by contributions of the participants in a series of consultations and seminars; the subjects and the names of participants are listed in the Acknowledg- ments. The consultations could not have occurred without financial cooperation from the following organizations, whose assistance is warmly acknowledged: the Canadian International Development Association, the Danish International Development Agency, the Edna McConnell Clark Foundation, the Norwegian Ministry of Foreign Affairs, the Rockefeller Foundation, the Swiss Development Cooperation, the U.S . Agency for International Development, the Overseas Development Adminis- tration of the United Kingdom, and the Environmental Health Division and the Special Programme for Research and Training in Tropical Diseases of the World Health Organization. The World Health Organization and the United Nations Children's Fund contributed to the preparation of the statistical appendices. Three academic institutions-the Harvard Center for Population and Development Studies, the London School of Hygiene and Tropical Medicine, and the Swiss Tropical Institute- provided important support for the preparation of the Report . iv Contents Definitions and data notes x Overview 1 Health systems and their problems 3 The roles of the government and of the market in health 5 Government policies for achieving health for all 6 Improving the economic environment for healthy households 7 Investing in public health and essential clinical services 8 Reforming health systems: promoting diversity and competition 11 An agenda for action 13 1 Health in developing countries: successes and challenges 17 Why health matters 17 The record of success 21 Measuring the burden of disease 25 Challenges for the future 29 Lessons from the past: explaining declines in mortality 34 The potential for effective action 35 2 Households and health 37 Household capacity: income and schooling 38 Policies to strengthen household capacity 44 What can be done? · 51 3 The roles of the government and the market in health 52 Health expenditures and outcomes 53 The rationales for government action 54 Value for money in health 59 Health policy and the performance of health systems 65 4 Public health 72 Population-based health services 72 Diet and nutrition 75 Fertility 82 Reducing abuse of tobacco, alcohol, and drugs 86 Environmental influences on health 90 AIDS: a threat to development 99 The essential public health package . 106 5 Clinical services 108 Public and private finance of clinical services 108 Selecting and financing the essential clinical package 112 Insurance and finance of discretionary clinical services 119 Delivery of clinical services 123 Reorienting clinical services and beyond 132 v 6 Health inputs 134 Reallocating investments in facilities and equipment 134 Addressing imbalances in human resources 139 Improving the selection, acquisition, and use of drugs 144 Generating information and strengthening research 148 7 An agenda for action 156 Health policy reform in developing countries 156 International assistance for health 165 Meeting the challenges of health policy reform 170 Acknowledgments 172 Bibliographical note 176 Appendix A. Population and health data 195 Appendix B. The global burden of disease, 1990 213 World Development Indicators 227 Boxes 1 Investing in health: key messages of this Report 6 2 The World Summit for Children 15 1.1 Controlling river blindness 19 1.2 The economic impact of AIDS 20 1.3 Measuring the burden of disease 26 1.4 The demographic and epidemiological transitions 30 2.1 Progress in child health in four countries 38 2.2 Teaching schoolchildren about health: radio instruction in Bolivia 48 2.3 Violence against women as a health issue 50 3.1 Paying for tuberculosis control in China 58 3.2 Cost information and management decisions in a Brazilian hospital 60 3.3 Cost-effectiveness of interventions against measles and tuberculosis 63 3.4 Priority health problems: high disease burdens and cost-effective interventions 64 4.1 Women's nutrition 76 4.2 The Tamil Nadu Integrated Nutrition Project: making supplementary feeding work 80 4.3 World Bank policy on tobacco 89 4.4 After smallpox: slaying the dragon worm 92 4.5 The costs and benefits of investments in water supply and sanitation 93 4.6 Environmental and household control of mosquito vectors 94 4.7 Air pollution and health in Central Europe 97 4.8 Pollution in Japan : prevention would have been better and cheaper than cure 98 4.9 Coping with AIDS in Uganda 104 4.10 HIV in Thailand: from disaster toward containment 105 5.1 Making pregnancy and delivery safe 113 5.2 Integrated management of the sick child 114 5.3 Treatment of sexually transmitted diseases 115 5.4 Short-course treatment of tuberculosis .1 16 5.5 Targeting public expenditure to the poor 119 5.6 Containing health care costs in industrial countries 122 5.7 Health care reform in the OECD 125 5.8 Traditional medical practitioners and the delivery of essential health services 129 5.9 "Managed competition" and health care reform in the United States 132 6.1 International migration and the global market for health professionals 141 6.2 Community health workers 143 vi 6.3 Buying right: how international agencies save on purchases of pharmaceuticals 146 6.4 The contribution of standardized survey programs to health information 149 6.5 Evaluating cesarean sections in Brazil 150 6.6 An unmet need: inexpensive and simple diagnostics for STDs 154 7.1 Community financing of health centers: the Bamako Initiative 159 7.2 Health sector reforms in Chile 162 7.3 Reform of the Russian health system 164 7.4 Health assistance and the effectiveness of aid 168 7.5 World Bank support for reform of the health sector 169 7.6 Donor coordination in the health sector in Zimbabwe and Bangladesh 170 Text figures 1 Demographic regions used in this Report 2 2 Burden of disease attributable to premature mortality and disability, by demographic region, 1990 3 3 Infant and adult mortality in poor and nonpoor neighborhoods of Porto Alegre, Brazil, 1980 7 1.1 Child mortality by country, 1960 and 1990 22 1.2 Trends in life expectancy by demographic region, 1950-90 23 1.3 Age-standardized female death rates in Chile and in England and Wales, selected years 24 1.4 Change in female age-specific mortality rates in Chile and in England and Wales, selected years 24 1.5 Disease burden by sex and demographic region, 1990 28 1.6 Distribution of disability-adjusted life years (DALYs) lost, by cause, for selected demographic regions, 1990 29 1.7 Trends in life expectancy and fertility in Sub-Saharan Africa and Latin America and the Caribbean, 1960-2020 30 1.8 Median age at death, by demographic region, 1950, 1990, and 2030 32 1. 9 Life expectancy and income per capita for selected countries and periods 34 2.1 Mutually reinforcing cycles: reduction of poverty and development of human resources 37 2.2 Child mortality in rich and poor neighborhoods in selected metropolitan areas, late 1980s 40 2.3 Declines in child mortality and growth of income per capita in sixty-five countries 41 2.4 Effect of parents' schooling on the risk of death by age 2 in selected countries, late 1980s 43 2.5 Schooling and risk factors for adult health, Porto Alegre, Brazil, 1987 44 2.6 Deviation from mean levels of public spending on health in countries receiving and not receiving adjustment lending, 1980-90 46 2.7 Enrollment ratios in India, by grade, about 1980 47 3.1 Life expectancies and health expenditures in selected countries: deviations from estimates based on GDP and schooling 54 3.2 Benefits and costs of forty-seven health interventions 62 4.1 Child mortality (in specific age ranges) and weight-for-age in Bangladesh, India, Papua New Guinea, and Tanzania 77 4.2 Total fertility rates by demographic region, 1950-95 82 4.3 Risk of death by age 5 for fertility-related risk factors in selected countries, late 1980s 83 4.4 Maternal mortality in Romania, 1965-91 86 4.5 Trends in mortality from lung cancer and various other cancers among U.S. males, 1930-90 88 4.6 Population without sanitation or water supply services by demographic region, 1990 91 4.7 Simulated AIDS epidemic in a Sub-Saharan African country 100 4.8 Trends in new HIV infections under alternative assumptions, 1990-2000: Sub-Saharan Africa and Asia 101 5.1 Income and health spending in seventy countries, 1990 110 5.2 Public financing of health services in low- and middle-income countries, 1990 117 6.1 The health system pyramid: where care is provided 135 6.2 Hospital capacity by demographic region, about 1990 136 6.3 Supply of health personnel by demographic region, 1990 or most recent available year 140 7.1 Disbursements of external assistance for the health sector, 1990 166 vii Tables 1 Population, economic indicators, and progress in health by demographic region, 1975-90 2 2 Estimated costs and health benefits of the minimum package of public health and essential clinical services in low- and middle-income countries, 1990 10 3 Contribution of policy change to objectives for the health sector 14 1.1 Burden of disease by sex, cause, and type of loss, 1990 25 1.2 Burden of five major diseases by age of incidence and sex, 1990 28 1.3 Evolution of the HIV-AIDS epidemic 3~ 2.1 Poverty and growth of income per capita by developing region, 1985 and 1990, and long- and medium-term trends 42 3.1 Global health expenditure, 1990 52 3.2 Actual and proposed allocation of public expenditure on health in developing countries, 1990 66 3.3 Total cost and potential health gains of a package of public health and essential clinical services, 1990 68 4.1 Burden of childhood diseases preventable by the Expanded Programme on Immunization (EPI) by demographic region, 1990 73 4.2 Costs and health benefits of the EPI Plus cluster in two developing country settings, 1990 74 4.3 Direct and indirect contributions of malnutrition to the global burden of disease, 1990 76 4.4 Cost-effectiveness of nutrition interventions 82 4.5 Estimated burden of disease from poor household environments in demographically developing countries, 1990, and potential reduction through improved household services 90 4.6 Estimated global burden of disease from selected environmental threats, 1990, and potential worldwide reduction through environmental interventions 95 4.7 Costs and health benefits of public health packages in low- and middle-income countries, 1990 106 5.1 Rationales and directions for government action in the finance and delivery of clinical services 109 5.2 Clinical health systems by income group 111 5.3 Estimated costs and health benefits of selected public health and clinical services in low- and middle- income countries, 1990 117 5.4 Social insurance in selected countries, 1990 120 5.5 Strengths and weaknesses of alternative methods of paying health providers 124 5.6 Policies to improve delivery of health care 126 6.1 Annual drug expenditures per capita, selected countries, 1990 145 6.2 Some priorities for research and product development, ranked by the top six contributors to the global burden of disease 152 7.1 The relevance of policy changes for three country groups 157 7.2 Official development assistance for health by demographic region, 1990 167 Appendix tables A.1 Population (midyear) and average annual growth 199 A.2 GNP, population, GNP per capita, and growth of GNP per capita 199 A.3 Population structure and dynamics 200 A.4 Population and deaths by age group 202 A.5 Mortality risk and life expectancy across the life cycle 203 A.6 Nutrition and health behavior 204 A.7 Mortality, by broad cause, and tuberculosis incidence 206 A.8 Health infrastructure and services 208 A.9 Health expenditure and total flows from external assistance 210 A.10 Economies and populations by demographic region, mid-1990 212 B.1 Burden of disease by age and sex, 1990 215 B.2 Burden of disease in females by cause, 1990 216 B.3 Burden of disease in males by cause, 1990 218 viii B.4 Burden of disease by age and the three main groups of causes, 1990 220 B.S Burden of disease by consequence, sex, and age, 1990 221 B.6 Distribution of the disease burden in children in demographically developing economies, 3howing the ten main causes, 1990 222 B.7 Distribution of the disease burden in the adult and elderly populations in demographically developing economies, showing the ten main causes, 1990 223 B.8 Deaths by cause and demographic group, 1990 224 ix Definitions and data notes Selected terms related to health, as used in this calculated as the present value of the future years Report of disability-free life that are lost as the result of the premature deaths or cases of disability occurring in Child mortality. The probability of dying between a particular year. (See Box 1.3 and Appendix B for birth and age 5, expressed per 1,000 live births. further details.) The term under-five mortality is also used . Population-based health services. Services, such as Median age at death. The age below which half of immunization, that are directed toward all mem- all deaths in a year occur. This measure is deter- bers of a specific population subgroup . mined both by the age distribution of the popula- Tertiary care facility. A hospital or other facility tion and by the age pattern of mortality risks . It that offers a specialized, highly technical level of does not represent the average age at which any health care for the population of a large region. group of individuals will die, and it is not directly Characteristics include specialized intensive care related to life expectancy. units, advanced diagnostic support services, and Total fertility rate. The number of children that highly specialized personnel. would be born to a woman if she were to live to the end of her childbearing years and bear children at Country groups each age in accordance with prevailing age-specific fertility rates. For operational and analytical purposes the World Externality. A spillover of benefits or losses from Bank's main criterion for classifying economies is one individual to another. gross national product (GNP) per capita . Every Intervention (in health care). A specific activity economy is classified as low-income, middle-in- meant to reduce disease risks, treat illness, or palli- come (subdivided into lower-middle and upper- ate the consequences of disease and disability. middle), or high-income. Other analytical groups, A/locative efficiency. The extent of optimality in based on regions, exports, and levels of external distribution of resources among a number of com- debt, are also used. peting uses. Because of changes in GNP per capita, the coun- Technical efficiency. The extent to which choice try composition of each income group may change and utilization of input resources produce a spe- from one edition to the next. Once the classifica- cific health output, intervention, or service at low- tion is fixed for any edition, all the historical data est cost. presented are based on the same country group- Cost-effectiveness (in health care). The net gain in ing. The income-based country groupings used in health or reduction in disease burden from a this year's Report are defined as follows. health intervention in relation to the cost. Mea- • Low-income economies are those with a GNP per sured in dollars per disability-adjusted life year capita of $635 or less in 1991. (see next two entries) . • Middle-income economies are those with a GNP Global burden of disease (GBD) . An indicator de- per capita of more than $635 but less than $7,911 in veloped for this Report in collaboration with the 1991. A further division, at GNP per capita of World Health Organization that quantifies the loss · $2,555 in 1991, is made between lower-middle- of healthy life from disease; measured in disabil- income and upper-middle-income economies. ity-adjusted life years. • High-income economies are those with a GNP Disability-adjusted life year (DALY) . A unit used per capita of $7,911 or more in 1991. for measuring both the global burden of disease • World comprises all economies, including and the effectiveness of health interventions, as economies with sparse data and those with less indicated by reductions in the disease burden. It is than 1 million population; these are not shown X separately in the main tables but are presented in The regional grouping of economies in the WDI Table 1a in the technical notes to the World Devel- differs from that used in the main text of this Re- opment Indicators (WDI). port. Part 1 of the table "Classification of econ- omies" at the end of the WDI lists countries by the Demographic regions WDI's income and regional classifications. For purposes of demographic and epidemiological Low-income and middle-income economies are analysis, this year's Report (including its health sometimes referred to as developing economies. data appendices but not the WDI) groups econ- The use of the term is convenient; it is not in- omies into eight demographic regions, defined as tended to imply that all economies in the group are follows: experiencing similar development or that other • Sub-Saharan Africa comprises all countries economies have reached a preferred or final stage south of the Sahara including Madagascar and of development. Classification by income does not South Africa but excluding Mauritius, Reunion, necessarily reflect development status. (In the and Seychelles, which are in the Other Asia and WDI, high-income economies classified as devel- islands group. oping by the United Nations or regarded as devel- • India oping by their authorities are identified by the • China symbol t.) The use of the term "countries" to refer • Other Asia and islands includes the low- and to economies implies no judgment by the Bank middle-income economies of Asia (excluding India about the legal or other status of a territory. and China) and the islands of the Indian and Pa- Analytical groups cific oceans except Madagascar. • Latin America and the Caribbean comprises all For some analytical purposes, other overlapping American and Caribbean economies south of the classifications that are based predominantly on ex- United States, including Cuba. ports or external debt are used, in addition to in- • Middle Eastern crescent consists of the group of come or geographic groups. Listed below are the economies extending across North Africa through economies in these groups that have populations the Middle East to the Asian republics of the for- of more than 1 million. Countries with sparse data mer Soviet Union and including Israel, Malta, and those with less than 1 million population, al- Pakistan, and Turkey. though not shown separately, are included in • Formerly socialist economies of Europe (FSE) in- group aggregates. cludes the European republics of the former Soviet • Fuel exporters are countries for which exports Union and the formerly socialist economies of of petroleum and gas accounted for at least 50 per- Eastern and Central Europe. cent of exports in the period 1987-89. They are • Established market economies (EME) includes all Algeria, Angola, Brunei, Congo, Gabon, Islamic the countries of the Organization for Economic Co- Republic of Iran, Iraq, Libya, Nigeria, Oman, operation and Development (OECD) except Tur- Qatar, Saudi Arabia, Trinidad and Tobago, Turk- key, as well as a number of small high-income menistan, United Arab Emirates, and Venezuela. economies in Europe. • Severely indebted middle-income economies (ab- These eight regions fall into two broad demo- breviated to "Severely indebted" in the WDI) are graphic groups. The first consists of the FSE and twenty-one countries that are deemed to have en- EME, where relatively uniform age distributions countered severe debt-servicing difficulties. These are leading to older populations. The other six re- are defined as countries in which, averaged over gions are referred to as demographically developing, 1989-91, either of two key ratios is above critical in the sense that their age distributions are youn- levels: present value of debt to GNP (80 percent) ger but aging. The demographically developing or present value of debt to exports of goods and all economies correspond approximately to the low- services (200 percent). The twenty-one countries and middle-income economies. Figure 1 of t_he are Albania, Algeria, Angola, Argentina, Bolivia, Overview depicts these regional groups. Table Brazil, Bulgaria, Congo, Cote d'Ivoire, Cuba, Ec- A.10 of Appendix A lists all economies by demo- uador, Iraq, Jamaica, Jordan, Mexico, Mongolia, graphic region and indicates their mid-1990 popu- Morocco, Panama, Peru, Poland, and Syrian Arab lation. Appendix tables A.3 through A.9 provide Republic. demographic and health data by economy within • In the WDI, DECO members, a subgroup of these regions for economies with populations high-income economies, comprises the members greater than 3 million. of the OECD except for Greece, Portugal, and Tur- xi key, which are included among the middle-income Acronyms and initials economies. In the main text of the Report, the term "OECD countries" includes all OECD mem- AIDS Acquired immune deficiency syn- bers unless otherwise stated. drome ARI Acute respiratory infection Data notes BCG Bacillus of Calmette and Guerin vac- cine (to prevent tuberculosis) • Billion is 1,000 million. DALY Disability-adjusted life year • Trillion is 1,000 billion. DPT Diphtheria, pertussis, and tetanus vac- • Tons are metric tons, equal to 1,000 kilograms, cine or 2,204.6 pounds. EPI Expanded Programme on Immuniza- • Dollars are current U.S. dollars unless other- tion (immunization against diphtheria, wise specified. pertussis, tetanus, poliomyelitis, mea- • Growth rates are based on constant price data sles, and tuberculosis) and, unless otherwise noted, have been computed EPI Plus EPI with additional components: im- with the use of the least-squares method. See the munization against hepatitis B and yel- technical notes to the WDI for details of this low fever and, where appropriate, vi- method. tamin A and iodine supplementation • The symbol I in dates, as in "1988/89," means GBD Global burden of disease that the period of time may be less than two years GDP Gross domestic product but straddles two calendar years and refers to a GNP Gross national product crop year, a survey year, or a fiscal year. HIV Human immunodeficiency virus • The symbol .. in tables means not available. HMO Health maintenance organization • The symbol - in tables means not applicable. NGO Nongovernmental organization (In the WDI, a blank is used to mean not OECD Organization for Economic Coopera- applicable.) tion and Development (Australia, Aus- • The number 0 or 0.0 in tables and figures tria, Belgium, Canada, Denmark, Fin- means zero or a quantity less than half the unit land, France, Germany, Greece, shown and not known more precisely. Iceland, Ireland, Italy, Japan, Lux- The cutoff date for all data in the WDI is April embourg, Netherlands, New Zealand, 30, 1993. Norway, Portugal, Spain, Sweden, Historical data in this Report may differ from Switzerland, Turkey, United Kingdom, those in previous editions because of continuous and United States) updating as better data become available, because STD Sexually transmitted disease of a change to a new base year for constant price UNDP United Nations Development Pro- data, or because of changes in country composi- gramme tion of income and analytical groups. UNICEF United Nations Children's Fund Economic and demographic terms are defined in UNPF United Nations Population Fund the technical notes to the WDI. WHO World Health Organization xii Overview Over the past forty years life expectancy has im- ratios are, on average, thirty times as high in de- proved more than during the entire previous span veloping countries as in high-income countries. of human history. In 1950 life expectancy in devel- Although health has improved even in the poor- oping countries was forty years; by 1990 it had est countries, the pace of progress has been un- increased to sixty-three years. In 1950 twenty-eight even. In 1960 in Ghana and Indonesia about one of every 100 children died before their fifth birth- child in five died before reaching age 5-a child day; by 1990 the number had fallen to ten. Small- mortality rate typical of many developing coun- pox, which killed more than 5 million annually in tries. By 1990 Indonesia's rate had dropped to the early 1950s, has been eradicated entirely. Vac- about one-half the 1960 level, but Ghana's had cines have drastically reduced the occurrence of fallen only slightly. Table 1 provides a summary of measles and polio. Not only do these improve- regional progress in mortality reduction between ments translate into direct and significant gains in 1975 and 1990. (Figure 1 illustrates the demo- well-being, but they also reduce the economic bur- graphic regions used in Table 1 and frequently den imposed by unhealthy workers and sick or throughout this Report.) absent schoolchildren. These successes have come In addition to premature mortality, a substantial about in part because of growing incomes and in- portion of the burden of disease consists of disabil- creasing education around the globe and in part ity, ranging from polio-related paralysis to blind- because of governments' efforts to expand health ness to the suffering brought about by severe psy- services, which, moreover, have been enriched by chosis . To measure the burden of disease, this technological progress. Report uses the disability-adjusted life year Despite these remarkable improvements, enor- (DALY), a measure that combines healthy life mous health problems remain. Absolute levels of years lost because of premature mortality with mortality in developing countries remain unac- those lost as a result of disability. ceptably high: child mortality rates are about ten There is huge variation in per person loss of times higher than those in the established market DALYs across regions, mainly because of differ- economies. If death rates among children in poor ences in premature mortality; regional differences countries were reduced to those prevailing in the in loss of DALYs as a result of disability are much rich countries, 11 million fewer children would die smaller (Figure 2) . The total loss of DALYs is re- each year. Almost half of these preventable deaths ferred to as the global burden of disease. are a result of diarrheal and respiratory illne'Ss, The world is facing serious new health chal- exacerbated by malnutrition . In addition, every lenges. By 2000 the growing toll from acquired im- year 7 million adults die of conditions that could be mune deficiency syndrome (AIDS) in developing inexpensively prevented or cured; tuberculosis countries could easily rise to more than 1.8 million alone causes 2 million of these deaths. About deaths annually, erasing decades of hard-won re- 400,000 women die from the direct complications ductions in mortality. The malaria parasite's in- of pregnancy and childbirth. Maternal mortality creased resistance to available drugs could lead to 1 The first six regions named in the key are at intermediate stages of the demographic transition. Figure 1 Demographic regions used in this Report ., ·: · \:· ;·. '" Sub-Saharan Africa Latin America and the Caribbean India Middle Eastern crescent China Formerly socialist economies of Europe Other Asia and islands Established market economies Table 1 Population, economic indicators, and progress in health by demographic region, 1975-90 Income per capita Growth rate, Population, Deaths, 1975-90 Life expectancy at 1990 1990 Dollars, (percent per Child mortality birth (years) Region (millions) (millions) 1990 year) 1975 1990 1975 1990 Sub-Saharan Africa 510 7.9 510 -1.0 212 175 48 52 India 850 9.3 360 2.5 195 127 53 58 China 1,134 8.9 370 7.4 85 43 56 69 Other Asia and islands 683 5.5 1.320 4.6 135 97 56 62 Latin America and the Caribbean 444 3.0 2,190 -0.1 104 60 62 70 Middle Eastern crescent 503 4.4 1,720 -1.3 174 111 52 61 Formerly socialist economies of Europe (FSE) 346 3.8 2,850 0.5 36 22 70 72 Established market economies (EME) 798 7.1 19,900 2.2 21 11 73 76 Demographically developing group• 4,123 39.1 900 3.0 152 106 56 63 FSE and EME 1.144 10.9 14,690 1.7 25 15 72 75 World 5,267 50.0 4,000 1.2 135 96 60 65 Note: Child mortality is the probability of dying between birth and age 5, expressed per 1,000 live births; life expectancy at birth is the average number of years that a person would expect to live at the prevailing age-specific mortality rates . a. The countries of the demographic regions Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent. Source: For income per capita, World Bank data; for other items, Appendix A. 2 a doubling of malaria deaths, to nearly 2 million a role in the developing world is not in doubt. Public year within a decade . Rapid progress in reducing health measures brought about the eradication of child mortality and fertility rates will create new smallpox and have been central to the reduction in demands on health care systems as the aging of deaths caused by vaccine-preventable childhood populations brings to the fore costly noncommuni- diseases. Expanded and improved clinical care has cable diseases of adults and the elderly. Tobacco- saved millions of lives from infectious diseases and related deaths from heart disease and cancers injuries. But there are also major problems with alone are likely to double by the first decade of the health systems that, if not resolved, will hamper next century, to 2 million a year, and, if present progress in reducing the burden of premature smoking patterns continue, they will grow to more mortality and disability and frustrate efforts to re- than 12 million a year in developing countries in spond to new health challenges and emerging dis- the second quarter of the next century. ease threats. • Misallocation. Public money is spent on health Health systems and their problems interventions of low cost-effectiveness, such as surgery for most cancers, at the same time that Although health services are only one factor in ex- critical and highly cost-effective interventions, plaining past successes, the importance of their such as treatment of tuberculosis and sexually The disease burden is highest in poor countries, but disability remains a problem in all regions. Figure 2 Burden of disease attributable to premature mortality and disability, by demographic region, 1990 []] Prem,1turc mortality I2Ll Disability DALYs lost per 1,000 population Sub-Saharan India Middle Otner Asia Latin China Formerly Established Africa Eastern and islands America socialist market crescent and the economies economies Caribbean of Europe Source: Appendix B. 3 transmitted diseases (STDs), remain under- lose out in health because public spending in the funded. In some countries a single teaching hospi- sector is heavily skewed toward high-cost hospital tal can absorb 20 percent or more of the budget of services that disproportionately benefit better-off the ministry of health, even though almost all cost- urban groups. In Indonesia, despite concerted effective interventions are best delivered at lower- government efforts in the 1980s to improve health level facilities. services for the poor, government subsidies to • Inequity. The poor lack access to basic health health for the richest 10 percent of households in services and receive low-quality care. Government 1990 were still almost three times the subsidies spending for health goes disproportionately to the going to the poorest 10 percent of Indonesians. affluent in the form of free or below-cost care in In middle-income countries governments fre- sophisticated public tertiary care hospitals and quently subsidize insurance that protects only the subsidies to private and public insurance. relatively wealthy-a small, affluent minority in the case of private insurance in South Africa and • Inefficiency. Much of the money spent on Zimbabwe and, in Latin America, the larger indus- health is wasted: brand-name pharmaceuticals are trial labor force covered by compulsory public in- purchased instead of generic drugs, health surance (so-called social insurance). The bulk of workers are badly deployed and supervised, and the population, especially the poor, relies heavily hospital beds are underutilized. on out-of-pocket payments and on government • Exploding costs. In some middle-income devel- services that may be largely inaccessible to them. oping countries health care expenditures are grow- In Peru, for example, more than 60 percent of the ing much faster than income. Increasing numbers poor have to travel for more than an hour to obtain of general physicians and specialists, the availabil- primary health care, as compared with less than 3 ity of new medical technologies, and expanding percent of the better-off. The quality of care is also health insurance linked to fee-for-service pay- low: drugs and equipment are in short supply; ments together generate a rapidly growing de- patient waiting times are long and medical consul- mand for costly tests, procedures, and treatments. tations are short; and misdiagnoses and inap- World health spending-and thus also the po- propriate treatment are common. tential for misallocation, waste, and inequitable In the formerly socialist economies, where govern- distribution of resources-is huge. For the world ments have historically been responsible for both as a whole in 1990, public and private expenditure the financing and the delivery of health care, on health services was about $1,700 billion, or 8 health care is free in principle, and wide coverage percent of total world product. High-income coun- of the population has been achieved. This has led tries spent almost 90 percent of this amount, for an to greater apparent equity. But in reality, better-off average of $1,500 per person. The United States consumers make informal out-of-pocket payments alone consumed 41 percent of the global total- to get better care: about 25 percent of health costs more than 12 percent of its gross national product in Romania and 20 percent in Hungary, for exam- (GNP). Developing countries spent about $170 bil- ple, are under-the-table payments for phar- lion, or 4 percent of their GNP, for an average of maceuticals and gratuities to health care providers. $41 per person-less than one-thirtieth the amount Inefficiency is also widespread because the gov- spent by rich countries. ernment-run health system is highly centralized, In the low-income countries government hospitals bureaucratic, and unresponsive to citizens. Gov- and clinics, which account for the greatest part of ernments have been slow to regulate workplace the modern medical care provided, are often ineffi- safety and environmental pollution and have cient, suffering from highly centralized decision- failed to mount effective campaigns against un- making, wide fluctuations in budgetary alloca- healthy personal behaviors-especially alcohol tions, and poor motivation of facility managers consumption and cigarette smoking. In recent and health care workers. Private providers-. years real government spending for health has mainly religious nongovernmental organizations fallen dramatically in the course of the transition to (NGOs) in Africa and private doctors and un- more market-oriented economies. The public sec- licensed practitioners in South Asia-are often tor has suffered from serious shortages of drugs more technically efficient than the public sector and equipment and a lack of skills to manage and offer a service that is perceived to be of higher changing health institutions. The consequences quality, but they are not supported by government have been declining staff morale and falling qual- policies. In low-income countries the poor often ity of care. 4 The roles of the government and of the market ill. A second has to do with "moral hazard": in- in health surance reduces the incentives for individuals to avoid risk and expense by prudent behavior and Three rationales for a major government role in the can create both incentives and opportunities for health sector should guide the reform of health doctors and hospitals to give patients more care systems. than they need. A third has to do with the asym- • Many health-related services such as informa- metry in information between provider and pa- tion and control of contagious disease are public tient concerning the outcomes of intervention; goods. One person's use of health information does providers advise patients on choice of treatment, not leave less available for others to consume; one and when the providers' income is linked to this person cannot benefit from control of malaria- advice, excessive treatment can result. As a conse- carrying mosquitoes while another person in the quence of these last two considerations, in unregu- same area is excluded. Because private markets lated private markets costs escalate without appre- alone provide too little of the public goods crucial ciable health gains to the patient. Governments for health, government involvement is necessary have an important role to play in regulating pri- to increase the supply of these goods. Other health vately provided health insurance, or in mandating services have large externalities: consumption by alternatives such as social insurance, in order to one individual affects others. Immunizing a child ensure widespread coverage and hold down costs. slows transmission of measles and other diseases, If governments do intervene, they must do so conferring a positive externality. Polluters and intelligently, or they risk exacerbating the very drunk drivers create negative health externalities. problems they are trying to solve. When govern- Governments need to encourage behaviors that ments become directly involved in the health sec- carry positive externalities and to discourage those tor-by providing public health programs or fi- with negative externalities. nancing essential clinical services for the poor- • Provision of cost-effective health services to policymakers face difficult decisions concerning the poor is an effective and socially acceptable ap- the allocation of public resources. For any given proach to poverty reduction. Most countries view amount of total spending, taxpayers and, in some access to basic health care as a human right. This countries, donors want to see maximum health perspective is embodied in the goal, "Health for gain for the money spent. An important source of All by the Year 2000," of the conference held by guidance for achieving value for money in health the World Health Organization (WHO) and the spending is a measure of the cost-effectiveness of United Nations Children's Fund (UNICEF) at different health interventions and medical pro- Alma-Ata in 1978, which launched today's pri- cedures-that is, the ratio of costs to health bene- mary health care movement. Private markets will fits (DALYs gained). not give the poor adequate access to essential clini- Until recently, little has been done to apply cost- cal services or the insurance often needed to pay effectiveness analysis to health. This is, in part, for such services. Public finance of essential clini- because it is difficult. Cost and effectiveness data cal care is thus justified to alleviate poverty. Such on health interventions are often weak. Costs vary public funding can take several forms: subsidies to between countries and can rise or fall sharply as a private providers and NGOs that serve the poor; service is expanded. Some groups of interventions vouchers that the poor can take to a provider of are provided jointly, and their costs are shared. their choice; and free or below-cost delivery of Nonetheless, cost-effectiveness analysis is already public services to the poor. demonstrating its usefulness as a tool for choosing • Government action may be needed to com- among possible health interventions in individual pensate for problems generated by uncertainty and countries and for addressing specific health prob- insurance market failure. The great uncertainties sur- lems such as the spread of AIDS. rounding the probability of illness and the effica_cy Just because a particular intervention is cost- of care give rise both to strong demand for insur- effective does not mean that public funds should ance and to shortcomings in the operation of pri- be spent on it. Households can buy health care vate markets. One reason why markets may work with their own money and, when well informed, poorly is that variations in health risk create incen- may do this better than governments can do it for tives for insurance companies to refuse to insure them. But households also seek value for money, the very people who most need health insurance- and governments, by making information about those who are already sick or are likely to become cost-effectiveness available, can often help im- 5 Box 1 Investing in health: key messages of this Report This Report proposes a three-pronged approach to provements and a productive asset-better health - to government policies for improving health. the poor. • Improve management of government health ser- Foster a11 cnvir011111ellf that e11ables llousellolds vices through such measures ,,s decentralization of ad- lo improve Ilea/Ill ministrati\'l' and budgl'lary authority and contracting Household decisions shape health, but these decisions out of services. are constrained by the income and education of house- Prcmrofe diversity alilt competition hold members. In addition to promoting overall L 'Co- nomic growth, governments can help to improvL' those Government finance of public health and of a nation- decisions if they : ally defined package of essential clinical services would • Pursue economic growth policies that will benefit leave the remaining cl inical services to be financed pri- the poor (including, where necessary, adjustment poli- vately or by social insurance within the context of a cies that preserve cost-effective health expenditures) policy framework establishL•d by the government. Gov- • Expand investment in schooling, particularly for ernments Ciln promote diversity and compl'lition in girls provision of health services and insurance by adopting • Promote the rights and status of women through policies that: political and economic empowerment and legal protec- • Encourage social or private insurance (with regula- tion against abuse. tory incentives for equitable access and cost contain- ment) for clinical services outside the ess~ntial Improve govem11renf spe11di11g 011 health p<1ckage. • Encourage suppliers (both public and private) to The challenge for most governments is to concentrate compL•te both to deliver clinical services and to provide resources on compensating for market failures and effi- inputs, such as drugs , to publicly and privatL·Iy fi" ciently financing services that will particularly benefit nanced health services. Domestic suppliers should not the poor. Several directions for policy respond to this be protected from international competition. challenge: • Generate and disseminate information on pro- • Reduce government expenditures on tertiary facil- vider performance, on essential equipment and drugs, ities, specialist training, and interventions that provide on the costs and effectiveness of interventions, and on little health gain for the money spent. the accredit,1tion status of institutions and providers. • Finance and implement a package of public health interventions to deal with the substantial externalities Increased scientific knowledge has accounted for much surrounding infectious disease control, prevention of of the dramatic improvement in health that hils oc- AIDS, environmental pollution , and behaviors (such as curred in this century-by providing information that drunk driving) that put others at risk. forms the basis of household and government action • Finance and ensure delivery of a package of essen- and by underpinning the development of preventive, tial clinical services. The comprehensivenc·ss and com- curative, and diagnostic technologies. lnvestml•nt in position of such a package can only be defined by each continued scientific advance will amplify the effective- country, taking into account epidemiological condi- ness of e<1ch element of the three-pronged <1pproach tions, local preferences, and income. In most countries proposed in this Report. Because thL' fruits of science public finance, or publicly mandated finance, of the benefit all countries, internationaily collabor<1tive ef- essential clinical package would provide a politically forts, of which there are several excellent exampll's, acceptable mechanism for distributing both welfare im- will often be the right way to proceed . prove the decisions of private consumers, pro- portance of continued investment in scientific viders, and insurers. advance. • Since overall economic growth-particularly Government policies for achieving health for all poverty-reducing growth-and education are cen- tral to good health, governments need to pursue This Report focuses primarily on the relation be- sound macroeconomic policies that emphasize re- tween policy choices, both inside and outside the duction of poverty. They also need to expand basic health sector, and health outcomes, especially for schooling, especially for girls, because the way in the poor. Box 1 summarizes the Report's three key which households, particularly mothers, use in- messages for government policy and notes the im- formation and financial resources to shape their 6 dietary, fertility, health care, and other life-style The poor suffer far higher levels of mortality at all choices has a powerful influence on the health of ages than do the rich. household members. • Governments in developing countries should spend far less-on average, about 50 percent less- Figure 3 Infant and adult mortality in poor than they now do on less cost-effective interven- and nonpoor neighborhoods of tions and instead double or triple spending on ba- Porto Alegre, Brazil, 1980 sic public health programs such as immunizations and AIDS prevention and on essential clinical ser- Infant mortality Adult mortality vices . A minimum package of essential clinical ser- (ages 45-64) vices would include sick-child care, family plan- ning, prenatal and delivery care, and treatment for Deaths per 1,000 Deaths per year per 100,000 tuberculosis and STDs. Low-income countries live births persons in age group would have to redirect current public spending for 50 2,000 health and increase expenditures (by government, donors, and patients) to meet needs for public 40 health and the minimum package of essential clini- 1,500 cal services for their populations; less reallocation would be needed in middle-income countries. Ter- 30 tiary care and less cost-effective services will con- 1,000 tinue, but public subsidies to them, if they mainly benefit the wealthy, should be phased out during a 20 transitional period. • Because competition can improve quality and 500 drive down costs, governments should foster com- 10 petition and diversity in the supply of health ser- vices and inputs, particularly drugs, supplies, and 0 0 equipment. This could include, where feasible, Infants Males Females private supply of health care services paid for by governments or social insurance. There is also con- • Poor • Nonpoor siderable scope for improving the quality and effi- ciency of government health services through a Note: Poor neighborhoods were defined according to combination of decentralization, performance- specific criteria. They are, broadly, squntter settlements based incentives for managers and clinicians, and with substandard housing and infrustructure. Source: Barcellos und others 1986. related training and development of management systems . Exposing the public sector to competition with private suppliers can help to spur such im- countries in which average incomes rose by more provements. Strong government regulation is also than 1 percent a year. Economic policies conducive crucial, including regulation of privately delivered to sustained growth are thus among the most im- health services to ensure safety and quality and of portant measures governments can take to im- private insurance to encourage universal access to prove their citizens' health. coverage and to discourage practices-such as fee- Of these economic policies, increasing the in- for-service payment to providers reimbursed by a come of those in poverty is the most efficacious for "third-party" insurer-that lead to overuse of ser- improving health. The reason is that the poor are vices and escalation of costs. most likely to spend additional income in ways that enhance their health: improving their diet, ob- Improving the economic environment taining safe water, and upgrading sanitation and for healthy households housing. And the poor have the greatest remain- Advances in income and education have allowed ing health needs, as Figure 3 illustrates for Porto households almost everywhere to improve their Alegre, Brazil. Government policies that promote health. In the 1980s, even in countries in which equity and growth together will therefore be better average incomes fell, death rates of children under for health than those that promote growth alone. age 5 declined by almost 30 percent. But the child In the 1980s many countries undertook macro- mortality rate fell more than twice as much in economic stabilization and adjustment programs 7 designed to deal with severe economic imbalances tions are currently saving an estimated 3 million and move the countries onto sustainable growth lives a year. Social marketing of condoms to pre- paths. Such adjustment is clearly needed for long- vent transmission of human immunodeficiency vi- run health gains. But during the transitional pe- rus (HIV) has proved highly successful in Uganda, riod, and especially in the earliest adjustment pro- Zaire, and elsewhere. Information on the risks of grams, recession and cuts in public spending smoking, and taxes on both tobacco and alcohol, slowed improvements in health. This effect was are changing behavior in some countries-al- less than originally feared, however-in part be- though mostly, so far, in the richer countries. cause earlier expenditures for improving health Governments need to expand these efforts and and education had enduring effects. As a result of to move forward with other promising public this experience, most countries' adjustment pro- health initiatives. Several activities stand out be- grams today try to rationalize overall government cause they are highly cost-effective: the cost of spending while maintaining cost-effective expen- gaining one DALY can be remarkably low-some- ditures in health and education. Despite these im- times less than $25 and often between $50 and provements, much is still to be learned about more $150. Activities in this category include: efficient ways of carrying out stabilization and ad- • Immunizations justment programs while protecting the poor. • School-based health services Policies to expand schooling are also crucial for • Information and selected services for family promoting health. People who have had more planning and nutrition schooling seek and utilize health information more • Programs to reduce tobacco and alcohol effectively than those with little or no schooling. consumption This means that rapid expansion of educational • Regulatory action, information, and limited opportunities-in part by setting a high minimum public investments to improve the household standard of schooling (say, six full years) for all-is environment a cost-effective way of improving health. Educa- • AIDS prevention. tion of girls and women is particularly beneficial to Intensified government support is required to household health because it is largely women who extend the Expanded Programme on Immuniza- buy and prepare food, maintain a clean home, care tion (EPI), which currently protects about 80 per- for children and the elderly, and initiate contacts cent of the children in the developing world with the health system. Beyond education, gov- against six major diseases at a cost of about $1.4 ernment policies that support the rights and eco- billion a year. Expanding EPI coverage to 95 per- nomic opportunities of women also contribute to cent of all children would have a significant impact overall household well-being and better health. on children in poor households, who make up a disproportionately large share of those not yet Investing in public health reached by the EPI. Other vaccines, particularly and essential clinical services those for hepatitis B and yellow fever, could be added to the six currently included in the EPI, as The health gain per dollar spent varies enormously could vitamin A and iodine supplements. In most across the range of interventions currently fi- developing countries such an "EPI Plus" cluster of nanced by governments. Redirecting resources interventions in the first year of life would have from interventions that have high costs per DALY the highest cost-effectiveness of any health mea- gained to those that cost little could dramatically sure available in the world today. reduce the burden of disease without increasing A second high priority for governments should expenditures. A limited package of public health be to provide inexpensive and highly efficacious measures and essential clinical interventions is a medications to treat school-age children afflicted top priority for government finance; some govern- with schistosomiasis, intestinal worm infections, ments may wish, after covering that minimum for and micronutrient deficiencies. Treatment of these everyone, to define their national essential pack- conditions through distribution of medications age more broadly. and micronutrient supplements in schools would greatly improve the health, school attendance, and Public health learning achievement of hundreds of millions of children, at a cost of $1 to $2 per child per year. In Government action in many areas of public health addition to treatment, schoolchildren can be has already had an important payoff. lmmuniza- taught by their teachers or by radio about the hu- 8 man body and about avoiding risks to health-for gun to spread through human populations, it has example, from smoking or unsafe sex. so far caused 2 million deaths and infected about Governments need to encourage healthier be- 13 million individuals. Some parts of the develop- haviors on the part of individuals and households ing world are already heavily infected: in Sub- by providing information on the benefits of breast- Saharan Africa an average of one in forty adults feeding and on how to improve children's diets. has the virus, and in certain cities the rate is one in Programs in Colombia, Indonesia, and elsewhere three. In Thailand one adult in fifty is infected. show the potential for success. Information on the More than 90 percent of the infected individuals benefits of family planning and on the availability are in their economically most productive years, of family planning services is also critical. Govern- ages 15-40. They will be developing AIDS and ment dissemination of this information can take a dying over the next decade. Projections of the fu- number of creative forms, as the effective use of ture course of the epidemic are gloomy: conserva- radio drama and folk theater in Kenya and Zim- tive estimates from WHO are that by 2000, 26 mil- babwe demonstrates. lion individuals will be HIV-infected and 1.8 Measures to control the use of tobacco, alcohol, million a year will die of AIDS. By destroying indi- and other addictive substances-through informa- viduals' immune systems, HIV will also vastly tion campaigns, taxes, bans on advertising, and, worsen the spread of other diseases, especially tu- in certain cases, import controls-can help sub- berculosis. In highly affected areas demand for stantially to reduce chronic lung disease, heart dis- AIDS treatment will overwhelm capacity for clini- ease, cancer, and injuries. Unless smoking behav- cal treatment and cause a deterioration of care for ior changes, three decades from now premature other illnesses. deaths caused by tobacco in the developing world What governments need to do is clear: intervene will exceed the expected deaths from AIDS, tuber- early, before a major epidemic gets under way. culosis, and complications of childbirth combined. Countries as diverse as Bangladesh, Ghana, and Governments must do more to promote a Indonesia share the preconditions for rapid trans- healthier environment, especially for the poor, mission of HIV-substantial numbers of pros- who face greatly increased health risks from poor titutes and high rates of prevalence of other STDs, sanitation, insufficient and unsafe water supplies, such as syphilis, gonorrhea, and chancroid, which poor personal and food hygiene, inadequate gar- facilitate the spread of the AIDS virus. Strong pub- bage disposal, indoor air pollution, and crowded lic action is required to reduce HIV transmission. and inferior housing. Collectively, these risks are Particularly important are efforts targeted to high- associated with nearly 30 percent of the global bur- risk groups: information to promote change in den of disease. To help the poor improve their sexual behavior; distribution of condoms; and household environments, governments can pro- treatment for other STDs. Early reduction in HIV vide a regulatory and administrative framework transmission by high-risk individuals is very cost- within which efficient and accountable providers effective, but later in an AIDS epidemic the cost- (often in the private sector) have an incentive to effectiveness of interventions declines substan- offer households the services they want and are tially. Current expenditures on AIDS prevention willing to pay for, including water supply, sanita- in developing countries-totaling less than $200 tion, garbage collection, clean-burning stoves, and million a year-are woefully inadequate. Five to housing. The government has a vital role in dis- ten times this level of spending is needed to deal seminating information about hygienic practices. with the emerging epidemic. It can also improve the use of public resources by eliminating widespread subsidies for water and Essential clinical services sanitation that benefit the middle class. Govern- ment legislation and regulations to increase secu- The components of a package of essential clinical rity of land tenure for the poor would encourage services of high cost-effectiveness will vary from low-income families to invest more in safer, country to country, depending on local health healthier housing. needs and the level of income. At a minimum, the A special challenge for concerted public health package should include five groups of interven- action is to reduce the spread of AIDS. The AIDS tions each of which addresses very large disease epidemic has already become a dominant public burdens. The five groups are: health concern in many countries. Although HIV, • Services to ensure pregnancy-related (prena- the virus that causes AIDS, has only recently be- tal, childbirth, and postpartum) care; strength- 9 ened efforts could prevent most of the almost half- broader range of interventions than this mtm- million maternal deaths that occur each year in mum. At modest increases in spending, relatively developing countries. cost-effective measures for the treatment of some • Family planning services; improved access to common noncommunicable conditions could be these services could save as many as 850,000 chil- included. Examples are low-cost protocols for dren from dying every year and eliminate as many treatment of heart disease using aspirin and anti- as 100,000 of the maternal deaths that occur hypertensive drugs; treatment for cervical cancer; annually. drug treatment of some psychoses; and removal of • Tuberculosis control, mainly through drug cataracts. therapy, to combat a disease that kills more than 2 Many health services have such low cost-effec- million people annually, making it the leading tiveness that governments will need to consider cause of death among adults. excluding them from the essential clinical package. • Control of STDs, which account for more than In low-income countries these might include heart 250 million new cases of debilitating and some- surgery; treatment (other than pain relief) of times fatal illness each year. highly fatal cancers of the lung, liver, and stom- • Care for the common serious illnesses of ach; expensive drug therapies for HIV infection; young children-diarrheal disease, acute respira- and intensive care for severely premature babies. tory infection, measles, malaria, and acute malnu- It is hard to justify using government funds for trition-which account for nearly 7 million child these medical treatments at the same time that deaths annually. much more cost-effective services which benefit These clinical interventions are all highly cost- mainly the poor are not adequately financed. effective-often costing substantially less than $50 Widespread adoption of an essential clinical per DALY gained. package would have a tremendous positive impact A minimal package of essential clinical services on the health of people in developing countries. If would also include some treatment for minor in- 80 percent of the population were reached, 24 per- fection and trauma and, for health problems that cent of the current burden of disease in low- cannot be fully resolved with existing resources, income countries and 11 percent of that in middle- advice and alleviation of pain. The provision of income countries could be averted (Table 2). The hospital-based emergency care other than the in- estimated impact of implementing the minimum terventions mentioned above would depend on clinical services is more than twice that for the day-to-day capacity and availability of resources. public health package outlined above; when com- This emergency care includes, for example, treat- bined with the public health package, the share of ment of most fractures, as well as appendec- current illness that could be eliminated rises to tomies. Depending on resource availability and so- perhaps 32 percent for low-income countries and cial values, some countries may define their 15 percent for middle-income countries. This re- essential clinical package to include a much duction in disease is equivalent, in terms of DALYs Table 2 Estimated costs and health benefits of the minimum package of public health and essential clinical services in low- and middle-income countries, 1990 Cost Cost Approximate (dollars per capita as a percentage reduction in burden Group per year) of income per capita of disease (percent) Low-income countries (Income per capita = $350) Public health 4.2 1.2 8 Essential clinical services a 7.8 2.2 24 Total 12.0 3.4 32 Middle-income countries (Income per capita = $2,500) Public health 6.8 0.3 4 Essential clinical services a 14.7 0.6 11 Total 21.5 0.9 15 a. The estimated costs and benefits are for a minimum essential package of clinical services, as defined in the text. Many countries may wish, if they have the resources, to define their essential clinical package more broadly. Source: World Bank calculations. 10 gained, to saving the lives of more than 9 million groups can lead to erosion of political support for infants each year. the essential package and to decreased funding and lower quality of care. Furthermore, problems Paying for the package of cost escalation and access to insurance on the part of high-risk groups can complicate private fi- The most sophisticated facility required to deliver nance. For these reasons, in most member coun- the minimum elements of the essential clinical tries of the Organization for Economic Coopera- package is a district hospital. Providing services in tion and Development (OECD), governments lower-level facilities allows costs to be contained at finance (or mandate the financing of) comprehen- modest levels for minimal versions of the essential sively defined essential packages for virtually all clinical package. The cost is about $8 per person their citizens. each year in low-income countries and $15 in In low-income countries, where current public middle-income countries. The cost differences are spending for health is less than the cost of an es- the result of distinct demographic structures, epi- sential package, some degree of targeting is inevi- demiological conditions, and labor costs in the two table. If the wealthy are already opting out of gov- settings. When the cost of the public health inter- ernment-financed services because of the higher ventions described above is added, total costs rise quality and convenience of privately financed ser- to $12 per capita in low-income countries and $22 vices, targeting is fairly easy. Community-financ- per capita in middle-income countries. ing schemes, whereby patients at local health cen- Adoption of the package in all developing coun- ters and pharmacies pay modest fees, are another tries would require a quadrupling of expenditures option that can help both to improve the quality of on public health, from $5 billion at present to $20 care and, when fees are retained and managed lo- billion a year, and an increase from about $20 bil- cally, to sustain services. A large number of coun- lion to $40 billion in spending on essential clinical tries in Africa have had some early success with services. In the poorest countries governments community financing as part of the Bamako Initia- typically spend about $6 per person for health and tive led by UNICEF and WHO. Nonetheless, expe- total health expenditures are about $14 per person. rience to date suggests that introduction of user There, paying for an essential package will require fees at levels that do not discourage the poor is a combination of increased expenditures by gov- likely to be more useful for improving technical ernments, donor agencies, and patients and some efficiency (for example, by facilitating drug sup- reorientation of current public spending for ply) than for raising substantial revenues on ana- health. In middle-income countries, where public tionwide basis. spending for health averages $62 per person, the $22 cost of the package is financially feasible if the Reforming health systems: promoting diversity political commitment exists for shifting existing re- and competition sources away from discretionary services with lower cost-effectiveness toward public health pro- Ensuring basic public health services and essential grams and essential clinical care. These major clinical care while the rest of the health system changes cannot be made overnight, but it is impor- becomes self-financed will require substantial tant to start and complete them as swiftly as possi- health system reforms and reallocations of public ble, before interest groups and bureaucratic inertia spending. Only by reducing or eliminating spend- undermine reform. ing on discretionary clinical services can govern- A critical question in designing an essential clini- ments concentrate on ensuring cost-effective clini- cal package is the extent of government financing. cal care for the poor. One way to do so is by Should governments pay for everyone, or only for charging fees to affluent patients who use govern- the poor? The main problem with universal gov- ment hospitals and services. In Chile, Kenya, ernment financing is that it subsidizes the wealthy, Lesotho, and other countries governments are who could afford to pay for their own services, increasing user fees for the wealthy and for those and thus leaves fewer government resources for covered by insurance and are strengthening the the poor. A policy requiring those who can pay all legal and administrative systems for billing pa- or part of their own costs to do so may make sense tients and collecting revenues. on equity grounds, but it also has disadvantages. Promoting self-financed insurance, thus elim- Often, the administrative costs of targeting are inating large and inequitable subsidies to the more high, and exclusion of wealthy and middle-income affluent groups who are covered by insurance, 11 would also help to free government funds for pub- keep patients longer than necessary and are lic health programs and essential clinical care. Sub- poorly organized and managed. Countries pay too sidies in the form of tax relief for contributions to much for drugs of low efficacy, and drugs and sup- private insurance are equal to nearly a fifth of total plies are stolen or go to waste in government ware- government spending for health in South Africa. houses and hospitals. In Latin America subsidies to the social insurance In the short term, reforms in pharmaceutical us- systems are widespread and include tax relief, di- age offer the greatest gains in efficiency. Govern- rect transfers to cover the operating deficits of so- ments that have introduced competition in the cial security health funds, and matching govern- procurement of drugs have typically achieved sav- ment funds for employee payroll contributions. ings of 40 to 60 percent. Governments can also Where these subsidies benefit only the better-off in develop national essential drug lists, consisting of society, they need to be scaled back. a limited number of inexpensive drugs that ad- Reforms entail shifting new government spend- dress the important health problems of the popu- ing for health away from specialized personnel, lation. Many countries have such lists, but not all equipment, and facilities at the apex of health sys- use them to guide the selection and procurement tems and "down the pyramid" toward the broad of drugs for the public sector. New treatment pro- base of widely accessible care in community facili- tocols and alternative uses of facilities can also ties and health centers. Very few cost-effective in- raise efficiency. Outpatient surgery can replace terventions depend on sophisticated hospitals and some procedures customarily performed on an in- specialized physicians-all the services contained patient basis, at considerable savings. in the minimum essential clinical package pro- In the long run, decentralization can help to in- posed in this Report can be provided by health crease efficiency when there is adequate capacity centers and district hospitals. Yet specialized facili- and accountability at lower levels of the national ties everywhere absorb a large amount of public health system. Some countries, such as Botswana resources, a problem that has frequently been ex- and Ghana, have delegated a wide range of man- acerbated by donor investments in tertiary care fa- agement responsibilities to regional and district- cilities. In the 1980s Papua New Guinea, to correct level offices of the ministry of health; others, overconcentration of resources on higher-level fa- including Chile and Poland, have devolved au- cilities, limited public spending on hospitals to 40 thority and resources to local government agen- percent of the recurrent budget of the Ministry of cies. Their experience provides evidence that Health-well below the level in most developing success is possible-but also that hasty and countries. unplanned decentralization, sometimes purely in Governments need to use more effective policies response to political pressures, can create new for financing training (including use of national problems. service mechanisms) to help meet the need for pri- Greater reliance on the private sector to deliver mary care providers, particularly nurses and mid- clinical services, both those that are included by a wives, and for public health, health policy, and country in its essential package and those that are management personnel. At the same time, gov- discretionary, can help raise efficiency. The private ernments should limit or eliminate subsidies for sector already serves a large and diverse clientele specialist training. Increased government support in developing countries and often delivers services for health information systems and operations re- of higher quality without the long lines and inade- search would help to guide public policies for quate supplies frequently found in government fa- health. Estimates of the national burden of disease cilities. In many countries private doctors and along the lines of the global burden of disease pharmacies face unnecessary legal and administra- methodology used in this Report, and local infor- tive barriers, and these need to be removed. But mation on the cost-effectiveness of different inter- the tendency for profit-making providers to over- ventions, would enable governments to establish prescribe drugs, procedures, and diagnostics health priorities. needs to be countered; encouraging the for-profit In every developing country decisive steps are sector to move away from fee-for-service to pre- needed to correct the pervasive inefficiency of clin- paid coverage (through, for example, encouraging ical health programs and facilities and especially of health maintenance organizations) is one feasible government services. Clinics and outreach pro- approach. grams operate poorly because of shortages of Governments could also subsidize private drugs, transport, and maintenance. Hospitals health care providers who deliver essential clinical 12 services to the poor. This is already beginning to private health maintenance organizations and in happen and needs to go further. In many African the British National Health Service. Another is for countries, including Malawi, Uganda, and Zam- insurers jointly to negotiate uniform fees with doc- bia, governments subsidize the operating expendi- tors and hospitals, as is done in Japan's social in- tures of church hospitals and clinics in rural areas surance system and Zimbabwe's private medical and the training of their health personnel. In Ban- aid insurance system; or insurers themselves can gladesh, Kenya, Thailand, and other countries . set fixed payments for specified medical diag- governments, with assistance from donors, are noses, as in Brazil. Yet a third approach, which has supporting the work of traditional birth attendants been tested on a limited scale in the United States, in safe pregnancy and delivery care and of tradi- is "managed competition." This scheme pursues tional healers in controlling infectious diseases the three objectives of cost-effective health spend- such as malaria, diarrhea, and AIDS. ing, universal insurance coverage, and cost con- Regulation is an essential element of govern- tainment simultaneously through tightly regulated ment efforts to encourage private health care sup- competition among companies that provide a spe- pliers. In most countries, governments have an cified package of health care for a fixed annual fee. important role to play in ensuring the quality of Each of these approaches has proved workable, private sector health care-through accreditation but each also has its limits and disadvantages. of hospitals and laboratories, licensing of medical There are no simple answers for health schools and physicians, regulation of drugs, and policy makers. reviews of medical practices. Some countries in which the government's ability to regulate is par- An agenda for action ticularly weak could explore self-regulation for health care providers, while building up govern- Adoption of the main policy recommendations of ment capacity. In Brazil experiments with self- this Report by developing country governments regulation for local hospital associations and medi- would enormously improve the health status of cal ethics boards are now under way. their people, especially poor households, and Government regulation of insurance is equally would also help to control health care spending important. In some countries part of the popula- (Table 3). Millions of lives and billions of dollars tion is denied insurance because of selection bias could be saved. Implementation of the public under private voluntary insurance. In the United health and essential clinical care packages, pursuit States millions of people with high health risks- of economic growth strategies that reduce poverty, and thus high need for health insurance-are un- and increased imrestment in schooling for girls able to obtain affordable coverage. Some types of would have the largest payoffs in averting deaths insurance schemes also seem to contribute to and reducing disability. Scaling back public spend- pushing up health care costs; this is particularly ing for tertiary care facilities, specialist training, true of third-party systems and of systems that and clinical care with lower cost-effectiveness reimburse hospitals and physicians item by item would help to increase the effectiveness of health for any and all services performed. In both the spending. So would encouragement of competi- Republic of Korea, which relies on universal social tion in delivery of health services and regulation of insurance, and the United States, which uses insurance and of provider payment systems. mostly private insurance, health care already ab- These recommendations will facilitate progress sorbs an unusually high share of GNP-and costs toward the goal contained in the declaration from are still rising. During the 1980s, for example, the historic 1978 Alma-Ata conference: "The at- health expenditures in Korea increased from 3.7 to tainment of all peoples of the world by the year almost 7 percent of GNP, in large part because of 2000 of a level of health that will permit them to expansion of third-party insurance coverage com- lead a socially and economically productive life." bined with fee-for-service provider compensation. Continued momentum toward this goal was pro- To eliminate selection bias and expand insur~ vided by the 1990 World Summit for Children. Al- ance coverage, governments can require insurers most 150 countries have now signed commitments to pool risks across large numbers of people. To to specific goals for their countries to improve the control costs, governments have a number of op- health of children and women (Box 2). These goals tions for limiting payments to health providers. include reduction of child mortality rates by one- One approach is to encourage prepayment of a third (or to 70 per 1,000 births, whichever would fixed amount for each person, as is now done in be less) over the course of the decade of the 1990s, 13 Table 3 Contribution of policy change to objectives for the health sector Contribution to goals Government objectives and policies Foster au enabling environme11t for llousellolds to improve healt/1 Pursue economic growth policies that benefit the poor Exp11nd investment in educ11tion, particularly for females Promote the rights 11nd status of women through political and economic empowerment and legal protection against abuse Improve govemment investments i11 l1ealt11 Reduce government expenditures for tertiary care facilities, specialist training, and discretionary services Finance and ensure delivery of a public health pack11ge, including AIDS prevention Finance and ensure delivery of essential clinical services, at least to the poor Improve the man<1gement of public health services Facilitate involveme11t by tl1 e private sector Encourage private finance and provision of insurance (with incentives to contain costs) for all discretionary clinical services Encourage private sector delivery of clinical services (including those that arc publicly financed) Provide information on performance and cost • Very favorable II F11vorable D Somewhat favorable D No impact expected reduction of maternal mortality rates by half, erad- nancing of essential clinical services should be at ication of polio, and major reductions in morbidity the top of the policy agenda . In most middle- and mortality from several other diseases . Com- income countries these policies are still germane, mitments to specific improvements in education, but reducing public subsidies for insurance and nutrition, water supply, and sanitation were also discretionary care would also yield large benefits made. These commitments underscore the politi- and should therefore be a key element of policy cal potential of action agendas for improving change . In the formerly socialist economies there health . are two particularly crucial policy areas-improv- The relevance of the main recommendations of ing the management of government health ser- this Report varies from one setting to another. In vices and developing sustainable health-financing low-income countries renewed emphasis on basic systems that maintain universal coverage while schooling for girls, strengthening of public health encouraging competition among cost-conscious programs, and support for expanded public fi- suppliers. 14 Box 2 The World Summit for Children The declaration and plan of action adoptt>d at the gets and external aid if priorities .for human devdop- World Summit for C hildrL' n, held in New York in 1990, ment are to be met. The health goals of the summit's incorporate a politically salient ,,genda for health. The plan of action include: summit focused, in p.uticular, on the needs of children • The L'radication of polio by 2000 and women but w.1s sl't in the bro.1der context of hu- • The elimination of neonatal tetanus by 1995 man and community goals. The seventy-one heads of • A 90 percent reduction in measles cases and a 95 state who attended and the seventy-seven more who percent reduction in measles deaths subsequently signed the declaration committed their • Achievement (by 2000) and maintenance of at least countries to developing national progr.1ms of action 90 percent immunization cover,1ge of one-year-old chil- (NPAs) for achieving these goals. To date, ,,bout dren, as well as universal tetanus immuniza tion for eighty-five countries h.we drawn up NPAs, and ,,n- women of childbearing ,,ge other sixty are in thL' process of prep.1ring them . • A halving of child deaths caused by diarrhea and. a NPAs typically cover, ,,mung other concerns, pri- one-quarter reduction in the incidence of diarrheal mary health care, family planning, safe water, environ- disease mental sanitation, nutrition, and basic education. Be- • A reduction by one-third in child deaths caused by cause oftheir concentration on the welfare of children, acute respiratory infections NPAs are able to transcend political differences. They • Virtual elimination of vitamin A deficiency and io- offer a means of mobiliz ing the whoiL' of civil society- dine deficiency disorders neighborhood and civic associations, religious groups • A reduction in the incidence of low · birth weight and professional bodies, businesses, voluntary agen- (2.5 kilograms or less) to no more than 10 percent cies, organized labor, and universities-in the cause of • A one-third reduction fn1m 1990 levels in iron d efi- investment for health. ciency anL'mia among women NPAs are being integrated into national develop- • Access for all women to prenatal care, trained at- ment planning . They set forth measurable, attainable tendants during childbirth, and referral for high-risk goals- to be met by 2000 or earlier- that are adapted to pregnancies and obstetric emergencies . the realities of the country. By quantifying the re- The agenda for action of the children's health summit sources required to achieve these goals, NPAs help to is broadly consistent with the messages of this Report. identify the changes that are needed in national bud- At first glance, it might appear that adoption of central hospitals and has concentrated on improv- this Report's major recommendations will be easy. ing health centers and other district-level infra- To reach most people living in the developing structure. Tunisia has converted eleven large gov- world with the minimum package of cost-effective ernment hospitals to semiautonomous institutions public health and essential clinical services, about with strong incentives for improved performance. half of current government expenditures on other, During the 1980s Chile delegated responsibility for more discretionary care would have to be redi- its entire primary clinical care system to local gov- rected. But in reality, change will be difficult, since ernments and fostered more public and private an array of interest groups may stand to lose- competition in health service delivery and in in- from suppliers of medical services to rich benefici- surance. Costa Rica and Korea achieved universal aries of public subsidies to protected drug com- health coverage through social insurance. panies. Many of the changes will take years to im- The international community can do more to plement because they mean a major redirection of support health policy reforms. In 1990 donors dis- public resources and require the development of bursed about $4.8 billion of assistance for health, new institutional capabilities. or about 2.5 percent of all health spending in de- A number of developing countries have already veloping countries. The share of total develop- shown in recent years that broad reforms in the ment aid for health declined slightly in the 1980s, health sector are possible when there is sufficient from 7 to 6 percent, despite widespread calls for political will and when changes to the health sys- increased investment in human resource develop- tem are designed and implemented by capable ment, including health . As an immediate first planners and managers . Zimbabwe has imposed a step, donors need to restore this share to its former decade-long moratorium on new investments in level. A more substantial increase can be easily 15 justified, given the importance of health in reduc- essential clinical care-especially for tuberculosis ing poverty and the large gap between current and control, the EPI Plus program, AIDS prevention, needed spending for public health programs and and reduction of tobacco consumption-would be minimum clinical services. An additional $2 billion a significant contribution to policy reform. So a year from donors would meet about one-quarter would support for capacity-building. Countries of the costs of stabilizing the AIDS epidemic ($500 that are willing to undertake major changes in million) and one-sixth of the extra resources health policy should be strong candidates for in- needed to provide the public health and clinical creased aid, including donor financing of recurrent care package for low-income countries ($1.5 billion costs. An increasing number of donors, among of the $10 billion required). them the World Bank, are now supporting this Increased external assistance for health research kind of broad sectoral reform. Stronger donor co- that focuses on the major health problems of de- ordination, especially at the level of individual de- veloping countries-such as the search for new an- veloping country clients, would improve the posi- timalarial drugs and new or improved vaccines- tive impact of aid on health, as shown by the could have a very high payoff and would build on experience of Bangladesh, Senegal, and the comparative advantage of donor countries in Zimbabwe. conducting scientific research. That most health The benefits to the developing world from research benefits many countries further justifies adopting sound policies for health are enormous. donor support, particularly through such effective There is great potential for change during the internationally collaborative mechanisms as the closing years of this decade as more countries en- Special Programme for Research and Training in courage broad political participation and public ac- Tropical Diseases. countability, as levels of education and knowledge Donors and developing country governments improve, and as understanding of human biology, can also do much to improve the effectiveness of public health, and health care systems increases. If aid for health. This is especially important in low- the right policy choices are made, the payoff will income Africa, where aid already accounts for an be high. The momentum of past reductions in the average 20 percent of health spending-and for burden of infectious disease in developing coun- over half in Burundi, Chad, Guinea-Bissau, tries can be maintained and accelerated. The AIDS Mozambique, and Tanzania. Even in other devel- epidemic can be slowed or reversed. The emerging oping regions, where aid amounts to 2 percent or problems of noncommunicable disease in aging less of health expenditures, better targeting and populations can be managed without rapid in- management of this assistance can catalyze policy creases in health expenditures. In the end, this will change. translate into longer, healthier, and more produc- Redirecting donor money from hospitals and tive lives for people around the world, especially specialist training to public health programs and the more than 1 billion now living in poverty. 16 Health in developing countries: successes and challenges On October 22, 1977, Ali Maow Maalin, a twenty- Few investments of any kind generate human three-year-old cook living in the town of Merca, and financial benefits on that scale . Yet in many Somalia, developed a fever and rash that was sub- ways the Intensified Smallpox Eradication Pro- sequently diagnosed as smallpox. Vaccination gramme exemplifies the potential of today' s medi- teams immediately descended on Merca and cine. Around the world, the past half century has within three weeks had vaccinated more than seen startling improvements in health . Progress in 50,000 people. They also began an intensive search drugs, vaccines, epidemiological knowledge, and for other cases in Merca and along the road and organizational experience continually expands the footpaths leading to it. By December 29 the World range of options for tomorrow. Tools and methods Health Organization (WHO) had removed Merca for combating and eliminating much of the re- from its list of potential outbreaks of smallpox and maining burden of disease are now affordable, had initiated a two-year surveillance for the dis- even by the poorest countries. Good policy, how- ease throughout the Horn of Africa. It turned out, ever, is essential for achieving good health . Some however, that Mr. Maalin had experienced the countries have made full use of the potential of world's last case of smallpox. He survived, and medicine; others have barely tapped it, despite WHO's twelve-year-long Intensified Smallpox heavy spending. This Report draws from this var- Eradication Programme was brought to a trium- ied experience lessons that will assist policymakers phant end . in realizing the enormous potential returns from In 1967, the year when the program began, their countries' investments in health . somewhere between 1.5 million and 2 million peo- ple died from smallpox. Perhaps half a million Why health matters more were blinded, and more than 10 million were seriously and permanently disfigured . In the early Good health, as people know from their own ex- 1950s the toll from smallpox had been three or four perience, is a crucial part of well-being, but spend- times greater. Then more and more countries un- ing on health can also be justified on purely eco- dertook vaccination programs, and by the time the nomic grounds. Improved health contributes to global program began, the disease had been vir- economic growth in four ways: it reduces produc- tually eradicated in 125 countries. Even so, the cos_t tion losses caused by worker illness; it permits the of smallpox vaccination, quarantine programs, use of natural resources that had been totally or and treatment totaled more than $300 million in nearly inaccessible because of disease; it increases 1968 alone. The eradication program, by contrast, the enrollment of children in school and makes cost $300 million over the whole of its twelve-year them better able to learn; and it frees for alterna- life and has therefore saved hundreds of millions tive uses resources that would otherwise have to of dollars a year in direct, measurable costs, as well be spent on treating illness . The economic gains as unquantifiable amounts of human suffering. are relatively greater for poor people, who are typ- 17 ically most handicapped by ill health and who ability in their income. In Paraguay, for example, stand to gain the most from the development of farmers in malarious areas choose to grow crops underutilized natural resources. that are of lower value but that can be worked outside the malaria season. Gains in worker productivity Improved utilization of natural resources The most obvious sources of gain are fewer work days lost to illness, increased productivity, greater Some health investments raise the productivity of opportunities to obtain better-paying jobs, and land. In Sri Lanka the near-eradication of malaria longer working lives. To take a classic example, during 1947-77 is estimated to have raised national leprosy is a disease that affects people in the prime income by 9 percent in 1977. The cumulative cost of life, with peak incidence rates among young was $52 million, compared with a cumulative gain adults. As many as 30 percent of those affected in national income over the thirty-one years of $7.6 may be seriously deformed, and their working billion, implying a spectacular benefit-cost ratio of lives will be shortened as well. A study of lepers in more than 140. Areas previously blighted by mos- urban Tamil Nadu, India, estimates that the elim- quitoes became attractive for settlement; migrants ination of deformity would more than triple the moved in, and output increased. In Uganda mas- expected annual earnings of those with jobs. The sive migration to fertile but underexploited land prevention of deformity in all of India's 645,000 followed the partial control of river blindness (on- lepers would have added an estimated $130 mil- chocerciasis) in the 1950s. The Onchocerciasis lion to the country's 1985 GNP. This amount is the Control Programme, conducted in eleven coun- equivalent of almost 10 percent of all the official tries of the Sahel, is a more recent example of the development assistance received by India in 1985. same benefits (see Box 1.1). Yet leprosy accounted for only a small proportion of the country's disease burden, less than 1 per- Benefits in the next generation through education cent in 1990. Healthier workers earn more because (as re- There is no question that schooling pays off in search in Bangladesh has demonstrated) they are higher incomes. Four years of primary education more productive and can get better-paying jobs. In boosts farmers' annual productivity by 9 percent Cote d'Ivoire daily wage rates are estimated to be on average, and workers who do better at school 19 percent lower, on average, among men who are earn more. Studies in Ghana, Kenya, Pakistan, likely to lose a day of work per month because of and Tanzania indicate that workers who scored 10 illness than among healthier men. percent above the sample mean on various cogni- When illness strikes, an individual's lost output tive tests have a wage advantage ranging from 13 and earnings often go undetected in economic sta- to 22 percent; in Nepal farmers with better mathe- tistics because they are borne by the household. In matical skills are more likely to adopt profitable many developing countries unemployment (or new crops. disability) insurance is rare, and healthier mem- Poor health and nutrition reduce the gains of bers of the household work harder or longer to schooling in three areas: enrollment, ability to make up for the loss in income. In a sample of 250 learn, and participation by girls. Children who en- Sudanese households, each of which lost, on aver- joy better health and nutrition during early child- age, forty working hours per year because of ma- hood are more ready for school and more likely to laria alone, this extra work made up for 68 percent enroll. A study in Nepal has found that the proba- of the lost agricultural labor. Similar findings have bility of attending school is only 5 percent for nu- come from research in Paraguay and Colombia. tritionally stunted children, compared with 27 per- In the long run, the benefits of improved health cent for those at the norm. are also likely to influence the way work is orga- Health and nutrition problems affect a child's nized and carried out. With a healthy work force, ability to learn. Nutritional deficiencies in early employers can reduce the costs of building slack childhood can lead to lasting problems: iron defi- into their production schedules, invest more in ciency anemia reduces cognitive function, iodine staff training, and exploit the benefits of specializa- deficiency causes irreversible mental retardation, tion. Similar gains are likely among farmers, who and vitamin A deficiency is the primary cause of often hedge against sickness by being risk-averse; blindness among children. Older children are sub- they forgo higher output in return for less vari- ject to other kinds of disease. In a recent study in 18 Box 1.1 Controlling river blindness Onchocerciasis, or river blindness as it is more com- a team of entomologists, epidemiologists, field staff, monly known, is caused by a parasitic worm which and pilots; 97 percent of the staff are nationals of the produces millions of larvae that move through the participating countries. The World Bank organizes the body, causing intense itching, debilitation, and eventu- finances and manages them through a trust fund. It ally blindness. The disease is spread by a small, fiercely also supports socioeconomic development in the areas biting blackfly that transmits the larvae from infected to affected by the disease . uninfected people. The program is widely regarded as a great success. It The goals of the Onchocerciasis Control Programme protects from river blindness about 30 million people, (OCP), set up in 1974 and covering eleven Sahelian including more than 9 million children born since the countries, are to control the blackfly by destroying its ocr began, at an annual cost of less than $1 per per- larvae with insecticides sprayed from the air. The envi- son. More than 1.5 million people who were once seri- ronmental impact of the insecticides is continuously ously infected have completely recovered. It is esti- monitored by an independent ecological committee, in mated that the program will have prevented at least cooperation with the national governments. The com- 500,000 cases of blindness by the time it is wound up mittee has full authority to screen insecticides and to around the end of the century. And it is already freeing approve or reject their use. The program has also col- approximately 25 million hectares of previously laborated with the pharmaceutical industry to develop blighted land for resettlement and cultivation, boosting for human use a drug, ivermectin,.that safely and effec- agricultural production. tively kills the larvae in the body. Ivermectin, however, The estimated cost of the OCP during the whole of has little impact on the adult worm and so must be its existence, from 1974 to 2000, is about $570 million. supplemented with vector .control by aerial spraying. Its estimated internal rate of return is in the range of 16 The producer of ivermectin, Merck & Co., has commit- to 28 percent (depending on the pace at which the ted itself to provide the drug free of charge as long as it newly available land is settled, the incremental output is needed to combat river blindness. added by the new land, the income level of the ocr The OCP's four sponsoring agencies-the Food and area, and the productivity growth rate that is pro- Agriculture Organization, the United Nations Devel- jected). These estimated benefits do not include the opment Programme (UNDP), the World Bank, and program's favorable effects on income distribution; its WHO-through a steering committee chaired by th e main beneficiaries are subsistence farmers whose in- World Bank, make broad policy decisions and oversee comes are well below average. operations. WHO has executive responsibility through Jamaica children with moderate whipworm infec- Girls are particularly liable to suffer from iodine tion scored 15 percent lower before treatment than or iron deficiency-reasons why fewer of them uninfected children in the same school. When re- complete primary school. Other health-related tested after treatment, those same children did al- reasons include dropping out as a result of preg- most as well as the uninfected children. nancy and parental concern about sexual violence. In a sample of children in a poverty-stricken In societies where girls' education is given lower area of northeast Brazil, inadequately nourished priority than boys', girls miss school because they children lagged 20 percent behind the average gain have to stay home to look after sick relatives. in achievement score over a two-year period. The same study also shows the harm done by a simple Reduced costs of medical care and easily remedied handicap: children with bad eyesight lagged 27 percent behind the average Spending that reduces the incidence of disease can gain over the two years. Both groups had below- produce big savings in treatment costs. For some average promotion rates and above-average drop- diseases the expenditure pays for itself even when out rates. In China a child at the twentieth percen-- all the indirect benefits-such as higher labor pro- tile in height-for-age (a sign of poor health) aver- ductivity and reduced pain and suffering-are ig- ages about one-third of a year behind the grade nored. Polio is one example. Calculations for the normally reached by children of that age. In Thai- Americas made prior to the eradication of polio in land children whose height-for-age is 10 percent the region showed that investing $220 million over below average are 14 percent lower in grade fifteen years to eliminate the disease would pre- attainment. vent 220,000 cases and save between $320 million 19 Box 1.2 The economic impact of AIDS The AIDS epidemic, through its effects on savings and That AIDS kills so many skilled adults adds to its productivity, poses a threat to economic growth in economic impact. At a large hospital in Kinshasa, for many countries that are already in distress . World Bank example, more than 1 percent per year of the health simulations indicate a slowing of growth of income per personnel, including highly trained staff, become in- capita by an average 0.6 percentage point a year in the fected (through sexual rather than occupational con- ten worst-affected countries in Sub-Saharan Africa . In tact) . Among the (largely male) employees at a Kin- Tanzania, where income per capita has already fallen shasa textile mill, managers had a higher infection rate 0.2 percent a year in recent years, the estimated slow- than foremen, who in turn had a higher rate than down ranges between 0.1 and 0.8 percentage point, workers. The cost of replacing skilled workers will be depending on the assumptions used. In Malawi, which substantial. A study of Thailand estimates that through has had a recent growth rate of 0.9 percent a year, the 2000 the cost of replacing long-haul truckers lost to simulated reduction ranges from 0.3 to 0.5 percentage AIDS will be $8 million, and another study, of Tan- point. These calculations include the effect of the epi- zania, projects the cost of replacing teachers at $40 mil- demic on population growth, which will slow slightly lion through 2010. in severely affected countries. The death of an adult can tip vulnerable households The heavy macroeconomic impact of AIDS comes into poverty. Even in Tanzania, where the government partly from the high costs of treatment, which divert pays a large share of health costs, a World Bank study resources from productive investments . Tanzanian cli- shows that affected rural households in 1991 spent nicians estimate that, on average, an HIV-infected $60-roughly the equivalent of annual rural income per adult suffers 17 episodes of HIV-related illnesses prior capita-on treatment and funerals . The study also to death and a child suffers 6.5 episodes. Depending showed that the effects of losing an adult persist into on how much medical care a patient gets, in the typical the next generation as children are withdrawn from developing country the total cost per adult death school to help at home. School attendance of young ranges from 8 to 400 percent of annual income per cap- people ages 15-20 is reduced by half if the household ita; the average is about 150 percent of annual income has lost an adult female member in the previous year. per capita. and $1.3 billion (depending on the number of peo- previously uninfected person every four years . At ple treated) in annual treatment costs. The pro- this rate, there will be six HIV-positive persons in gram's net return, after discounting at even as 2000 for every one today. If the transmission rate much as 12 percent a year, was calculated to be could be slowed to one every five years, that num- between $18 million and $480 million. ber could be reduced to only four infected persons AIDS is another example. Although it remains in 2000 for every one today. The corresponding much less common in the developing world than reduction in medical costs, after discounting at 3 diseases such as malaria, its economic impact per percent a year, amounts to $750 by 2000 for each case is greater for two reasons: it mainly affects currently HIV-positive person in India, for a total adults in their most productive years, and the in- saving of $750 million . Similar calculations for fections resulting from it lead to heavy demand for Thailand suggest savings of $1,250 per currently expensive health care (Box 1.2). For example, be- HIV-positive person, for a potential total of $560 cause individuals with AIDS are typically more million. prone to pneumonia, diarrhea, and tuberculosis, Health investments and poverty the cost of medical care is high even though there is no effective treatment as yet for the disease it- The goal of reducing poverty provides a different self. Research in nine developing and seven high- but equally powerful case for health investments. income countries suggests that preventing a case The adverse effects of ill health are greatest for of AIDS saves, on average, about twice GNP per poor people, mainly because they are ill more of- capita in discounted lifetime costs of medical care; ten, but partly because their income depends ex- in some urban areas the saving may be as much as clusively on physical labor and they have no sav- five times GNP per capita. Calculations for India ings to cushion the blow . They may therefore find show that, given prevailing transmission patterns, it impossible to recover from an illness with their each currently HIV-positive person infects one human and financial capital intact . 20 The health consequences of poverty are severe: of growth in income per capita between 1960 and the poor die younger and suffer more from disabil- 1990 in about seventy countries to the initial level ity. In Porto Alegre, Brazil, adult mortality rates in of national income, the initial educational level, poor areas in the late 1980s were 75 percent higher and an indicator of initial health status (the child than in rich areas, and in Sao Paulo rates were two mortality rate, used in this Report to mean the risk to three times higher for nonprofessionals than for of dying by age 5 per 1,000 live births). The health professionals. In the late 1970s among Kenyan status indicator is found to be a highly significant families in which the mother had no schooling, the predictor of economic performance. For the aver- probability of dying by age 2 averaged 184 per age country in the sample, the annual growth rate 1,000 in regions where half of the families lived of income per capita is 1.40 percent and the child below the poverty line but 100 per 1,000 in regions mortality rate is 116 per 1,000. An otherwise aver- where only one-fifth of the families lived in pov- age country with a child mortality rate of 106 erty. The poor are exposed to greater risks from would have a growth rate of income per capita of unhealthy and dangerous conditions, both at 1.55 percent, whereas one with a child mortality home and at work. Malnourishment and the leg- rate of 126 would have a growth rate of 1.26 acy of past illness mean that they are more likely to percent. fall ill and slower to recover, especially as they Not surprisingly, the health status variable is have little access to health care. strongly correlated with educational stock, but the When a family's breadwinner becomes ill, other significant association between income growth members of the household may at first cope by and health remains strong and of similar magni- working harder themselves and by reducing con- tude across time periods and for a range of model sumption, perhaps even of food. Both adjust- formulations. Although it is possible that unob- ments can harm the .health of the whole family. If served factors such as government capacity to im- free health care is not available, the costs of treat- plement effective policies could explain the appar- ment may drive a household deeper into debt. Al- ent association, the data do suggest that better though ill health is only one of many factors that health means more rapid growth. can cause financial distress, its potential for disas- ter means that it should be explicitly recognized in The record of success formulating policies. Investments to reduce health risks among the poor and provision of insurance Mortality started to decline in Europe, North against catastrophic health care costs are impor- America, and Australasia about two centuries ago, tant elements in a strategy for reducing poverty. but slowly at first. A century ago life expectancy in Spending on health is a productive investment: the United States, then the world's richest coun- it can raise incomes, particularly among the poor, try, was only forty-nine years, and child mortality and it reduces the toll of human suffering from ill was about 180 per 1,000. The rate of improvement health. Good health, however, is a fundamental accelerated in the first half of this century; by 1950 goal of development as well as a means of acceler- life expectancy in the United States had increased ating it. Targeting health as part of development to sixty-six years, and child mortality had fallen to efforts is an effective way to improve welfare in 34 per 1,000. Progress was also being made in de- low-income countries. Evidence gathered over the veloping countries: in Chile, for example, life ex- past thirty years indicates that in health, unlike pectancy increased from thirty-seven years in 1930 income, the gap between poor and rich countries to forty-nine in 1950, and child mortality fell from has been narrowing. 350 to 209 per 1,000. Putting the effects together Mortality transitions since 1950 The detrimental effects of poor health on individ- Health conditions around the world have im- uals and households and on the use of resources proved more in the past forty years than in all suggest that better health should lead to better previous human history. Life expectancy at birth economic performance at the national level. A in developing countries increased from forty to number of analyses have found a positive relation- sixty-three years, and child mortality fell from 280 ship between growth of income per capita and the to 106 per 1,000. In a high-income country life ex- initial national educational stock. A similar analy- pectancy is more than seventy-five years; in a low- sis carried out for this Report examines the relation mortality developing country it is seventy years or 21 Child mortality has fallen sharply in the past thirty years, with particularly rapid declines in parts of Asia and Latin America. Figure 1.1 Child mortality by country, 1960 and 1990 ,, Under-five 111ortality rate • 175 or more ·• 125- 174 D 75-124 D 50-74 D 25-49 • Less than 25 ,, Source: Appendix A. 22 more; and in Sub-Saharan Africa, the region Life expectancy has increased substantially where least progress has been made, it is about everywhere over the past fo'rty years. fifty years. Much of what is known about the decline in mortality in the developing world since 1950 is Figure 1.2 Trends in life expectancy limited to the mortality of children and has come by demographic region, 195Q-90 from a series of standardized, internationally funded demographic surveys . Enormous reduc- Life expectancy at birth (years) tions in child mortality occurred almost every- 80 where around the world between 1960 and 1990 (Figure 1.1). For example, child mortality in Chile dropped from 155 to 20 per 1,000, in Tunisia from 245 to 45, and in Sri Lanka from 140 to 22. 70 The statistics for adult mortality in the develop- ing world are much less satisfactory than those for child mortality. Approximate estimates for all de- 60 veloping countries suggest that the adult mortality rate (defined as the probability of dying between ages 15 and 60 per 1,000 persons reaching age 15) fell from about 450 in 1950 to about 230 in 1990. In 50 Chile, a country with excellent statistics, the rate dropped from 466 in 1930 to 152 in 1990. The decline in mortality has accelerated over the 40 past thirty years. In the 1960s child mortality fell 1950 1960 1970 1980 1990 by approximately 2 percent a year in about seventy developing countries for which estimates are avail- able. The annual decline increased to more than 3 Sub-Saharan Africa percent in the 1970s and to more than 5 percent in India the 1980s. This result could be skewed by changes China in the mix of countries with reliable data; there Other Asia and islands were, however, twenty-one countries with a con- Latin America and the Caribbean tinuous series of acceptable estimates of child mor- Middle Eastern crescent tality from the early 1%0s to the late 1980s, and for Formerly socialist economies of Europe this group as a whole the fall in child mortality Established market economies averaged 3 percent a year in the 1960s but 6 per- cent a year in the 1980s. In seventeen of the Source: Appendix A. twenty-one the pace of decline increased over the period. Regional patterns There are strong parallels between the pattern of The extent of success has varied significantly be- mortality decline in the high-income countries and tween regions. Between 1950 and 1990 all eight the accelerated progress of developing countries demographic regions used for this Report enjoyed over the past forty years. In both groups the con- increases in life expectancy at birth, but China and trol of communicable diseases, particularly those the Middle Eastern crescent did particularly well of childhood, accounts for most of the gains. (The (see Figure 1.2). Sub-Saharan Africa showed the term ''communicable diseases,'' in the analyses slowest improvement, with life expectancy in- for this Report, includes deaths from maternal and creasing only from thirty-nine to fifty-two years- perinatal causes.) Progress against noncommuni- although even this compares well with European cable diseases-primarily those of the circulatory experience in the nineteenth century. (It took En- and respiratory systems, which principally affect gland and Wales more than half a century to raise adults-has been much slower. In both Chile (from life expectancy by a similar amount.) The formerly 1930 to 1987) and England and Wales (over the socialist economies of Europe showed a rapid im- longer period 1891 to 1990) mortality from commu- provement in the 1950s and 1960s, but the rise was nicable disease fell to less than 5 percent of its much slower in the 1970s and 1980s. initial level, whereas mortality from noncommuni- 23 Mortality from communicable diseases has fallen Similar patterns in Chile and in England and Wales much faster than that from noncommunicable show how mortality rates have declined much more diseases or injuries. sharply for the young than for the old. Figure 1.3 Age-standardized female death Figure 1.4 Change in female age-specific rates in Chile and in England and Wales, mortality rates in Chile and in England selected years and Wales, selected years • Earlier period" 0 Later period b Ratio of mortality rates at two time periods Chile 10 Communicable 0.8 I diseases< Noncommunicable diseasesd 0.6 ~ I Chile " ./ 0.4 ,.- v ' ,r-- Injuries I 0.2 b l--" ~ II' England and England and Wales o I" ~ - vyales~ I 0 5 15 25 35 45 55 65 75 85 Communicable Age diseases< a. Ratio of 1990 rates to 1930 rates. b. Ratio of 1981 rates to 1891 rates. Noncommunicable Source: For Chile 1930 and England and Wales 1891 ; diseasesd Preston, Keyfitz, and Schoen 1972; for Chile 1990, World Health Organization datil; for England and Wales 1990, United Nations, Dc111ogmpl1ic Ycarlook, 1991. Injuries 0 2 4 6 8 10 12 14 adults. In Chile, for example, mortality risks up to Age-standardized death rate age 30 fell by more than 90 percent between 1930 per 1,000 population and 1990; the decline was at least 60 percent at ages 30-70, but above age 70 the gains were much .a. For Chile, 1930; for England and Wales, 1891 . smaller (Figure 1.4). The age pattern of mortality b. For Chile, 1'1H7; for England and Wales, 1990. decline in Chile over sixty years is strikingly simi- c. Includes ;naternal and neonatal mortality . d . For earlier period, includes "other and unknown" lar to the pattern in England and Wales during the category. ninety years from 1891 to 1981. Source: For C hile 1'130 and England and Wales 1891 , Preston, Keyfitz, and Schoen 1'172; for 1987 and 1990! The only exception to this broad similarity be- WHO, World /lcai/IJ Stat istics Aunua/, 1989 and 1991. tween industrial and developing countries has been in the formerly socialist economies. In these countries child mortality has continued to decline, as has the mortality of women, albeit more slowly. cable disease fell much less rapidly (Figure 1.3). The mortality of adult men, on the other hand, has One result of this change is that mortality risks stopped declining in the past two decades and has have fallen much faster for children than for actually started to increase. This excess male mor- 24 tality is largely the result of extremely high death Table 1.1 Burden of disease by sex, cause, rates from cardiovascular disease, associated with and type of loss, 1990 (millions of DALYs) heavy smoking and drinking. Disease category Measuring the burden of disease Sex and outcome Communicable" Noncommunicable Injuries Male The health improvements of the past few decades Premature death 259 152 70 Disability 47 146 39 have done much to enhance human welfare, both directly and indirectly. But much more remains to Female Premature death 244 135 33 be done. Communicable (and largely preventable) Disability 74 142 20 diseases are still common. Health systems also Note: DALY, disability-adjusted life year. have to cope with the aging of populations, which a. Includes maternal and perinatal causes. leads to an increased burden of the more expen- Source: Appendix B. sive noncommunicable diseases. New illnesses, such as AIDS, have emerged. One simple statistic gives a sense of the remaining burden of disease: arrhea, childhood diseases such as measles, respi- about 12.4 million children under age 5 died in ratory infections, worm infections, and malaria ac- 1990 in the developing world. Had those children count for one-quarter of the GBD. The burden of faced the mortality risks of children in the estab- these largely preventable or inexpensively curable lished market economies, the number of deaths diseases of children is far larger in Sub-Saharan would have been cut by more than 90 percent, to Africa (43 percent of all DALYs lost) than any- 1.1 million. where else, although it is still substantial in India Any discussion of health policy must start with (28 percent), Other Asia and islands (29 percent), a sense of the scale of health problems. These and the Middle Eastern crescent (29 percent). For problems are often assessed in terms of mortality, adults too, communicable diseases are far from but that indicator fails to account for the losses that trivial: sexually transmitted diseases (STDs) and occur this side of death because of handicap, pain, tuberculosis together contribute 7 percent of the or other disability. A background study for this GBD. Report, undertaken jointly with the World Health Even as broad a measure as the GBD does not Organization, measures the global burden of dis- capture all the consequences of disease or injury. It ease (GBD) by combining (a) losses from prema- excludes the social costs of disfigurement, such as ture death, which is defined as the difference be- that arising from river blindness or leprosy, and of tween actual age at death and life expectancy at dysfunction-for example, marital breakups re- that age in a low-mortality population, and (b) loss sulting from obstetric fistula (permanent damage of healthy life resulting from disability. The GBD is to the reproductive tract incurred during delivery). measured in units of disability-adjusted life years And some health-related factors are likely to be (DALYs). Worldwide, 1.36 billion DALYs were lost underreported. A clear example is violence against in 1990, the equivalent of 42 million deaths of new- women, much of which goes undetected-but not born children or of 80 million deaths at age 50. unsuffered. Premature deaths were responsible for 66 percent Comparisons of absolute numbers of DALYs lost of all DALYs lost and disabilities for 34 percent. In may be misleading because the sizes and age struc- the developing world 67 percent of all DALY loss tures of the populations at risk are not the same. was a result of premature death; in the established The effects of population size can be allowed for by market economies and the formerly socialist econ- expressing the 1990 burden per 1,000 population. omies of Europe the figure was only 55 percent. Figure 1.5 shows the resulting rates by sex and Table 1.1 shows the GBD broken down by sex, regional group. This index is 259 for the world as a category of disease, and type of loss (premature whole, but it varies widely among regions. Sub- death or disability). The three categories of disease Saharan Africa loses 574 DALYs for every 1,000 used are the group of communicable diseases, population, more than twice the global average. noncommunicable diseases, and injuries. India, the Middle Eastern crescent, and Other Asia The derivation and interpretation of the GBD and islands all have values between 250 and 350. are explained in Box 1.3. The results of research on For China, the formerly socialist economies of Eu- the GBD challenge the belief that the war against rope, and Latin America and the Caribbean, the infectious and parasitic diseases has been won. Di- figures are between 150 and 250. The burden per 25 Box 1.3 Measuring the burden of diseas·e Most assessments of the relative importance of differ- healthy life lost was then obtained by multiplying the ent diseases are based on how many deaths they cause. expected duration of the condition (to remission or to This convention has certain merits: death is an unam- death) by a severity weight that measured the severity biguous event, and the statistical systems of many of the disability in comparison with loss of life. Dis- countries routinely produce the data· required . There eases were grouped into six classes of severity of dis- are, however, many diseases or conditions that are not ability; for example, class 2, ..yhich includes most cases fatal but that are responsible for great loss of healthy of leprosy and half· the cases of pelvic inflammatory life: examples are chronic depression and paralysis disease, was g~ven a severity weight of 0.22, and class caused by polio. These conditions are common, can 4, which includes 30 percent of cases of dementia and last a long ti!Tie, and frequently lead to significant de- 50 percent-·of those of blindness, was assigned a sever- mands on health systems . ity weight o£.0.6. The death and disability losses were To quantify the full loss of healthy life, the World then combined, and allowance was made for a dis- Bank and the World Health Organization undertook a coun't rate of 3 percent (so that future years of healthy joint exercise for this Report. Diseases were classified life were valued at progressively lower levels) and for into 109 categories on the basis of the li1lernational C/as- age weights (so that years of life lost at different ages sificatio'' of Diseases (ninth revision). These categories were given different relative values). The value for cover all possible causes of death and about 95 percent each year of life lost, shown in the left-hand panel of of the possible causes of disability. Using the recorded Box figure 1.3, rises steeply from zero at birth to a peak cause of death where available, and expert judgment at age 25 and then declines gradually with increasing when records were not available, the study assigned all age. These age ·weights reflect a consensus judgment, deaths in 1990 to these categories by age, sex, and .de- but other patterns could be used-for example, uni- mographic region . For each death, the number of years form age weights, with each year of life having the of life lost was defined as the difference between the same value, which would increase the relative impor- actual age at death and the expectation of life at that, tance of childhood diseases. age in a low-mortality population . For disability, the The combination of discounting and age weights incidence of cases by age, sex, and demographic region· produces the pattern of DALYs (disability-adjusted life was estimated on the basis of community surveys or, years) lost by a death at -each age . As the right-hand failing that, expert opinion; the number of years of panel of Box figure 1.3 shows, the death of a newborn Box figure 1.3 Age patterns of age weights and DALY losses Value of a year of life DALYs lost by death at given year (females) Relative value of a year of life at age x Disability-adjusted life years (DALYs) 1.6 40 1.2 / 1---- ~ 30 v r--..1'-. If ' 0.8 1 I " "" !'...r---.... 20 "' ~ ' "" "" ~ 10 ~ ........ 0 I 0 .......... 0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90 Agex Age at death in years 1-------- Sou rce: World Bank data . 26 Box table 1.3 Distribution of DALY loss by cause and demographic region, 1990 (percent) Fo rmerlv Sub' Latin Middle sodalisi Established Saharan Oth er Asia America and Eusft•n1 ccmowics of t.• market Cause World Africa India C!zi>m and islands the Caribbean cresCe11t Europe economies Population (millions) 5,267 510 850 1,134 683 444 503 346 798 Communicable diseases 45.8 71.3 50.5 25 .3 48.5 42.2 51.0 8.6 9.7 Tuberculosis 3.4 4.7 3.7 2.9 5.1 2.5 2.8 0.6 0.2 STDsand HIV 3. 8 8.8 2.7 1.7 1.5 6.6 0.7 1.2 3.4 Di.1rrhe~ 7.3 10.4 9.6 2.1 8.3 5.7 10.7 0.4 0.3 Vaccine-preventable childhood infections 5.0 9.6 6.7 0.9 4.5 1.6 6.0 0.1 0.1 Malaria 2.6 10.8 0..3 1.4 0.4 0.2 Worm infections 1.8 1.8 0.9 3.4 3.4 2.5 0.4 Respiratory infections 9.0 10.8 10.9 6.4 11 .1 6.2 11.5 2.6 2.6 Maternal causes 2.2 2.7 2.7 1.2 2.5 1.7 2.9 0.8 0.6 Perin.1tal causes 7.3 7.1 9.1 5.2 7.4 9.1 10.9 2.4 2.2 Other 3.5 4.6 4.0 1.4 3.3 5.8 4.9 0.6 0.5 Noncommunicable diseases 42.2 19.4 40.4 58.0 40.1 42.8 36.0 74.8 78.4 Cancer 5.8 1.5 4.1 9.2 4.4 5.2 3.4 14.8 19.1 Nutritional deficiencies 3.9 2.8 6.2 3.3 4.6 4.6 3.7 1.4 1.7 Neuropsychiatric disease 6.8 3.3 6.1 8.0 7.0 8.0 5.6 11 .1 15.0 Cerebrovascular disease 3.2 1.5 2.1 6.3 2.1 2.6 2.4 8.9 5.3 Ischemic heart disease 3.1 0.4 2.8 2.1 3.5 2.7 1.8 13.7 10.0 Pulmonary obstruction 1.3 0.2 0.6 5.5 0.5 0.7 0.5 1.6 1.7 Other 18.0 9.7 18.5 23.6 17.9 19.1 18. 7 23.4 25.6 Injuries 11.9 9.3 9.1 16.7 113 15.0 13.0 16.6 11.9 Motor vehicle 2.3 1.3 1.1 2.3 2.3 5.7 3.3 3.7 3.5 Intentional 3.7 4.2 1.2 5.1 3.2 4.3 5.2 4.8 4.0 Other 5.9 3.9 6.8 9.3 5.8 5.0 4.6 8.1 4.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Millions of DALYs 1,362 293 292 201 177 103 144 58 94 Equivalent infant deaths (millions) 42 .0 9.0 9.0 6.2 5.5 3.2 4.4 1.8 2.9 DALYs per 1,000 population 259 575 344 178 260 233 286 168 11 7 • Less than 0.05 percent. Note: DALY, disability-adjusted life year; STD, sexually transmitted disease; HIV. human immunodeficiency virus. Sou rce: World Bank data. baby girl represents a loss of 32.5 DALYs; a female one .year per person. A baby who died in 1990 contrib- death at age 30 means the loss ~f 29 DALY~; and a uted about thirty-two years (the discount ed value of female death at age 60 represents 12 lost DALYs. about eighty years of expected life) to the burden but (Values are slightly lower for males .) The sum across all counted as one in the population . To take an extreme ages, conditions, and regions is referred to as the case, if the entfre population of the world were to be global burden of disease (GBD) . More details on the killed in one year, the burden per 1,000 population in GBD are presented in Appendix B. that year wottld exceed 20,000 DALYs. There is there- The global burden measures the present value of the ·fore no absolute scale w ith which the GBD per 1,000 future stream of disability-free life lost as a result of population can be compared; the only co mparisons death, disease, or injury in 1990. It is thus based on t\1at make sense are those between categories-of re- events that occurred in 1990 but includes the loss of gions, risk factors, disease groups, or sex. Box table 1.3 disability-free life in future yea.rs. This Report ex- shows the GBD by cause and demographic region. presses the burden in three distinct ways: as the num- The approach used to compute the GBD can also be ber of DALYs, as a percen.tage of some larger aggregate used to track improvements in a nation ' s health over (such as the percentage of total loss attributable to a time by following changes in the national burden of specific disease), and in relation to population size in disease. Preliminary plans for initial national assess- 1990. This last measure tails for careful interpretation ments have been developed for Costa Rica, South Af- because all future loss is expressed in relation to the rica, and Andhra Pradesh State in India . .current population, and the measure can easily exceed 27 Disease burdens per 1,000 population vary substantially by region. Figure 1.5 Disease burden by sex and demographic region, 1990 Disease burden per 1,000 population 0 100 200 300 400 500 600 650 Sub-Saharan Africa India China Other Asia and islands Latin America and the Caribbean Middle Eastern crescent Formerly socialist economies of Europe Established market economies World liZ! Females [] Males Sottrce: Appendix B. 1,000 population for the established market econ- bers of "missing" (and presumed dead) women in omies is easily the lowest, at 117. It turns out that relation to the expected population balance be- these broad rankings are not significantly affected tween the sexes. In China illegal female infanticide by differences in age distributions between (and, in the recent past, illegal sex-selective abor- regions. tion) is thought to be the main reason. Females have abcut a 10 percent lower disease burden per 1,000 population than males for the world as a whole. They lose fewer DALYs from Table 1.2 Burden of five major diseases by age of incidence and sex, 1990 premature mortality, but their DALY loss from dis- (millions of DALYs) ability is about the same as for males. Within the Age (years) disability category, however, the female disease Disease and sex 0-4 5- 14 15-44 45-59 60+ Total burden from the group of communicable causes is Diarrhea considerably higher than that for males, partly be- Male 42.1 4.6 2.8 0.4 0.2 50.2 cause of a substantial toll from maternal causes but Female 40.7 4.8 2.8 0.4 0.3 48.9 also because of the much greater female burden Wonn infection associated with STDs. Effective interventions exist Male 0.2 10.6 1.6 0.5 0.1 13.1 for much of this excess female burden. For all Female 0.1 9.2 0.9 0.5 0.1 10.9 causes together, the female advantage ranges from Tuberculosis more than 30 percent in the formerly socialist · Male 1.2 3.1 13.4 6.2 2.6 26.5 economies, where adult mortality is much higher Female 1.3 3.8 10.9 2.8 1.2 20.0 for men than for women, to negative in India, Ischemic heart disease where females suffer a disadvantage of 8 percent. Male 0.1 0.1 3.6 8.1 13.1 25.0 Both India and China show a female disadvantage Female 1.2 3.2 13.0 17.5 •• Less than 0.05 m illion. in disease burden per 1,000 population, and, not Note: DALY, disability-adjusted life year. coincidentally, both countries also have large num- Source: World Bank data. 28 The figures on disease burden by age (Table 1.2) The share of communicable diseases in the disease suggest how health officials should target their burden declines as mortality rates fall. programs. More than 80 percent of the DALY loss from diarrhea is a result of infections in children under age 5. Worm infections are concentrated Figure 1.6 Distribution of disability- among children ages 5-14. More than half the bur- adusted life years (DALYs) lost, by cause, den of tuberculosis is borne by the 15-44 age for selected demographic regions, 1990 group. More than 60 percent of the burden of isch- (percentage of total DALYs lost) emic heart disease falls on the population over age 60. Sub-Saharan Latin America The higher the disease burden, the higher the Africa and the Caribbean proportion attributable to the communicable group of causes (Figure 1.6). Sub-Saharan Africa has the highest disease burden per 1,000 popula- tion, and 71 percent of this is from the communica- ble disease group, whereas in Latin America (a medium-burden region) the figure is 42 percent and in the established market economies it is only 10 percent. Noncommunicable diseases show the opposite pattern, accounting for 19 percent of the total burden in Sub-Saharan Africa, 43 percent in Latin America, and 78 percent in the established Established market market economies. Despite these marked differ- economies ences in relative burden, however, the absolute rates of loss for both groups are highest in Sub- Saharan Africa and lowest in the established mar- ket economies. The pattern is plain: as health im- proves, the burden from all types of disease declines, but the distribution of the burden shifts dramatically from a preponderance of communica- ble disease to a preponderance of noncommunica- ble disease. Despite the sharp improvements in health around the world, the GBD calculations show that D Communicable diseases" a large burden of premature mortality and disabil- D Noncommunicable diseases ity still remains, particularly in the world's poorer • Injuries regions. There are inexpensive and effective ways to eliminate the share caused by communicable a. Includes maternal and perinatal causes. Source: Appendix B. diseases (other than maternal and perinatal condi- tions), which is roughly 35 percent of the world burden and more than 60 percent in Sub-Saharan Africa. The remaining 65 percent of the world bur- den is less responsive to such measures, and re- ber of drug-resistant disease strains; and the con- ducing it will require changes in the behavior and tinued use of health-damaging substances such as life-styles of adults. tobacco. Although nobody can forecast the impact of these challenges with any precision, reasonable Challenges for the future projections are possible. For example, outside the established market economies the number of New health challenges will emerge over the next deaths attributable to smoking is expected to in- few decades. Some are certain: these involve the crease from 1.7 million in 1990 (40 percent of which significant increase in noncommunicable diseases were in the formerly socialist economies of Eu- arising from the continuing demographic transi- rope) to more than 3 million by 2005 and to about tion. Others are less certain: the spread of HIV and 4.5 million by 2015. Other challenges are poten- the increase in AIDS deaths; the increasing num- tially important but not forecastable: possible ex- 29 Worldwide, fertility and mortality declines go hand in hand. Box .1.4 The demographic and epidemiological transitions Figure 1.7 Trends in life expectancy and Changes in the pattern of diSl'.!Sl' pnKL'l'd in two steps. fertility in Sub-Saharan Africa and Latin The first is the demographic transition, wlwn mort.1lity America and the Caribbean, 1960-2020 from infectious diseaSL' declines ,md, p<~rtly as" n•stiit, fertility decreases as well. Thl' second, ,1 consequl·nCl' of declining fertility and diffen•nti,tl r·'•-•• ... .. "' : .;.;~ 5-9 ·;;, 0-4 10 8 6 4 2 0 10 20 30 40 50 10 8 6 4 2 0 10 20 30 40 50 Percentage of Percentage of Percentage of Percentage of total population total deaths' total population total deaths +- Median age af deathb a . Projected. ·b. The age below which half of all deaths in a year occur. Source: For England and Wales 1891, Preston, Keyfitz, and Schoen ·1972; for England and Wales 1966, United Nations, Dcllwgmplric Yearbook, 1978; for Latin America and the Caribbean,. 1955 and 2030, World Bank data. 31 in tandem beginning in the late nineteenth cen- eases will increase sharply, both absolutely and tury, and population growth rates rarely exceeded proportionately. At the same time, the challenge of 2 percent a year.) communicable diseases of the young will persist. These demographic changes are having, and Despite declines in fertility, the number of births will continue to have, dramatic effects on age dis- each year in developing countries will rise some- tributions in developing countries. As fertility de- what, from 127 million in 1990 to 145 million in clines, age structures in these countries are evolv- 2020, before decreasing to 142 million in 2030. The ing toward the existing patterns in the established number of children under age 5 will increase more market economies and the formerly socialist econ- rapidly, from 552 million in 1990 to 682 million in omies (see Box 1.4) . The proportion of the popula- 2030. These changes, which took a century to com- tion age 65 and over is expected to increase from 4 plete in today's high-income countries, are occur- percent in 1990 to 9 percent by 2030 (in absolute ring within fifty years or less in parts of the devel- numbers, from 184 million to 678 million). As a oping world. result, the burden from noncommunicable dis- In judging the importance of the health prob- lems of the young in comparison with those of the elderly, one useful guide is the median age at death. For all six regions of the developing world, All regions will experience the health transition, but the median was below 20 in 1950, indicating the the timing will differ. dominance of the health problems of children . By 1990 the median had risen close to age 60 in China Figure 1.8 Median age at death, by and in Latin America but was scarcely above 5 in demographic region, 1950, 1990, and 2030 Sub-Saharan Africa and was still below 25 in the Middle Eastern crescent (see Figure 1.8) . By 2030, assuming that current trends continue, the median Median age at death (years)" age at death will have risen above 60 in all regions 80 except Sub-Saharan Africa, where it will still be close to 40 . The message from these population projections is that health services must plan for a modest in- crease in child-related demands over the next forty years. At the same time, the numbers of the el- derly, with very different health needs, will be ris- ing sharply. The pace of demographic change has been, and is expected to continue to be, faster in the developing world than it was in the high- income countries, and the problems of adaptation are therefore greater. Because treatments for non- communicable diseases are often expensive, there is a danger that these diseases will absorb re- sources needed to combat communicable diseases 1950 1990 2030 b (which will still be widespread). This kind of di- lemma has already been noted in World Bank studies in Brazil and China . • Sub-Saharan Africa • Middle Eastern crescent HIVandAIDS D India D Other Asia and islands More than 80 percent of the estimated 8.8 million • Latin America and the Caribbean people infected with HIV in 1990 lived in develop- • China ing countries. There the disease is primarily one of heterosexual adults, with substantial perinatal in- a. The age below w hich half of all deaths in a year occur. fection of young children . Of the eight demo- b. Projected . Source: World Bank data. graphic regions used in this Report, only the for- merly socialist economies, the Middle Eastern crescent, and China have little recorded spread of 32 Table 1.3 Evolution of the HIV-AIDS epidemic HIV incidence HlV prevalence AIDS-related (millions) (millimzs) deaths (millions) Region 1990 2000·' 1990 2000• 1990 2000·' Demographically developing groupb 1.6 2.5 7.4 25 0.3 1.7 Sub-Saharan Africa 1.1 1.0 5.8 12 0.3 0.9 Asia' 0.3 1.3 0.4 9 0.6 EME and FSEd 0.1 1.4 0.1 0.1 Total 1.7 2.5 8.8 26 0.4 1.8 ** Less than 0.05 million. Note: Incidence refers to new infections in a given year; prevalence refers to the total number of persons infected. a. Conservative estimates. b. The countries of the demographic regions Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent. c. India, China, and the demographic region Other Asia and islands. d. EME, established market economies; FSE, formerly socialist economies of Europe. Source: World Health Organization data. the virus. Spread of the virus may be about to mission, incubation periods, survival times, and occur even in these three regions. It takes six to ten the role of such factors as STDs-are not well years, on average, for an HIV-infected adult to de- quantified, and accurate projections are therefore velop AIDS. Thus, regardless of future changes in impossible. In the African communities that are transmission of the virus, there will certainly be an most severely affected, early assessments pre- increasing number of AIDS cases over the next few dicted absolute declines in population. Later views years. suggest that population growth will continue, al- It is difficult to predict the future course of the beit at a reduced rate. Trial projections for Sub- epidemic because so little is known about the dy- Saharan Africa, based on a high assumption of namics of HIV transmission. WHO has projected HIV prevalence of 60 million infections worldwide that in 2000, 2.5 million people will be newly infec- in 2000, suggest a reduction in life expectancy by ted with HIV, HIV prevalence will have reached 26 2010 of about six years, in comparison with a low- million, and AIDS deaths will total 1.8 million (see HIV model, and a 25 percent increase in adult mor- Table 1.3). These estimates are conservative, since tality. The effect on population growth would still they assume that the rate of new infections in Af- be modest: a reduction of about 0.25 percentage rica will slow somewhat and that new transmis- point a year, to an annual rate of 2.7 rather than sion will be concentrated in India and in the Other 2.95 percent in 2005-10. In areas such as Thailand Asia and islands region. If no effective interven- where fertility and mortality rates are much lower tions to slow transmission are introduced, the total than in Sub-Saharan Africa, AIDS may well con- number of deaths may be twice as large, in which tribute to actual population declines over a period case AIDS would be responsible for 8 percent of of thirty years or more. the global burden of disease by 2000 instead of the 3.5 percent implied by the estimates in Table 1.3. Drug-resistant diseases If, however, sexual behavior changes dramatically over the next decade, even the conservative pro- Microbes evolve as a result of natural mutation, jections given here may prove too pessimistic. which throws up new threats, and of drug thera- Relatively modest reductions in numbers of casual py-induced selection, which fosters drug resis- sexual partners, or in the prevalence of STDs-or, tance. Two major new threats have arisen in this alternatively, substantial increases in condom century: the influenza virus responsible for the use-could reduce transmission significan!ly. 1918-19 worldwide epidemic, and HIV. Early (and still tentative) findings from Thailand The evolution of drug resistance, partly driven are encouraging; perhaps behavior really will by incomplete or inadequate treatment, is more change. gradual and less dramatic but no less serious. The Opinions differ concerning the effects of AIDS everyday bacteria responsible for pneumonias and on population growth. The variables needed to diarrheas have become resistant to the older anti- model the epidemic-including baseline rates of biotics and will gradually do the same with the infections, behavioral risk factors, efficacy of trans- newer antibiotics developed over the past few dec- 33 Life expectancy is related to income, but the emerging. In the developing world the diseases for relationship has shifted upward during the twentieth which drug resistance is already a major issue (re- century. spiratory infections, tuberculosis, STDs, and ma- laria) accounted for almost one-fifth of the GBD in 1990. Figure 1.9 Life expectancy and income per This steady evolution of drug-resistant microbes capita for selected countries and periods poses challenges for research and for health care. Better understanding of infectious agents is needed as a basis for the development of new ther- Life expectancy (years) apies. Health providers must consider the effects 80 of drug use on the evolution of resistant microbial strains. Basic scientific advances can contribute to tracking resistance, as recently shown by develop- 70 ments in identifying drug-resistant strains of tu- berculosis. There is little reason to hope for perma- nent success in humanity's struggle against 60 infection; investments in scientific research and vigilance on the part of public health authorities will remain indispensable. 50 These problems arising from microbial evolution are most severe in Sub-Saharan Africa. If efforts to control the spread of HIV fail, by 2000 an addi- 40 • tional 1 million people in the region will be dying from AIDS each year. Most of them will be young adults who would otherwise have gone on to live 30 ~----L-----~----~----~----~ healthy lives. If malaria develops resistance to all 0 5,000 10,000 15,000 20,000 25,000 available drugs, the number of people it kills every Income per capita year could increase sharply, from the expected 1.5 (1991 international dollars) million deaths in 2000 to 2.3 million. Sub-Saharan Africa might also suffer from a tuberculosis epi- demic, driven partly by drug resistance and partly Note: International dollars are derived from national currencies not by use of exchange rates but by assessment by the spread of the disease by people with HIV of purchasing power. The effect is to raise the relative Extrapolation of current trends indicates an annual incomes of poorer countries, often substantially. For total of 8.5 million premature deaths in Sub- illustrative country comparisons and a more detailed explanation, see Table 30 in the World Development Saharan Africa by the end of the century. But it is Indicators. all too easy to project a figure as high as 11.5 mil- Source: Preston, Keyfitz, and Schoen 1972; World Bank data. lion, accompanied l:>y a sharp reduction in life expectancy. Lessons from the past: explaining declines in mortality ades. Tuberculosis resistant to the standard mix of antibiotics is becoming more common in the in- Three factors have been important in the dramatic dustrial world, and it kills many of those who con- and unprecedented mortality declines of the past tract it. Chloroquine-resistant malaria has now hundred years and in the still more dramatic de- spread to practically all endemic areas, and al- clines in developing countries since World War II. though new drug therapies are available, wide- These factors are income growth, improvements in spread resistance even to several of these is being medical technology, and public health programs reported. Malaria has thus reemerged as a signifi- combined with the spread of knowledge about cant health risk in urban areas that had been free health. of it for several decades. Resurgence of the disease has been abetted by the reduced effectiveness of Income growth vector control, which is partly attributable to the increasing resistance of mosquitoes to standard in- Increased income allows people, particularly the secticides. Resistant strains of many STDs are also poor, to buy more food, better housing, and more 34 health care. Throughout the twentieth century life health measures were responsible for only a small expectancy has been strongly associated at the na- part of the progress made. In the late nineteenth tional level with income per capita, as seen in Fig- century Robert Koch showed that the bacterium ure 1. 9. Life expectancy rises rapidly with income M. tuberculosis causes tuberculosis, and people be- at low levels of income, particularly when income gan to understand about germs. They took simple per capita is less than $3,000 (1991 purchasing precautions-preparing food and disposing of power dollars). The figure shows, however, that waste hygienically, eliminating flies, and quaran- the relationship has shifted upward over each tining sick family members-that had far-reaching thirty-year period, so that more health is realized benefits. Recent research has shown that child for a given income. For example, in 1900 life expec- mortality differed little by education or even by tancy in the United States was about forty-nine income in the United States in the last decade of years and income per capita in 1991 dollars was the nineteenth century but that differences wid- about $4,800. In 1990 that income per capita would ened sharply as child mortality fell in the early be associated with a life expectancy of about sev- twentieth century. The implication is that afflu- enty-one years. This upward shift shows that ence and education made little difference until sci- health depends on more than income alone. entific knowledge showed households how to reduce the dangers to their health. Since better- Improvements in medical technology educated individuals acquire and use new infor- mation more quickly, this emphasis on knowledge Before the 1930s medical technology had little to helps to explain the large differences in child mor- offer humanity, with the exception of smallpox in- tality by mother's education observed in develop- oculation, the use of which was widespread in Eu- ing countries today. rope from the late eighteenth century onward, and diphtheria antitoxin, discovered in 1894. Starting The potential for effective action in the 1930s, with the introduction of antibacterial drugs and new vaccines, a wide range of effective The recent declines in mortality in the developing interventions has become available to counter world have been sharper than the earlier declines most communicable diseases. in the high-income countries and more influenced The effect of these technological improvements by technical advances. To take one example, Sri on health has depended on other factors, such as Lanka achieved a remarkable decrease in mortality income gains for the poor, increased schooling, after World War II; the crude death rate fell from and public policies that affect health systems. As a 21.5 per 1,000 in 1945 to 12.4 in 1950. Some 23 result, outcomes have varied widely by country, percent of that drop has been attributed to the ma- even within the same region. For example, in the laria eradication program, which mainly involved early 1980s child mortality was three times higher spraying of insecticide from the air. The same ap- in Mali than in Botswana, six times higher in Boli- proach also did much to control yellow fever, on- via than in Chile, and five times higher in Ban- chocerciasis, and many other diseases. Wide- gladesh than in Sri Lanka. Between the early 1960s spread use of newly available antibiotics against and the early 1980s child mortality fell 20 percent conditions such as yaws in Africa helped to reduce in Bangladesh but 65 percent in Sri Lanka, 10 per- STDs and (probably) acute respiratory infections. cent in Uganda but 50 percent in Kenya, and 10 Improvements in water and sanitation curbed the percent in Haiti but nearly 80 percent in Costa spread of disease, particularly in towns and cities. Rica. Some countries have clearly made better use Whereas at the beginning of this century child of the available technology than others. mortality rates in today's high-income countries were much higher in urban than in rural areas, the Public health and the spread of knowledge opposite has been true of the developing world since 1950. The introduction of public health measures-par- Vaccination, too, has produced dramatic results, ticularly clean water, sanitation, and food regula- including the eradication of smallpox and the elim- tion-certainly contributed to the decline in child ination of paralytic polio in the Western Hemi- mortality in the late nineteenth century and to the sphere. About 80 percent of the world's children accelerated decline in the early twentieth century. are now vaccinated against the main infectious dis- The geographic distribution of mortality declines eases of childhood, thanks largely to the Expanded suggests, however, that until people began to un- Programme on Immunization (EPI) sponsored by derstand the sources of poor health, such public WHO and UNICEF. It is estimated that the EPI 35 prevented the deaths of 2.6 million children in more than 20 percent of the GBD in 1990. Epide- 1990 alone. Substantial benefits have also come miological advances are giving governments and from simple curative measures such as oral re- households warning of the enormous health toll hydration to avert death from diarrhea and a short from smoking. But if the full benefits of scientific course of drugs for curing tuberculosis. But there advances are to be realized, parallel developments is much more still to be done: in 1990 childhood are needed to empower households so that they deaths from diarrhea and immunizable diseases can put the advances into practice. The key devel- alone accounted for 12 percent of the GBD. opments are schooling, particularly of girls; in- The march of science has increased both the come growth, particularly of the poor; and a flex- range of inexpensive clinical treatments and prac- ible, responsive health system able to provide the tices and the potential performance of health sys- necessary preventive and curative care. The poli- tems. It is now possible to treat at low cost tuber- cies needed to achieve these developments are the culosis, STDs, many respiratory infections, and subject of the remainder of this Report. risky deliveries, which together accounted for 36 Households and health What people do with their lives and those of their their effects on the conditions facing households children affects their health far more than any- and individuals, can be important to people ' s thing that governments do. But what they can do health. Especially in the poorest countries, policies is determined, to a great extent, by their income that accelerate income growth and reduce poverty and knowledge-factors that are not completely make it possible for people to afford better diets, within their control . In every society, moreover, healthier living conditions, and better health care . the capabilities, income, and status of women ex- Policies to expand educational opportunities, par- ert a powerful influence on health. Because of ticularly for girls, help households achieve health- these interrelations, government actions, through ier lives by increasing their access to information Economic growth and investments in human resources interact to improve well-being. Figure 2.1 Mutually reinforcing cycles: reduction of poverty and development of human resources Health /;mpm"'m''"~ \ Expansion of Empowerment ..___.. schooling of women opportunities Economic growth ~ policies that ~ benefit the poor 37 and their ability to make good use of it. The same Household capacity: income and schooling goes for policies that work to ensure effective and accessible health services for all. When all these Within the household, health improves rapidly as policies are combined, they create a virtuous cycle people escape from poverty and low education in which reduction of poverty and improvements (Box 2.1). Beyond the household, every society's in health reinforce each other (Figure 2.1). health services are affected by its national income, Box 2.1 Progress in child health in four countries In the 1960s a child born in the developing world had a only 60 percent as fast . Thus, improvements in school- 77 percent chance of surviving the first five years of ing were most significant in Cote d'lvoire, whereas in life . About thirty years later, the chances of survival Egypt growth in income per capita accounted for fully have improved to 89 percent. How much did income half of the gain in child health. growth and expansion of schooling contribute to this Costa Rica and japan followed the same pattern as gain? What was the role of other factors, such as prog- Egypt: growth of income per capita contributed sub- ress in science and medicine' Some answers to these stantially more to child health gains than did educa- questions emerge from data on child survival from sev- tional improvements. Technical progress (estimated enty-five industrial and developing countries for the using the passage of time as a proxy), however, was period between 1960 and 1987 (see note to Appendix important in.Japan, whereas in Costa Rica and Egypt it table A.3). This box reviews the results for four coun- mattered less than improvements in education. Except tries with different income levels-Costa Rica, Cote in japan, where people were already quite well edu- d ' lvoire, Egypt, and japan (see Box table 2.1 and Box cated in 1960, the analysis probably underestimated figure 2.1). the contribution of schooling because it dealt with the In all four countries, part of the gain in child health schooling of all adults rather than of women alone. depends on the initial levels in 1960 of schooling in the Child health is particularly affected by maternal educa- population and of income per capita. Because school- tion, and the number of years of schooling received by ing and income per capita produce health benefits that younger women is likely to have risen much faster be- often persist through time, health in a population may tween 1960 and 1987 than was the case for the adult be improved simply by maintaining initial levels of population as a whole. schooling and income. In Costa Rica, where in 1960 income per capita was relatively high and schooling Box table 2.1 Child health, income per capita, was already widespread, initial conditions accounted and schooling in Costa Rica, Cote d'lvoire, for 58 percent of the gain in child· health between 1960 Egypt, and Japan, 1960-87 and 1987. In Cote d'lvoire and Egypt, where the levels Cote of schooling and income per capita were modest in Indicator Costa Rica d'lvoire·' Egvrt Japan 1960, initial conditions contributed only about one- 1960 fifth to one-quarter of the gains. In japan, too, these Child survivaJb 0.89 0.72 0.74 0.96 initial conditions contributed a fifth of the gains in child Income per capita health, but this is not surprising in a ·country where a (1987 international baby's chance of survival was already very good in dollars)< 2,160 1,021 557 2,701 1960. Average schooling In reality, of course, income and schooling have im- of adults (years) 4.0 .0.2 3.0 10.7 proved in all these countries, and these improvements Average a111111al percmtage clra11ge, 1960-87 contributed to further gains in child survival. In Cote Child survivalb 0.4 0.8 0.6 0.1 d'lvoire ·educational improvements did the most for Income per capita 2.3 3.2 5.2 5.3 child health, accounting for 66 percent of the gains be- Schooling of adult tween 1960 and 1980. For .Egypt, by contrast, the .figure population 2.0 11.8 2.4 0.2 was only 21 percent. A comparison between Cote Elasticity of child sun>ival witlr respect to: d'lvoire and Egypt is illuminating. The probability of Income per capita" 0.04 0.06 0.06 0.02 surviving the first five years of life started at similaF Schooling of adult levels in both countries and improved at comparable populationd 0.03 0.04 0.04 0.02 rates. In both, too, the responsiveness of child survival a . Data refer to 1960-80. b. Child survival refers to the probability of surviving from birth to income per capita and to the schooling of adults was through age 5. comparable. In Cote d'Ivoire, however, adult schooling c. Income is adjusted for differences in purchasing power parity. started from much lower levels than in Egypt but in- d . Elasticities denote the percentage change in the probability of surviving from birth through age 5 corresponding to a 1 percent creased five times faster. Income per capita in Cote change in the indicated v.uiable. d'lvoire was nearly twice Egypt's in 1960 but then grew Soura: Lau and others, b.Kkground paper. 38 and its ability to acquire and apply new scientific the more likely its people are to live long and knowledge depends on the level of schooling in healthy lives. Of course, this effect tapers off as the population. income rises: a doubling of income per capita (ad- justed for purchasing power parity) from, say, The influence of income on health $1,000 in 1990 corresponds to a gain of eleven The higher a country's average income per capita, years in life expectancy, whereas a doubling from Box figure 2.1 Gains in child health, 1960-87, and share contributed by various factors Costa Rica Cote d'Ivoire 5 Egypt 3 Japan 9 3 • Initial levels of schooling and income per capita 50 1!1 Increase in schooling • Increase in income per capita D Technicill progress Note: The <1rea of the circle is proportional to the absolute increase, over the period 1960-87, in the probability of surviving to age 5. For Cote d'Ivoire ch11 nges for the period are extrapolated from the observed change during 1960-80. Source: Lau and others, background paper. 39 Within the same city, health status is worse in poorer people in poverty matter as well. In industrial areas. countries life expectancy depends much more on income distribution than on income per capita, and it has been rising faster in countries with im- Figure 2.2 Child mortality in rich and poor proving income distribution. Japan and the United neighborhoods in selected metropolitan Kingdom had similar income distributions and life areas, late 1980s expectancies in 1970, but they have diverged since then. Japan now has the highest life expectancy in 1!!11 Poor neighborhoods the world and a highly egalitarian income distribu- 0 Rich neighborhoods tion. In the United Kingdom, where income dis- parity has widened since the mid-1980s, life expec- tancy is now more than three years shorter than in Percentage deviation from national mean Japan. 0.50 In developing countries the number of people in poverty is an especially important reason for dif- ferences in health. One study looked at twenty- 0.25 two developing countries with comparable data on poverty (defined as the share of the population consuming less than $1 a day at 1985 purchasing power parity prices) and found that variation in 0 the prevalence of poverty and in per capita public spending on health is important in explaining cross-country variation in life expectancy. Differ- -0.25 ences in income per capita became unimportant once those two factors were taken into account. This does not mean that income growth is irrele- vant to increased life expectancy; rather, its main -0.50 effect lies in how much it reduces poverty and Cairo, Egypt Colombo, Lima, Peru supports public health services. In the twenty-two Sri Lanka countries, roughly one-third of the effect of eco- nomic growth on life expectancy came through Note: Child mortality indexes for each neighborhood are poverty reduction and the remaining two-thirds calculated by dividing the observed number of deaths through increased public spending on health. In among children of women in the sampled households of a neighborhood by the expected number (given the Sri Lanka an increase in per capita public spending distrioution of women by the length of time they have on health was twenty-two times more effective in been bearing children and the national average child mortality levels at each duration of childoearing); reducing infant mortality than was the same in- Percentage deviations from the national average are crease in average income. obtained by subtracting 1 from a neighborhood's index and multiplying the result by 100. Neighborhoods in Within countries, too, health correlates strongly each city were ranked according to the proportion of with poverty. In India, Indonesia, and Kenya child houses with concrete floors. Poor neighborhoods were , the lowest 25 percent in this ranking; rich.neighborhoods' ' mortality is higher in states or provinces with were the top 23 percent. ,, larger proportions of poor people. Within cities, Source: Calculated from data from national Demographic there are large differences in child survival be- and Health Surveys. tween rich and poor neighborhoods (Figure 2.2). And children in poor families are less healthy. In Madurai, the second largest city in India's Tamil $4,000 is matched by a gain of only four years (see Nadu State, children ages 2-9 in the poorest Figure 1.9 in Chapter 1). Income growth has more. households were more than twice as likely to suf- impact in poor populations because additional re- fer from serious physical or mental disabilities as sources buy basic necessities, particularly food children from slightly better-off families. and shelter, that yield especially large health Poor people are vulnerable to disease not only benefits. because of poor living conditions but often also for Because poverty has a powerful influence on work-related reasons. In Adana, Turkey, the risk health, it is not just income per capita that is rele- of malaria is significantly greater among migrant vant; the distribution of income and the number of workers than for the local population; the average 40 number of anopheline mosquito bites per person Child mortality falls faster in countries where income was five times greater in the tents of these workers per capita is growing rapidly. than in the houses of village residents. In Sri Lanka one of the commonest causes of pesticide poisoning is leaky knapsack sprayers; surveys Figure 2.3 Declines in child mortality show that although farmers are aware of the risks and growth of income per capita involved, they continue to use broken equipment in sixty-five countries because they cannot afford to replace or repair it. The distribution of income within households Annual rate of change in child also affects health. Increasing women's access to mortality, 1970-88 (percent) income can be especially beneficial for the health 1 of children. In Brazil income in the hands of the 0 r• • ~-+----------------------------__, mother has a bigger effect on family health than income controlled by the father. In Jamaica house- • • -2 • holds headed by women eat more nutritious food •• • •• • than those headed by men; they also spend more of their income on child-centered goods and signif- • • -4 • • icantly less on alcohol. In Cote d'Ivoire a doubling of household income under women's control re- •• -6 duces the share of alcohol in the family budget by • • 26 percent and the share of cigarettes by 14 per- • ••• • • cent. In Guatemala it takes fifteen times more -8 • • spending to achieve a given improvement in child • nutrition when income is earned by the father than -10 • -1 0 2 3 4 5 6 7 when it is earned by the mother. Although a work- Annual income per capita growth ing mother may breastfeed less and have less time rate, 1960-88 (percent) for child care-both of which could be detrimental to her children's health-evidence from numerous developing countries suggests that this harm can Note: Child mortality refers to the probability of dying be offset by the health benefits that her earnings between birth and age 5; the period over which the rate of- mortality decline is averaged differs from the period bring. used for income per capita growth to take account of lags Because fewer people live in poverty as average in the relation between the two rates. Source: World Bank data. incomes rise, there is generally a strong link be- tween incomes and health status. Across coun- tries, more than 75 percent of the difference in health is associated with income differences. In- deed, this relation is not merely associative but countries and by as much as 4 to 8 percent in Cote causal and structural: income growth leads di- d'Ivoire and Ghana. rectly to better health. In a sample of fifty-eight These findings highlight the costs to health of developing countries, a 10 percent increase in in- slow economic growth. Child health has been im- come per capita, all else being equal, reduced in- proving everywhere, but gains are much less rapid fant and child mortality rates by between 2.0 and in countries with slow income growth (Figure 2.3). 3.5 percent and increased life expectancy by a During the 1980s the economic performance of de- month. This estimate reflects the total impact of veloping countries was mixed, with income per income on health; it includes effects working di- capita constant or falling, and in some countries rectly through income (such as food consump- the incidence of poverty rose (Table 2.1). Had eco- tion), as well as indirectly through factors that an; nomic growth been as fast in the 1980s as in the themselves mainly determined by income (access period between 1960 and 1980, in 1990 alone an to safe water and sanitation, availability of physi- estimated 350,000 infant deaths, or 6 percent of cians, and so on). Studies based on individual total infant deaths, would have been averted in households corroborate the cross-country results. developing countries (excluding India and China). A 10 percent advantage in income reduces infant In Africa and Latin America, where average mortality by between 1 and 2 percent in Nigeria, growth was 2.5 percentage points slower during Sri Lanka, Thailand, and several Latin American the 1980s, the saving in babies' lives in 1990 would 41 Table 2.1 Poverty and growth of income per Bangladesh, for example, found that over a period capita by developing region, 1985 and 1990, of two years following the death of a mother, mor- and long- and medium-term trends tality rates, in comparison with those of children Annual percentage with living mothers, were twice as high for boys Head-count index change in income ofpovertya per capita and three times as high for girls. Region 1985 1990 1970-92 1982-92 Education greatly strengthens women's ability All developing to perform their vital role in creating healthy countries 30.5 29.7 1.7 0.8 households. It increases their ability to benefit Sub-Saharan from health information and to make good use of Africa 47.6 47.8 -0.2 c-1.1 health services; it increases their access to income East Asia 13.2 11.3 5.3 6.3 and enables them to live healthier lives. It is not South Asia 51.8 49.0 2.0 3.0 surprising, therefore, that a child's health is af- Eastern Europe 7.1 7.1 1.2 1.7 fected much more by the mother's schooling than Middle East and North Africa 30.6 33.1 0.1 -1.6 by the father's schooling. Data for thirteen African Latin America countries between 1975 and 1985 show that a 10 and the percent increase in female literacy rates reduced Caribbean 22.4 25.2 1.1 -0.2 child mortality by 10 percent, whereas changes in Note: Regional data on annual change in income per capita refer to male literacy had little influence. Demographic unweighted country averages. The regions used in this table are as defined in the World Development Indicators, except for Eastern and Health Surveys in twenty-five developing Europe, which includes Albania, Bulgaria, Hungary, Poland, Ro- countries show that, all else being equal, even one mania, the former Czechoslovakia, and the former Socialist Federal Republic of Yugoslavia. Disaggregated data for the last two are not to three years of maternal schooling reduces child yet available. mortality by about 15 percent, whereas a similar a. Estimated share of the population consuming less than $32 per person per month at 1985 purchasing power parity prices. level of paternal schooling achieves a 6 percent Source: For poverty index, World Bank 1993c; for change in income reduction. The effects increase when mothers have per capita, World Bank data. had more education; in Peru, for example, seven or more years of maternal schooling reduces the have been as much as 7 and 12 percent, respec- mortality risks nearly 75 percent, or about 28 per- tively. Latin America's recession in 1983 is esti- cent more than the reduction for the same level of mated to have caused 12,000 additional deaths of paternal schooling (Figure 2.4). Countries that in babies, or 2 percent of all infant deaths in that year. 1965 had achieved near-universal enrollment for And because slow economic growth hampers pov- boys but much less for girls had about twice the erty reduction and constrains spending on health, infant mortality in 1985 of countries with a smaller schooling, and other services, it is highly likely boy-girl gap. that the health of the poor suffered dispropor- The advantages that a mother's schooling con- tionately in the 1980s. fers on her children's health are felt even before birth. In developing countries better-educated The influence of schooling on health women marry and start their families later, dimin- ishing the risk to child health associated with early Households with more education enjoy better pregnancies. Educated women also tend to make health, both for adults and for children. This result greater use of prenatal care and delivery assis- is strikingly consistent in a great number of tance. In a study in Lima that controlled for service studies, despite differences in resec:rch methods, availability and socioeconomic status, 82 percent time periods, and population samples. of women with six or more years of education sought prenatal care, compared with only 62 per- MATERNAL SCHOOLING AND CHILD HEALTH. In cent for women with no education. most households women have the main responsi- Following birth, the children of educated bility for a broad range of activities that affest mothers continue to enjoy other health-enhancing health. They manage household chores, keep the advantages: better domestic hygiene, which re- house clean, process foods and prepare meals, duces the risk of infection; better food and more feed and care for young children, and look after immunization, both of which reduce susceptibility the sick. Women's own health and their efficiency to infection; and wiser use of medical services. A in using available resources have an important study of women in Bangladesh documented how bearing on the health of others in the family, par- educated women kept their homes and children ticularly children. A study of children under 10 in tidier and cleaner than uneducated women and 42 Child health depends more on the mother's than on the father's schooling. Figure 2.4 Effect of parents' schooling on the risk of death by age 2 in selected countries, late 1980s 4-6 years of schooling 7 or more years of schooling Percentage reduction in child mortality Percentage reduction in child mortality (in relation to parents with no schooling) (in relation to parents with no schooling) 80 60 40 20 0 0 20 40 60 80 Indonesia Kenya Morocco Peru • Mother Ell Father Source: Hobcraft 1993. expressed a preference for water from tanks or example, 24 percent of the children of mothers tubewells at home rather than from canals or with no education were stunted, compared with rivers. In Brazil, India, and Nigeria better-edu- only 11 percent of children of mothers with some cated households are willing to pay 6 to 50 percent elementary schooling. Educated women are an im- more than other households for improved water portant part of the reason for the impressive supplies. health achievements of China, Costa Rica, India's Educated mothers are also better at getting in- Kerala State, and Sri Lanka, despite relatively low formation on health and acting on it. In Brazil the incomes. child health benefits of a father's education work mostly through his income, whereas almost all the SCHOOLING AND ADULT HEALTH. Personal habits effect of maternal education comes from learning and life-style choices affect adult health enor- about health through newspapers, television, and mously. Because educated people tend to make radio. In Thailand mothers with primary educa- choices that are better for their health, there is of- tion were 30 percent more likely than mothers ten a strong relation between schooling and with no education to treat childhood diarrhea with health. A study of U.S. life expectancy at age 25 oral rehydration therapy or a homemade solution found that between the highest and the lowest of salt and sugar; this figure rose to 90 percent for levels of education, the difference was about six mothers with secondary or higher education. Sim- years for white men and about five years for white ilar results have been reported in countries as di- women. These differences-which may partly re- verse as Burundi, Colombia, Ghana, Morocco, and flect differences in income associated with educa- Nigeria. And well-educated mothers often manage tion-have persisted since the 1960s. to reduce the damage that poverty does to health. The same pattern occurs in developing coun- Among poor rural households in Cote d'Ivoire, for tries. Surveys in Cote d'Ivoire, Ghana, Pakistan, 43 Schooling reduces the risk of adult ill health. ample, when the AIDS epidemic began, infection was initially concentrated among well-educated elites, but these same groups were the first to Figure 2.5 Schooling and risk factors for change their life-styles as information became adult health in Porto Alegre, Brazil, 1987 available about the disease and its prevention. In Brazil between 1982 and 1985, 79 percent of those infected had completed postsecondary education; Prevalence of risk (percent) by the late 1980s this group's share of cases had 100 fallen to 33 percent. Even more striking is the way that well-educated people have changed their be- havior on smoking. In the United Kingdom the proportion of smokers among adults declined be- tween 1958 and 1975 by 50 percent among the most educated but hardly changed among the least educated. In the United States between 1974 and 1987 the smoking habit declined nine times faster in the highest education group than in the lowest. The corresponding difference was twofold in Canada and threefold in Norway. Policies to strengthen household capacity Because people's ability to improve their own D No schooling health depends so much on income and educa- • Postsecondary schooling tion, the policy conclusions are clear: governments should work to boost economic growth, reduce poverty, expand schooling (particularly for girls), Source: Achutti and others 1988. and help strengthen women's ability to care for their families. This section deals with each of these points in turn. and Peru show that respondents whose parents Promoting growth and reducing poverty were educated were more likely to have living par- ents than those with uneducated parents. In Peru During the 1980s the pattern of economic growth 72 percent of the educated fathers of respondents in developing countries was very uneven. Income ages 25-29 were still alive at the time of the survey, per capita grew at more than 6 percent a year in compared with only 55 percent of the uneducated East Asia but remained constant or fell in many fathers. In Jamaica education had a bigger influ- other countries. The disappointing record re- ence on adult health than did income, particularly flected the impact of adverse external shocks as before age 50. Death rates for specific diseases also well as poor domestic policies. Nonetheless, some show educational differentials. In Russia death economies grew rapidly despite the external from coronary heart disease was two to three shocks, showing that a great deal can be done by times more common for the poorly educated than developing countries themselves. for those with higher education. In Brazil those Because it is difficult to reduce poverty without who were illiterate or who had only primary economic growth, establishing sound economic schooling were about five times more likely to policies for growth is one of the most valuable have high blood pressure than those with post- things a government can do. Development strate- secondary schooling. The first group was also sub- gies also need to emphasize broadly based growth stantially more inclined to obesity, alcohol and to- to give the poor better income-earning oppor- bacco consumption, and lack of exercise (Figure tunities and better access to a range of social ser- 2.5). vices. To protect the most vulnerable members of The advantages of education continue to show society, it is appropriate for governments to make up when new types of health risk appear. For ex- transfers and other special arrangements. 44 POLICY REFORM AND ADJUSTMENT LENDING. As a equal, middle-income countries in the "intensive" consequence of the economic crisis of the early group boosted their growth rates during 1986-90 1980s, many developing countries changed their by an estimated average of about 4 percentage economic policies. They adopted macroeconomic points a year over what would probably otherwise reforms intended to achieve price stability and sus- have occurred. The low-income countries, espe- tainable internal and external monetary balance cially in Sub-Saharan Africa, did less well; for and made microeconomic and institutional re- them, the benefit was 2 percentage points. forms to promote the efficient use of resources and Since health is helped by economic recovery and faster economic growth. These changes typically faster long-term growth, adjustment lending, by involved cuts in public spending, the opening of facilitating economic progress, benefits health in the economy to competition, liberalization of the long run. When a government has to adjust- prices, measures to improve the efficiency of pub- in response to economic shocks or to rectify mis- lic expenditure, and the development of a sound taken past policies-the whole society, poor and financial system and other institutions needed in a nonpoor, may suffer short-run reductions in em- well-functioning market economy. ployment and wages. But the resulting fall in in- To support these reforms, the World Bank and come is caused not so much by policies associated the International Monetary Fund have extended with adjustment lending as by the necessity for the adjustment lending. The purpose of this lending is country to curb its consumption; without adjust- to cushion an economy during the transitional ment loans, even greater decreases in consump- phase to its new growth path. Adjustment lending tion would probably have been necessary. None- is therefore essentially an investment in a more theless, adjustment lending can take five or more productive future. It has been central to the re- years to bear fruit, and the transition can be pain- forms in Latin America and Sub-Saharan Africa ful because incomes may fall in the short run. Evi- and important in other regions as well. Its role will dence from Sub-Saharan Africa and Latin America continue in the 1990s: it is already a major channel suggests that economic downturns are associated of assistance for the formerly socialist economies; _ with less favorable child mortality outcomes than it is being used for the first time in India; and it has would be predicted from long-term trends. In both old and new clients in other parts of the countries where child mortality rates are declining world. over time, for example, adjustment lending would Nonetheless, adjustment lending remains con- be associated, in the short run, with a slower rate troversial. Does it really raise long-term growth? of decline. To minimize such adverse effects, some Do the poor suffer as a consequence of such ad- countries have begun to use resources, including justment policies as cuts in public spending and adjustment loans, to support nutrition programs liberalization of food and other prices? How is for vulnerable children, as well as basic health and health affected? The answers to these questions other social services targeted to the poor. are complicated because adjustment lending is nei- ther necessary nor sufficient for policy reform. ADJUSTMENT LENDING AND PUBLIC EXPENDITURE Some of the most dramatic "adjustment" reforms ON HEALTH. Because cuts in government spending took place without adjustment lending (as in Chile are usually central to an adjustment program, and Viet Nam), and some countries that received health spending is likely to be reduced. In many adjustment loans did little or nothing to pursue countries early cuts were indiscriminate and failed reforms (for example, Tanzania and Zambia). In to preserve those elements of the health system addition, because a country's economic perfor- with the strongest long-term benefits for health. mance is affected by many factors, it is hard to Drugs were often cut more heavily than personnel isolate the part played by adjustment lending. because it is difficult to lay off public employees. Despite these difficulties, World Bank studies on Cote d'Ivoire's experience illustrates the mistakes the impact of adjustment lending are reveali!lg. that occurred in some early programs of economic The research looked at countries in the "intensive adjustment. With real income per capita falling 19 adjustment lending" group (which includes coun- percent between 1980 and 1984, the government tries that received at least two structural adjust- cut public spending, among other measures. ment loans or three sectoral adjustment loans by Health expenditure dropped in real terms by 12 1990, with the first loan started by mid-1986) and percent between 1981 and 1984. But personnel found that in general they did achieve faster costs were not cut; instead public expenditures on growth than in other countries. All else being medicines and materials absorbed the reduction, 45 Public spending on health recovered faster in countries that received adjustment lending. Figure 2.6 Deviation from mean levels of public spending on health in countries receiving and not receiving adjustment lending, 1980-90 Deviation in share of country income spent Deviation in per capita spending (percentage points) (1991 dollars) 0.35 8 0.30 6 0.20 I\ I . 1. NonreCipients I 4 \ Recipients _.... ~' - / .\ [.)\ ~ Nonrecipients I "" I 0.10 2 \ 1\ I 0 Recipients 0 \ ......... -I - -0.10 -0.20 1980 81 82 " ~ I 83 84 ' 85 1 / 86 87 """" -v88 89 v 1-- 90 -2 -4 -6 " "" \/ / / 1980 81 82 83 84 85 86 87 88 89 90 Nole: Recipients of adjustment lending include countries that received two structural adjustment loans or three or more adjustment operations, all effecti ve by June 1990, with the first operation effective in or before June 1986. Data were available for the following countries in this group: Bolivia, Brazil, Chile, Costa Rica, Kenya, Republic of Korea, Mauritius, Mexico, Pakistan, Philippines, Turkey, and Uruguay. Nonrecipients are countries that had not received adjustment lending by June 1990. Data were availabll' for Burkina Faso, Dominic,m Republic, Egypt, El Salvador, Guatemala, Liberia, Malaysia, and Papua New Guinea . Source: Yazbeck, Ta n, <1lld Tanzi, background paper. shrinking in real terms by more than one-third ticularly in rural areas, have been improving as a during the first half of the 1980s. In rural clinics, result . already precarious supplies of basic consumables Various studies have assessed the effect of ad- became even scarcer. justment programs on public spending on health. The implications for child health looked grim. Most have found that central government expen- Cross-sectional data show that the nutritional sta- diture on health in countries with adjustment tus of Ivorian children is strongly related to the lending programs did not suffer more than else- availability of drugs in the community. All else be- where; this result, however, is not definitive be- ing the same, the difference in height-for-age (a cause state and local governments are often re- measure of long-run nutritional status) of children sponsible for a substantial share of public in communities lacking basic medicines and those spending on health . More comprehensive data in well-supplied communities was equivalent to available for twenty countries during 1980-90 more than one-third of the difference between the show that in both countries with and without ad- average child in Cote d'lvoire and in the United justment loans, public spending on health as a States. The health of children from poor families percentage of total country income declined in the suffers even more when drugs are unavailable. early 1980s in relation to the average for the dec- Since 1990 the government has begun putting ade. In 1985-90, however, health spending recov- more resources into nonwage health inputs: their ered much faster in countries with adjustment pro- share of the health budget rose from 20 percent in grams . Similarly, per capita public spending on 1991 to 24 percent in 1993. Health services, par- health also recovered faster in such countries (Fig- 46 ure 2.6). Unfortunately, the data are not good they complete even the first few years of basic edu- enough to allow any judgment on whether adjust- cation. Fewer than 60 percent of first-graders in ment programs directly helped to ensure that pub- the lowest-income countries and about 70 percent lic spending on health was efficient. (And, as this of those in the lower-middle-income countries Report will show, not all health spending deserves reach the last year of primary school. to be protected; some of it is inefficient and Enrollments are particularly low in isolated rural regressive.) areas, for lower socioeconomic groups, and for girls. In developing countries as a group, about 10 Expanding and improving schooling percent of boys ages 6-11 do not enroll; for girls in the same age group the figure is 40 percent. Espe- In general, developing countries have made much cially in poor countries, the gaps can be substan- progress in expanding schooling since the 1960s, tial, as Figure 2.7 illustrates for India. But Sri but the trends conceal some shortcomings. In the Lanka's experience shows that this gender gap is poorest countries, especially in Africa, many chil- not an inevitable consequence of poverty. dren never go to school at all. In Mali, for example, Leaving aside the gaps in enrollment, education fully 77 percent of all school-age children never go in many countries is inadequate. Even children to school-a figure that has remained largely un- who complete primary school fail to acquire basic changed since 1980. Of those who do go to school, literacy and numeracy skills and scientific under- many often enroll late-thus missing the benefits standing. These weaknesses in the education sys- of early learning opportunities-and leave before tem reduce the potential impact of schooling on Substantial male-female gaps in schooling persist in some low-income countries. ' Figure 2.7 Enrollment ratios in India, by grade, about 1980 Enrollment ratio (percent) 100 1 2 3 4 5 6 7 8 9 10 11 12 Primary grades Middle grades Secondary grades fill Boys • Girls Note: The enrollment ratio is the share of children in the age group corresponding to a given grade who are enrolled in that grade. Source: Tan 1md Mingat 1992. 47 Box 2.2 Teaching schoolchildren about health: radio instruction in Bolivia Many personal habits and life-style choices that have basic health concepts and practices at a young age are important consequences for health are formed early in more likely to maintain them as parents. life. Health education in schools can help young people It is still too early to assess the long-term health im- make informed choices and so reinforce the effect of pact of teaching health lessons through radio. None- schooling on health. theless, pupils already show significant gains in several Bolivia has had success with health education areas, including ability to recognize symptoms of dehy- through radio lessons. Radio instruction was first intro- dration and knowledge of the proper mixture and ap- duced in 1987 for teaching mathematics and proved to plication of homemade oral rehydration solution. be both inexpensive (with costs per pupil averaging There is also evidence of increased hand-washing, and less than $1 a year) and effective. In 1989 the health and more households are using simple water filters . education ministries began to ·try out the use of radio The radio program is now being expanded. In 1993 for teaching disease prevention to schoolchildren, more than 1,000 third- and fourth-grade classrooms starting with a module on diarrhea prevention and oral will receive broadcasts of a new curriculum that in- rehydration. Children ages 8-13 were targeted because cludes lessons on cholera, personal and dental hy- they often take care of younger siblings and perform giene, acute respiratory infections, immunizations, in- household chores involving food preparation and sani- fectious diseases, and accident prevention. Nutrition, tation . They thus have a strong influence on their own environmental health, and self-esteem are to be added health and that of younger siblings. The radio health in 1994. In response to parents' requests, a comple- program emphasizes actions that a child can do for ·mentary community-based radio program is also being himself or herself or can do for or teach to a younger developed and tested. sibling. It rests on the belief that children who learn health. More important, they also reduce parents' - than a third of girls reach this level, and in China willingness to enroll their children, thus perpetu- and Latin America only 60 percent do . Achieving ating a vicious cycle of poor schooling and poor 5,000 hours of schooling for all children will thus health. In India, for example, more than 40 percent require significant and sustained policy effort in of parents in a nationwide survey cited either "not large parts of the world. To reinforce the effects of interested in education/further study" or "fail- school expansion on health, it may be useful to ure'' as the main reason for not sending their chil- include health topics in school curricula (see Box dren to school. 2.2). Much more needs to be done to extend educa- Incomplete enrollments reflect the combined in- tion in developing countries. Government support fluence of weak demand for education (which is for schooling at the lower levels and for girls is partly caused by low achievement) and inadequate especially justified: the benefits for society are schooling opportunities. To overcome these obsta- large, and poor families in low-income countries cles requires a combination of policies . Govern- typically undervalue the benefits of sending chil- ments can do more to ensure that lower levels of dren to school or are unaware of them. In addi- schooling receive priority in the allocation of pub- tion, for such families the opportunity costs of lic spending. In some countries, current levels of sending children to school are often high. A policy resources for primary schooling are insufficient to priority is to ensure that every child receives a support even minimal conditions for instruction. minimum quantity of schooling-say, 5,000 hours, In India public spending per primary pupil in rela- or roughly six full years of schooling. This would tion to income per capita averages only one-third be consistent with the aims of the 1990 World Con- that in Korea, basically because much more of In- ference on Education for All, sponsored by the dia's public expenditure on education goes to United Nations Educational, Scientific, and Cul: higher education. In Burkina Faso, Mali, Mauri- tural Organization (UNESCO), UNICEF, the tania, and Niger one-quarter of the education bud- UNDP, and the World Bank. Most pupils in devel- get is for higher education, and between 60 and 80 oping countries currently receive much less than percent of that quarter is devoted to scholarships 5,000 hours of schooling in the primary grades (be- and other forms of student aid. This, it can be cause of pupil and teacher absences caused by argued, is inefficient, and it is also extremely re- sickness, among other factors). In India no more gressive because most of the benefits of higher ed- 48 ucation are captured privately in the form of in- and that of their families. Removing discrimina- creased earnings and because students tend to tion-in the labor market, in access to credit, in come from higher-income families. property law, and so on-can boost women's On its own, spending more resources for pri- earnings and financial security, which (as an ear- mary schools is not enough. Whatever is spent lier section has shown) can promote family health. must also be used efficiently. And women need to be healthy themselves to ful- Although the health and nutrition of school-age fil their roles as mothers and household managers. children are not normally thought of as education They have specific health needs, including protec- issues, in fact they do affect a child's school atten- tion against violence. More than one-third of the dance and performance. Allocating resources to global burden of disease for women ages 15-44, address health problems in this population can of- and over one-fifth of that for women ages 45-59, is ten be an efficient way to improve schooling. (Spe- caused by conditions that afflict women exclu- cific interventions are discussed in Chapter 4.) sively (maternal mortality and morbidity and cer- Teachers and pedagogical materials are the main vical cancer) or predominantly (anemia, sexually school inputs at lower levels of schooling (with transmitted diseases, osteoarthritis, and breast teacher salaries absorbing the bulk of spending). cancer). Most of these problems can be addressed Efficient use of these inputs is thus crucial, par- cost-effectively, but health services in many devel- ticularly in countries where rapid population oping countries have typically focused narrowly growth threatens to reverse progress in expanding on women as mothers. enrollments. In Mauritania, for example, if spend- What is lacking is a strategy for engaging ing per pupil and the share of primary schooling in women in health care from adolescence onward. total government expenditure remain constant, Often this failing occurs because health services the enrollment ratio in primary schools is pro- are insensitive to the cultural needs of women: in jected to drop from 51 percent in 1988 to 45 percent many Middle Eastern countries, for example, most by 2000. To forestall such regress, maximizing the physicians are men, but there is a strong belief that learning gain per unit cost and making the correct women should not be seen after puberty by men tradeoffs between unit costs and coverage are of who are not part of their family. Inconvenience is particular importance. A recent World Bank-spon- another deterrent; in many countries individual sored review found expenditures on nonsalary in- health services (for example, prenatal care and im- puts such as textbooks and interactive radio in- munizations) are offered on different days, mean- struction to be most cost-effective in improving ing that women have to return repeatedly with learning outcomes. their children. The solution is often as feasible as it In some circumstances it may also be right to is clear: to provide child health services, prenatal spend more to lower the barriers to schooling for care, treatment of sexually transmitted diseases, girls and other disadvantaged groups. This can be and family planning services jointly at convenient done in many ways: through scholarships (used in times. The Bangladesh Women's Health Coalition Bangladesh to encourage girls to go to secondary and the Chilean Institute of Reproductive Medi- school); by offering free textbooks or fee-exemp- cine, for example, offer integrated family planning tions; or by siting schools close to pupils' homes so services at the same time as child health services, that parents are less worried about their daugh- and Thailand is experimenting with mobile health ters' safety. In Pakistan, for example, girls are as clinics to reach women in their homes. likely to enroll as boys when there is a school in The design of health services must also be sensi- the village but are 10 percent less likely to do so tive to the stigmas surrounding certain diseases, when the school is nearby but not in the village. In especially any that are sexually transmitted or several African countries distance education- physically disfiguring. Women are more likely to whereby radio and correspondence materials re- seek treatment for sexually transmitted diseases if place classroom teachers as the principal medium health centers offer multiple services, with privacy of instruction-has sometimes helped to overcome in consultations, so that it is not obvious why a the physical barriers to schooling for girls. person is visiting the center. Diseases that damage the skin (such as leprosy, onchocerciasis, and Empowering women leishmaniasis) have severe psychological implica- tions for girls and women, reduce their marriage In addition to education, other policies can en- prospects, and may lead to marital separation. In hance women's capacity to improve their health Colombia and India women tend to seek treatment 49 Box 2.3 Violence against women as a health issue Data from many industrial and developing countries woman's age ()r unhealthy habits (such ·as smoking). In reveal that anywhere between one-fifth and more than addition to physical injury and emotional trauma, rape half of women surveyed say they have been beaten by Yictims run the risk of becoming pregnant or contract- their partners. Often, this abuse is systematic and dev- ing sexually transmitted diseases, including AIDS. A astating. In Papua New Guinea, for example, 18 per- rape crisis center in Bangkok reports that 10 percent of cent· of all urban wives surveyed had sought hospital its clients contract STDs as a result of rape and 15 to 18 treatment for injJJries inflicted by their husbands . In percent become pregnant, a figure consistent with data the United States domestic violence is the leading from Korea and Mexico. In countries where abortion is cause of injury among women of reproductive age; be- restricted or illegal, rape victims often resort to unsafe tween 22 and 35 percent of women who visit emer- abortions, greatly increasing the danger of infertility or gency rooms are there for that reason. even death. Research has shown that battered women run twice Another form of violence against women and girls is the risk of miscarriage and four times the risk of having fe!Jla)e genital mutilation, popularly known as female a baby that is below average weight. In some places circumcision. An estimated 85 million to 114 million violence also accounts for a sizable portion of maternal women in the world today have experienced genital deaths. In Matlab Thana, Bangladesh, for example, in- mutilation. The practice. is reported in twenty-six Afri- tentional injury during pregnancy-'-motivated by can countries, among minorities in India, Malaysia, dowry disputes or shame over a rape or a pregnancy and Yemen, and among some immigrant populations outside wedlock- caused 6 percent of all ~aternal in Western countries. If current trends continue, more deaths between 1976 and 1986. Research from the than 2 million girls will be at risk of genital mutilation United States indicates that battered women ;~re four to every y.e ar, five times as likely to require psychiatric treatment as Clitoridectomies account for 80 to 85 percent of nonbattered women and are five times as likely to at- cases worldwide. Infibulation, which involves removal tempt suicide. They are also more prone to alcohol of more tissue, is more common in eastern Africa. abuse, drug dependence, chronic pain, and These initiation rituals pose a health risk to girls and depression . women and are a threat to their psychological, sexual, Rape and sexual abuse also damage women' s health and reproductive well-being . The consequences of and are widespread in all regions, classes, and cul- both procedures can include hemorrhage, tetanus, in- tures. In Seoul 17 percent of women report being vic- fection, urine retention, and shock. Infibulation carries tims of attempted or actual rape. In one study of U.S: the added risk of long-term complications because of women a history of rape or assault was a stronger pre- the repeated cutting and stitching at marriage and with dictor of how many times women sought medical help each childbirth, and it can limit a woman's choice of and of the severity of their health problems than was a contraceptive method . for leprosy later than men do, when patches have cent of the total disease burden among women already reached the face and hands; they are reluc- ages 15-44 in developing countries, where the bur- tant to ask for help when the first patches appear, den from maternal and communicable causes still on the buttocks. Again, sensitivity is needed to overwhelms that from other conditions . In indus- encourage women to come forward. trial countries, where the total disease burden is The same is true of another category of danger much smaller, this share rises to 19 percent. By to women's health: domestic violence and rape. damaging a woman's physical, mental and emo- Violence against women is widespread in all coun- tional capacity to care for her family, domestic vio- tries in which it has been studied (see Box 2.3). lence and rape also hurt the health of other family Although this has only recently been viewed as a members, particularly young children. public health issue, it is a significant cause of fe- This is an issue with complex economic, cul- male morbidity and mortality, leading to psycho- tural, and legal roots, and it is therefore not easily logical trauma and depression, injuries, sexually dealt with by public policies. Prevention will re- transmitted diseases, suicide, and murder. Rape quire a coordinated response on many fronts . In and domestic violence cause a substantial and the short to medium term, the right measures in- roughly comparable level of disease burden per clude training health workers to recognize abuse, capita to women in developing and industrial expanding treatment and counseling services, and countries . These problems account for about 5 per- enacting and enforcing laws against battering and 50 rape. In the long term, much depends on changing achievements of the past point to the requirements cultural beliefs and attitudes toward violence of the future-above all, to economic growth and against women. In Africa women's groups have the expansion of schooling and health services. worked to break the practice of female circumci- According to World Bank projections, income per sion, partly by informing people of its severe con- capita in Sub-Saharan African countries will grow sequences for health. In the United States the by only 0.8 percent a year over the next ten years. American Medical Association launched a major Even this modest increase will bring about a de- campaign in 1991 to educate the public and physi- cline in the infant mortality rate of between 2 and 4 cians about family violence. Research shows that percent. In South Asia, where faster growth-3.3 even health professionals often fail to identify percent a year-is projected, infant mortality de- cases of battering. Recently, the U.S . Joint Com- clines of 15 percent can be expected. mission on Hospital Accreditation issued new These benefits can be powerfully reinforced by standards requiring all hospitals to develop proto- better education and health services. In Africa in- cols and train their staffs to respond to different creasing female literacy rates by 10 percent is likely forms of abuse. In Colombia the Ministry of to lower the infant mortality rate by an estimated Health has begun to document the scale of the 10 percent. In India and Kenya two maternal problem in its most recent Demographic and deaths and about forty-five infant deaths would be Health Survey. These efforts come on the heels of averted for every 1,000 girls provided with one almost two decades of organizing efforts by extra year of primary schooling . Even in poor women around the world; in Latin America alone countries governments can enhance people's abil- there are now nearly 400 separate organizations ity to improve their own health by expanding working to reduce violence against women. schooling opportunities for all children-with spe- cial efforts to encourage parents to enroll their What can be done? daughters-and by widening access to health ser- vices, particularly for women and children. Such Around the world, much has already been done to investments pay off in better health and provide a enable people to live longer, healthier lives. The foundation for future economic growth. 51 The roles of the government and the market in health World spending on health totaled about $1,700 bil- tion, and other sectors important for health, as lion in 1990, or 8 percent of global income. Of this, well as through regulation of health systems, governments spent more than $1,000 billion, or health providers, and insurers. Governments fur- nearly 60 percent. Of the $170 billion spent on ther affect health by their impact on household health in the developing countries of Africa, Asia, income and educational levels (as discussed in and Latin America, governments spent half the Chapter 2), by financing public health services, total amount-2 percent of those regions' GNP. In and by providing care directly. What governments the established market economies, where total . do varies enormously from country to country, but health spending was almost $1,500 billion, govern- every government plays an important role. ments spent just over $900 billion-more than 5 Three economic rationales justify and guide a percent of GNP (Table 3.1) . The sheer size of these government role in health . They are discussed in expenditures on health makes it critical to under- greater detail in ''The rationales for government stand the impact of government policies on peo- action,'' below. ple's health. But governments profoundly influ- • The poor cannot always afford health care ence health in less direct ways, through their that would improve their productivity and well- policies toward education, water supply, sanita- being. Publicly financed investment in the health Table 3.1 Global health expenditure, 1990 Public sector health Total health Health expenditure Percentage Per capita Ratio of Percentage expenditure expenditure as percentage of GNP health per capita of world (billions of as percentage of regional spent on expenditure spending Demographic region population dollars) of world total total health (dollars) (SSA ~ 1) Established market economies 15 1,483 87 60 9.2 1,860 78.9 Formerly socialist economies of Europe 7 49 3 71 3.6 142 6.0 Latin America 8 47 3 60 4.0 105 4.5 Middle Eastern crescent 10 39 2 58 4.1 77 3.3 Other Asia and islands 13 42 2 39 4.5 61 2.6 India 16 18 22 6.0 21 0.9 China 22 13 59 3.5 11 0.5 Sub-Saharan Africa 10 12 55 4.5 24 1.0 Demographically developing countries 78 170 10 50 4.7 41 1.7 World 100 1,702 100 60 8.0 329 13 7 Note: SSA, Sub-Saharan Africa. Source: Appendix table A.9. 52 of the poor can reduce poverty or alleviate its education bring. The second factor is the amount consequences. and effectiveness of expenditure in the health sys- • Some actions that promote health are pure tem. The third factor is the range of diseases pre- public goods or create large positive externalities. sent, which is determined largely by climate and Private markets would not produce them at all or geography. Effective health policy takes account of would produce too little. different disease prevalences but is not simply de- • Market failures in health care and health in- termined by them. surance mean that government intervention can Differences in health spending are an obvious raise welfare by improving how those markets starting point in the search for an explanation of function. differences in health. In 1990 total annual health Any potential benefits from greater public sector spending ranged from less than $10 per person in involvement in health must be weighed against several African and Asian countries to more than the risk that governments will in fact make matters $2,700 in the United States. There was also consid- worse. For example, to satisfy special interest erable variation within regions. In Africa, Tanzania groups, governments may adopt policies that re- spent only $4 per capita for health in 1990, while duce the general welfare. Even when they choose Zimbabwe spent $42 per person. In Asia, Ban- correct policies, they may fail to implement them gladesh spent $7 per person each year, as against properly. $377 in Korea. Since the share of GNP devoted to Governments have a responsibility to spend health tends to rise with income, rich countries well, to get "value for money," whenever they differ from poor ones even more in health expen- devote public resources to health. This means al- diture than in income. locating resources so as to obtain the most im- But health spending alone cannot explain all the provement in health per public dollar, taking into variation in health among countries. Nor can in- account the private market's response to public come and education, or even spending, income, sector spending. Because private health care mar- and schooling taken together. Figure 3.1 illustrates kets can also fail to achieve value for money, gov- the discrepancies. The vertical axis shows how far ernment policy has a role in providing information life expectancy in a country differs from the value and incentives to improve the allocation of re- predicted on the basis of that country's income sources by the private sector. In most of the world and average schooling. France, Haiti, Singapore, a great deal of additional health could be obtained and Syria have almost exactly the life expectancy from a relatively small number of cost-effective in- predicted. China, Costa Rica, Honduras, and Sri terventions that could be delivered at modest cost Lanka, in the top half of the figure, all achieve five and with little need for high-level facilities or years or more of life beyond what would be ex- medical specialties. pected. Egypt, Ghana, Malawi, Uganda, the United States, and Zambia, in the bottom half of Health expenditures and outcomes the figure, all have a life expectancy about five years lower than expected, given their levels of Chapter 1 showed how greatly health status dif- income and education. fers among populations. Life expectancy ranges The horizontal axis of Figure 3.1 shows how far from forty years or less in some countries of Sub- total health spending differs from the value pre- Saharan Africa to seventy-five or more in the es- dicted by income and education. Egypt, Morocco, tablished market economies. In Sub-Saharan Af- Paraguay, Singapore, and Syria, in the left half of rica half of all deaths occur under age 5; in the the figure, spend relatively little. France, Haiti, In- established market economies half occur after age dia, Mozambique, and the United States, in the 74. Child mortality rates exceed 200 per 1,000 in right half, spend more than expected. several African countries but are below 20 in the At any level of income and education, higher richest countries. The burden of disease is five health spending should yield better health, all else times higher, per capita, in the worst-off than in being equal. But there is no evidence of such a the healthiest regions. relation. Countries are scattered in all quadrants of Three factors help to explain these huge differ- the figure. The countries that appear in the upper- ences. The first is human behavior. Chapter 2 left quadrant obtain better health for less money. showed that both health and the capacity to im- China, for instance, spends a full percentage point prove health are related to income and education less of its GNP on health than other countries at and to the changes in behavior that wealth and the same stage of development but obtains nearly 53 Health expenditure, income, and schooling only partly explain variation in life expectancy. Figure 3.1 Life expectancies and health expenditures in selected countries: deviations from estimates based on GOP and schooling Deviation from predicted life expectancy (years) f- 0 10 Better outcome, China Better outcome, lower expenditure higher expenditure Costa Rica Sri Lanka 0 0 0 Honduras 5 I- Greece 0 • • • • ... . Paraguay 0 India o o• • • • .. • 0 0 Singapore O Morocco syria • • • • • -- • - • • . ..., • ... • •• .._~ • • ''· O France Haiti - ..., 0 Mozambique \ 0 • United States 0 -5 . I- Egypt Malawi Ghana 0 • 0 0 O uganda Worse outcome, Zambia Worse outcome, lower expenditure higher expenditure ,, I I I I I. i it _ -10 -5 -4 -3 -2 I -1 0 1 I 2 3 ' 4 5 6 Deviation from predicted percentage of GDP spent on health So11rce: World Bank data. ten years of additional life expectancy. Singapore than predicted to achieve several years less of life spends about 4 percent less of its income on health expectancy than would be typical for its high in- than others at the same level of development but come and high educational level. achieves the same life expectancy. Other coun- Analyses using other measures of health status, tries, of which Costa Rica and India are examples, such as child mortality, yield similar results. This obtain relatively good health results but also spend raises obvious and important questions. What ac- relatively more. (In the case of India health spend- counts for these large deviations? How much is ing is low and health status is poor, but even lower attributable to the characteristics of health sys- spending and worse status would be expected for tems? How can public policy help to provide better a country with such low levels of income and health outcomes for a given national effort? schooling.) Egypt and Zambia, by contrast, get poor health for a lower-than-predicted level of The rationales for government action spending. Finally, it is possible both to spend more than predicted on health care and still achieve un- Public policy in health is successful if it leads to expectedly poor results. The United States is an increased welfare through better health outcomes, extreme case, spending 5 percent more of GNP greater equity, more consumer satisfaction, or lower 54 total cost than would occur in the absence of public finance care: spending more can translate into action. Of course, the pursuit of one or more of more services for the poorest or the same services these objectives does not by itself justify govern- for more people, including the less poor. In prac- ment intervention. There must be a basis for be- tice, very poor countries must target if they are to lieving that the government can achieve a better offer the poor any meaningful health care. outcome than private markets can. There are three Public goods and externalities are forms of market broad reasons why that belief may be true: one failure that may justify government intervention. centers on poverty and the equitable distribution The key characteristic of public goodswhich may of health care and the other two involve market be products or servicesis that one individual can failures. use them or benefit from them without limiting Reduction or alleviation of poverty provides a others' consumption or benefit. As long as some- straightforward rationale for public intervention in body pays, everybody benefitswhich makes it health. Success in reducing poverty requires two difficult or impossible to find anybody altruistic equally important strategies: promoting the use of enough to pay. Many public health interventions, the most important asset of the poortheir labor such as wide-area control of disease vectors and and increasing their human capital through access radio-based health information campaigns, are to basic health care, education, and nutrition. As nearly pure public goods for which only the gov- Chapter 1 showed, investment in the health of the ernment can ensure provision. Another public poor raises their educability and productivity. It good, new scientific information, has contributed gives them both the assets they need to lift them- enormously to the rapid improvements in health selves from poverty and the immediate welfare during this century. Its continued creation will de- gain of relief from physical suffering. Further- pend at least in part on governments. The right more, in most societies providing health and edu- choice of interventions and the proper level of pro- cation for the poor commands a degree of political vision of any public good require careful analysis assent that is altogether lacking for transfers of in- of the health benefits in relation to the costs. Prices come or of assets such as land. Investing in the provide no indication of what benefits are worth health of the poor is an economically efficient because private markets do not supply public and politically acceptable strategy for reducing goods. Nonprofit nongovernmental organizations poverty and alleviating its consequences, as World (NGOs) may supply such goods but cannot fully Development Report 1990 emphasized. substitute for government action. If "the poor" are all those living on less than $1 Externalities, or spillovers of benefits or losses (in real purchasing power) per day, they can typ- from one individual to another, characterize cases ically neither afford much health care nor borrow in which a private market might function but to pay for it. Simply transferring small amounts of would produce too much or too little. For example, income to poor people would create relatively little curing an individual of tuberculosis also prevents additional demand for health care. But because the transmission of the disease. But an individual's poor are more sensitive to the price of medical care demand to be cured of tuberculosis (or of mild or and also suffer a greater burden of disease than the asymptomatic sexually transmitted disease) is nonpoor, access to free or low-cost care can pro- probably not affected by consideration of the risk duce large increases in their consumption of health to others. If the externality is not taken into ac- care. count, treatment will be priced too high in private To ensure that subsidized health services actu- markets, and too little treatment will be given. ally reach the poor, however, may require restric- Subsidies for treatment are therefore justified. An tions, particularly on the kind of care that is paid example of negative externalities is a person's use for by the public sector. Offering free care of all of antibiotics, which may, by increasing microbial kinds to everybody typically leads to rationing of resistance to a drug, reduce the drug's value to servicesgeographically or according to quality. others and increase their risks. Such universal programs may not reach the poor Failures in markets for health care and health insur- or improve their health. They may, however, com- ance provide a third rationale for government ac- mand more political support than targeting, and tion to improve efficiency and, in the case of fail- they more easily address the problems of insur- ures in the market for health insurance, to improve ance markets that are discussed below. Who equity. One source of market failure, "adverse se- should receive free care depends on the preva- lection," arises because individuals face different lence of poverty and on the country's capacity to risks. Customers who know themselves to be at 55 high risk are motivated to buy more insurance and There is some moral hazard in the markets for are more likely to use it. So it is in the insurer's house and vehicle insurance. The extreme form is interest to find out who the high-risk customers when somebody burns down a house to collect the are and either to exclude them or to compensate insurance or abandons a car and reports it as for their greater risk by charging them higher pre- stolen. But unlike consumption of too much health miums. (Higher prices for all customers would re- care, these actions are crimes, with penalties that duce demand by low-risk people and therefore may greatly exceed the value of the asset. In any push prices still higher.) Defensive efforts to ob- case, the insurer's potential liability is limited to tain valuable information about risks add to the the (easily determined) market value of the asset. cost of insured health care without improving All the limitations on moral hazard and adverse health outcomes. selection are weaker in health insurance. It is Adverse selection presents a serious problem for harder to identify individual risks, and still harder risks existing at the time insurance is taken out, to attribute them to behavioral choices. There is no but an even more complex problem arises from the market value for the human body and no possi- fact that an initially low-risk person may become bility of abandoning one that is worn out and ac- high-risk later in life. In principle, there should be quiring a new one. The lack of a natural limit on insurance available specifically against this likeli- costs (since the asset being insured, the body, has hood of increased risk, or else insurance should no price with which costs can be compared) distin- cover a person's entire lifetime, with sharing of guishes health from other insurable risks. risks that may arise in the distant future, as well as The difficulties in insurance markets carry over of current ones. Neither solution is easy to imple- directly into markets for health care. If people have ment because of the extreme uncertainty; insur- "too much" health insurance, they will have an ance can cover known risks but not uncertainty incentive to use "too much" health care at too about risks. high prices. Unfortunately the difficulty of judging Another problem is the tendency of consumers health care risks and the impossibility of placing a to use more of a service when its marginal cost to value on a living body make it impossible to deter- them decreases. Insurance reduces or eliminates mine how much is "too much" in health care and the marginal cost of health care to consumers. So, health insurance. Nor is making a consumer pay providing insurance does not simply shift the way more for health care a sure way of reducing only a given amount of health care is paid for but in- "unnecessary" demand. creases the amount of care demanded. Failures of information make matters even Because the financial cost of disease is reduced, worse. A patient who knew the likely outcome people may take less care of their health, leading and the cost to him or her of every possible treat- to more illness and more subsequent demand for ment might yet be able to choose rationally be- care. Or they may protect their health more by tween gains and costs. But patients do not have way of health care, paid for by insurance, and less such knowledge, and the medical professional through their own behavior. Passing costs on to generally knows far more than the customer. This others such as insurers because one does not bear asymmetry of information means that the provider the full consequences of one's actions is called not only provides services but also decides what "moral hazard." It arises because of uncertainty services should be provided. The result is a poten- and because insurers cannot fully monitor con- tial conflict of interest between what the provider sumers' behavior and make them responsible for stands to gain from selling more services and his their decisions. Moral hazard also results when or her duty to do what is best for the patient. The providers induce demand for services that neither patient is at even more of a disadvantage when they nor consumers will pay for. sick and unable to make decisions or when deci- Both adverse selection and moral hazard have sions must be made quickly because of threats to more pernicious effects in markets for health in- life. surance than in markets for insurance on houses The same potential for consumption of unneces- or cars. Risks to houses are higher in areas prone sary services can arise any time a supplier is better to earthquakes or hurricanes, but they are easy to informed than a customer. It is a notorious prob- determine, and insurers respond by charging lem in car repair and home improvement services. higher premiums in those areas. Similarly, car in- But in these sectors the insurer has more oppor- surance premiums are higher for young drivers tunity to supervise the service provider, and the and other identifiable groups at greater risk of ve- insurer may decide simply to replace the item hicular accidents. rather than to repair it. Health insurers have no 56 replacement option. They may try to review pro- Since poor people typically cannot buy such care fessionals' recommendations before agreeing to for themselves, there is a straightforward case for pay for services, but health professionals often dis- public finance. Public health measures and essen- agree about expected medical outcomes, and wait- tial clinical care together constitute a package of ing for a second opinion may cause pain, suffer- health care that might justifiably be financed by ing, and increased risk for the patient. general revenues, with perhaps some contribution These problems constitute the market failure pe- from user fees. This strategy is also compatible culiar to health: expenditure on medical care can with the argument that basic health care is a fun- be extremely high, yet not all justified care is pro- damental right. Although most of the population vided and much care of doubtful value is paid for. may be able to pay for such care, the government Some people are denied insurance, while others has a responsibility to ensure that the poor, too, may be overprotected. Those who do not pay the can exercise their rightat least to the extent that full costs of treatment may take poorer care of their society can afford. own health than they could. Many of the extra Third, the rationale that the government should costs are paid by society as a whole. intervene in health care markets because of signifi- The market for health care goods and services cant market failures applies particularly to the reg- can also fail through imperfect competition among ulation of health care and health insurance. The providers, which allows excess profits, inefficient government cannot finance all medical care for use of resources, poor quality, and too little pro- which insurance might be desirable without wors- duction. Sometimes governments themselves arti- ening the tendency toward higher costs and risk- ficially stifle competition. For example, govern- ing de facto rationing of health care, which par- ments may prohibit or interfere unduly with the ticularly hurts the poor. Beyond a well-defined operation of private health care providers, particu- package of essential services, therefore, the role of larly NGOs. Governments often also protect do- the government in clinical services should be lim- mestic producers of drugs and vaccines. In Ban- ited to improving the capacity of insurance and gladesh tetanus vaccine produced domestically at health care markets to provide discretionary care government insistence had such low potency that whether through private or through social insur- its use in 1989-92 risked thousands of lives before ance (earmarked taxes such as social security or it was replaced with imported vaccine. other mandated arrangements). Of course, the Economies of scale in productionwhich occur range of services included in the nationally de- when a single large producer is much more effi- fined essential package will vary substantially cient than many small onesalso lead to noncom- from country to country. To provide equitable ac- petitive situations. In many parts of the world hos- cess for the poor, to address problems of adverse pitals and specialists face little or no competition selection, and to contain costs, the governments of because of economies resulting from large-scale almost all OECD countries have made available a operation. Such situations may call for regulation comprehensive essential package with public (or of the private market. publicly mandated) financing. Poorer countries The three rationales for government interven- must, of necessity, define their essential packages tion in the health sectorprovision of public more narrowly. goods, reduction of poverty, and market failure Governments can further improve how markets correspond roughly to three different kinds of ser- function by providing information about the cost, vices. First, the services classified as public goods, quality, and outcome of health care. Simply by de- and some of those characterized by large exter- fining an essential clinical package, the public sec- nalities, constitute what is known as "public tor provides valuable guidance on what is and health." Public health includes those services pro- what is not cost-effective. This distinction may vided to the population at large or to the environ- then influence the design of private or social insur- ment, such as spraying to control malaria. It also ance packages and the behavior of individual pro- typically includes some services such as immun- viders or patients. Information on the relative cost- zations that are not public goods but that carry effectiveness of different discretionary procedures substantial externalities. is similarly valuable and might be used by insurers Second, the inclusion of health care as part of a and providers to reduce costs and attract clients. strategy for combating poverty justifies public fi- Neither theory nor experience points to a gen- nancing of "essential" clinical or individual ser- eral rule on the extent to which the public sector vices. These are highly cost-effective services that should provide health care directly, as distinct would greatly improve the health of the poor. from financing it. Governments might have to 57 Box 3.1 Paying for tuberculosis control in China Tuberculosis kills or debilitates more adults than any Charging tuberculosis patients had perverse effects. other single infectious agent. Without appropriate When doctors and institutions expected to be reim- treatment, 60 percent of those with the full-blown dis- bursed by insurance, they provided excessive diagnos- ease will die. In China, it is estimated, more than tic tests and examinations during treatment and dis- 360,000 peoplemost of them poor peasantsdie of pensed higher-cost antibiotics that should have been tuberculosis every year. Tuberculosis is best prevented reserved for the most resistant cases. Daunted by the by curing infectious persons early in the course of dis- costs, many low-income victims failed to enter treat- ease, thus interrupting transmission to others. Well- ment or dropped out early. There were no incentives to run programs can cure 80 to 90 percent of patients; ensure that patients completed treatment or were poorly administered programs cure 30 percent or less, cured. Because health system records showed very leading to larger numbers of sustained cases of infec- high rates of cure for those who completed treatment, tion and related deaths and to new infections. the government remained largely unaware of the dete- China made substantial progress against tuber- riorating situation. The direct cost to the health system culosis during the 1960s and 1970s, using standard of a poorly functioning program was nil, but the indi- long-term (twelve to eighteen months) antibiotic ther- rect costs to the economyand the personal costs to apy that was essentially free of charge. Since the early patients and their familieswere enormous. 1980s, however, infection rates in about half the coun- An estimated I million to 1.5 million additional tu- try's provinces have stagnated or increased, despite berculosis cases remained infectious during the 1980s the adoption of an improved short-course (six to eight because treatment was no longer free. Tens of millions months) therapy. Much of the trend is attributable to of new infections were produced, and many of those changed health-financing policies and, in particular, to infected will fall victim to the disease later in life. the government's decision that health facilities should The development of drug-resistant strains was also ac- be encouraged to charge patients for virtually all ser- celerated. Given appropriate policies, many of the vices. Starting in 1981, health institutions had to earn more than 3 million persons who died of tuberculosis much of their operating costs from sales of drugs and in China during the decade could have been saved, services. Although base salaries were still funded from and the risk of infection for society could have been the public budget, workers' bonuses, housing, and re- halved. tirement benefits depended in part on institutional in- China, having recognized the problems caused by come from service provision. Managers' investment charging for tuberculosis therapy, has begun a major budgets were also linked to revenues from fees. A few national tuberculosis-control effort that provides sub- public health services such as immunizations remained sidies for treatment and appropriate incentives for pro- partially subsidized, but tuberculosis diagnosis and viders of care. Early results of this policy show dra- treatment were not, despite drug costs of $30 to $80 per matic increases in the number of cases cured. treatment. -I supply a package of essential health services di- In some circumstances market failure may im- rectly where private care would not be feasible pose only slight welfare losses, and the benefit of without high subsidiesfor example, in lightly correcting it may be outweighed by the costs of populated, very poor areas. (In many parts of the government action. In other cases the losses from developing world an alternative method of provid- failure to take account of positive externalities and ing such services is to subsidize an NGO.) In most supplier-induced demand can be enormous. Pol- circumstances, however, the primary objective of icy toward tuberculosis control in China provides public policy should be to promote competition an example: elimination of some free health care among providersincluding between the public and introduction of profit incentives in the provi- and private sectors (when there are public pro- sion of health services dramatically reduced treat- viders), as well as among private providers, ment rates, reversing progress against the disease whether nonprofit or for-profit. CompetitionS and causing much needless suffering (Box 3.1). should increase consumer choice and satisfaction Failures of government intervention can arise, how- and drive down costs by increasing efficiency. ever, even when government action might be Government supply in a competitive setting may sound policy. improve quality or control costs, but noncompeti- Governments may misjudge how an interven- tive public provision of health services is likely to tion will work in practice. Governments have only be inefficient or of poor quality. partial control over how private actors respond, 58 and those responses can undermine the intended ments are most appropriate for affecting the objective. Since 1971 physicians' fees in all pro- behavior of insurers, providers, and patients. This vinces of Canada have been set by negotiation raises the question of how far the government with provincial governments, and fees are no should itself act as an insurer, through social insur- longer rising faster than the general price level. To ance, and how far it should regulate private in- protect their incomes, particularly during the infla- surers. Each of these decisions involves tradeoffs tionary period 1971-75, physicians carried out a among the objectives of health policy: better greater number of procedures. This reaction was health outcomes, lower costs, more equity, and strongest where real fees fell the most. The saving greater consumer satisfaction with the health sys- in government expenditure was therefore much tem as a whole and with individual care. less than had been anticipated. Governments may not have the capacity to Value for money in health administer or implement policies well. Indeed, they may suffer from corruption and from sheer No matter how health services are organized and incompetence. The examples of two donor- paid for, what they actually provide are health in- financed public hospitals, each with 500 to 600 terventions: specific activities meant to reduce dis- beds, in two Latin American countries illustrate ease risks, treat illness, or palliate the conse- the problem. One was simply too large to adminis- quences of disease and disability. Debates about ter and operate and therefore could not be used at whether health services should concentrate on more than 60 percent of capacity. The other was so "vulnerable groups" such as children, pregnant badly designed that it could not accommodate women, or the elderly, or about the relative roles more than one-third the planned number of of hospitals versus health centers, or about pre- patients. ventive versus curative activities, are at bottom de- Governments are vulnerable to special inter- bates concerning the proper mixture of interven- ests both within and outside the health system. By tions. In health, as in every other sector, customers financing the training of unneeded physicians, by want value for the money spent on such interven- paying for low-value discretionary services for bet- tionswhether they pay directly or indirectly, in ter-off patients, and by protecting domestic indus- their roles as taxpayers or as buyers of health tries, governments help create the interests that insurance. later impede good policy, especially when quick Knowing the cost-effectiveness of a health inter- responses are needed to meet changing circum- ventionthe net gain in health (compared with stances or new opportunities. Even when society doing nothing) divided by the costcan be ex- as a whole would gain, public action may fail be- tremely useful for both public and private deci- cause it does not overcome the resistance of those sions. Governments can generate such informa- who would lose as a result. tion, and they can use it in two ways. First, they Perhaps the most fundamental problem facing can use it in determining whether a particular pub- governments is simply how to make choices about lic sector intervention is cost-effective: this means health care. Too often, government policy has con- judging the improvement in health compared with centrated on providing as much health care as pos- what would have happened through private deci- sible to as many people as possible, with too little sions in the absence of public action. (Chapter 4 attention to other issues. If governments are to addresses these issues for public health measures finance a package of public health measures and and Chapter 5 for the public finance of essential clinical services, there must be a way to choose clinical services.) Second, they can supply infor- which services belong in the package and which mation about the outcomes and costs of different will be left out. (The next section describes a mea- health interventions to consumers, providers, and sure of cost-effectiveness for health interventions insurers, and this knowledge can increase the that helps with this choice.) If financing is public value per health dollar spent in the private sector, but provision is private, governments must decide including what is spent on discretionary services. how to subsidize private care. The question of in- Private providers have no more incentive than centives to providers raised by that issue also ap- public providers to measure health outcomes, but plies to paying for publicly provided careto the they do face greater incentives to know their costs. "internal market" in the public sector. And if gov- Cost information alone can promote allocative effi- ernments are to influence the market for discre- ciency, as the experience of a Brazilian nonprofit tionary services, they must decide what instru- maternal and child hospital demonstrates (Box 59 Box 3.2 Cost information and management decisions in a Brazilian hospital The Instituto Materno-Infantil de Pernambuco (IMIP) is lations, it was evident that closing the intensive care a private, nonprofit hospital founded in 1960 to serve unit (except for newborns) and strengthening other the metropolitan area of Recife. In 1992 it received the services would save a greater number of children's first UNICEF award to a "child-friendly" hospital in lives. In particular, since the children who died in the Brazil in recognition of its work, particularly in the pro- hospital typically arrived very sick and often severely I motion of breastfeeding. IMIP depends for 95 percent malnourished, it appeared more cost-effective to try to of its revenue on contracts from the Instituto Nacional find high-risk children and treat them earlier. The strat- de Assistência Médica e Previdência Social (INAMPS) egy used was to expand the network of small commu- of Brazil's Ministry of Health. Annual spending runs nity health posts in the slum neighborhoods of Recife. about $6 million. IMIP opened the first such posts in 1983; by 1986 infant Starting in 1989, IMIP organized an accounting sys- mortality in those neighborhoods had fallen from 147 tem that divides services according to eleven cost cen- to 101 per 1,000 births. ters for final output. Administrative, laundry, food, ra- The experience of IMIP illustrates three lessons diology, laboratory, transport, and other nonfinal about cost-effective delivery of essential care. One is services were assigned to these final outputs in propor- that allocative efficiency can be improved without com- I tion to their measured or estimated use. plete information: medical professionals know much IMIP must match its average costs to average reve- about outcomes and often need only to know more nues determined by the price schedule of INAMPS, about costs. A second lesson is that autonomy facili- which is organized by treatment groups rather than by tates such changes: since private facilities generally individual services. Losses in any cost center must be have much more autonomy than public ones, this is an I offset by surpluses elsewhere. Gravely ill children are argument for more public finance of private provision referred to the hospital from all over northeast Brazil, or for decentralization of public systems. The third les- and there are three infant deaths per day among them. son is that even prices that are not based Ofl cost-effec- reduce mortality, IMIP created a pediatric intensive tiveness criteria can guide decisions about what care to care unit. The treatments provided, however, cost provide. It is more useful for government to set those much more than INAMPS would pay. And mortality prices correctly than to try to make all the allocative did not decline. Even without cost-effectiveness calcu- choices. Lb 3.2). By estimating costs for "cost centers" and value to human life, as would be necessary if costs relating them to outputs, the hospital discovered and gains were to be put in the same units. that its pediatric intensive care unit would drain Only in the past decade have costs and effective- resources from other departments, given the ness been systematically estimated for a wide prices the government paid for various services. range of health care interventionsalthough the The decision was made to limit the intensive care first such calculations had been made many years unit to newborns; community-level health posts earlier. Only a small share of the thousands of appeared more cost-effective for other cases. known medical procedures has been analyzed, but the approximately fifty studied would be able to Measuring the cost-effectiveness of health interventions deal with more than half the world's disease bur- den. Just implementing the twenty most cost- Given a common currency for measuring cost and effective interventions could eliminate more than a unit for measuring health effects, different inter- 40 percent of the total burden and fully three-quar- ventions can be compared by what it costs to ters of the health loss among children. achieve one additional year of healthy life. Out- The cost and effectiveness estimates used in this comes are measured in the same unit of disability- Report are based, as far as possible, on actual con- adjusted life years (DALYs) used to estimate the ditions in developing countries. Some fixed costs burden of disease. Nonhealth burdens, such as in- of operating a health system that cannot be attrib- come lost because of disease, are not included in uted to particular interventions are not consid- the measure. The ratio of cost and effect, or the ered, but the costs of intervention-specific capacity unit cost of a DALY, is called the cost-effectiveness are taken into account. Costs are assessed at mar- of the intervention; the lower that number, the ket prices. For inputs that cannot be traded inter- greater the value for money offered by the inter- nationally (such as semiskilled labor), costs will be vention. This approach avoids assigning a dollar lower in developing countries. For drugs, most 60 equipment, and high-level manpower, costs are either the public or the private sector costs lives. likely to be equal across countries, leaving aside An expenditure of $100,000 on chemotherapy for the effects of tariffs or other barriers. Indirect tuberculosis could directly save about 500 patients. costs, such as patients' costs of travel to treatment It would also prevent them from infecting others, or the income they forgo, can be substantial for for a total gain of about 35,000 DALYs. The same some interventions and perhaps particularly for expenditure on management of diabetes would women. Because these costs are difficult to deter- also benefit 500 patients but would save only 400 mine, they were largely ignored; more study is DALYs; each patient would gain less than one needed of how these barriers affect the utilization healthy year from a year of treatment, and there of health services. would be no benefit from reducing incidence. In- For some common health service packages such sisting on value for money is not only fully con- as immunizations, costs are computed on a joint sistent with compassion for the victims of disease, basis rather than separately for each intervention it is the only way to avert needless suffering. in the package. The estimates do not represent an The results of cost-effectiveness analysis confirm unattainable ideal; they assume that medically cor- the value of the primary health care interventions rect procedures are followed and that reasonable included in programs to reduce childhood malnu- care is taken as to quality, but they also allow for trition and mortality, chiefly from infectious dis- incomplete coverage or compliance. Whenever eases. Several hitherto neglected interventions are possible, actual experience is used to guide esti- also very cost-effective: chemotherapy against tu- mates of such things as how many patients will fail berculosis, integrated prenatal and delivery care, to complete a course of treatment. Future gains mass programs to deworm children, provision of from current interventions are discounted at 3 per- condoms along with information and education to cent per year, which has little effect on the ranking combat AIDS, and measures against smoking, of interventions the effects of which are felt such as education, consumer taxes on tobacco quickly, although it does reduce measured gains products (an effective deterrent for adolescents from interventions when the health effects are felt who are not yet addicted), and prohibition of only in the long run. smoking in public places. Many of the most cost- This Report found huge differences in both the effective health interventions are preventive in cost and the effectiveness of various health inter- character. But not all preventive measures are cost- ventions. Figure 3.2 presents both dollar costs and effective: spraying to control the mosquitoes that gains in DALYs for each of forty-seven different carry dengue is an example of relatively poor value interventions. Higher points represent interven- for money. At the same time, a small number of tions that are more effective in improving health; neglected but cost-effective clinical (mostly cura- points farther to the right represent lower-cost in- tive) interventions could eliminate a substantial terventions. Some interventions cost more than fraction of the burden of disease in many $10,000 per person benefited, while others cost countries. less than $1. Some interventions add more than In general, most cost-effective interventions can ten years of healthy life; for others the gain is be performed outside hospitals. By treating a small equivalent to only a few hours or days of full number of severe cases of disease, however, hos- health. Both axes are scaled in logarithms so that pitals can sometimes improve health at a lower the diagonal lines show equal cost-effectiveness cost per DALY than lower-level facilitiespro- ratios in dollars per DALY. These ratios vary vided that clinics or health posts treat most cases widely, from as little as $1 to as much as $10,000. and refer to hospitals only those requiring more Higher lines represent more cost-effective inter- sophisticated care. ventions. Four specific interventions illustrate ex- treme combinations of cost and health gain: vita- Complications in the use of cost-effectiveness min A supplementation in areas where the risk of blindness from vitamin deficiency is high (very Both the cost and the effectiveness of an interven- low cost, high gain), chemotherapy for tuber- tion can be affected by the incidence and preva- culosis (high cost, very high gain), environmental lence of the disease and the probability of dying control of dengue (low cost, low gain), and treat- from it. Preventive interventions are less cost- ment of childhood leukemia (very high cost, mod- effective for relatively rare diseases because more erately high gain). people have to be reached to prevent one case. The Because interventions can differ so much in cost- fatality rate matters because preventing or control- effectiveness, making allocative decisions badly in ling a disease saves more lives if there is a high 61 The costs and effectiveness of health interventions vary greatly. Figure 3.2 Benefits and costs of forty-seven health interventions Increase in DALYs (log scale) 100 Chemotherapy for tuberculosis . 10 Vitamin A supplementation . . 0.1 Treatment of $1/DALY leukemia S 0.01 $10/DALY Greater effectiveness Environmental 0.001 Lower control of dengue $100/DALY cost $10,000/DALY $1,000/DALY 0.0001 10,000 1,000 100 10 1 0.10 Cost per intervention or per intervention-year (dollars, log scale) Target: Children under age 15 Adults age 15 or older a Note: DALY, disablility-adjusted life year. Interventions are specific activities intended to reduce disease risks, treat illness, or palliate the consequences of disease and disability; an intervention-year is an intervention repeated throughout the year rather than provided only once. a. Jncludes some interventions that benefit all age groups. Source: Jamison and others forthcoming; Worlc Bank data. probability of dying. Immunization in an environ- information is needed to judge which interven- ment in which children are undernourished and tions should have priority. National or regional as- many die from preventable diseases is more cost- sessments are also important for estimating the effective than if children are otherwise healthy and expenditure levels required and the probable im- face little risk of dying. (Box 3.3, on measles and pact on the national burden of disease. tuberculosis, illustrates these issues.) Fortunately, If providing an intervention did not impose any differences in cost-effectiveness between one in- fixed costs in infrastructure and program adminis- tervention and another are often much larger than tration, then a low cost per DALY saved would either the variation from one locale to another or suffice to justify the intervention. In practice, there the uncertainty in the estimates. Where this is not may be substantial fixed costs to share over a num- the case, as exemplified by malaria, detailed local ber of interventions, and administrative capacity 62 Box 3.3 Cost-effectiveness of interventions against measles and tuberculosis The costs and effects of measles vaccination were esti- average cost of the lower-coverage strategy, but the mated for a model urban area based on data from higher-coverage approach continues to be extremely Lagos and Kinshasa. Data from Matlab, Bangladesh, cost-effective. Similar calculations were made for che- were used to model measles in rural areas. In each motherapy for tuberculosis, for both a standard course area, 36,400 cases of measles were assumed to occur in of treatment (twelve to eighteen months) and a short the absence of vaccination, with 1,452 urban deaths course (six to eight months), each with and without and 806 rural deaths. All the health damage from mea- hospitalization. Using data from Malawi, Mozam- sles comes from deaths, each of which costs thirty bique, and Tanzania, the average incremental cost DALYs. The simulations considered three different (marginal cost plus the average cost attributable to the strategies: no vaccination, immunization at nine fixed costs of the tuberculosis control program but ex- months (the earliest age at which the standard vaccine cluding other fixed costs of the health system) was esti- is effective) with 60 percent coverage, and nine-month mated at about $80 to $110 per cure for ambulatory vaccination plus efforts to increase coverage to 80 per- treatment, and $160 to $300 when hospitalization was cent, which raises supervision costs by 10 percent. For required. Cost per death directly averted was in the the last strategy, incremental as well as average costs range of $75 to $275, but cost per total death averted, were calculated to highlight the effect of raising cover- taking account of the interruption of transmission, can age. Costs were related to each of three effects: cases be as low as $20 and never exceeds $100. These very averted, deaths averted, and DALYs gained. Box table low costs translate into costs per DALY saved of about 3.3 shows the results. $1 to $3, making chemotherapy for tuberculosis one of Measles strikes later in childhood in rural areas, so the most cost-effective of all interventions. These costs cases are easier to prevent. But because earlier cases (in do not vary with the annual rate of infection. The cost- urban areas) cause more deaths, the cost per death effectiveness of BCG vaccine, by contrast, is extremely averted or per DALY is higher. In both urban and rural sensitive to infection rates; the vaccine is cost-effective areas, the marginal cost of raising coverage exceeds the only when the risk of infection is high. Box table 3.3 Health costs and gains from measles immunization Urban, by percentage vaccinated Rural, hr percentage vaccinated Item 60 80 60 80 Cases prevented (thousands) 10.7 14.2 16.2 22.6 Total cost per case prevented (dollars) 17 18 11 12 Incremental cost (dollars) - 22 - 15 Deaths averted (thousands) 0.4 0.6 0.4 0.5 Total cost per death averted (dollars) 432 462 525 561 Incremental cost (dollars) 552 670 DALYs gained (thousands) 12.3 16.4 10.2 13.5 Total Cost per DALY (dollars) 15 16 18 19 Incremental cost (dollars) - 19 - 23 Source: Foster, McFarland, and John forthcoming. may be limited. Spending on interventions that Since only relative ranking is possible, the at- are very cost-effective but resolve very small dis- tractiveness of an intervention also varies accord- ease burdens could waste resources by making it ing to what other health problems and treatments difficult to deal with diseases that impose much are locally prevalent. Oral rehydration therapy is larger burdens. Priority should go to those health an example: in environments in which child mor- problems that cause a large disease burden and for tality is low, it is much less cost-effective than im- which cost-effective interventions are available munization because it may have to be given re- (Box 3.4). If a particular health problem causes the peatedly during a child's first few years, but as loss of many healthy life years but there are no mortality rises it becomes more cost-effective. In meansor only very costly meansfor dealing general, the cost-effectiveness ratio varies not only with it, then it should be a priority not for health with local conditions but also with the degree to care but for research on development of cost-effec- which an intervention penetrates or covers a tive interventions. population. Cost-effectiveness is also influenced 63 r Box 3.4 Priority health problems: high disease burdens and cost-effective interventions It is easy to determine which health problems among ease burden in women ages 45-59 and for smaller children under age 5 deserve priority. As Appendix shares in other age and sex groups. Among communi- table B.6 shows, nine diseases each account for more cable diseases tuberculosis, AIDS, and respiratory in- than 1 percent of the total disease burden in this age fections deserve priority, but they cause less than 10 group. These diseases range from acute respiratory in- percent of all ill health after age 45 and only 20.1 per- fections (more than 17 percent in both boys and girls) cent in men ages 15-44. to iodine deficiency (1.2 to 1.3 percent). Of these prob- Large disease burdens and cost-effective interven- lems, which cause fully 80 percent of young children's tions coincide for only one group of adults, women ill health, eight can be addressed by interventions cost- ages 15-44. Six of the ten main sources of ill health can ing less than $100 per DALY saved. The only exception be prevented or treated for less than $100 per DALY. is congenital problems, which are responsible for more These range from maternal health problems (18.0 per- than 6 percent of the disease burden but for which no cent of the burden) to respiratory infections and ane- cost-effective interventions are known. mia (2.5 percent of the burden each) and account in The situation is much more complicated for adults total for 44 percent of ill health among women of repro- (Appendix table B.7). For example, cerebrovascular ductive age. Two other problemsdepression and self- disease is the leading cause of healthy life years lost in inflicted injuryeach cause at least 3 percent of the both sexes after age 60 and in women ages 45-59, but disease burden, but dealing with them is much more interventions to deal with it cost $1,000 or more per problematic. DALY saved. Ischemic heart disease is the second or These calculations illustrate the chief problem a third leading cause of ill health in both sexes after age health care system faces as the population ages: the 45, but the cost per DALY of dealing with it is $250 to marginal cost of a year of healthy life gained rises $1,000. Among the ten principal noncommunicable sharply, leading to difficult choices between increased causes of ill health in this age group, interventions spending and lower health gains. However, many costing less than $100 per DALY saved exist only for health problems of the elderly that cannot be fully re- I cataracts, anemia, and cancers of the respiratory sys- solved may be palliated at low cost. And much can be tem (through reduction of smoking) and the cervix. done at earlier ages to improve the health of future These problems account for only 7.9 percent of the dis- generations of old people. by the presence of other interventions that might An important limitation on the use of cost-effec- affect costs (through sharing of joint costs) or out- tiveness analysis of resource allocation in health is comes. Sometimes combining two interventions, that a number of interventions with important one preventive and one curative, is the appropri- health consequences also affect income or welfare ate way to deal with a particular disease, as is the in other ways. Chemotherapy for tuberculosis has case for tuberculosis and malaria. Some treatment no value beyond the DALY gain associated with for malaria is necessary because preventive mea- curing tuberculosis, but investing in girls' school- sures do not protect everyone; even if treatment is ing has both important consequences for health more costly, both interventions should be applied. (as documented in Chapter 2) and for income and In exceptional circumstances it may be worth status later in life. Similarly, family planning, in paying high marginal costs to extend coverage of addition to its health benefits, permits family an intervention to the entire population because choice about the number and spacing of children; the disease can be eradicated permanently (as has improved water supply and sanitation create ame- occurred with smallpox and may now be possible nity and time-saving benefits; increased food con- with polio). The gains in such cases include not sumption allows higher levels of physical activity; only the DALYs saved at the margin from the last and improved road safety reduces property dam- people immunized but all the healthy years that age and saves lives. For some of these interven- would otherwise be lost to the disease in the tions (for example, family planning and girls' future. A similar argument holds if a low-cost schooling) the cost per DALY is sufficiently low to intervention that has to be applied continually make them attractive on health grounds alone; is replaced by one with large initial costs other benefits only strengthen the case. For other but permanent effects, such as sanitation to reduce interventions the cost per DALY gained may be the need for treating many fecally borne diseases. too high to justify investment on health grounds 64 alone, but consumer willingness to pay for non- has the most easily measured output. Coverage in health benefits means that costs to the health sys- many regions remains incomplete. Immunization tem can be low. Many water and sanitation invest- against measles and against diphtheria, pertussis, ments are in this category. and tetanus has reached 90 percent or more in Using cost-effectiveness to select health inter- Chile, China, Cuba, Korea, and Saudi Arabia, but ventions for public financing does not necessarily it is still below 50 percent in some Asian and many mean spending the most resources where the bur- Sub-Saharan African countries. Many of the other den of disease is greatest. Instead, it means con- components of an adequate public health package centrating on the interventions that offer the great- scarcely exist. est possible gain in health per public dollar spent. As far as clinical services are concerned, the The relevant comparison is usually not with a situ- principal government failing in most countries is ation in which nothing is done but with the situa- the attempt to provide everything to everybody, tion created by privately financed health interven- with no distinction between more and less essen- tions. The most justified public measures will tial care and more or less needy patients. For some therefore combine a strong rationale for public ac- health services provided by the public sector, the tion with a cost-effective health intervention. Be- system of provision is so grossly inefficient that it cause individuals differ in how they value the is unlikely to be cost-effective no matter what in- present in relation to the future and in how they terventions the system tries to provide. Such inef- judge the seriousness of different health condi- ficiencies have been criticized so clearly and for so tions, a uniform ceiling on what the government long that it is evident they will only be overcome pays to gain one DALY may leave some people by radical changes in the organization of health with more publicly financed care than they would caresuch as a shift in the government's role from choose and some with less. But of all possible uni- providing care to financing care and stimulating form criteria by which to judge what interventions competition among providers. These changes will to pay for, cost-effectiveness appears to yield the in turn require a clear distinction between essential most efficient distribution of health resources. and discretionary spending and a new determina- tion by governments to achieve value for money in Data limitations health services. There is no other equitable way to control government spending. Cost-effectiveness analysis requires data on expen- Most governments also perform poorly in regu- ditures for particular interventions and on out- lating markets for private services, including in- comesinformation that health facilities and sys- surance. As recent research in Brazil has shown, tems, particularly those in the public sector, the quality of medical care could be substantially typically do not collect. Such information could improved at low cost if government discharged promote substantial gains in efficiency, but it will this role better. The rapid growth and almost total take considerable time and effort for most public lack of regulation of private insurance in such systems in developing countries to learn how to countries as Brazil and Korea present another chal- gather and use it. Budgets often disaggregate only lenge for which governments are ill-prepared. by inputs, not by programs, and costs per consul- tation or per bed-day mix many different interven- Allocation of spending to cost-effective services tions. Outcomes are seldom quantified. For inter- ventions the cost-effectiveness of which vary There are no calculations of how many years of greatly with local conditions, there is no substitute healthy life are currently saved by health systems. for information on both costs and results. Nonetheless, it is clear that many of them perform much worse than they might. Many governments Health policy and the performance of health spend too much on sophisticated hospital services systems of low cost-effectiveness and too little on essential public health and clinical services. The share of All governments subscribe to the view that the public expenditures for health absorbed by tertiary state must ensure the provision of certain basic and secondary care hospitals, for example, is as public health services. But few achieve this goal high as 70 to 75 percent in Jordan and Venezuela. even for immunization, which is probably the Tertiary care hospitals alone may consume 30 to 50 health intervention that has received the greatest percent of the health budget. Only a quarter of government attention and donor support and that government spending, and often less, is devoted 65 Table 3.2 Actual and proposed allocation of public expenditure on health in developing countries, 1990 (dollars per capita) Estimated Spending under the proposed package actual Low- Middle- All spending, all income income developing developing Package component count ries count ri cs' countries countries Contents Public health 4 7 5 1 EPI Plus; school health programs; tobacco and alcohol control; health, nutrition, and family planning information; vector control; STD prevention; monitoring and surveillance Essential clinical services 8 15 10 4-6 Tuberculosis treatment; (minimum package) management of the sick child; prenatal and delivery care; family planning; STD treatment; treatment of infection and minor trauma; assessment, advice, and pain alleviation Total, public health and 12 22 15 5-7 minimum essential clinical services Discretionary clinical services' 6 40 6 13-15 All other health services, including low-cost- effectiveness treatment of cancer, cardiovascular disease, other chronic conditions, major trauma, and neurological and psychiatric disorders Total 6 62 21 21 Note: Current spending on essential clinical services is estimated to be 20-30 percent of total public expenditure on health on the basis of estimates in World Bank health sector reports. The numbers reported should be regarded as approximations. Estimated for an income level of $350 per capita. Estimated for an income level of $2,500 per capita. Estimated residually. The negative number for low-income countries reflects total spending below the cost of the package. Source: World Bank calculations; World Bank sector reports on Ghana (1989), India (1992), Indonesia (1991), Jamaica (1989), Jordan (1989), Mexico (1989), Nigeria (1991), Pakistan (1992), Turkey (1990), and Venezuela (1992). to cost-effective public health measures and essen- propriate package of cost-effective care would ad- tial clinical care, delivered mainly in health centers dress. Even the best-designed care package could and communities. In many countries the share of not prevent all the health damage from these dis- public spending devoted to these basic services eases because of the low cost-effectiveness of some has been falling in recent years. In Brazil 64 per- interventions and the increasing marginal costs of cent of public spending on health in 1965 was for even the best ones. Still, because of the size of the preventive and public health activities, but by the burden and the low cost per DALY of the interven- mid-1980s the share had dropped to 15 percent, tions, it is reasonable to conclude that public ex- and hospitals absorbed fully 70 percent of expendi- penditure on health should initially be concen- ture. The resulting weakness of the primary care trated on those conditions. What this implies for network leads patients to seek care in hospitals; Up the distribution of spending by type of input or to 80 percent of the cases crowding hospital emer- level of facility is less clear, but it probably means gency rooms could be treated as effectively, but that facilities above the district hospital level more cheaply, at the primary level. should account for only a small share of the total, In the world as a whole, almost half the existing primarily for dealing with referrals. disease burden is from communicable diseases, Table 3.2 illustrates the degree of misallocation nutritional disorders, and maternal and perinatal of health spending by comparing estimated actual causes. It is primarily these problems that an ap- expenditure with what would be spent for a pack- 66 age of health services designed to address most cluding private spending, is about $14, about the effectively the burden of disease in the developing same as the proposed package. This means that world. This package consists of public health ser- either substantial private resources will have to be vices that would cost just over $4 per capita in poor used or additional government resources will be countries (with average income per capita of $350) needed; even if all public expenditure on discre- and a minimum package of essential clinical ser- tionary services were eliminated, current govern- vices that would cost about $8 per capita more. In ment spending on health would not meet the costs middle-income countries (with average income per of the package. capita of $2,500), the same package of public Since the minimum package would cost only health measures and essential clinical services about $60 billion for all developing countries to- would cost about 80 percent more, or $22 per per- gether, the task is to reallocate resources in mid- son. This difference partly reflects different epi- dle-income countries and to find additional re- demiological conditions, but input costs, particu- sources of about $10 billion in low-income larly salaries, would also be higher in middle- countries. The $8 to $10 per capita needed in extra income countries. Many countries will define the spending on public health measures and essential essential clinical package much more broadly than clinical services is less than the $13 to $15 per cap- the minimum discussed here. Even in relatively ita now spent, on average, on discretionary or poor countries, targeting public finance of essen- nonessential clinical services. In fact, spending on tial services to the poor would allow creation of a these less cost-effective services is now roughly broader and more generous package. Such a pack- double the amount that countries spend on the age can be built up by adding interventions in or- recommended package of public health measures der of decreasing cost-effectiveness until the addi- and essential clinical care. The right combination tional health gain is judged not to be worth the of reallocation and additional expenditure would cost, given the country's resources. To ensure po- allow governments to achieve a large improve- litical support and to deal with problems of market ment in overall health. failure and equity, countries may choose to finance Table 3.3 indicates how large this gain could be. the essential package universally from public, or Properly allocated, an expenditure of only $12 per publicly mandated, sources. person in low-income countries (excluding China) Governments in developing countries spend an would be enough to reduce the disease burden by estimated $21 per capita on health, for a total of almost one-third. This is 226 million DALYs, about $84 billion. It is estimated that only a little equivalent to 7.0 million infant deaths per year. In more than $1 per person, or a total of $5 billion, middle-income countries the proposed package goes for cost-effective public health measures. To could deal with only 15 percent of the disease bur- buy the package described here, countries would den, despite the higher expenditure per person. need almost to quintuple what they spend on pub- The total reduction in ill health in middle-income lic health. About $4 to $6 per capita, or $17 billion countries would be about 45 million DALYs, the to $25 billion total, goes for clinical services deliv- equivalent of 1.4 million infant deaths per year. ered through lower-level facilities or classified as The smaller gain in these countries reflects the fact primary health care. These services commonly in- that they have already eliminated much of the bur- clude many of those in the essential clinical care den from easily controlled communicable diseases. package, but they are usually not delivered to the A large part of the remaining burden is caused by entire population. And this expenditure includes chronic disease and disability. some less cost-effective services that should be re- It is assumed that the disease burden would de- garded as discretionary. Paying for a minimum cline by the same share in China as in a middle- package of essential clinical care would require ex- income country because China has already sub- penditure of an additional $16 billion to $24 billion, stantially reduced the burden from the diseases doubling the current expenditure level. If total addressed by the package. About 30 million spending did not change, this would imply a re- DALYs could be gained, the equivalent of 930,000 duction by about half in what is now spent on infant deaths prevented. The cost per capita would discretionary services. be the same as in low-income countries. For some countries, paying for the proposed Full coverage with the minimum package would package of services poses a severe challenge. In cost an estimated $22 billion in low-income coun- fact, in the poorest countries total current public tries, $14 billion in China, and $26 billion in spending of $6 per person is about $6 short of the middle-income countries. The total cost would be cost of the package. Total per capita spending, in- about $62 billion, or $15 per person in the develop- 67 Table 3.3 Total cost and potential health gains of a package of public health and essential clinical services, 1990 Cost per ca pit a Total cost (billions Reduction in disease burden Country group and package component (dollars) of dollars) Percent Millions of DALYs Low-income countrjes 12 22 32 226 Public health 4 8 8 57 Essential clinical services 8 14 24 170 Chinab 12 14 15 30 Public health 4 5 4 8 Essential clinical services 8 9 11 22 Middle-income countrjes 22 26 15 45 Public health 7 8 4 12 Essential clinical services 15 18 11 33 All developing countries 15 62 25 301 Public health 5 21 6 77 Essential clinical services 10 41 19 225 Estimated from data for Bangladesh, Egypt, India, Indonesia, Pakistan, and Sub-Saharan Africa. China is shown separately because its cost per capita is assumed to be that of a low-income country but its percentage reduction in disease burden is assumed to be that of a middle-income country. Estimated from data for Latin America and the Caribbean, Other Asia and islands except for Indonesia and Bangladesh, and the Middle Eastern crescent except for Egypt and Pakistan. Source: World Bank calculations. ing world as a whole. This figure includes what covered by a combination of greater public spend- countries are already spending on the services in ing, increased donor contributions, and more pri- the package, estimated at $20 billion to $30 billion. vate expenditure by those able to pay. Shifting The incremental cost would therefore be only $30 some part of the cost to higher-income con- billion to $40 billion a year. sumersfor example, through private or social in- The gain in health would be about 300 million surancewould allow for an expansion of the DALYs, which is equivalent to 9.3 million infant package or a reduced burden of public deaths. Universal application of the package expenditure. would therefore yield about the same health gain There are several reasons why developing coun- as eliminating nearly all the infant deaths in the tries fail to allocate sufficient resources to cost- world today. These gains could be achieved for an effective health interventions. Health providers average cost of about $50 per DALY for the public often lack incentives to provide cost-effective ser- health measures and about $100 per DALY for the vices. Doctors' pay, promotion, and professional minimum package of essential clinical services. recognition are enhanced by specialization and by The cost per DALY of the interventions in the the use of expensive new medical technologynot package ranges from less than $5 to more than by serving as public health doctors or district $200; average costs also reflect those public health medical officers in poor rural areas. Badly de- measures that do not improve health directly but signed government salary schedules and price sys- that are essential to the functioning of a health tems may exacerbate this trend. In China hospitals system. currently have a strong incentive to use new diag- In both low-income and middle-income coun- nostic and therapeutic technologies, for which full tries the marginal cost per DALY would be less costs can be charged, instead of older and less ex- than the average cost because all the fixed costs of pensive technologies for which government-set infrastructure are included in these estimates and prices are far below actual costs. there would be spare capacity for producing small Consumer demand for cost-effective services is additional amounts of most services. Because of often weak. This may reflect lack of information. joint costs, it is difficult to separate the cost per In rural Africa, for example, goiter and impaired DALY for every intervention in the package. mental abilities from iodine deficiency have in In middle-income countries the package could many places become accepted as the normal state be entirely financed by reallocating current public of affairs. Low demand may also reflect deficien- spending. In poor countries there would be a cies in supply. Most cost-effective interventions shortfall of about $10 billion a year, which could be can be delivered at primary care sites, but in poor 68 countries such as Burkina Faso and Mali, more health care center or doctor's office. In Indonesia than half of the population lives more than 10 kilo- in 1991, for example, rural households in the top meters from the nearest primary care center. income decile were three times more likely to live At a more fundamental level, the distribution of in a village with a health center than those in the political power explains much of the misallocation bottom decile. of government resources for health. The urban Partly because of difficulties in access, the poor population is better organized than rural groups in developing countries generally consume fewer and more vocal in demanding health facilities and health services. Household surveys from Sub- services. Similarly, middle-class workers in wage Saharan Africa and Latin America demonstrate employment, who frequently belong to powerful clearly that among people who report themselves trade unions, are more effective than self-em- to be sick, those in urban areas seek and obtain ployed farmers and workers in the informal sector medical care more often than those in rural areas, in lobbying for government-subsidized health and the wealthy contact a care provider more often benefits. Health professionals are also often better than do the poor. The differences can be large: in organized than the population they serve, and in Côte d'Ivoire in the mid-1980s, for example, an promoting their own interests they may make the urban household was nearly twice as likely to seek health system less efficient. Despite these prob- care as a rural household (60 versus 36 percent), lems, many countries have succeeded in dramati- and within the rural population a family in the top cally improving the health of their people. This income quintile was almost twice as likely to seek success can be accelerated, as newly available in- care as a family in the bottom quintile (44 versus 23 formation makes it clear how costly misallocation percent). is and how much health can be gained for rela- A study of Peru showed similar inequalities tively modest levels of spending. among geographic regions and educational groups. There was little variation in self-declared Equity in health status, utilization, and finance illness, but the likelihood of obtaining medical care when sick was nearly three times higher in some Data on health status, physical access to health parts of the country than in others. Regional dif- services, consumption of health care, the distribu- ferences in immunization rates were highest for tion of the financial burden of health care spend- uneducated mothers, whose children were only ing, and public expenditures for health all tell the one-third as likely to be fully immunized as the same story of severe inequities in developing children of women with secondary schooling. The countries. In Bangladesh, for example, the infant 36 percent of all the self-declared sick who lived in mortality rate for the urban poor (13.4 percent) is the capital city accounted for 53 percent of all Min- nearly twice the urban average (6.8 percent) and istry of Health ambulatory consultations, 41 per- about 50 percent higher than the average rate for cent of hospital admissions, and 47 percent of all the entire country (10 percent). In China, despite public expenditure attributable to care for individ- remarkable overall progress in health (infant mor- ual patients. At the other extreme was Piura, a tality fell from 20 percent in 1950 to 4.6 percent in poor mountainous region with 10 percent of the 1982), there is considerable geographic variation, sick but only 4 percent of public spending and of which is strongly related to income. Poor regions consultations. such as Yunnan, Xinjiang, and Tibet have infant Inequity in public spending for health both ac- mortality rates of more than 7 percent, compared counts for and reflects marked inequalities in ac- with less than 2 percent in more affluent Beijing, cess to and utilization of care. In Indonesia, for Guangdong, Shanghai, and Tianjin. To take an- example, despite significant investments in lower- other example, in Kenya the probability of a level health facilities in the 1980s, only 12 percent child's dying before age 2 varied among ethnic of public spending for health in 1990 went for ser- groups from 7.4 to 19.7 percent, and in Cameroon vices consumed by the bottom 20 percent of these probabilities ranged from 11.6 to 20.5 households, while the top 20 percent obtained 29 percent. percent of the government subsidy. This bias in The poor also have considerably worse access to favor of the wealthy was mainly a result of the health care. A number of surveys show that low- distribution of government spending for hospital income households, especially in rural areas, have inpatient and ambulatory care, services that were to travel considerably farther or longer to reach the used more frequently by the rich. Much more un- first level of referral services, usually a primary equal situations can be found in many countries 69 that concentrate government spending even more Household surveys systematically show that on high-level facilities. people choose whether to seek care and which The few countries in which public spending on provider to consult on the basis of many factors health is biased toward the poor show that govern- hours of service, travel time or cost, waiting time, ment policies can help reduce inequities in access availability of doctors or of drugs, and how pa- and health status. In Malaysia the government has tients are personally treated. The time required to followed a pro-poor policy since the 1970s, with get care can be valued according to local wages the lowest-income groups receiving a larger share and treated as a cost of service together with of public subsidies for health than the middle class money payments. On this basis, free public medi- and the wealthy. Similarly, in Costa Rica govern- cal care often is more costly than unsubsidized pri- ment spending for health has continued to favor vate care for which patients do not have to travel the poor, despite economic shocks and a major so far or wait so long. It is not surprising that, adjustment program in the 1980s that entailed cuts under these circumstances, even poor people ex- in public expenditure. In 1988 about 30 percent of press their dissatisfaction with public services by government spending for health went to the poor- paying for a great deal of private outpatient care. est 20 percent of households and just over 10 per- In both El Salvador and the Dominican Republic cent of spending to the richest 20 percent. The residents of the poorest quintile of the capital city poverty-oriented pattern of public spending for obtain more than half their ambulatory care from health in Costa Rica can be explained largely by private physicians. Although the price of private the high degree of coverage by the social security care to the poorest quintile is, on average, half that health system (the entire population is covered in for patients in the richest quintile, it is still fifteen principle, even though only 63 percent of the times higher than Ministry of Health fees. Differ- working population contributes) and the relatively ences in waiting timeone hour for private pa- equal access to and quality of care enjoyed by all tients as against two and a half hours at Ministry Costa Ricans. It also helps that the wealthy get of Health facilitiesaccount for much of this effect. most of their outpatient care from the private Sensitivity to price and travel time is also found sector. in rural Peru and Côte d'Ivoire. But private hospi- Consumer satisfaction with health care tal care is still much too expensive for the poor; even those who use private doctors go to public How satisfied people are with their own health hospitals. The excessive use of hospital care in re- and their health care can be only partly explained lation to ambulatory services often seen in public by objective criteria; subjective expectations mat- health systems partly reflects dissatisfaction with ter. People can also be pleased with their own the high cost in time and the perceived poor qual- health care and dissatisfied with their country's ity of ambulatory care. In the absence of incentives health system as a whole. A comparison of ten to improve lower-level facilities and service, this OECD countries with different health systems overuse reinforces the tendency to concentrate re- found that in eight countries public satisfaction sources on hospitals, urban areas, and less cost- was related to the level of spending. Canada, with effective interventions. the second-highest expenditure, had the highest The importance of public satisfaction with a satisfaction rating, and people were generally bet- health care system raises two issues for the pack- ter satisfied with the costlier health systems of age of publicly financed services proposed here. France, Germany, and the Netherlands than with First, it suggests that quality can be maintained the lower-spending systems in Australia, Italy, Ja- only if coverage is broad enough. Services de- pan, and the United Kingdom. Both very high ex- signed only for the poor will almost inevitably be penditure and great dissatisfaction were found in low in quality and will not receive the political sup- the United States. The study also showed that port necessary for adequate provision. This is a having a unified national health system did not difficult political issue because it may be hard to guarantee a high level of satisfaction. In most maintain equity and control costs if coverage is countries 30 to 50 percent of those polled sup- universal. The proper balance between more care ported "fundamental changes" in the health sys- for fewer people and the same amount of care for tem. In Italy and the United States many people more people depends on ensuring that the poor thought that such changes would not suffice and have access to the same quality of care as everyone that the health system should be "completely else and on limiting public finance to cost-effective rebuilt." services for which there is a sound rationale. Sec- 70 ond, reform of public provision alone, important more cost-effective care, but they also cannot as it may be, may have much less effect on health provide equitably what care they do offer be- outcomes, costs, and satisfaction than reforms that cause facilities will inevitably be geographically also try to stimulate competition and improve peo- concentrated. ple's access to a variety of providers. One of the principal responsibilities of govern- ment is to match the available instruments of poi- Matching means and ends icythe levers the public sector actually controls to the objectives. Much of governments' failure to The objectives of a health system are to improve achieve better health outcomes derives not from outcomes, control costs, increase equity, and sat- the wrong choice of objectives but from the wrong isfy users. Policy instruments, however, do not choice of instrumentsin particular, from too correspond to individual objectives. What govern- much reliance on direct provision of care and cen- ments actually do is build facilities, buy equipment tral control of health facilities and too little use of and supplies, hire and train people, set fees or the financial, informational, and regulatory instru- other service conditions, regulate providers and ments at the disposal of the government. These insurers, disseminate information, determine instruments are particularly important for improv- overall policy, and maintain surveillance of disease ing performance in the private market. When gov- conditions or other variables. Misallocation and in- ernments pay for health care in addition to regulat- equity are caused by mistakes in deciding what ing it, they have a further responsibility to provide facilities to build, where to locate them, how to value for money by ensuring that public resources staff them, and what services to provide. If gov- go first to cost-effective public health and essential ernments spend too much on tertiary care, for ex- clinical services so as to buy the largest health gain ample, not only can they not adequately finance possible. 71 Public health Health services interact with households in two ated, in collaboration with UNICEF, a global effort fundamentally different ways. Public health pro- to prevent a range of childhood diseases by immu- grams strike against health problems of entire nization. The EPI now reaches about 80 percent of populations or population subgroups. Their objec- children in developing countries and averts an es- tive is to prevent disease or injury and to provide timated 3.2 million deaths a year at a cost of $1.4 information on self-cure and on the importance of billion a year. seeking care. Clinical services respond to demand Population-based health services such as the EPI from individuals. They generally seek to cure or to rely on personnel with limited training to provide ease the pain of those already sick. This chapter drugs, vaccines, or specific health services directly discusses public health; Chapter 5 turns to clinical to specific populationsin schools, at worksites, services. or in households. Government finance for such Public health programs work in three ways: they programs is justified because the objective is usu- deliver specific health services to populations (for ex- ally to provide services to all in a community, be- ample, immunizations), they promote healthy be- cause the services create externalities or indirect havior, and they promote healthy environments. benefits, and because the diseases they typically Governments play a leading role, and provision of combat are particular problems for the poor. Three information through public education is a central types of interventions are immunization, mass feature of most programs, especially those de- treatment for worms and other infections, and signed to change behavior. But difficult choices screening and referral. Information, education, have to be made about the best use of public and communication are critical to many such pro- money. The Expanded Programme on Immuniza- grams, both to attract participation and to achieve tion (EPI), described below, is highly cost-effec- durable change in behavior. tive, at about $25 per DALY gained, but not all programs offer such good value for money. This Immunization chapter examines six particularly cost-effective public health services in the realms of population- Vaccines to prevent tuberculosis, measles, diph- based services (including immunization), nutri- theria, pertussis, tetanus, and polio have revolu- tion, fertility, tobacco and other drugs, the house- tionized preventive medicine over the past two hold and external environment (including control decades. Costs are less than $10 per DALY gained of insect vectors of disease), and AIDS. Public for measles immunization and less than $25 for a health packages in developing countries should in- combination of polio plus DPT (diphtheria, per- clude components in most or all of these six areas. tussis, and tetanus). These vaccines, together with BCG immunization against tuberculosis and lep- Population-based health services rosy and immunization of pregnant women against tetanus, form the EPI. In 1979 the World Health Organization declared As a result of the EPI, the proportion of children that smallpox had been eradicated. It then initi- immunized rose from less than 5 percent in 1977 to 72 20 to 30 percent by 1983. By 1990 coverage with paigns continue to be justified. In areas with better polio, DPT, and measles vaccines had reached ap- infrastructure, routine services make more sense. proximately 80 percent of all children, and about An ambitious current goal, established in 1988 35 percent of pregnant women were receiving tet- by WHO's governing body, the World Health As- anus toxoid. The lowest vaccine coverage is re- sembly, is to eradicate polio by 2000. Current ported in Sub-Saharan Africa. trends suggest that even if eradication is not Had vaccination coverage remained at the low achieved on that schedule, it will be soon there- levels of the 1970s, as many as 120 million DALYs a after. And substantial success has already been year (the equivalent of 23 percent of the global achieved: there has been no naturally occurring burden of disease among children under age 5 in case of polio in the Western Hemisphere since Au- 1990) would be lost to diseases preventable by the gust 1991. EPI. At current levels of vaccination coverage, Two extensions to the EPI appear to be justified. these diseases cause a loss of 55 million DALYs, or First, coverage should be extended, probably to 95 10 percent of the disease burden among children percent of all children born. The costs of expand- under age 5 (Table 4.1). ing coverage are relatively high, but so are the The cost of fully immunizing a child in low- gains. Those not covered at present often lack any income countries is about $15, with a range of $6 to health services and are disproportionately vulner- more than $20, depending on the prices of labor able to the diseases. Second, it makes sense to and other local inputs. Reducing the number of include additional items in the package: hepatitis B contacts needed to immunize each child fully and yellow fever vaccines for selected countries could cut costs dramatically, by as much as 70 per- and vitamin A and iodine supplements in regions cent if only one contact instead of the current five where deficiency of these micronutrients is highly were needed. This prospect depends on the suc- prevalent. If micronutrients are not delivered cess of ongoing research efforts. Technical im- through the EPI, some other vehicle must be provements in the cold chain (by which vaccines found for reaching very young children. Adding are kept refrigerated until use), good administra- these two vaccines and two micronutrients to the tion, widespread deployment of delivery teams, EPI (EPI Plus) would improve health substantially, and effective social mobilization efforts can also particularly in the poorest households, for a mod- contribute to dramatic cost reductions. In the est increase of about 15 percent in the cost of Gambia the cost of immunizations fell from $19 in reaching each child with complete services (vac- 1982 to $6 in 1988. Costs also depend on the immu- cine and micronutrients). Table 4.2 summarizes nization strategy: campaigns achieve high initial the estimated costs and health benefits of the EPI coverage, but routine services are more cost-effec- Plus cluster in two different settings. Total annual tive. In Ecuador campaigns cost $66 per DALY costs range between $2.2 billion and $2.4 billion for gained compared with $30 for routine services. Be- EPI Plus, or less than 2 percent of the public health cause many countries lack the infrastructure to de- expenditure of developing countries. Expanding liver vaccines routinely in remote rural areas, cam- coverage from 80 to 95 percent would probably Table 4.1 Burden of childhood diseases preventable by the Expanded Programme on Immunization (EPI) by demographic region, 1990 Share of the Burden per 1,000 Burden (millions total burden in children children under age 5 Region of DALYS per year) under age 5 (percent) (DALYs) Sub-Saharan Africa 23 15 242 India 16 12 137 China I 3 8 Other Asia and islands 7 10 81 Latin America and the Caribbean .1 3 18 Middle Eastern crescent 7 10 86 * Formerly socialist economies of Europe I * Established market economies I World 55 10 87 * Less than 1. Note: The EPI includes immunizations for pertussis, polio, diphtheria, measles, tetanus, and tuberculosis. These estimates exclude the burden from tuberculosis because most of it falls on adults. Source: Calculated from Murray and Lopez, background paper. 73 Table 4.2 Costs and health benefits of the EPI Plus cluster in two developing country settings, 1990 Middle-income countries Low-income countries (low mortality and Costs and benefits (high mortality and fertility) medium fertility) Cost per capita (dollars) 0.5 0.8 Cost per fully immunized child (dollars) 14.6 28.6 Cost per DALY gained (dollars) 12-17 25-30 Cost per DALY gained as a percentage of income per capita 0.14 0.03 Potential health gains as a percentage of the global burden of disease 6.0 1.0 Note: Figures are based on 95 percent coverage. a. Income per capita in 1990 was assumed to be $350 for low-income countries and $2,500 for middle-income Countries. Source: World Bank data and authors' calculations. increase annual costs by between $500 million and tion can be reduced by environmental improve- $750 million. In low-income countries the increase ments, especially sanitation, but where this is im- in coverage would reduce by about 6 percent the practical or unaffordable, it is cost-effective to global disease burden. repeat the therapy at regular intervals. The benefits of individual treatment can be sig- Mass treatment for parasitic worm infection nificantly enhanced by community-wide treatment which, by lowering the overall levels of contam- The most common intestinal wormsround- ination of the environment with infective stages of worms, hookworms, and whipwormseach infect the worms, slows the rate of reinfection. Treat- between 170 million and 400 million school-age ment programs targeted at the most heavily infec- children annually. Schistosomiasis infection, also ted group (school-age children) reduce infection caused by parasitic worms, affects almost 100 mil- immediately both among those treated and in the lion school-age children annually. The immediate rest of the population. Treatment through schools effects of infectionincluding failure to thrive, also allows delivery at relatively low cost: a pro- anemia, and impaired cognitioncan now be rap- gram in Montserrat was estimated to cost less than idly reversed by low-cost, single-dose oral ther- $1.50 per person for a cycle of eight treatments. A apy. Studies of single-course treatment of school- program managed by a nongovernmental organi- children with hookworm or schistosomiasis in zation in Jakarta initially cost $0.74 per capita per Kenya, with worm-induced disease in India, and year, but after expansion to almost 1,000 schools with trichuriasis in the West Indies showed re- the costs fell to $0.26. Such programs are ex- markable spurts in growth and development in all tremely cost-effective, at $15 to $30 per DALY the populations studied, including the large per- gained. In light of this cost-effectiveness and the centages of children with asymptomatic infections. burden of disease addressed, the Rockefeller And treatment also appears to have improved cog- Foundation and the UNDP are initiating a major nitive development. program to document and explore the potential of Curing worm infections is simple with inexpen- school-based health interventions that focus on sive modern drugs such as albendazole and prazi- deworming and provision of micronutrient quantel because it is not necessary to determine supplements. which species are present. Furthermore, the high level of safety of these drugs has led WHO to de- Mass screening and referral velop protocols for their use on a mass basis (where a high prevalence of infection exists) and Mass screening for disease control involves the ex- by providers who are not medically traineda amination of asymptomatic individuals to identify combination that makes for high cost-effective- and treat those affected by disease. Although this ness. Treatment usually cures the current infec- method has been used to control some infectious tion, but in endemic areas children will inevitably diseases, such as tuberculosis, it is mostly used become reinfected. A return to pretreatment levels for noncommunicable diseases. Mass screening of infection typically takes about twelve months makes sense for highly prevalent diseases that can for roundworm and whipworm and twenty-four be cured by early treatment, especially when la- months or more for hookworm. Rates of reinfec- tency periods span many years. An example is cer- 74 vical cancer, which is the leading cause of death of adding energy or protein to an initially poor from cancer among women in developing coun- diet, the effect of deficiencies in either or both tries, accounting for 150,000 deaths each year. components is combined under the term "protein- Screening with Papanicolaou (Pap) smears is com- energy malnutrition." Foods rich in protein, such mon in industrial countries, but attempts to repli- as soybeans and animal products, tend to be rela- cate those efforts in developing countries have tively costly per unit of energy, and low-cost rarely been successful. Such programs could, sources of energy such as cassava tend to be ex- however, be made cost-effective by the use of a pensive per unit of protein. Because food takes up simplified design that targets women over 35, much of a poor household's budget, choosing the screens only every five to ten years, and uses inex- protein-energy balance that is right for health can pensive outpatient treatment (such as freezing ab- be difficult. normal cells) for severe precancerous conditions. Protein-energy malnutrition raises the risk of When backed up with good follow-up services, death and may reduce physical and mental capac- such interventions are cost-effective, at $150 to ity. Worldwide, about 780 million people are esti- $200 per DALY gained. mated to be energy deficient according to WHO standards. It is not known how many of them are Diet and nutrition also protein deficient, or how many people who get enough energy may still suffer from shortage Eating well is necessary for good health. Either of protein. Exploratory studies of the determinants directly or in association with infectious diseases, of human growth suggest that at the margin, the inadequate diets account for a large share of the importance of additional protein may be greater world's disease burden, including as much as a than is recognized. Malnutrition is not syn- quarter of that among children. Much of this suf- onymous with hunger because people who have fering stems from poverty-related underconsump- become accustomed to a deficient diet may not tion of protein and energy, but equally important consider themselves hungry. If malnutrition is are deficiencies of key micronutrientsiodine, widespread in the community, underweight and vitamin A, and ironfrom which children and lethargic children look normal to parents who do women suffer disproportionately. Increasing the not know how healthy children behave. incomes of the poor is the most effective means of Iron deficiency is the most common micro- reducing protein-energy malnutrition, but govern- nutrient disorder. It reduces physical productivity ments can play an effective direct role through nu- and children's capacity to learn in school. By re- trition education, measures to increase consump- ducing appetite, it may diminish children's intake tion of micronutrients, and reduction in diarrheal and growth. Women suffer especially because and parasitic infections among children. Public ac- menstruation and childbearing raise their need for tion is also essential for preventing crop failures iron, and anemia, a shortage of iron in the blood, from leading to famines. increases the risk of death from hemorrhage in childbirth. The problem is worst in India, where 88 Malnutrition and ill health percent of pregnant women are anemic. Almost 60 percent of women are anemic in other parts of Low height for a given age, or stunting, is the most Asia, but the proportion does not exceed 40 per- prevalent symptom of protein-energy malnutri- cent in China, Africa, or Latin America. Anemia tion; approximately 40 percent of all two-year-olds affects 15 percent of pregnant women in the estab- in developing countries are short for their age (see lished market economies. Appendix table A.6). The prevalence of stunting Iodine deficiency causes mental retardation, de- may be as high as 65 percent in India; it is more layed motor development, and stunting, as well as than 50 percent in Asia other than India and China neuromuscular, speech, and hearing disorders. It and about 40 percent in China and Sub-Sahran is the leading preventable cause of intellectual im- Africa. Stunted children are often also under- pairment in the world. Cretinism from iodine defi- weight or have low weight for their age. Wasting ciency affects about 5.7 million people, and lack of (low weight for a given height) is less prevalent- iodine causes another 20 million to be mentally 11 percent or less worldwide except in India, retarded. where it reaches 27 percent. Vitamin A deficiency causes varying degrees of Diets must contain both energy and protein. Be- vision loss and is the primary cause of acquired cause little is known about the relative importance blindness in children. It also increases the severity 75 Table 4.3 Direct and indirect contributions of malnutrition to the global burden of disease, 1990 (millions of DALYs, except as specified) Latin America Middle Formerly Established Sub-Saha ran Other Asia and the Eastern socialist market Type of malnutrition Africa India China and islands Caribbean crescent economies economies World Direct effects Protein-energy malnutrition 2.2 5.6 1.7 0.9 1.0 1.0 0.2 0.2 12.7 Vitamin A deficiency 2.2 4.1 1.0 2.5 1.4 0.5 0.0 0.0 11.8 Iodine deficiency 1.7 1.4 1.0 1.3 0.5 1.4 0.0 0.0 7.2 Anemia 1.0 4.5 2.7 2.3 1.0 1.5 0.4 0.6 14.0 Total direct 7.0 15.5 6.3 7.0 3.9 4.5 0.6 0.9 45.7 Total DALYs per 1,000 population 13.8 18.3 5.6 10.3 8.9 8.9 1.7 1.1 8.7 Indirect effects (minimum estimate) Mortality from other diseases attributed to mild or moderate underweight' 23.6 14.9 3.3 8.0 2.4 8.0 0.0 0.0 60.4 Mortality from other diseases attributed to vitamin A deficiencyb 13.4 14.0 1.0 7.0 1.8 2.0 0.0 0.0 39.1 Based on the global burden of disease (GBD) attributable to deaths from tuberculosis, measles, pertussis, malaria, and diarrheal and respira- tory diseases in children under age 5; in developing countries 25 percent of those deaths are attributed to mild or moderate underweight. Based on estimated deaths attributable to vitamin A deficiency in the age groups 6-Il months and 1-4 years. These account for, respectively, 10 and 30 percent of all such deaths in high-risk countries and for 3 and 10 percent of all such deaths in other countries. Thirty lost DALY5 are attributed to each child death; losses are redistributed to the regional classification used in this Report. Source: For GBD calculations, Appendix B; for estimate of mortality from underweight, Pelletier 1991; for estimate of mortality from vitamin A deficiency, Humphrey, West, and Sommer 1992. Box 4.1 Women's nutrition Women suffer more than men from iron deficiency anemia, from stunting caused by protein-energy mal- normalities. A significant proportion of pregnancies end in poor maternal or infant health as a direct conse- 1 nutrition, and from iodine deficiency. The largest gap quence of maternal malnutrition. is for iron deficiency anemia, which affects 458 million Iodine and vitamin A deficiencies tend to be lo- adult women but 238 million men. About 450 million calized rather than widely distributed and could be vir- women are stunted because of protein-energy malnu- tually eliminated through targeted, sporadic interven- trition, compared with 400 million men. Iodine defi- tions, given a reasonable health infrastructure and a ciency also affects substantial numbers of women, high level of political will. Anemia and protein-energy probably more than for men. Corneal lesions and malnutrition, by contrast, affect much larger numbers blindness caused by vitamin A deficiency afflict both of women and require more continuous intervention. sexes equally, but deficiency as such is twice as com- Distribution of a regular supply of ferrous sulfate tab- mon for girls as for boys. Women's nutritional prob- lets can prevent or cure anemia among pregnant and lems are worst in South Asia, where prevalences of lactating women. Such efforts should include all anemia, protein-energy malnutrition, and vitamin A women of reproductive age, certainly where the preva- deficiency are the highest in the world and where, as a lence of anemia among women in general exceeds 50 result of widespread discrimination, girls and women percent. To reduce protein-energy malnutrition, much suffer disproportionately. must be done outside the health sector toward making Small pelvic size among stunted women increases more food available to households, increasing employ- the risk of maternal and infant mortality, as does ma- ment opportunities for women, decreasing the time ternal anemia. Iodine-deficient mothers give birth to and energy costs of women's home production, and more infants with cretinism and other congenital ab- reducing discrimination against women and girls. 76 of and mortality from a variety of infections, espe- Childhood mortality drops sharply as ,iutritional cially measles and diarrhea. WHO calculates that status improves. 13.8 million children have some degree of eye damage because of vitamin A deficiency; of these, 250,000 to 500,000 go blind every year, and two- Figure 4.1 Child mortality (in specific age thirds of the blinded children die. Both vitamin A ranges) and weight-for-age in Bangladesh, and iodine deficiency are particularly common in India, Papua New Guinea, and Tanzania Asia and Sub-Saharan Africa. These four diseases of malnutrition caused a di- rect loss of almost 46 million DALYs in 1990, or 3.4 percent of the global burden of disease (Table 4.3). (The estimates do not include the health damage from deficiencies of other micronutrients. Calcium deficiency may be the most important of these; it causes bone deformities and slows skeletal growth 500 4, in children, and it may contribute to osteoporosis in the elderly.) The estimated burden is slightly 269 larger for females than males because anemia af- fects mostly women ages 15-44 (Box 4.1); anemia accounts for 1.3 percent of the total female disease ,J 119 burden but for 24 percent among women in the reproductive ages. Children under 5 are the princi- 47 pal victims of vitamin A deficiency, iodine defi- ciency, and protein-energy malnutrition. The nu- o 18 tritional disease burden for young children is 32 0 million DALYs, or 6 percent of their total burden of illness. 6.7 The total impact of malnutrition on health is much larger, however, because mild or moderate 2.5 protein-energy malnutrition and micronutrient de- 55 65 75 85 ficiencies (as well as overcorisumption of energy, Weight-for-age fat, salt, and sugar) are risk factors for illness and (percentage of NCHS reference median) death. Studies in Asia and Africa consistently show that mild to moderate stunting or under- - Tanzania, 6-30 months weight in children raises the risk of death (Figure - Papua New Guinea, 6-30 months 4.1), contributing to 25 to 50 percent of childhood Matlab, Bangladesh, 12-59 months mortality. The greatest risk occurs for children in Punjab, India, 12-36 months their second year, after they are weaned. Malnour- ished children die principally from measles, diar- Note: NCHS, (U.S.) National Center for Health Statistics. rheal and respiratory disease, tuberculosis, per- The vertical axis shows the child mortality rate, CMR, in tussis, and malaria. Child deaths from these log-odds ratio form, log[CMR/(1000 - CMR)l. Thus, for a child in Punjab weighing 60 percent of the reference diseases cost 231 million DALYs, making the total median, the probability of dying between 12 and 36 burden attributable to malnutrition at least one- months is 18 per 1,000. fourth that amount, or 60 million DALYs. Source: Pelletier 1991. Vitamin A deficiency, too, raises the risk of death from other causes. Of the 8 million deaths of children with vitamin A deficiency that occur ech year, between 1.3 million and 2.5 million might be prevented by eliminating the vitamin deficiency, the total direct and indirect damage from malnutri- for a gain of 39 million to 74 million DALYs. Dam- tion is at least 20 to 25 percent of the disease bur- age from being underweight and from vitamin A den in children. deficiency cannot be added together because many Links between nutrition and growth in child- children suffer from both problems. Nonetheless, hood persist into adulthood. Both height and 77 weight affect the risk of adult mortality. For men part of the year when malnutrition is most preva- and women at all ages, greater height is associated lent. Small variations in diet can be fatal to chil- with greater survival. Stunted adults are at par- dren already at risk. ticularly high risk of cardiovascular disease and The extreme form of this risk is widespread fam- obstructive lung disease. If the relative risk of ine as a result of a breakdown in food production, death associated with stunting is the same in low- food distribution, or the flow of income with income countries as for richer populations, ap- which people buy food. Famines occurred in proximately 300,000 adult female deaths between China in 1959-61, in Bangladesh in 1974, in Ethio- the ages of 15 and 59 can be attributed to stunting. pia and the Sahel in 1973-74, in Ethiopia and Somalia in the 1980s, and in Somalia and Sudan in Sources of malnutrition the early 1990s. As many as 30 million people are believed to have died in the Chinese famine and Both food consumption and communicable dis- hundreds of thousands in the recent famines in ease affect nutritional status by way of a "malnu- Sub-Saharan Africa. A relatively small number of trition-infection complex." Food consumption de- people die from outright starvation; many die of pends both on people's capacity to acquire food infectious diseases, to which people weakened by and on their knowledge of how to choose a nutri- hunger are especially susceptible. tious diet. For infants the chief determinant of nu- Public action is critical in preventing a food crisis tritional status is whether they are exclusively from becoming a famine. A combination of actions breastfed for at least the first four to six months of is required to ensure that food is available in fam- life. In southern Brazil infants who were not ine areas (through both market and nonmarket breastfed were eighteen times as likely to die from mechanisms) and to sustain the incomes of vulner- diarrhea and three times as likely to die from respi- able households (through public employment or ratory illness as breastfed babies, both because other transfers). This is particularly difficult when they got less to eat and because of increased risk of there is a breakdown in order: the major African infection. After six months children need solid famines of the past decade were mostly associated food even if they are still breastfed. The composi- with war. tion and hygiene of this food are crucial to contin- Hunger and crowding into refugee camps facili- ued good health. tate the spread of infectious disease and raise the risk of death from it, particularly when such INCOME AND FOOD SECURITY. Chronic malnutri- camps are first established. Control of communica- tion is mostly a consequence of poverty. Higher ble disease is as crucial as the provision of food or income allows people to buy a more balanced diet, of money to buy food. Even when refugee popula- as well as better hygiene and medical care. In In- tions are protected from starvation, they are often donesia during 1984-87 rising incomes translated exposed to micronutrient deficiencies because they into reduced malnutrition in nearly all fifty-two are dependent on just a few foodstuffs. In recent regions of the country. The fraction of families eat- years there have been outbreaks of scurvy (vitamin ing less than 2,200 calories per person per dayan C deficiency) in Ethiopia and Somalia, pellagra energy intake adequate for only light physical ac- (niacin deficiency) among Mozambican refugees in tivitydeclined only 2 percentage points, but the Malawi, and ben-ben (thiamine deficiency) among decline was 9 percentage points at 1,800 calories Cambodian refugees in Thailand. In nonfamine and 26 percentage points at 1,400 calories. Con- conditions these diseases make no contribution to versely, increases in food prices in Côte d'Ivoire the world's burden of disease. during the 1980s reduced the weight of both chil- Beyond ensuring food distribution and control- dren and adults. Nutrition is also affected by who ling the diseases that can easily become epidemics in the household controls the money; women's in conditions of social and sanitary breakdown, income is more likely than men's to be spent on governments have two overriding responsibilities better nutrition. in famines. The first is to recognize the early signs Chronic food insecurity for poor people is often of trouble and act before large numbers of people made worse by seasonal fluctuations in availability have become destitute. The second is to allow free and prices. In India and the Philippines temporal flow of information about conditions during the variation in children's food intake is greatest famine so that relief agencies and others can react. among poor households, and severely malnour- Hiding the extent of a disaster only makes it ished children are more likely to die during that worse. 78 DIET AND DISEASE. Some children receive so lit- nicable diseases but can also increase diseases of tle to eat that getting more food is by far the most dietary excess. As diets change, usually to include important means of improving their growth. But a smaller proportion of complex carbohydrates for those with a barely adequate diet, controlling and more sugar and animal fat, people become infectious disease can be as important as getting more susceptible to cardiovascular disease and to more food. (This is the most common situation in cancers of the colon, prostate, and breast. Obesity poor countries.) In the absence of diarrhea, studies becomes more prevalent and increases the risk of have found little difference in growth in children premature death, particularly from cardiovascular up to thirty-six months of age despite significant diseases and diabetes. Increased sugar consump- differences in energy intake. Children with both tion contributes to dental caries and may raise the low energy intake and diarrhea grow less. Diar- risk of diabetes. High salt intake increases hyper- rhea has little effect on the growth of adequately tension, raising the risks of stroke and cardio- fed children. However, a diet adequate for healthy vascular disease. For people eating a modern children may be inadequate under the additional "Western" diet, a 60 percent reduction in lifetime demand imposed by infection. salt intake would reduce the risk of death from Control of parasitic worms can also help im- coronary disease at age 55 by 16 percent and the prove nutrition for older children. Hookworm and risk of death from stroke by 23 percent. The share other intestinal worm infections cause anemia, of the disease burden attributable to these dietary and roundwormsthe most prevalent of all changes is unknown but may be quite large. wormscompete with the host for food. All these The diseases of overnutrition, which strike peo- infections may suppress appetite and reduce food ple later in life, may seem unimportant compared intake. More than 1 billion people are infected with those of undernutrition. The burden of car- with one or more of these parasites, and about 100 diovascular disease and of some cancers, however, million of them suffer from stunting or wasting. is already important in Brazil, China, and other Recent intervention studies show rapid spurts in developing countries; the demographic and epi- height after children are treated for worm infec- demiological transitions documented in Chapter 1 tions, suggesting that these infections may be sig- will accelerate this trend. Treatment of chronic dis- nificant contributors to malnutrition and that it ease is often expensive or ineffective, so modifying may be easier to reverse stunting in older children diet and other risks is the best way of avoiding than was previously believed. growth in the burden from these diseases and in Malaria is the other major infection leading to unnecessary health care expenditure. malnutrition, particularly anemia. It is an impor- Other interventions for better nutrition tant risk factor in much of Africa. Control of ma- laria has led to substantial decreases in the preva- Six other interventions, in addition to control of lence of anemia. infectious diseases, may help to reduce malnutri- tion: nutrition education, control of intestinal LACK OF NUTRITIONAL KNOWLEDGE. People may parasites, micronutrient fortification of food, mi- eat more poorly than their incomes allow because cronutrient supplementation, food supplementa- of ignorance. This is most true for vitamin A; defi- tion, and food price subsidies. Control of parasites ciencies persist although almost everywhere in the was discussed above; the other five interventions world foods rich in vitamin A can be grown at low are reviewed here. cost in family gardens or commercially The effect of ignorance cannot be quantified, but it is striking NUTRITION EDUCATION. Inducing behavioral that it often increases the gender or age bias of changethus enabling families to improve their malnutrition. In a number of Asian and African diets even without additional incomeis often the countries, children and women, especially preg- most cost-effective way to improve nutritional sta- nant women, are discouraged from eating eggs tus. In Indonesia a large-scale program to teach and fruit. Ignorance also interacts with economic mothers about child feeding has reduced malnutri- factors. When the price of leafy vegetables rich in tion among children at very low cost. In a Colom- vitamin A rises in the Philippines, people switch to bian program using food supplements and "ma- vegetables containing much less of the vitamin. ternal tutoring," the effects of education on children's height and weight were as large as the THE NUTRITION TRANSITION AND CHRONIC effects of extra food. Education about feeding chil- DISEASE. Better nutritional status reduces commu- dren adequately during illness is particularly im- 79 portant in dealing with the interaction of malnutri- and may require efforts to compensate for the cost tion and infection. It may also ensure that in time that breastfeeding imposes on women. additional food is actually consumed by the family Breastfeeding may be incompatible with some oc- members who need it most. cupations, but in most cases modification of work- Probably the most valuable form of nutrition ed- place practices can facilitate nursing by working ucation is promotion of breastfeeding. The princi- mothers. pal gain is improved child health, but the nursing mother also benefits from conservation of iron MICRONUTRIENT FORTIFICATION. Fortifying the stores (because menstruation is suppressed), bet- foods people already eat raises micronutrient in- ter spacing of births, decreased risk of breast or takes even without a change in eating habits. Bra- ovarian cancer, and possibly less postpartum zil's national salt iodization program, which began bleeding. The economic benefits to families and in 1978, greatly reduced endemic goiter in areas of health facilities can be substantial. Breast-milk sub- iodine deficiency. Many experimental programs stitutes would cost an estimated $15 billion a year have also shown the value of fortification. In Chile for the 120 million infants now relying on mother's the addition of iron to powdered milk and soy- milk. Promotion of breastfeeding in a large Philip- based infant formula decreased anemia in nine- pine hospital saved 8 percent of the budget by re- month-old babies from 32 to 12 percent and in fif- ducing the cost of substitute foods and the time teen-month-olds from 30 to 6 percent. Iron is also spent in feeding by nursery staff. Such programs often added to flour. Vitamin A can be added to a require education of both mothers and health pro- variety of foods, including sugar, milk, cereals, fessionals (who often discourage breastfeeding) and monosodium glutamate (MSG). Fortification Box 4.2 The Tamil Nadu Integrated Nutrition Project: making supplementary feeding work In the late 1970s the government of the state of Tamil feeding was continued for up to 180 days. Intensive Nadu in south India was operating twenty-five differ- nutrition education was directed at mothers of at-risk ent supplementary feeding programs. Evaluation children. Food supplementation was also offered to showed these programs to be ineffective and identified women whose children were being fed, to those who several reasons. The programs were not directed to- had numerous children, and to those who were nurs- ward malnourished children; they provided food that ing while pregnant. was often not suitable for small children and was eaten The project cut severe malnutrition in half and pre- by other family members; they replaced rather than vented many at-risk children from becoming malnour- supplemented home consumption of food; they did ished. Of those receiving food supplementation, 67 not educate mothers; and they failed to provide percent gained enough weight to graduate in ninety needed nutrition-related health care. The Tamil Nadu days; all except the severely malnourished graduated Integrated Nutrition Project, the first phase of which within 150 days. Because participants were fed only ran from 1980 to 1989, was accordingly designed to when required, food was only 13 percent of the proj- target services more effectively, to improve family nu- ect's total cost, much less than is typical in supplemen- trition and health practices, and to improve maternal tation programs. (The initial share dropped during the and child health services. course of the project as the number of children who Children ages 6-36 months were weighed each needed feeding declined.) When the program began, month. Of every 100 children selected for feeding, 44 in 1980, 45 to 50 percent of the children required feed- were normal in weight but faltering in growth, 34 were ing; by 1988 the project had brought the share down to moderately malnourished and faltering, and 22 were 24 percent. Selective, limited-duration supplementary severely malnourished. Supplementary feeding was feeding worked in Tamil Nadu because the community I provided immediately to those who were severely mal- nutrition workers were well trained and highly moti- nourished, and feeding for children with faltering vated and because mothers came to understand the growth was provided after one month (for children importance of feeding for healthy growth and were ages 6-12 months) or three months (for children ages pleased when their children grew well. The experience I 12-35 months). The children selected were fed for at of Tamil Nadu suggests that appropriate supplemen- least ninety days. If they failed to gain at least 500 tary feeding is both an inexpensive and an effective grams in weight, they were referred to health care, and form of nutrition education. I 80 of MSG in Indonesia cut child mortality by 30 per- nomic stagnation. There was rio improvement in cent. Whenever a food consumed by the target districts not participating in the program. A large- population can be fortified at reasonable cost, for- scale program in Chile substantially reduced child- tification can provide the same benefits as promot- hood malnutrition while increasing the use of the ing changes in diet and may be quicker and easier. public health system. In many countries free meals for schoolchildren may have little effect on their MICRONUTRIENT SUPPLEMENTATION. Supplying nutritional status but improve school attendance micronutrients separately from food requires regu- and performance. In general, food supplementa- lar, sometimes frequent, contact with the target tion works best when it is used to motivate and population. This may make it more difficult to sus- educate mothers to care for their children's health, tain high coverage. It may also make supplements when it can be concentrated within a crucial inter- more costly than fortification of foodsalthough val (during pregnancy, for example), or when it micronutrient supplementation can be added at provides additional, rionnutritional benefits. very low marginal cost to immunization programs or school-based dewormirig programs. Vitamin A FOOD PRICE SUBSIDIES. Letting people buy basic can be given in capsules at intervals of one week to foodstuffs more cheaply can, in theory, increase six months, reducing the risk of blindness substan- intake of particular foods, but there are often prac- tially. Vitamin A supplementation can reduce mor- tical problems in targeting subsidies to needy tality from measles and diarrheal disease by about households. Targeting by locale or by commodities 30 percent but has little effect on deaths from res- eaten primarily by poor people is more efficient piratory disease. than wasteful general subsidies but less precise Iodine can also be provided as a supplement to than targeting according to specific needs. Ineffi- diets. Oral doses of iodized oil protect for two to ciencies in administration can eat up much of the four years, and injectable oil protects for three to potential benefit. One large urban subsidy pro- five years. Side effects are usually not serious and gram in Brazil has often sold food for nearly the occur mostly in older adults. Supplements for same price as private markets, despite a nominal women of reproductive age prevent mental retar- 20 percent price reduction. When such waste is dation in their children and reduce the risks of avoided, targeted subsidies can effectively transfer infant mortality. Iron deficiency is the most diffi- income to poor households. As with direct trans- cult micronutrient shortage to combat by supple- fers of income or of food, subsidies are more likely mentation: tablets must be taken every day, and to improve nutrition and health when they are they often cause side effects. Because these prob- combined with nutrition education and related lems limit compliance, supplements are usually health interventions. Unless that is done, sub- given only to pregnant women, who suffer most sidies are not cost-effective. from anemia. When the principal cause of iron de- There is a strong case for government interven- ficiency is infection with hookworm and other tion to improve health by improving nutrition, but parasites, however, iron supplements are also not for interfering generally in food markets, ex- given to all those treated for a limited period after cept in extraordinary conditions such as famine. deworming. Government action in nutrition has often been wasteful because it has duplicated what private FOOD SUPPLEMENTATION. Programs that provide markets do and has paid too little attention to the food instead of micronutrient supplements are causes of poverty and to cost-effective measures harder to implement effectively. Inadequate tar- that improve families' knowledge and capacity to geting, replacement of food from the normal diet, feed themselves adequately. Reductions in mortal- or lack of attention to other causes of malnutrition ity, blindness, mental impairment, and anemia can often mean that the food is wasted. With proper make fortification and supplementation extremely targeting and attention to changing behavior, cost-effective, comparable to the best control mea- however, some supplementation programs have sures for other diseases (Table 4.4). A year of been made to work. A program in Tamil Nadu, healthy life can be bought for less than $10 with India, achieved remarkable gains by distributing some micronutrient interventions and for less than food only when children's growth faltered, while $100 with programs that provide food supple- providing information to mothers continuously ments sparingly and combine them with behav- through highly motivated community nutrition ioral change and health care. Improved adult workers (Box 4.2). This success came despite eco- health, more productive schooling, higher in- 81 Table 4.4 Cost-effectiveness of nutrition interventions Approximate cost (dollars) Intervention Target group Per death averted Per DALY saved Iron supplementation Pregnant women 800 13 Iron fortification Entire population 2,000 4 Iodine supplementation Women of reproductive age 1,250 19 Iodine supplementation Entire population 4,650 37 lodization of salt or water Entire population 1,000 8 Vitamin A supplementationa Children under age 5 50 I Vitamin A fortification Entire population 154 4 Food supplementation Children under age 5 1,942 63 Food supplementation Pregnant women" 733 24 Semiannual mass dose. Deaths averted and DALYs saved are for fetal deaths. Source: Pinstrup-Andersen and others forthcoming; Levin and others forthcoming. Fertility has been declining worldwide, but at comes, and other benefits that come with good different paces. childhood nutritional status strengthen the case for appropriate nutrition interventions. Fertility Figure 4.2 Total fertility rates by demographic region, 1950-95 All pregnancies and births carry some health risks to the mother and the child. But the risks are higher when women have health problems (such Total fertility rate as high blood pressure, heart disease, malaria, or 7 diabetes) that could be aggravated by pregnancy, when pregnancies come too early or too late in a woman's reproductive life, when they are too closely spaced or are unwanted, and when they occur to high-parity women (for example, those who have already had four or more babies). The use of family planning services by couples is an effective means of avoiding many of these fer- tility-related health risks, and it enables families to achieve their fertility goals. In many parts of the world, fertility has been falling over time as the use of family planning spreads (Figure 4.2). Gov- ernments can do much to help couples by promot- ing family planning as a socially acceptable prac- tice, by providing information on the health effects 1950 60 70 80 90 95 of fertility regulation, by teaching couples about effective methods of contraception, and by remov- ing restrictions on the marketing of contraceptives. - Sub-Saharan Africa India Subsidies may be justified in low-income popula- tions, in rural areas, and for programs targeted to - China young people. Nongovernmental organizations Other Asia and islands and the private sector will often have a large role in - Latin America and the Caribbean service provision. Ensuring access to safe abortion - Middle Eastern crescent can complement family planning services in im- - Formerly socialist economies of Europe proving health. Established market economies Fertility patterns and health Note: Dotted lines represent projected values. Source: World Bank data. Births to very young women elevate the health risks to both mother and child. Births that are too 82 Child mortality risks are higher for babies born shortlij after a zious hi ung mothers. Figure 4.3 Risk of death by age 5 for fertility-related risk factors in selected countries, late 1980s Age of mother Length of previous birth interval Percentage increase in child mortality Percentage increase in child mortality (in relation to children (in relation to children born 24 to 48 born to women ages 20-34) months after the previous birth) 220 200 160 120 80 40 0 0 40 80 120 160 200220 Egypt Guatemala Indonesia Kenya Under 18 0 o Less than 18 months 18-19 0 18-24 months Source: Hobcraft 1991. closely spaced increase the risk of child mortality; to avoid such births. If the closely spaced births births at older ages and higher parities are riskier were delayed until mothers wanted them, overall to mothers, as are unwanted pregnancies that lead child mortality in those countries might be re- to unsafe abortion or to neglect of prenatal care. duced by more than 20 percent. The reduction could be as large as 30 percent in Brazil, Colombia, THE TIMING OF BIRTHS AND CHILD HEALTH. Short Ecuador, Peru, and Tunisia, where between 40 and birth intervals pose substantial risks to child health 50 percent of births are spaced too closely. throughout the first five years of life (Figure 4.3). In Kenya, for example, infants born within eigh- CHILDBEARING AND MATERNAL HEALTH. Each teen months of the birth of a previous child are year about 430,000 women in developing countries more than twice as likely to die as those born after die from complications associated with childbear- a longer interval. In Egypt the risks are more than ing. In the absence of obstetric care, women who triple. Babies born to teenage women are also at have a birth before age 18 are three times as likely greater risk of dying. In Indonesia, for example, to die in childbirth as those who have a birth be- babies born to mothers age 18 and under are 50 tween ages 20 and 29; for women over 34, the risk percent more likely to die than those born to of maternal mortality is five times as high. First women ages 20-24. births are often riskier than second or third births, Surveys taken in twenty-five developing coun- but as parity rises thereafter, the risk of maternal tries in the 1980s show that, on average, nearly 35 mortality also climbs. In Jamaica the risk increases percent of births occur within twenty-four months by 65 percent after the third birth. In Kenya, the of the previous birth and that many women wish Philippines, and Zimbabwe 30 to 60 percent of 83 pregnant women are estimated to be at elevated tion of contraceptives, and abandonment of un- risk of death on account of either age or parity. necessary prescription requirements. Experience Maternal deaths also arise from unsafe abortion. in the formerly socialist economies of Europe has Almost 30 percent of pregnancies end in abortion, shown that all such constraints reduce contracep- for a total of about 55 million induced abortions in tive use and often damage maternal and child the world each year; 25 million of these are per- health. Demand for many contraceptive supplies formed under unsafe conditions. The damage to and services can be met by private doctors and maternal health arises mainly from infection (the commercial outlets, especially in towns and, for long-run consequences of which include ectopic some methods (such as condoms), in rural areas as pregnancy, chronic pain, and infertility), hemor- well. rhage, damage to the cervix or uterus, and reaction Use of contraceptives is the best way to avoid to anesthesia and the drugs used to induce abor- unwanted pregnancies, but it is not foolproof. For tion. About 60,000 women a year are estimated to women who wish to terminate their pregnancies, die from unsafe abortions (see Appendix table access to safe abortion as a complement to contra- B.8); other estimates range as high as 200,000. ceptive services is also important to women's Treatment of abortion-related complications can health. consume significant resources. In Brazil in 1988 about 2 percent of all hospital admissions in the REACHING LOW-INCOME AND RURAL POPULATIONS. publicly financed, privately provided health sys- The health infrastructure in poor countries is often tem were for abortion-related complications, and limited in its ability to reach highly dispersed the costs amounted to about 6 percent of all spend- populations in rural areas. In rural Uganda, for ing on obstetrics and 1 percent of all hospital example, travel time to the nearest family planning spending in that system. facility averages one hour, whereas it is only fif- Better health through family planning services teen minutes in rural Thailand. Long waits at the facility are another problem; a study of clinics in Family planning services can help women reduce several Latin American countries found that wait- the health risks from mistimed and unwanted ing times for initial visits averaged one hour and pregnancies. In low-income populations and in twenty minutes. In many countries rural women rural areas there is a strong case on equity grounds have no access to family planning fieldworkers for the government to subsidize and organize the who can provide information and simple services. provision of family planning services, using public In Guatemala, for example, 86 percent of rural as well as nongovernmental and private channels women live in communities without a family plan- as appropriate. In these settings subsidized family ning fieldworker; in Egypt the figure is only planning services are often the most effective way 33 percent. Community-based strategies have of transmitting family planning information to the been used with success in some countries to reach poor. They can also be an efficient means of im- low-income women. In Colombia, Zaire, and proving the welfare of poor families, especially Zimbabwe community-based-distribution (CBD) when private medical care is unavailable. For both workers serve the dual purpose of spreading infor- reasons, family planning services are part of the mation about family planning and providing the minimum essential clinical package discussed in most isolated populations with family planning the next chapter. Special efforts are also appropri- methodsprimarily barrier methods, such as con- ate for addressing the needs of adolescents, both doms and foaming tablets, but also oral because they tend to be particularly uninformed contraceptives. about reproductive health risks and because Family planning services provided through they often misjudge the consequences of early community-based distribution are a highly cost- childbearing. effective means of improving maternal and child Beyond providing subsidized services to specific health. In countries where both mortality and fer- populations, the government also has a role in en- tility are still relatively high, the cost per child suring access to family planning services for those death averted is extremely low. In Mali, for exam- able and willing to pay. Encouraging better ser- ple, it averages about $130, which corresponds to a vices and availability of more contraceptive mere $4 to $5 per DALY gained. In other countries, methods requires various changes, including re- such as Colombia, Mexico, and Thailandwhere moval of price controls and bans on contraceptive mortality and fertility are substantially lower- advertising, easing of restrictions on the importa- CBD family planning services cost no more than 84 $25 per DALY gained and thus remain highly cost- women to discontinue contraceptive use. In Thai- effective. land and Colombia, where the programs offer good services, about 6 percent of users quit each REACHING YOUNG PEOPLE. In developing coun- year for these reasons; in Paraguay as many as 18 tries childbearing among teenage women (ages 15- percent of users discontinue annually. 19) is common. Surveys in the 1980s in Liberia, There is considerable scope for broadening the Mali, and Uganda show that more than one in five range and quality of contraceptive methods. India, teenage women had had at least one child or was for example, has for a long time heavily empha- pregnant at the time of the interview. In Latin sized sterilization and offered attractive financial American and Caribbean countries 16 percent of incentives to both clients and providers. A nation- all births in 1992 were to teenage mothers. Adoles- wide survey in 1986-87 showed that among non- cent pregnancies are often unintended and tend to sterilized couples seeking a temporary method of be more prevalent among low-income women. In contraception, nearly 75 percent of those who both Ghana and Kenya, for example, about 40 per- wanted intrauterine devices (IUDs) reported fail- cent of married teenagers who have had children ure to get them, 67 percent reported failure to get said their first pregnancies were unintended; contraceptive pills, and 40 percent reported not among unmarried teenagers the proportion of un- being able to get condoms. In China, where steel- intended births rose to 58 percent in Ghana and 77 ring IUDs have been in widespread use, the gov- percent in Kenya. A 1986 study of Brazilian ernment, because of concern about the risk to women showed that 65 percent of those who be- women's health, recently decided to switch to the came mothers before age 20 came from poor fami- safer Copper-T IUDs. In some countries the range lies (that is, those with household income below of available methods is constrained because public the national median), in contrast to 48 percent for sector providers are required to use products on an women who delayed childbearing. essential drug list and the list mistakenly excludes Family life education in schools and other some contraceptive methods. Other constraints on venues can help teenagers make informed choices method availability include excessively restrictive about sexual behavior and the prevention of sexu- medical screening requirements, unnecessary or ally transmitted disease (STD). Family planning duplicative approval procedures, packaging and services are needed to help sexually active adoles- labeling requirements that perform no useful func- cents prevent pregnancies. And programs to help tion but increase costs, and import restrictions or teenagers cope with unintended pregnancies, es- tariffs. A study in Indonesia that surveyed a group pecially premarital ones, can be especially valu- of women eighteen months after they started able. In Jamaica the Women's Center Program has using contraceptives found that, all else being the had some success in helping young mothers to same, women who failed to get their contraceptive complete their schooling after childbirth and to method of choice were more than three times as avoid another mistimed pregnancy. likely to have discontinued use as women who did receive their preferred method. IMPROVING SERVICES AND ENCOURAGING GREATER Providing an appropriate mix of contraceptive VARIETY IN METHODS. The quality of family plan- methods can also help to reduce the spread of ning services in developing countries has been im- STDs and human immunodeficiency virus (HIV). proving, but more can be done. Providing good Linking the provision of family planning services counseling to clients is important, in part because with screening programs for STDs requires a clini- women's contraceptive needs change over the re- cal setting in which positive diagnoses may be fol- productive life cycle. Temporary methods are lowed up with treatment. The discussions of AIDS more appropriate earlier in the cycle, while perma- in this chapter, below, and in Chapter 5 address nent methods are more appropriate toward the this point in more detail. end. And certain methods are more or less suitable depending on the duration of protection desired ENSURING ACCESS TO SAFE ABORTION. In 1990 and on whether the woman is breastfeeding. Com- about 40 percent of the world's population lived in petent advice offered with sensitivity can help cli- countries where induced abortion was permitted ents choose the right method at each stage and use on request, 25 percent lived where it was allowed it effectively while also addressing their concerns only if the woman's life was in danger, and the about possible side effects. Dissatisfaction with remaining 35 percent lived in places where abor- services and contraceptive failure often cause tion laws varied in strictness between these ex- 85 In Roman ía, maternal deaths shot up when abortion maternal mortality rate had risen by nearly 40 per- was banned and fell sharply when it was legalized. cent above the level in 1965 (Figure 4.4). Before 1966 Romania's maternal mortality rate was simi- lar to the rates in other Eastern European coun- Figure 4.4 Maternal mortality in Romania, tries. By 1989 it was at least ten times the rate of 1965-91 almost any other European country. In 1990 Ro- mania's new government legalized abortion, and Maternal deaths per 100,000 live births the decline in maternal mortality was immediate 180 and even sharper than its rise following the ban: only one year after abortion was legalized, mater- nal mortality had fallen to just 40 percent of the A 1989 level. The percentage of all maternal deaths All causes caused by abortion dropped from nearly 90 per- cent before the ban on abortion was lifted to just IA over 60 percent in 1990. COSTS AND POTENTIAL GAINS IN HEALTH. Family planning services, particularly when delivered through community-based distribution, are among the most cost-effective means of improving maternal and child health. There is much scope for improving services in developing countries, where more than one women in five who wants to avoid pregnancy is not using contraception. In Bolivia, Ghana, Kenya, Liberia, and Togo at least one in three women ages 15-49 falls into this category. Lack of access to family planning services is one reason for not using themalthough it is certainly Abortion legalized not the only one. The cost of supplying family planning services to women without access (num- Abortion made illegal bering an estimated 120 million in the developing world) is estimated at about $2 billion annually for developing countries as a whole. Selective alloca- N tion of public resources to address the needs of these women, particularly those in poor families, Source: Adapted from Stephenson and others 1992, which would be a cost-effective means of promoting their used Romanian Ministry of Health data. well-being, as well as that of their children. Satis- fying the expressed wish of women to space or limit future births might each year avert as many as 100,000 maternal deaths and 850,000 deaths tremes. In countries where abortion is illegal, among children under 5. women resort to clandestine, and often unsafe, abortions at high risk to their health. Legalizing Reducing abuse of tobacco, alcohol, and drugs abortion is inadequate for protecting maternal health when problems with access continue. In In- Decisions about the use of tobacco, alcohol, and dia, for example, abortion is legal but not readily other drugs are among the most important health- available, and many women continue to rely on related choices that individuals can make. Because unsafe abortion, with detrimental effects to their individual options are limited by the strongly ad- health. dictive character of these substances, and because Romania's experience is the most striking exam- addiction is often established in adolescence, deci- ple of the impact of abortion laws on maternal sions about the control of tobacco and other addic- health. In 1966 the government banned abortion tive substances are among the most important and contraception and took steps to enforce the health-related choices that societies can make col- law. The consequences were dramatic: by 1970 the lectively. In many populations prolonged cigarette 86 smoking is already the greatest single cause of pre- crease in consumption from 500 billion cigarettes mature death. Alcohol and other drugs also con- in 1978 to 1,700 billion in 1992 has produced smok- tribute to disease and disability. The damage from ing patterns that, if they persist, will eventually substance abuse is not limited to the individuals result in about 2 million deaths a year from to- involved; others also suffer indirectly because of bacco. Similar consumption patterns exist in sev- drunk driving, fires, passive smoking, and drug- eral other countries. If, as now, about one-third of related crime and violence. the world's young adults become regular cigarette Several sorts of government policy can be used smokers and, as in industrial countries, more than to discourage consumption of tobacco, alcohol, one-third of them die prematurely because of the and other drugs. Educating the public about the habit, then, of the 120 million who reach adult life harmful effects of these substances is essential. each year, more than 10 percentmore than 12 Appropriate action will often involve special em- million a yearwill die prematurely because of to- phasis not only on reaching school-age children bacco. On current smoking patterns, the chief un- but also on helping adults to escape from addic- certainty is not whether mortality from tobacco tion. Tax policies on tobacco and alcohol have also will reach 12 million a year in the second quarter of reduced consumption, especially by discouraging the next century, but exactly when it will do so. use by young adults before they become addicted. Largely because of the long delay between cause Governments can ban all direct or indirect adver- and full effect, people tend to misjudge the haz- tising or promotion of tobacco goods or trade- ards of tobacco. When a generation of young marks and could do the same for alcohol. adults begins to smoke, they do not witness the Tobacco high mortality associated with their behavior until they reach middle age. The best-documented ex- Tobacco is in legal use everywhere in the world, ample of this delay is that of men in the United yet it causes far more deaths than all other psycho- States, among whom the main increase in smok- active substances combined. About 3 million pre- ing took place before 1945. In 1945 smoking was mature deaths a year (6 percent of the world total) common but lung cancer rare, as in developing are already attributable to tobacco smoking. If cur- countries today. Over the next forty years (1945- rent trends continue, deaths from tobacco world- 85) the smoking habit did not change greatly wide are projected to reach 10 million a year, or among young men in the United States, but lung more than 10 percent of total deaths, by the second cancer in this population rose sharply (Figure 4.5). quarter of the next century. Tobacco is already re- Among U.S. nonsmokers lung cancer remained sponsible for 30 percent of all cancer deaths in de- approximately constant at a low level during 1965- veloped countries, including deaths from cancers 85, but among smokers the rates increased twen- of the lung, oral cavity, larynx, esophagus, blad- tyfold. In 1985 tobacco caused the large majority der, pancreas, and kidney. Even more people die (110,000) of all lung cancer deaths, among both from tobacco-related diseases other than cancer, males and females, in the United States, as well as including stroke, myocardial infarction, aortic an- an even larger number (290,000) of deaths from eurysm, and peptic ulcer. In countries where other diseases, for about 20 percent of 2 million smoking has long been widespread, tobacco use is U.S. deaths. About half of those killed by tobacco now responsible for about 30 percent of all male were still in middle age (35-69) and thereby lost deaths in middle age. Smoking also harms the almost twenty-five years of nonsmoker life health of others. Among nonsmokers, exposure to expectancy. environmental tobacco smoke increases the risk of Effective discouragement of addiction to tobacco lung cancer. And the babies of mothers who involves slow social changes that take place over smoke weigh, on average, 200 grams less at birth many years. Public education is central to this pro- than those of nonsmokers. cess. In China, the United Kingdom, and the Per capita consumption of tobacco is decreasing United States, autonomous national action groups slowly in industrial countries and has remained such as Action on Smoking and Health have relatively unchanged in the formerly socialist helped sustain serious efforts to alert people to the economies. By contrast, per capita tobacco con- hazards of tobacco consumption and, through le- sumption is rising in many developing countries gal action, to protect the public from the harmful among both men and women and is expected to health effects of the habit. Governments can con- increase by about 12 percent between 1990 and tribute to the efforts of citizen groups by, for exam- 2000 (see Appendix table A.6). In China the in- ple, requiring prominent health warnings on ciga- 87 The spread of cigarette smoking among U.S. males free zones. Tobacco consumption per adult ap- was followed by mounting lung cancer ratesbut pears to have fallen between 1975 and 1990. China, only after a decades-long delay. with 300 million smokers, is following a similar path: in 1992 it banned most tobacco advertising, mandated health education for youths, prohibited Figure 4.5 Trends in mortality from lung smoking in many public places, and required pro- cancer and various other cancers among gressive reduction of tar levels. In support of U.S. males, 1930-90 countries' efforts to discourage tobacco consump- tion, the World Bank in 1992 set forth a new policy Deaths per 100,000 males a on tobacco (Box 4.3). 80 Tobacco has traditionally been taxed, although probably because it is a good source of revenue rather than for the health gains. Taxation reduces consumption, especially among the young. In in- dustrial countries a 10 percent price increase re- duces consumption by about 4 percent in the gen- eral population and about 13 percent among adolescents. Besides having few resources, most adolescent smokers probably have not been smok- ing long enough to be fully addicted and so tend to be more price sensitive than other smokers. In In- dia cigarette sales declined by 15 percent after the excise tax on most of the popular cigarette brands more than doubled in 1986. In Papua New Guinea a 10 percent increase in the tobacco tax reduced consumption by 7 percent. 1930 40 50 60 70 80 85 Alcohol and illegal drugs Alcohol-related diseases affect 5 to 10 percent of Lung cancer the world's population each year and accounted -. Various other cancers for about 3 percent of the global burden of disease in 1990. Of the 2 million alcohol-related deaths Note: Other cancers shown include leukemia and cancers that occur worldwide each year, about 50 percent of the bladder, esophagus, pancreas, liver, prostate, stem from cirrhosis of the liver, about 35 per- stomach, and colon and rectum. cent from cancer of the liver or esophagus, 10 a. Adjusted to the age structure of the U.S. population in 1970. percent from alcohol dependence syndrome, and Source: Boring, Squires, and Tong 1993. 5 percent from injuries caused by motor vehicles. The problems caused by alcohol abuse consume scarce medical resources and extend beyond the damage that drinkers do to themselves. In many Latin American countries in the 1980s, 20 percent rette packages and advertisements, as well as by of all hospital and emergency room admissions targeting of clear messages not only to school-age were alcohol-related. In Papua New Guinea more children but also to adults. Reaching adults is im- than 85 percent of fatal road accidents in the 1980s portant because over the next few decades it is involved either drunk drivers or drunk pedes- those who are already smoking who will account trians. Within households, drinking often leads to for nearly all of the tens of millions of deaths per assault and injury, although the scale of the prob- decade caused by tobacco. Some countries go fur- lem is hard to quantify. ther by banning commercial promotion of tobacco Alcohol consumption is stable in the industrial goods and tobacco trademarks and by placing re- world but is on the rise in many developing coun- strictions on public smoking. Singapore has been tries. Between 1960 and 1981 annual beer con- in the forefront of public activism in Asia: it has sumption per capita rose from 12 liters to 135 liters prohibited advertising (since 1971), issued strong in Gabon and from 3 liters to 20 liters in Côte warnings on health effects, and created smoke- d'Ivoire. Total world production of beer nearly 88 Box 4.3 World Bank policy on tobacco In 1992, in recognition of the adverse effects of tobacco requirements. The World Bank seeks to help these consumption on health, the World Bank articulated a countries diversify away from tobacco. formal policy on tobacco. The policy contains five main To the extent practicable, the World Bank does not points. lend indirectly for tobacco production activities, al- The World Bank's activities in the health sector though some indirect support of the tobacco economy including sector work, policy dialogue, and lending may occur as an inseparable part of a project that has a discourage the use of tobacco products. broader set of objectives and outcomes (for example, The World Bank does not lend directly for, invest rural roads). in, or guarantee investments or loans for tobacco pro- Unmanufactured and manufactured tobacco, to- duction, processing, or marketing. However, in the bacco-processing machinery and equipment, and re- few countries that are heavily dependent on tobacco as lated services are included on the negative list of im- a source of income and of foreign exchange earnings ports in loan agreements and so cannot be included (for example, those where tobacco accounts for more among imports financed under loans. than 10 percent of exports) and especially as a source of Tobacco and tobacco-related producer or con- income for poor farmers and farmworkers, the World sumer imports may be exempt from borrowers' agree- Bank treats the subject within the context of respond- ments with the Bank to liberalize trade and reduce tar- ing most effectively to these countries' development iff levels. doubled between 1970 and 1989, far surpassing (that is, about one to two months' wages) are esti- population growth, with much of the increase oc- mated to reduce traffic fatalities by about 5 per- curring in developing regions. cent. Mandatory jail sentences for drunk driving As with alcohol, abuse of illegal drugs causes have also been weakly effective. The effect of in- serious health and social problems. Individuals formation campaigns concerning alcohol con- run the risk of death from infectious, circulatory, sumption has not been quantified, but there is evi- respiratory, and digestive diseases, as well as from dence that in countries where alcohol is legal but violence, overdose, and AIDS. Users of cocaine, commercial promotion is not, per capita alcohol especially in the form of "crack," often suffer consumption is 30 percent lower than elsewhere acute cardiovascular problems that require emer- and deaths from motor vehicle accidents are 10 gency room services, and the babies of pregnant percent fewer. As a successful alcohol rehabilita- users of cocaine are often born with severe health tion program in south India demonstrates, com- problems. munity efforts are generally more effective than Reliable data on trends and patterns in illegal medical interventions in helping individuals to drug use are scarce. Users typically fall in the age overcome alcohol dependence, in part because of group 15-44, although most are in their mid-twen- the importance of sustained encouragement, ties. In the past decade the production and con- which is more easily offered by the community sumption of illicit drugs, especially cocaine, ap- than by health service institutions. pear to have increased considerably worldwide. In Prohibition is a common approach to drug some developing countries the use of psychoactive abuse. In the United States prohibition as an ap- drugs such as inhalants is also a serious problem. proach to control of alcohol failed early in this cen- Taxes and judicial penalties have been used to tury. It appears to be having, at best, limited suc- discourage alcohol abuse. A 1982 U.S. study indi- cess in controlling use of other drugs now. In other cated that an increase in the liquor tax of about settings, including Malaysia and Singapore, prohi- $3.50 (at 1991 prices) per gallon equivalent of pure bition coupled with a mandatory death penalty for alcohol would lower demand enough to reduce drug trafficking appears to have been more effec- the incidence of liver cirrhosis by 5 percent in the tive. The successes achieved in controlling the use short run and perhaps twice as much in the long of alcohol and tobaccothrough restrictions on run. In industrial countries mandatory license promotion and access, high taxation, rehabilitation sanctions on drunk drivers are estimated to de- of addicts, and public educationmay also be rele- crease traffic fatalities by about 10 percent; the im- vant for efforts against other drugs. For alcohol position of a minimum legal drinking age and the and tobacco, past successes with these measures assessment of relatively large mandatory fines should spur efforts toward full implementation. 89 Environmental influences on health compounded by poor hygiene), inadequate gar- bage disposal and drainage, heavy indoor air pol- The environment in which people live has a huge lution, and crowding. The diseases associated influence on their health. For poor people and with poor household environments occur mainly poor regions, it is the household environment that in developing countries, where they account for carries the greatest risks to health. By providing nearly 30 percent of the total burden of disease information, reducing poverty, and facilitating and (Table 4.5). Modest improvements in household stimulating private sector action, governments can environments would avert almost a quarter of this deploy potent mechanisms to improve this envi- burden, mostly as a result of reductions in diar- ronment. Potential health gains from these efforts rhea and respiratory infections. total nearly 80 million DALYs a year in developing countries. Other government actions, designed to WATER AND SANITATION. About 1.3 billion people ameliorate or remedy unsafe conditions in the in the developing world lack access to clean and workplace and pollution of the ambient environ- plentiful water, and nearly 2 billion people lack an ment, could save 36 million and 8 million DALYs a adequate system for disposing of their feces (Fig- year, respectively. Finally, feasible reductions in ure 4.6). Feces deposited near homes, contami- the toll taken by road traffic injuries could avert nated drinking water (sometimes caused by poorly the loss of 6 million DALYs a year. designed or maintained sewerage systems), fish The household environment from polluted rivers and coastal waters, and agri- cultural produce fertilized with human waste are Poor households generally live in a domestic envi- all health hazards. Water quantity is as important ronment with high health risks caused by poor as water quality. Washing hands after defecation sanitation and inadequate water supply (often and before preparing food is of particular impor- Table 4.5 Estimated burden of disease from poor household environments in demographically developing countries, 1990, and potential reduction through improved household services Burden from these diseases in Reduction Burden averted Burden averted developing achievable by feasible per 1,000 Principal diseases related countries through feasible interventions population to poor household (millions of interventions (millions of (DALYs per environmentsa Relevant environmental problem DALYs per year) (percent)b DALYs per year) year) Tuberculosis Crowding 46 10 5 1.2 Diarrhea' Sanitation, water supply, hygiene 99 40 40 9.7 Trachoma Water supply, hygiene 3 30 1 0.3 Tropical clusterd Sanitation, garbage disposal, vector breeding around the home 8 30 2 0.5 Intestinal worms Sanitation, water supply, hygiene 18 40 7 1.7 Respiratory infections Indoor air pollution, crowding 119 15 18 4.4 Chronic respiratory Indoor air pollution diseases 41 15 6 1.5 Respiratory tract Indoor air pollution cancers 4 10 * 0.1 All the above 338 - 79 19.4 * Less than one. Note: The demographically developing group consists of the demographic regions Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent. The diseases listed are those for which there is substantial evidente of a relationship with the household environment and which are listed in Appendix B. Examples of excluded conditions are violence related to crowding (because of lack of evidence) and guinea worm infection related to poor water supply (not listed in Appendix B). Estimates derived from the product of the efficacy of the interventions and the proportion of the burden of disease that occurs among the exposed. The efficacy estimates assume the implementation of improvements in sanitation, water supply, hygiene, drainage, garbage disposal, indoor air pollution, and crowding of the kind being made in poor communities in developing countries. Includes diarrhea, dysentery, cholera, and typhoid. Diseases within the tropical cluster most affected by the domestic environment are schistosomiasis, South American trypanosomiasis, and Bancroftian filariasis. Based on very inadequate data on efficacy. Source: Appendix tables B.2 and B.3 and authors' calculations. 90 Many people worldwide still lack safe water and adequate sanitation. Figure 4.6 Population without sanitation or water supply services by demographic region, 1990 Percentage of population without services Millions of people without services 100 80 60 40 20 0 0 500 1,000 1,500 2,000 2,500 Sub-Saharan Africa IIIIIIIIIIIUIL I ndia Ilijilh!!! China Other Asia and islands 11111 Latin America and the Caribbean II !MII Middle Eastern crescent D Formerly socialist economies of Europe I Established market 11 economies World Sanitation W Water Note: Coverage is defined in accordance with local standards. Source: World Health Organization data. tance in reducing disease transmission, but with- suspended particulates, smoky houses in Nepal out abundant water in or near the home, hygiene and Papua New Guinea have peak levels of 10,000 becomes difficult or impossible. The lack of water or more. Rural people in developing countries may supply and sanitation is the primary reason why receive as much as two-thirds of the global expo- diseases transmitted via feces are so common in sure to particulates. Women and young children developing countries. The most important of these suffer the greatest exposure. diseases, diarrhea and intestinal worm infections, Indoor air pollution contributes to acute respira- account for an annual burden of 117 million tory infections in young children, chronic lung dis- DALYs, or 10 percent of the total burden of disease ease and cancer in adults, and adverse pregnancy in developing countries. In addition, an inade- outcomes (such as stillbirths) for women exposed quate water supply increases the risk of schis- during pregnancy. Acute respiratory infections, tosomiasis, skin and eye infections, and guinea principally pneumonia, are the chief killers of worm disease (Box 4.4). young children, causing a loss of 119 million DALYs a year, or 10 percent of the total burden of INDOOR AIR POLLUTION. Indoor air pollution, disease in developing countries. Data from the which World Development Report 1992 identified as Gambia, Nepal, South Africa, the United States, one of the four most critical global environmental and Zimbabwe suggest that reducing indoor air problems, probably exposes more people world- pollution from very high to low levels could poten- wide to important air pollutants than does pollu- tially halve the incidence of childhood pneumonia. tion in outdoor air. Whereas air in such cities as Adults can suffer chronic damage to the respira- Delhi, India, and Xian, China, contains a daily av- tory system from indoor pollution. Studies in erage of 500 micrograms per cubic meter of total China, India, Nepal, and Papua New Guinea have 91 Box 4.4 After smallpox: slaying the dragon worm Guinea worm disease, or dracunculiasis ("infection out of sources of drinking water when they have with a little dragon"), was endemic from ancient times guinea worm blisters on their legs and to filter their in a belt stretching from West Africa through the Mid- water with a cloth if they do not have a safe water dle East to India and Central Asia. It has been suc- source. cessfully eliminated from the Central Asian republics Eradication of dracunculiasis by the end of 1995 has and from Iran, where the last case was seen in the been adopted as an international goal. Pakistan may 1970s, and it has spontaneously disappeared from have achieved eradication in 1992, and Cameroon, In- most of the Middle East and from several African coun- dia, and Senegal may do so in 1993. Between 1987 and tries, such as the Gambia and Guinea. There are now 1992 cases reported per year fell from 653,000 to probably fewer than a million cases worldwide. 201,000 in Nigeria, from 180,000 to 33,000 in Ghana, The disease does not kill people, but it causes pain and from 17,000 to 900 in India. In general, these ad- and disability to its victims for several weeks in the vances have been achieved through "vertical" pro- year as the 60-centimeter-long female worm emerges gramsthat is, programs specific to dracunculiasis. from a blister, usually on the leg. In some cases the The eradication of the disease from the poor, sparsely disability is permanent. The worms usually emerge in populated endemic countries in West Africa will, how- the early rainy season, the time when the incapacitated ever, require integrated programs in which the re- victims would otherwise be planting and weeding their sources available for guinea worm control are shared crops. Children whose parents are stricken by guinea with other activities, such as immunization. A by-prod- worm are more likely to be malnourished in the follow- uct of guinea worm eradication will be community- ing year. based surveillance systems, which can be used by com- Because dracunculiasis can only be caught by drink- munities to monitor and improve their own health and ing infected water, improving the water supply is an by public health workers to combat other diseases, important preventive measure. Health education is such as polio. also essential. Villagers need to be persuaded to stay shown that up to half of adult women (few of ford the household improvements, including bet- whom smoke) suffer from chronic lung and heart ter water and sanitation services, that they desire. diseases. Nonsmoking Chinese women exposed to As people acquire more education, their hygiene indoor coal smoke (which is especially harmful) improves, and their responsiveness to public in- have a risk of lung cancer similar to that of men formation programs increases. To support house- who smoke lightly. Comprehensive improvement holds' efforts, governments have an important in indoor air quality in the developing countries role in setting and enforcing appropriate environ- might avert a loss of 24 million DALYs each year mental standards and disseminating information by reducing the burden of acute respiratory infec- on, for example, the health benefits of good hy- tions and chronic respiratory diseases by 15 per- giene and the effects of exposure (especially of cent and of respiratory tract cancers by 10 percent babies) to smoke. Governments should also con- (Table 4.5). centrate on strengthening security of tenure HousiNG. In many cities 30 to 60 percent of the (which is essential for encouraging households to population live in overcrowded and deteriorating invest in their housing) and on establishing a legal, shanties, tenements, and boardinghouses. Crowd- regulatory, and administrative framework that fa- ing is associated with increased airborne infection cilitates responsive, accountable, and efficient pro- and personal violence. Poor structures lead to vision, often by private suppliers, of services that greater exposure to heat, cold, noise, dust, rain, people want and are willing to pay for. And they insects, and rodents. And housing locations ate should refrain from supplying services directly often unhealthy because of, for example, poor and from granting indiscriminate, widespread drainage. subsidies. Such subsidies are often captured by wealthier consumers, go for improvements that POLICIES FOR IMPROVING THE HOUSEHOLD ENVI- households would make anyway, or encourage RONMENT. The most powerful forces for reducing consumption patterns that are detrimental to domestic risks to health are rising incomes and health. (For example, subsidies for coal used in increased education for household members. cooking lead to more indoor air pollution than Higher incomes make it possible for people to af- would be the case with cleaner liquid or gas fuels.) 92 Past experience in water and sanitation illus- poorer class that receives little or no service; and a trates the limitations of direct government provi- ripe environment for political patronage. sion of household services. Despite technical prog- The poor usually miss out on both services and ress in developing affordable engineering solu- subsidies. They suffer the substantial health con- tions to the problems of water, sanitation, drain- sequences described in Table 4.5 and pay high age, and housing, the delivery and maintenance of prices for inadequate services. In Lima poor peo- these services, especially by governments, have ple may pay $3 for a cubic meter of contaminated been disappointing. At the end of the Interna- water collected by bucket from a private vendor, tional Drinking Water Supply and Sanitation De- while the middle class pays 30 cents per cubic me- cade (the 1980s), most people in the poorer regions ter for treated water provided on tap in their of the world still lacked sanitation, and the num- houses by the publicly subsidized water company. ber of urban residents without water had not been Broadly based subsidies are not necessary for reduced. ensuring access to safe water and sanitation. In Supply-side failures are largely caused by ineffi- most urban communities households are willing to cient and unresponsive public sector monopolies pay the full costs of water service and often the full which, in the water sector, typically provide subsi- cost of sanitation services. Willingness to pay for dized services at between one-third and two-thirds water may be high in rural areas as well, but what of the full economic cost. Massive public invest- people can afford is commonly not enough to ments, often supported by the donor community cover the high costs of supply. Subsidy may be and the World Bank, have been made in public or justified in such situations. But the rationale quasi-public agencies responsible for the delivery should be primarily one of redistribution: a society and maintenance of household services. The net re- may choose to provide cheap water or other ser- sult has often been bloated public agencies with low vices to the poor as one of many alternative means accountability to their customers and few incen- of improving their welfare. Health benefits alone tives for improving efficiency; a middle class that is do not generally provide a rationale for public sub- increasingly well served with subsidized services; a - sidy of water and sanitation (see Box 4.5). Box 4.5 The costs and benefits of investments in water supply and sanitation People want safe water and good sanitation and are discharge. The cost of water and sanitation services can willing to pay for these services, especially for plentiful range from $15 per person per year for simple rural water in or very near the home. Improvements in wa- systems to $200 for full-fledged urban systems. Poor ter supply raise productivity through savings in the households cannot afford the design standards of in- fuel used to boil polluted water and, even more impor- dustrial countries, but such standards are not neces- tant, through the time and energy savings for women sary on health grounds. Completely eliminating fecal who have to collect water from distant sources. Provi- bacteria requires expensive chlorination, but low con- sion of public handpumps in Imo State, Nigeria, re- centrations present little health hazard and should be duced the median time that each household spent on tolerated. water collection in the dry season from six hours a day If households pay the total cost of water and sanita- to forty-five minutes. In Lesotho, not an especially dry tion services because of the productivity and amenity country, the benefits in time saved alone are sufficient benefits, substantial health gains are an added bonus to justify investments in rural water supply. Sanitation achieved at no cost per DALY gained. When willing- improvements have high amenity value, making possi- ness to pay is much less than costs, it is usually a mis- ble a cleaner and more pleasant environment. take to justify subsidies on the basis of health benefits The costs of water supply and sanitation services alone. First, such subsidies compromise the demand- vary by technology, population density, the hydrologic driven approach to service provision (that is, provision and geologic environment, and design standards..De- of services that people want and are willing to pay for); sign standards for water supply can range from one lack of accountability and inefficiency are the inevitable handpump per 250 people, supplying 20 liters per per- consequences. And second, if publicly financed invest- son per day, to multiple-tap in-house connections that ments in these services are being considered for health supply several hundred liters of fully treated water per reasons, it should be noted that such investments gen- person per day. Design standards for sanitation can erally cost more per DALY gained than other health vary from a pit latrine to flush toilets connected to a interventions recommended in this Report. sewerage system, with downstream treatment prior to 93 Box 4.6 Environmental and household control of mosquito vectors Diseases transmitted by insect vectors account for Polystyrene beads losses of 44 million DALYs worldwide each year (35 million in Sub-Saharan Africa), or 3 percent of the The application of polystyrene beads to pit latrines has world burden (12 percent in Sub-Saharan Africa). Al- proved successful in reducing the breeding of Culex though widespread application of insecticides is help- mosquitoes and the transmission of filariasis. The ing to control river blindness in West Africa (see Box beads form a floating layer that discourages egg laying 1.1) and Chagas' disease in South America, it is no and suffocates any mosquito larvae that do hatch. In longer the mainstay of vector control against other dis- the town of Makunduchi (population 12,000) in eases. Emphasis has shifted to a range of targeted bio- Zanzibar, Tanzania, a combination of polystyrene-bead logical, physical, and behavioral approaches supported application and mass drug treatment of the population by insecticides when necessary. Two examples are between January 1988 and June 1989 virtually elimi- given here. nated biting by infective mosquitoes, and the propor- tion of people infected by filariasis fell from 50 to 10 Impregnated bednets percent. By January 1993 the proportion of people in- fected had fallen to 3 percent. The polystyrene bead Bednets impregnated with a pyrethroid insecticide of layers remained intact and effective for several years low mammalian toxicity form lethal traps for mosqui- and were disrupted only by exceptional flooding. toes attracted by the carbon dioxide and body odor In Zanzibar Town researchers are studying whether emitted by the occupants. In Sichuan Province, China, it is better to make beads freely available so that house- up to 2.25 million netsalready owned by nearly all holders can apply them to their own pits or to have householdershave been treated each year since 1987. trained teams identify and treat all pits requiring treat- If nothing else, the cost is much lower than spraying ment. In Dar es Salaam polystyrene beads are being the same houses with DDT. In Emei County, Sichuan, used to control Culex nuisance biting and thereby in- the number of malaria cases had been steady at about crease public acceptance of house spraying against the 4,000 between 1980 and 1986. After bednet treatment Anopheles vectors of malaria. The effectiveness of the began, the number declined steadily, to 352 in 1991. In Culex control measures is evidenced by declining sales the Gambia a combination of net treatment and chemo- of mosquito coils in local shops. In Madras, India, poly- prophylaxis, carried out by primary health care person- styrene beads are being applied to water tanks to con- nel, reduced overall child mortality by 63 percent. trol the local vectors of malaria and dengue; the quality of the water is not affected. An important policy issue, on which there is an sanitation servicesas in urban areasthe de- apparent tension between health objectives and mand for improved sanitation has invariably risen the demand-driven approach advocated here and automatically as the demand for water services is in World Development Report 1992, concerns the se- satisfied. Second, where the demand-driven ap- quencing and packaging of investments in water proach has not been followed, service provision and sanitation. Given the patterns of household has almost always been characterized by ineffi- choice, a demand-driven approach will usually ciency and lack of accountability. For the provision mean that provision of water supply services pre- of water supply and wastewater collection ser- cedes that of sanitation services. It is frequently vices, therefore, the demand-driven approach argued that this sequence would produce few should be compromised only in rare health benefits because rapid increases in water circumstances. use can overwhelm existing waste disposal capac- Households are less willing to pay for the cost of ity and because health benefits are maximized only trunk sewers and treatment of excreta and waste- when households utilize both better water and water. Because these investments benefit the better sanitation services. These arguments are whole community and are important for environ- plausible, but experiences in many countries sug- mental quality and health, there is potentially a gest that close adherence to the demand-driven case for using public funds to finance them. A few approach remains appropriate in most places, in- other situations may also justify direct government cluding low-income settings. First, where rapid in- action or subsidies. Householders tend to under- crease in water use is likely to cause environmental value such investments as areawide pollution and health problems in the absence of household abatement, vector control involving actions within 94 households (see Box 4.6), and research and devel- The occupational environment opment. There may thus be grounds for public subsidy or other interventions in these areas. It Many women work in the home and thus suffer will often be difficult to disentangle environmental disproportionately from the health risks in the and health benefits, and judgments will be neces- household environment just described. Both men sary concerning the use of public funds. and women may also encounter health risks in Large institutional and cultural shifts are needed workplaces outside the home. A burden of 36 mil- to create an efficient system for allocating scarce lion DALYs, or 3 percent of the global burden of public and private resources to improve the house- disease, is caused each year by preventable inju- hold environment. Many developing countries ries and deaths in high-risk occupations and by have inheritedand then elaborated onthe for- chronic illness stemming from exposure to toxic mer colonial powers' worst traditions of public chemicals, noise, stress, and physically debilitat- sector inertia and professional inflexibility. En- ing work patterns (Table 4.6). couragingly, however, private sector involvement The International Labour Office has estimated is increasing rapidly in both industrial and devel- that the cost of occupational injuries and associ- oping countries. SODECI, the privately run utility ated production losses in a sample of industrial in Abidjan, is considered one of the best-run water countries is between 1 and 4 percent of GNP. In companies in Africa. EMOS, the utility that serves developing countries this proportion is likely to be Santiago, has used private sector contracts for greater because accident rates tend to be higher. such functions as meter reading, pipe mainte- Rates of fatal occupational injuries among con- nance, billing, and vehicle leasing and is one of the struction workers, for example, are more than ten most efficient utilities in Latin America. The role of times higher in Kenya and Thailand than in Fin- community organizations and NGOs may also be land. Agriculture, which employs more than half significant, particularly in drainage and sanitation of all adults in most developing countries, is improvements. In cities such as Karachi and São among the world's most dangerous occupations. Paulo, community groups have significantly accel- Not only do agricultural workers suffer injuries, erated the provision of low-cost water supply and but they are also exposed to disease-carrying ani- sanitation services to poor households, as well as mals and to poisonous agrochemicals. Health risks helping to maintain and manage local services. are high in other sectors as well. Miners, construc- Table 4.6 Estimated global burden of disease from selected environmental threats, 1990, and potential worldwide reductions through environmental interventions Reduction achievable Burden averted by Burden from these through feasible feasible interventions Burden averted per Type of environment and diseases (millions of interventionsb (millions of DALYs 1,000 population principal related diseases' DALYs per year) (percent) per year) (DALYs per year) Occupational 318 - 36 7.1 Cancers 79 5 4 0.8 Neuropsychiatric 93 5 5 0.9 Chronic respiratory 47 5 2 0.5 Musculoskeletal 18 50 9 1.8 Unintentional injury 81r 20 16 3.1 Urban air 170 - 8 1.7 Respiratory infections 123 5 6 1.2 Chronic respiratory 47 5 2 0.5 Road transport (motor vehicle injuries) 32 20 6 1.2 Alltheabove 473' - 50 10.0 The diseases shown are those for which there is substantial evidence of a relationship with the particular environment and which are listed in Appendix B. Estimates derived from the product of the efficacy of the interventions and the proportion of the global burden of disease that occurs among the exposed. All estimates of efficacy are speculative and assume the implementation of known, feasible, and affordable interventions in the circumstances encountered in developing countries. Computed by subtracting motor vehicle injuries (32 million DALY5) from all unintentional injuries (113 million DALY5). Adjusted for double counting. Source: Appendix tables B.2 and B.3 and authors' calculations. 95 tion workers, migrant workers, and child laborers able. Under the assumption that achievable reduc- all suffer increased risk of disease because of their tions in urban air pollution can prevent 5 percent occupations. Small workplaces may have espe- of all infectious and chronic respiratory disease, cially low standards of safety, yet such risks are these reductions could avert a burden of 8 million often overlooked by government agencies and DALYs each year, or 0.6 percent of the global bur- trade unions alike. A survey of companies in den of disease (Table 4.6). Local impacts and the Samud Prakhan, Thailand, found that smaller effects on especially vulnerable groups can be plants, with fewer than fifty workers, had substan- much greater (Box 4.7). tially lower levels of sanitation, health services, Lead poisons many systems in the body and is safety provisions, and environmental control mea- particularly dangerous to children's developing sures than larger enterprises. Workers suffered brains and nervous systems. Airborne lead con- more than twice as much noise and a third more centrations are high in polluted urban environ- lead fumes and vapors. And they experienced sig- ments, where lead comes mainly from the exhaust nificant work-related health problems: 22 percent of vehicles burning leaded gasoline. Elevated lead had lead poisoning or absorption, 27 percent had levels in children have been associated with im- upper respiratory symptoms, and 6 percent had paired neuropsychologic development as mea- chronic obstructive pulmonary disease, even sured by loss of IQ, poor school performance, and though most workers were below age 30. behavioral difficulties. Alleviating occupational risk depends on safety education for workers and managers, use of ap- WATER POLLUTION. Newly industrialized coun- propriate equipment and technology, and sound tries, as well as many industrial countries, have management practices. Governments can encour- polluted or are polluting their rivers, lakes, and age these initiatives through legislation and regu- coastal waters with a variety of chemical and bio- lation, financial incentives, investment in educa- logical wastes of both industrial and domestic ori- tion, and research and development. Where gin. The practice of letting raw wastewater from worker organizations are strong, they have played industry and residential areas flow into rivers or a major role in identifying and reducing occupa- the sea is common but unwise. Investment in pre- tional risks. Tripartite agreements between venting it may be justified because of the possibly workers, employers, and governments can lead to severe local health consequences (as illustrated in speedy progress. Box 4.8) and because generalized water pollution, The ambient environment by reducing the number of water sources available for domestic supply, can foreclose cost-effective Radiation and pollution of air and water are addi- options for responding to demand for domestic tional health hazards. Since there is no market for services. clean air and water, government action is fre- quently justified. RADIATION. Individuals are exposed to natural background ionizing radiation and to radiation AIR POLLUTION. Many cities suffer from air p01- used for medical and dental diagnosis. Only a tiny lution caused by industry, power plants, road amount of additional radiation comes from safely transport, and domestic use of coal. About 1.3 bil- operated nuclear power stations or other installa- lion urban residents worldwide are exposed to air tions (roughly one-thousandth of the background pollution levels above recommended limits. Air dose for those living within 50 kilometers of a nu- quality in the established market economies has clear power station). Current evidence suggests generally improved in the past two decades. But in that the health effects of this radiation on the gen- many developing countries and in the formerly so- eral population are extremely small or nonexistent. cialist economies, air quality has deteriorated be- Accidents and occupational risks to workers in nu- cause of rising industrial activity, increasing power clear industries and to miners of radioactive ores, generation, and the congestion of streets with however, are different matters. The consequences poorly maintained motor vehicles that use leaded of the nuclear power plant accident at Chernobyl, fuel. Ukraine, in 1986 have yet to be fully documented Air pollution damages the human respiratory but are undoubtedly large. (The risk of such acci- and cardiorespiratory systems in various ways. dents is particularly high in the formerly socialist The elderly, children, smokers, and those with economies because of their large number of poorly chronic respiratory difficulties are most vulner- designed nuclear facilities.) Standards and safe- 96 Box 4.7 Air pollution and health in Central Europe Contrary to expectations, public ownership and cen- mortality of 1 percent for every 10 micrograms per cu- trally planned economies have neither controlled poi- bic meter of particulates and sulfur dioxide. These esti- lution nor brought health benefits to the populations of mates suggest air pollution causes up to 3 percent of the formerly socialist economies. The countries of this total mortality in the Czech Republic and is responsible region face a variety of serious environmental health for roughly 9 percent in the gap in mortality rates be- threats, of which the greatest are particulates and gases tween the Czech Republic and Western Europe. Simi- in air, lead in air and soil, and nitrates and metals in lar estimates have been obtained for Silesia in Poland. water. A substantial gap in health status between these The effect of air pollution on mortality is greater for countries and those of Western Europe has opened up certain causes of death in specific age groups. A recent since the early 1960s: life expectancy is roughly five study of postneonatal respiratory mortality showed a years shorter in the formerly socialist economies, and rate 2.4 times higher in the most polluted districts of mortality rates in middle-aged males are roughly the Czech Republic than in the least polluted, after double. There has been considerable speculation adjusting for a battery of socioeconomic factors. An among scientists and the public in Central Europe increase in particulates of 25 micrograms per cubic me- about how much of this health gap is attributable to ter was associated with an increase in postneonatal res- environmental pollution. piratory mortality of 58 percent. Air pollution is the environmental factor that has had The contribution of air pollution to morbidity in the the greatest negative effect on health in Central Eu- Czech Republic is likely to be considerably greater than rope. Of the many air pollution 'hotspots" through- the effect on mortality and to have larger economic out the region, the worst-affected area is the 'Black consequences through health expenditures, lost Triangle," which covers northern Bohemia and Mo- schooling, and lost productivity. Children in heavily ravia, Silesia, and Saxony and has a population of polluted areas may suffer twice the rates of respiratory roughly 6.5 million. In August 1991 the three govern- morbidity of those in clean areas. Overall, air pollution ments involvedthe Czech Republic, Germany, and may be responsible for up to one-quarter of all respira- Polandand the Commission of the European Com- tory morbidity in Czech children. munities formed a Working Group for Neighbourly "Hotspots" of lead exposure exist throughout the Cooperation on Environmental Issues to deal with the formerly socialist economies. Average blood levels of extremely high levels of air pollutants in the area. more than 25 micrograms per deciliter in children have The overall effect of air pollution on mortality in the been reported in, for example, Pribram, Czech Repub- I Czech Republic has been estimated using data on the lic, and Katowice, Poland. In comparison with normal distribution of the population, the ambient levels of levels, these higher levels could double the proportion particulates and sulfur dioxide, and the relationship of children requiring special education and halve the between excess mortality and pollution. This relation- proportion in the exceptionally gifted group (IQ greater ship is derived from studies in Canada, the United than 130). Kingdom, and the United States that indicate excess guards against accidents and occupational hazards appropriate coping strategies) and healthier. The have been greatly improved, but risks may re- best preparation at the national level for these un- main, and continued research and vigilance are certain future events is therefore to pursue sound required. Putative links of certain cancers with ex- economic and health policies in the medium term. posure to radon in houses and with electromagne- tic fields created by high-voltage cables are being IMPROVING THE AMBIENT ENVIRONMENT. Improv- investigated in several industrial countries. ing health is only one of several reasons why soci- eties may choose to invest in a cleaner environ- GLOBAL THREATS. Depletion of the atmospheric ment. The policies and actions needed to clean up ozone layer and global warming pose potential the air in a given city or area will depend on the threats of unknown magnitude to health. Internã- origins of the pollution at that site. In most cities in tional agreements are limiting or will limit the re- developing countries motor vehicles are a signifi- lease of chlorine compounds that can harm the cant source of air pollution and need to be specifi- ozone layer and of the greenhouse gases that con- cally targeted. A few cities in the developing tribute to global warming. The societies that will world, among them Bangkok and Mexico City, are suffer least from these global changes are those pursuing systematic policies to reduce motor vehi- that are wealthier (and therefore able to invest in cle emissions, and their experience will be valuable 97 Box 4.8 Pollution in Japan: prevention would have been better and cheaper than cure In the 1950s and 1960s Japan experienced a period of in 1956 patients with a severe neurological affliction, rapid industrialization and economic growth, hut little later to be called Minamata disease, were observed. In attention was paid to the environmental consequences. 1968, following extensive research, the disease was The result was high levels of pollutants in the air, wa- linked to the ingestion of seafood containing high con- ter, and soil in some areas and several infamous out- centrations of methyl mercury, a compound dis- breaks of disease. Strong corrective action was taken in charged into Minamata Bay by the Chisso Corporation the 1970s and 1980s to redress the severest problems. as a by-product of the manufacture of acetaldehyde. Three conclusions emerge from the examples given be- The discharge of methyl mercury peaked in 1959; it low: allowing the release of toxic substances into the ended in 1968 when the company ceased production of environment can lead to serious health consequences acetaldehyde, but by then the floor of the bay and its and economic losses; prevention, as Japan is now do- aquatic life had become heavily contaminated. Starting ing, is less costly than cleaning up; and taking correc- in 1974, 1.5 million cubic meters of polluted sediment tive action now is less costly than allowing problems to were dredged and removed. persist. By 1991, 2,248 people (1,004 of whom had died) had been certified as suffering from Minamata disease and Case 1: sulfur dioxide in the air were eligible for compensation. An additional 2,000 people are pursuing claims for compensation. Had dis- Between 1956 and 1973 one of Japan's largest petro- charge of mercury continued, the estimated annual chemical complexes was constructed at Yokkaichi City. costs of the damage, including patient treatment and By 1960 air pollution was causing local concern, and by compensation, sediment dredging, and losses to fish- 1963 one-hour average sulfur dioxide levels exceeded eries, would have been $97 million a year. If 2,800 micrograms per cubic meter, far above WHO's acetaldehyde production had continued, pollution suggested maximum of 350 micrograms per cubic me- abatement through in-plant waste recycling would ter. In 1967 local residents successfully sued six com- have cost only $1 million a year. panies, claiming medical costs and compensation for lost income. Seven percent of the total population of Case 3: cadmium in the soil the district were certified to have been medically af- fected by ambient air pollution. Increasingly stringent In the late 1940s a disease characterized by extreme pollution measures were introduced starting in 1970, generalized pain, kidney damage, and loss of bone and by 1976 sulfur dioxide levels were in compliance strength appeared in the Jinzu River Basin. The dis- with local standards. ease, which primarily afflicted women, was called itai- Air pollution control costs since 1971including itai (''It hurts, it hurts!'') after the cries of the sufferers. technical installations and their operation, monitoring, Two decades of research led, in 1968, to the conclusion and creation of environmental buffer zoneshave been that the cause was chronic cadmium poisoning, which $114 million a year. Without this investment, however, was traced to the effluent from the Mitsui Mining and medical expenses and compensation would have been Smelting Company located in the upper reaches of the more than $160 million a year. basin. The route for the cadmium poisoning was from river water to irrigation water to soil to rice. By 1991, Case 2: mercury in the water 129 people had been certified as itai-itai sufferers, and 116 of them had died. At the turn of the century Minamata was a scenic A major program of soil restoration was initiated in coastal town of 12,000 people who made their living 1979. By 1992, 36 percent of the contaminated area of from wood products, oranges, and fish. In 1908 a fertil- 1,500 hectares had been treated. Had the further re- izer plant was established that eventually became the lease of cadmium not been prevented, the annual costs Chisso Corporation, one of Japan's largest manufac- from medical compensation, agricultural losses, and turers of chemicals. By the 1920s compensation for soil restoration would have been $19 million a year. The damage to fisheries had already become an issue, and costs of prevention were $5 million a year. in designing the next generation of programs. Suc- lead levels in gasoline and are using price differen- cessful policies include incentives and regulations tials to encourage consumers to switch to lead-free to improve fuel quality, enhance engine perfor- products. Lead concentrations in the air have mance and maintenance, and reduce traffic vol- fallen by 50 percent or more in response to these ume. Most industrial countries and an increasing measures, and average blood lead levels in urban number of developing countries have set limits on areas have also declined substantially. 98 Clean technologies and practices can reduce lo- the eradication of smallpox, and one great medical cal industrial pollution levels even as output ex- tragedy, AIDS. Unknown prior to 1981, AIDS now pands. To encourage adoption of such technolo- dominates public health programs and health ser- gies, governments need to pursue policies that vices in several countries and may come to domi- improve the efficiency with which energy is used. nate in many more. The human immunodeficiency Such policies include the elimination of subsidies virus (HIV) that causes AIDS is transmitted for power generation and, in many countries, for through sexual intercourse. Like other STDs, it can vehicle fuels and coal. Efficient reforms help re- also be transmitted by contact with contaminated duce pollution while raising a country's economic blood (notably from transfusions) and from output. Policy options are described in full in World mother to child during the perinatal period. Ca- Development Report 1992. sual transmission from person to person does not occur. In developing countries more than 85 per- The road transport environment cent of infections occur through heterosexual in- tercourse. There is no cure, and discovery of a Motor vehicle crashes are responsible for an in- vaccine is unlikely before 2000. Action is needed creasing burden of injury and death in developing now to combat the spread of the disease. countries. Each year throughout the world road traffic injuries cause a loss of 32 million DALYs, or Why AIDS is a special case more than 2 percent of the global burden of dis- ease (Table 4.6). Men suffer roughly twice the bur- AIDS deserves special attention because failure to den from road traffic injuries as women. The control the epidemic at an early stage will result in young and the old are particularly vulnerable, as far more damaging and costly consequences in the are drivers of nonmotorized vehicles and pedes- future. trians. The number of road fatalities and injuries in developing countries is rising rapidly with urban- The HIV epidemic is bad and is getting worse. An ization and growth in the volume of traffic. Road estimated 9 million people worldwide carried the fatalities in Africa increased fourfold between 1968 HIV virus in 1990; as many as 26 million could be and 1988, whereas in Europe they declined by infected by 2000, according to WHO estimates (see more than 20 percent during the same period. Table 1.3 in Chapter 1). AIDS will then contribute A multipronged approach to road safety can re- about 3.3 percent to the global burden of disease, duce crashes at reasonable cost. Public investment and 1.8 million people will die of AIDS each year. in improved road infrastructure and highway op- Given the short time it takes infection rates to eration systems, remedial action at known "black- double in many developing countries and the spots" with high accident rates, and expanded rapid spread of the disease to countries that previ- public transport systems all make a difference. ously had low numbers of infections, total figures Legislation, financial incentives, and programs of in 2000 may be two or three times higher than the road safety education can improve driver behav- above projections. ior, reduce traffic speeds, promote use of seat More than 80 percent of those infected lived in belts, improve vehicle safety, and reduce drunk developing countries in 1990; by 2000 this will in- driving. The insurance and legal liability systems crease to an estimated 95 percent. In Thailand one may also offer powerful incentives for road safety. in fifty adults is infected. In Sub-Saharan African A carefully designed package of measures such as one in forty adults is already infected, and in cer- those mentioned above can, over time, reduce tain cities of Africa the prevalence of infection is as road fatalities and injuries by at least one-fifth, high as one in three. In some of these high-preva- thus preventing the loss of at least 6 million lence communities AIDS is already starting to re- DALYs a year worldwide (Table 4.6). Several coun- verse long-term declines in child mortality. tries, including Kenya and Malaysia, have set The cost-effectiveness of interventions drops more ambitious targets for reductions in deaths sharply when infections cross from high-risk groups to and injuries over the next decade. the general population. Since there is no vaccine or cure for AIDS, primary prevention is the only way AIDS: a threat to development to fight the disease. In the absence of adequate preventive action, AIDS spreads rapidly in the Historians will look back on the latter half of this "core" groups (such as sex workers and their cli- century as having had one great medical triumph, ents), followed by a slower and then accelerating 99 Early intervention against AIDS prevents spread of mand for health care for AIDS patients will crowd the disease to the general population. out the needs of other patients. Furthermore, the number of tuberculosis cases is increasing dramati- cally as a direct result of HIV, and the presence of Figure 4.7 Simulated AIDS epidemic HIV worsens problems with other sexually trans- in a Sub-Saharan African country mitted diseases. (STDs both facilitate HIV trans- mission and are harder to treat in HIV-infected individuals.) AIDS cases per 1,000 sexually active population Prevention of AIDS involves sensitive and politi- 100 cally charged issues. Preventing HIV infection often necessitates working with socially marginalized groups (including, in many cultures, homosex- 80 uals), and with people who pursue illegal activities such as drug use or prostitution. In addition, an effective preventive program must reach out to in- 60 A form young people frankly about sexual practices and risks. These activities offer little political bene- 40 +-.- Spread into general population fit and may be highly controversial. Strong gov- ernment will and commitment are therefore essen- tial to effective programs, the more so because the seven-to-ten-year lag between HIV infection and 20 the development of AIDS makes it tempting for countries and individuals to put off dealing with Spread through AIDS issues until it is too late to avert a wide- core groups spread epidemic. 0 0 20 40 60 80 100 Prevention: an absolute necessity Time from start of epidemic (years) A combination of strategies, backed up with ade- Source: Adapted from Potts, Anderson, and Boily 1991. quate resources, is required for stemming the spread of AIDS. Crucial elements in these strate- gies are providing information on how to avoid infection, promoting condom use, treating other spread in the general population (Figure 4.7). sexually transmitted diseases, and reducing blood- Early and effective targeting of HIV interventions borne transmission. These measures are especially is critical because these interventions diminish in cost-effective when targeted at the relatively few cost-effectiveness as the infection moves out of the people in the core groups. Unless effective preven- high-risk, high-transmission core groups. The tive action is taken, the number of new HIV infec- large number of new sexual contacts in the core tions can be expected to grow, especially in parts groups means that each HIV case avoided in this of Asia. But a comprehensive AIDS prevention group can avert more than ten times as many addi- program could check the growth of the disease tional infections as can a case avoided in the gen- (Figure 4.8). eral population. Current annual worldwide expenditure on AIDS has catastrophically costly consequences. AIDS prevention is about $1.5 billion a year. Per- AIDS, affecting as it does mainly people in the haps less than $200 million of this is spent in devel- economically productive adult years, has powerful oping countries, where 85 percent of all infections negative economic effects on households, produc- occur. Among developing countries Thailand tive enterprises, and countries (see Box 1.2 in spends the most for AIDS prevention, with 1992 Chapter 1). Because so many of its victims are spending of $45 million, more than 75 percent of heads of households or parents, AIDS devastates which was from government funds. Total AIDS families. Heavily infected countries have found spending on prevention in all Sub-Saharan Africa their health systems burdened with costly cases of was only twice this amount, with a mere 10 per- AIDS-related opportunistic infections. If the AIDS cent from government funds. A recent study for epidemic continues unchecked, the accelerated de- WHO's Global Program on AIDS suggested that 100 comprehensive AIDS and STD prevention services Effective prevention can markedly slow the rate of for all developing countries would cost $1.5 billion new infection with HIV to $2.9 billion a year. This is ten to fifteen times current spending, but it would yield enormous benefits. The estimated number of new adult HIV Figure 4.8 Trends in new HIV infections infections averted by such spending between 1993 under alternative assumptions, 1990-2000: and 2000 would be about 9.5 million-4.2 million Sub-Saharan Africa and Asia in Africa, 4.2 million in Asia, and 1.1 million in Latin America. Sub-Saharan Africa Groups to be targeted New infections per year (millions) 3 Preventive efforts must reach populations with di- verse needs: people at particularly high risk of ac- quiring and transmitting HIV infection (core groups), young people, and women. Preventive programs for the population at large are less cost- case effective than targeted programs but are needed to increase awareness and understanding of AIDS and STDs, reduce discrimination against infected Optimistic case persons, and prepare the way for subsequent in- 0 terventions when levels of infection rise. Monog- 1990 1995 2001) amy might be encouraged as part of public infor- mation efforts to curb the spread of HIV, but it cannot be the only strategy; even where it is the Asia societal norm, not all individuals adhere. High-risk groups may include sex workers, mi- New infections per year (millions) grants, members of the military, truck drivers, and 4 drug users who share needles. For these groups, prevention of sexual transmission essentially Worstd means education on safer sex, promotion of con- dom use, and prevention and treatment of STDs. 2 It is important not simply to provide information on condoms but also to ensure their availability and to empower members of the core group, espe- I cially female sex workers, to use them. Brothel managers and clients must also be persuaded of the need to change their behavior; experience from 1990 1995 2000 Zaire and other countries shows that promotion of condoms to male clients substantially improves the success of programs targeted at sex workers. Areas of high STD prevalence warrant aggressive Note: Asia includes China, India, and Other Asia and attempts to control STDs through condom promo- islands. Source: World Health Organization and World Bank data. tion, case management and counseling, notifica- tion of partners, and surveillance. These can be provided through a wide spectrum of health insti- tutions such as family planning clinics and pri- mary health centers. particularly important because men so often domi- Young people, both in and out of school, need nate the sexual relationship. The curriculum comprehensive education on reproduction and re- should be sensitive to local cultural conditions but productive health issues. To be most effective, ed- should provide explicit, honest explanations of ucation must begin before the onset of sexual ac- sexuality, gender issues, safe sexual practices, tivity (ages 12-14 in many countries) and must be STDs and HIV, safe motherhood, and family plan- targeted at boys as well as girls. Reaching boys is ning. All potential behavioral choices, including 101 abstinence and condom use, should be presented. PROVIDING INFORMATION. Informing people There is no evidence to support the objection that about the steps they can take to protect themselves providing sex education encourages promiscuity. against HIV infection is central to any strategy for In societies in which it is unacceptable for teachers combating AIDS. Individuals need to know that to provide sex education, the task can be delegated the risk of infection can be minimized by reducing to qualified voluntary groups. the number of new sexual partners they have, by Women are biologically more susceptible to ac- choosing partners of lowest risk, by avoiding con- quiring HIV infection through heterosexual inter- tact with contaminated blood, by using condoms course than are men, and social factors often add and refraining from risky sexual practices such as to the risks. In Uganda, for example, more than 60 anal sex, and by avoiding or seeking treatment for percent of infected persons are women. Preventive cofactors such as STDs. Intravenous drug users efforts addressed to women, especially those of can lower their risks by using clean needles. In childbearing age, can protect both maternal and communities where the HIV virus is present, peo- child health. In many African countries AIDS and ple should be aware that unprotected sex is safe HIV-related illnesses are already among the top only with a person known unequivocally to be un- ten causes of childhood mortality. At present there infectedfor example, someone who has not yet is no way to prevent HIV transmission from an been sexually active and has no other risk factors infected woman to her fetus; about 30 percent of (such as intravenous drug use or transfusion) or the babies of infected women are born with the who has recently undergone HIV testing and has infection. Most such babies survive their first year been found to be uninfected. but succumb to opportunistic infections during their second or third year. The uninfected children ENCOURAGING CONDOM USE. Condom use is ef- of infected mothers are also at increased risk of fective in slowing the spread of both HIV and dying because they are likely to lose one or both of STDs and needs to be encouraged in all risky sex- their parents. The only strategy for fighting child- ual encounters. Programs to promote condom use hood AIDS is to target preventive efforts to in highly vulnerable groups such as clients of sex women of childbearing age. workers are cost-effective. One such program tar- Babies can contract HIV through breast milk, geted to low-income sex workers in Nairobi re- creating difficult tradeoffs between the risk of in- duced the mean annual incidence of gonorrhea fection and the benefits of breastfeeding for child from 2.8 episodes per woman in 1986 to 0.7 epi- health. Recent studies suggest that the risk is sub- sode in 1989. The program averted an estimated stantial (about 30 percent) for babies breastfed by 6,000 to 10,000 new HIV infections a year at an mothers who develop an HIV infection after child- approximate cost of $0.50 per DALY gained. Com- birth; the risk for babies of women who are infec- munity-wide interventions are also being tried ted prenatally is smaller, although still significant. with success. In Zimbabwe a community interven- Randomized controlled studies are under way in tion estimated to cost $85,000 successfully reached Haiti, Kenya, and Rwanda to determine the risks more than 1 million persons, distributed more more accurately. In the meantime, in areas where than 5.7 million condoms, and reduced STDs in the primary causes of infant deaths are infectious the general population by 6 to 50 percent in differ- diseases and malnutrition, breastfeeding should ent areas. The intervention also changed behavior probably continue to be recommended. In areas among sex workers: the proportion reporting con- where a safe alternative to breastfeeding exists, dom use with their last client rose to 72 percent, testing of pregnant women would provide an op- from only 18 percent before the intervention. portunity to advise those infected about the health Social marketing is another strategy for promot- risk of breastfeeding for their babies. ing condom use. In Zaire distribution outlets from pharmacies to traditional healers and from Specific preventive interventions nightclubs to street vendorswere saturated with condoms. Condom sales rose from 20,000 in 1987 Widespread experience with national AIDS con- to 18.3 million in 1991. Consumer research indi- trol programs in industrial and developing coun- cates that 90 percent of the condoms were bought tries is already on hand. It suggests some areas in by men and that about 60 percent were intended which action needs to be taken and provides im- for casual sex. Estimates suggest that the program portant lessons for programs to control AIDS. averted about 25,000 HIV infections in 1991 alone. 102 REDUCING BLOOD-BORNE TRANSMISSION. Blood ning clinics has revealed infection rates as high as transfusions account for less than 5 percent of HIV 20 percent. transmission worldwide, but transfusion with in- fected blood almost always leads to infection. Mea- TESTING AND SCREENING. Voluntary HIV testing sures for preventing transmission of HIV through provides individuals with useful information blood transfusions include reducing the need for about themselves and their partners. Studies sug- transfusions, eliminating payments for donated gest that counseling and testing can help individ- blood (because paid donors tend to have a higher uals and couples adopt safer sexual behavior. The risk of HIV), and screening donors. Effective early once-prohibitive cost of testing has been declining treatment of health problems, combined with edu- sharply; a couple can now be screened for less cation for health care providers about the proper than $2 (excluding the cost of counseling). The indications for transfusion, can cut the need for test, however, is not always reliable because there transfusions by more than 50 percent. Public is a short period during which HIV may not be health programs such as helminth control in detected in a newly infected person; furthermore, schools, iron supplementation, prenatal care, and a negative test result is no guarantee of continuing malaria control can reduce the severity of existing risk-free behavior. As a result, testing is currently anemia and thus diminish the need for transfu- most useful for couples within or planning a long- sions. When transfusion is unavoidable, blood do- term relationship. Governments need to ensure nors can be screened to ensure a supply of unin- that testing remains voluntary and anonymous, fected blood. The cost-effectiveness of blood meets quality standards, and is accompanied by screening varies dramatically depending on the appropriate counseling or information. prevalence of HIV. To maintain the population's overall confidence in the medical community, SURVEILLANCE. Public health surveillance for however, blood screening has been advocated HIV is critical in areas where extensive spread of even where HIV prevalence is low. When blood the virus has not yet occurred. Countries that es- banks exist, screening adds only about 5 percent to tablish a timely and reliable system of surveillance the total cost of each unit of transfused blood. are able to give policymakers early warning of an impending spread of the virus. Useful activities INTEGRATING AIDS PREVENTION AND STD SER- include regular surveillance for HIV and syphilis VICES. Wide availability of STD services is crucial in a few prenatal clinics and in centers serving cli- for fighting AIDS because HIV transmission is fa- ents at high risk of infection, such as sex workers cilitated by the genital lesions and inflammation and patients with STDs. associated with STDs. Treatment of STDs is often Several of the preventive interventions dis- highly cost-effective in its own right. It becomes cussed above have been incorporated in the Indian even more cost-effective when the benefits of re- government's National AIDS Control Project, duced HIV transmission are added. Curing each launched in 1992 with the assistance of the World case of gonorrhea in a core group saves 120 Bank and WHO. The Indian program emphasizes DALYs, at a cost well below $1 per DALY gained if promoting public awareness about AIDS, promot- the benefits of fewer secondary cases and reduced ing health in core groups through NGOs, control- risk of HIV transmission are included. ling STDs, improving the safety of blood supplies, Because many STDs are asymptomatic (espe- strengthening surveillance and institutional capa- cially in women), infected individuals may have bilities for control of HIV-AIDS, and encouraging little motivation to be tested and treated. Even for the humane treatment of people with AIDS or HIV those with symptoms, charges for clinical services infections. may reduce access to treatment and therefore in- crease the spread of STDs. In Nairobi, for exam- Care of AIDS patients ple, the introduction of fees at the main STD clinic reduced attendance by 60 percent among men and In 1992 developing countries spent about $340 mil- 35 percent among women. Subsidizing STD ser- lion to care for AIDS patients. Although this is vices therefore makes sense as part of an AIDS only a small fraction of the $4.7 billion spent by control program. Another good strategy is to com- industrial countries to care for their AIDS patients, bine STD and family planning services; screening it is still nearly twice the amount spent on AIDS for asymptomatic STD infections in family plan- prevention in the developing world. If spending 103 per patient remains constant, the amount spent on tial sum. Palliative home care, by contrast, costs the care of AIDS patients in developing countries between $30 and $75 per DALY gained, but it may will more than triple, to $1.1 billion in 2000. To be a substantial burden to the family. Uganda's date, many AIDS control programs have not de- innovative activities have made it possible to pro- veloped guidelines for the cost-effective provision vide caring responses, at modest cost, to those af- of care for AIDS patients. Antiviral drugs such as flicted (Box 4.9). azidothymidine (AZT) are enormously expensive, have severe side effects, and may, at best, delay The need for national and international action the onset of AIDS and prolong life to some extent. One year of AZT costs more than $3,000, a prohib- At present, most national AIDS programs are in- itively high figure. Treatment options in many adequate, despite international attention and the low-income nations are therefore limited to allevia- significant effort by WHO to help design and im- tion of pain and management of the opportunistic plement plans for controlling AIDS. Most pro- infectionsmost commonly, tuberculosis, diar- grams use only the resources available to minis- rhea, and candidiasisthat afflict HIV-infected tries of health, are too standardized, and neglect persons. Strategic planning can greatly reduce the control of STDs. AIDS has to be approached as costs through the use of a small number of less- a national development issue. National leadership expensive drugs and outpatient or community is crucial; the most effective programs, such as treatment where possible. Basic care, including Thailand's, pursue strategies that involve many outpatient treatment of opportunistic infections, agencies, in and outside government, in an atmo- can cost $200 to $400 per DALY gained, a substan- sphere of openness and frankness (Box 4.10). Box 4.9 Coping with AIDS in Uganda By June 1992 Uganda had reported 33,971 AIDS cases; TASO has grown to include ninety-seven counselors, the true number may be between 100,000 and 300,000, three supervisors, and six trainers in eight locations. and it is estimated that I million to 1.5 million Ugan- Services, which reach more than 30,000 people a year, dans are infected with HIV. In Kampala more than 30 include counseling, condom education and distribu- percent of all pregnant women are infected, and in tion, home care, income-generating activities, feeding many parts of the country AIDS is the most common programs, and payment of orphans' school fees. cause of admission and death among hospitalized In 1990, to address the demand for personal testing, adults. With this immense burden, care of infected in- Uganda's first anonymous HIV testing and counseling dividuals and management of the social consequences center was established. The enormous demand has of infection are perceived to be as important as preven- made individual pretest counseling impossible, but tion of further cases of HIV. group counseling has become popular. Individual post- In response, a variety of innovative activities have test counseling continues to be offered, and HIV-posi- been undertaken. In 1987 the first AIDS clinic was tive patients are referred to TASO for further support. opened, with a small staff, a few drugs, and little out- AIDS awareness in Uganda is so high that many peo- side support. The clinic recently enrolled its 8,000th ple assume they are infected. Couples who are tested patient. Patients regard the care they get there as much and found to be negative report they are more moti- higher in quality than that available elsewhere. The vated to be monogamous, and a small follow-up study founder of the clinic, Dr. Ely Katabira, and another found that such clients have fewer casual sex partners physician at the national teaching hospital have pro- and use condoms regularly. Additional centers have duced a 104-page manual on AIDS care that recom- been established in other areas, as well as an executive mends simple diagnostic and treatment strategies for testing center for businessmen and parliamentarians AIDS; for example, nine relatively inexpensive drugs uncomfortable about being served in the busy public used in combination with tuberculosis therapy can clinic. High demand indicates that Ugandans want to achieve a high degree of relief for patients with AIDS. know whether they are infected, particularly before Also in 1987 sixteen Ugandans who were personally embarking on important life events such as marriage. affected by AIDS (because of their own infection or that Uganda's experience demonstrates that an AIDS-test- of a family member) set up a new voluntary organiza- ing program in a country with a high prevalence of tion, The AIDS Support Organization (TASO), to pro- heterosexual transmission can have a more positive in- vide emotional support for AIDS sufferers. Twelve of fluence on behavior than results from the industrial the founding members have since died of AIDS, but world would indicate. 104 Box 4.10 HIV in Thailand: from disaster toward containment As late as 1988 Thailand and the test of Asia were con- dination Bureau in the office of the permanent secre- sidered to be relatively free of HIV infection, leading tary of the prime minister. The multisectoral bureau senior Thai health experts to conjecture that Asians coordinates the planning and budgeting of AIDS activ- might be less susceptible to the disease. That year, ities among fourteen ministries, international funding however, an explosive HIV epidemic started its march agencies, and local sources of support. The bureau also through Thailand, affecting all levels of society. Today facilitates the planning of joint activities with private it is estimated that 2 percent of sexually active adults, businesses and NGOs. or 400,000 to 600,000 people, are infected. Without ef- Thailand's strategy has led to a broad consensus fective prevention, by 2000 the number infected may be within the country on the importance of taking action. as high as 2 million to 4 million. Spending for AIDS prevention was $28 million in 1991 Faced with the HIV epidemic, Thai officials have and $45 million in 1992. To monitor the epidemic, Thai- moved quickly from complacency to action. Thailand, land has established the world's most comprehensive they realized, could not sustain its 10 percent annual national HIV surveillance system, which reports twice growth of GNP in the presence of a huge AIDS epi- a year on HIV prevalence in all risk groups in all pro- demic. Indeed, in 1991 researchers projected that the vinces of the country. Acknowledging that commercial aggregate direct and indirect cost of AIDS could be as sex is ingrained in Thai society and will remain so in high as $8 billion over the next decade and that AIDS the short run, the government has decided to mandate could have negative effects on tourism, foreign invest- and enforce a policy of 100 percent condom use at the ment, and labor remittances from abroad. They ar- brothels. This ensures that brothels cannot compete for gued, however, that a major preventive effort, with the customers seeking condom-free sex. Preliminary evi- goals of reducing numbers of sexual partners by at least dence shows very high rates of condom use, with de- one-half, doubling condom use, and treating STDs, mand increasing from 10 million a year to about 120 could mean 3.5 million fewer infections and more than million a year and reductions in the incidence of STDs. $5 billion in savings by 2000. The prime minister's office is launching national AIDS prevention is now being accorded the highest campaigns through the mass media in 1993-94 to pro- priority in Thailand, and a national AIDS prevention mote changes in the sexual culture and the sexual and control committee chaired by the prime minister norms of the population. Only time will determine has been formed. In 1992 the cabinet approved the es- whether intervention has been prompt and effective tablishment of the AIDS Policy and Planning Coor- enough to halt the further spread of HIV No single strategy in the fight against AIDS will Areas with an HIV epidemic but as yet little meet the needs of every country. Three main crite- disease (for example, Thailand, and urban areas of ria can guide the choice of priorities from the range India) need to develop AIDS prevention programs of HIV-AIDS interventions listed above. These cri- for the entire population while continuing to target teria are current HIV prevalence, risk of future high-risk groups. Voluntary HIV testing and coun- spread, and existing AIDS burden. Strategies for seling and preparation for the care of AIDS pa- different countries and regions within countries tients should also begin. fall into four main groups. Finally, areas with a major epidemic and a Areas with little HIV and few STDs (for ex- high disease burden (for example, Uganda and ample, rural areas of northern China and North Zambia) have to combine a broadly based preven- Africa) should emphasize comprehensive repro- tive strategy with attention to care for AIDS pa- ductive health education for youth, with some at- tients (see Box 4.9). tention to AIDS prevention among high-risk Nongovernmental organizations can play a vital groups, and should establish a sensitive HIV sur- role in prevention, care, and community support veillance system. programs, using their credibility and access to Areas at high risk of an epidemic from early reach those at highest risk, such as intravenous spread of HIV or having a high rate of STDs (for drug users and sex workers. Such groups have example, Yunnan Province in China and Sura- been highly effective in using social marketing to baya, Indonesia) should undertake massive, tar- reach individuals at the grass-roots level, particu- geted preventive activities for high-risk groups, in- larly by initiating peer education and media pro- cluding sex workers, supplemented by general grams that reinforce behavior change and work to education and by testing of the blood supply. modify the perceived social norms. A recent 105 Table 4.7 Costs and health benefits of public health packages in low- and middle-income countries, 1990 Annual cost (dollars) Disease burden Country groupand component of package averted Per participant Per capita Per DALY (percent) Low-income (income per capita = $350) EPI Plus 14.6 0.5 12-17 6.0 School health program 3.6 0.3 20-25 0.1 Other public health programs (including family planning, health, and nutrition information)b Tobacco and alcohol control program 2.4 0.3 1.4 - - 0.3 35-50 0.1' AIDS prevention program° 112.2 1.7 3-5 2.0 Total - 4.2 - 8.2 (1.2) Middle-income (income per capita = $2,500) EPI Plus 28.6 0.8 25-30 1.0 School health program 6.5 0.6 38-43 0.4 Other public health programs (including family planning, health, and nutrition information)" 5.2 3.1 Tobacco and alcohol control program 0.3 0.3 45-55 Q3d AIDS prevention programe 132.3 2.0 13-18 2.3 Total 6.8 (0.3) - 4.0 Note: Numbers in parentheses refer to per capita cost as a percentage of income per capita. Although costs are estimated for 100 percent coverage, the health benefits are based on 95 percent coverage for EN Plus and 80 percent coverage for the school health, AIDS prevention, and tobacco and alcohol programs. Includes information, communication, and education on selected risk factors and health behaviors, plus vector control and disease surveil- lance and monitoring. The health benefits from information and communication and from disease surveillance are counted in the other public and clinical services in the health package. The health benefits from vector control are unknown. Calculation of the potential disease burden averted through this program assumes no change in the prevalence of smoking and alcohol consumption; if such prevalence were to rise, the potential benefits would be larger. Exiudes treatment of STDs, which are in the clinical services package; see Table 5.3. Source: Authors' calculations. needs-assessment study conducted in a number of gains at modest cost. Local conditions vary, but an developing countries showed that the full poten- essential public health package is likely to include: tial of NGOs was not being realized for lack of The Expanded Programme on Immunization, financial, managerial, and technical support. Plan- including micronutrient supplementation ning is under way for a program to provide inno- School health programs to treat worm infec- vative mechanisms for simple and flexible assis- tions and micronutrient deficiencies and to pro- tance to nongovernmental groups working on vide health education AIDS. Programs to increase public knowledge about The world must do more to deal with the global family planning and nutrition, about self-cure or challenge of AIDS. No country is immune from a indications for seeking care, and about vector con- future HIV epidemic, and the costs of delay are trol and disease surveillance activities high. A global coalition is needed that will encour- Programs to reduce consumption of tobacco, age and assist governments to take bold action be- alcohol, and other drugs fore it is too late. Without a substantial increase in AIDS prevention programs with a strong STD political commitment and leadership-as well as component. additional resources to support the $1.5 billion to $2.9 billion needed annually for effective preven- This public health package would yield large bene- tion of AIDS-the HIV epidemic could cause a fits at low cost (Table 4.7). In low-income countries health disaster and an enormous setback for it would avert more than 8 percent of the burden development. of disease at a cost of just $4 per capita (1.2 percent The essential public health package of income per capita), while in middle-income countries it might avert 4 percent of the burden of Public health programs that address the problems disease at a cost of $7 per capita (0.3 percent of described above can produce substantial health income per capita). Because it is difficult to quan- 106 tify the health gains from the activities under services. These are included in the essential pack- "Other public health programs," the correspond- age of clinical measures discussed in the next ing cost per DALY is not estimated. chapter. Health will also be served if governments Provision of information is needed in every as- do less in a number of areasif they avoid inter- pect of the program. Information should cover the vening in food markets, cut indiscriminate sub- benefits of healthy eating, contraceptive use, and sidies for water and sanitation, remove most re- hygienic practices in the household; the health ef- strictions on contraceptive services, and abolish fects of smoking and of alcohol and drug abuse; subsidies on fuels. Appropriate government regu- and prevention of HIV infections. Some public latory action on the ambient environment, occupa- health measures will involve providing services in tional conditions, and road safety can also safe- clinics, including family planning and STD-related guard people's health. 107 Clinical services This chapter analyzes the roles of the public and Although both the public and the private sectors private sectors in paying for and delivering clinical have important roles in the delivery of clinical ser- services. It examines in depth an important con- vices, government-run health systems in many de- clusion of Chapter 3: that governments have a fun- veloping countries are overextended and need to damental responsibility for ensuring universal ac- be scaled back. This can be done through legal and cess to an essential package of clinical services, administrative changes designed to facilitate pri- with special attention to reaching the poor (Table vate (NGO and for-profit) involvement in provi- 5.1). The choice of services to be included in such a sion of health services, by public subsidies to package for each country will be strongly influ- NGOs for supplying the essential package, and by enced by information on the distribution of dis- curtailment of new investments in public tertiary ease and the relative cost-effectiveness of clinical hospitals. At the same time, the efficiency of pub- interventions. A minimum package of clinical ser- lic sector health services can be greatly enhanced vices could reduce the present burden of disease through decentralization and improved manage- by about one-quarter in low-income countries and ment of government hospitals and programs. by about one-tenth in middle-income countries. This package is affordablebut only if govern- Public and private finance of clinical services ments carry out significant health-financing re- forms that will affect the allocation of public funds Around the world, clinical services are financed and the roles of insurance and of user charges. through four main channels. Twoout-of-pocket Only by reducing or eliminating spending on payments and voluntary insuranceare private. clinical services that are outside the nationally de- The other two are public: compulsory insurance fined essential package can governments concen- (sometimes known as social insurance) that is trate on ensuring essential clinical care for the either publicly managed or heavily regulated by poor. Two key ways to reallocate government governments, and funding from general govern- spending are to increase cost recovery, especially ment revenues. by charging the wealthy for services in govern- In the poorest countries total health expenditure ment hospitals, and to promote unsubsidized in- may be as low as $2 per person a year, and more surance for middle- and upper-income groups. than half of this comes from private sources, Governments can avoid the explosive increases in mainly in the form of out-of-pocket payments. In- health expenditures that many countries are now surance mechanisms in those countries are weak, confronting by encouraging competition among and the amount of government revenues devoted providers and prepayment for care, generating to health is low. As incomes increase, so do both and disseminating information on providers' costs the percentage of income spent on health (as and insurers' products, and, in some cases, setting shown in the upper panel of Figure 5.1) and the limits on compensation of physicians and share of health spending that comes from public hospitals. sources (illustrated in the lower panel). In the for- 108 Table 5.1 Rationales and directions for government action in the finance and delivery of clinical services Conditions that may call for government action: market failure Area and poverty Directions for government action Essential clinical services Failure to treat, for example, Finance essential clinical services by reallocating current tuberculosis and STDs government spending. In low-income countries this may creates risks for the general mean increasing public expenditures for health. population. Public Require through legislation that social insurance or financing can help offset mandated private insurance cover an essential package. the additional external Encourage more private and NGO provision of essential costs to society. Poor services, through appropriate legislation and targeted people have limited ability public subsidies. to save or borrow to meet unexpected and uninsured health expenses. Families, including children, can fall into poverty because of ill health. Clinical services outside the In insurance markets Reduce or eliminate subsidization of clinical services outside essential package selection bias leads to lack the essential package. Subsidies for public provision of of coverage for high-risk services at less than cost and tax relief for employer and groups. "Moral hazard," employee health insurance payments often cover services by insulating patient and with low cost-effectiveness and primarily benefit the provider from the cost wealthy. implications of their Legislate compulsory social insurance or mandated private decisions, results in insurance, or define the national essential package overuse of services. The comprehensively. asymmetry of information Limit government involvement in delivery of nonessential between patient and services and encourage competition in service delivery by provider can cause government, NGOs, and the private sector. suppliers to induce excess demand. Regulate private insurance by, for example, requiring community risk-rating and forbidding the rejection of high-risk consumers. Define the exact content of prepaid packages of care to serve as the products bought and sold in the insurance market. Encourage the use of prepayment or salary-based approaches to provider compensation. Foster improvements in the quality of private provision by encouraging self-regulation of hospitals, medical schools, and physicians and by disseminating performance indicators. merly socialist economies and the established mar- surance pays for much care and, if so, what is the ket economies (excluding the United States) public dominant type of insurance (Table 5.2). spending accounts for a full three-quarters of total In low-income countries private out-of-pocket pay- health expenditure. ments account for more than half of the mere $2 to In addition to the four sources of health financ- $40 per person spent each year for health care. ing, there are three ways of organizing clinical Most of this sum goes for doctors' fees, payments health services: public, private nonprofit, and pri- to traditional healers, and drugs. NGOs, particu- vate for-profit. All national health systems use at larly those related to religious institutions, make least two of the twelve possible combinations of important contributions to the provision of health financing method and health service organization, services in many low-income countries. In Tan- and sometimes the different combinations serve zania and Haiti NGOs operate nearly half of the sharply differentiated populations. Even so, it is hospitals, and in Cameroon and Uganda they possible to group countries according to income manage 40 percent of health facilities. In Ghana level and the predominant system of providing and Nigeria about a third of all hospital beds are health care. A principal distinction is whether in- located in mission hospitals. Government spend- 109 As countries get richer, the i spend more of their income on health, and the public share grows larger. Figure 5.1 Income and health spending in seventy countries, 1990 Share of GDP spent on health Percent 13 United States 11 France . . Canada 9 - Lesotho 0 0 . 0 7 0 - India 0 0 0 Korea, Rep. of 00 . o 0 Spain S 0 Norway0 0 0 0 Japan 0 0 0 Jamaica 0 United Kingdom 5 0 enya S 00 - 000 000 0Brazil o Cyprus 0 0 3 0 0 - 0 Egypt I 0 5,000 10,000 15,000 20,000 25,000 GDP per capita (1991 international dollars) Public share of total health spending Percent Norway. 0 90 0 0 Hungary 5 0 United Kingdom. 0 0 0 0 5 Japan 0 0 Spain Greece 5Can da 70 - Brazil 0 0 0 o Jamaica S 0 France 0 Cyprus 00 -' 0 00 0 50 0 0 S Israel 0 0 0 Unit d States Kenya S S 0 'Egypt . Korea, Rep. of Lesotho 0 0 30 0 'Philippines 0 0 Hong Kong 'India 0 S sUganda 10 0 5,000 10,000 15,000 20,000 25,000 GDP per capita (1991 international dollars) Source: Murray, Govindaraj, and Chellaraj, background paper. 110 Table 5.2 Clinical health systems by income group Health expenditure, 1990 As share Dollars Country group and 1990 per of GNP per Main capita income (dollars) (percent) capita characteristics Examples Low-income (100-600) 2-7 2-40 High private spending for traditional Bangladesh, India, medicine and for drugs Pakistan, most Public services financed from general Sub-Saharan revenues African Little insurance countries Middle-income (600-7,900) Private insurance 2-7 20-350 Government services for middle- and low- South Africa, income groups financed from general Zimbabwe revenues Private insurance and private provision for affluent (less than 10 percent of population) Social insurance 3-7 20-400 Public health and clinical care for low-income Costa Rica, Republic groups financed from general revenues of Korea, Turkey Social insurance for wage labor force, with mixed provision Formerly socialist 3-6 30-200 Public services (which are low in quality or Czech Republic, economies of Europe collapsing) financed from general revenues Poland, Slovak (650-6,000) Large underground market in privately Republic, republics provided services of former U.S.S.R. Established market 6-10 400-2,500 Universal or near-universal coverage through France, Germany, economies, excluding general revenue financing or compulsory Japan (social United States social insurance insurance); Norway, (5,000-34,000) Use of capped third-party payments and Sweden, United global budgets Kingdom (general tax revenues) United States (22,000) 12 2,800 Combination of private voluntary insurance United States and use of general revenue from taxes Unregulated and open-ended fee-for-service compensation High administrative costs associated with health provision and insurance a. Although China is a low-income country, its health system is closer to that of a middle-income country with social insurance. Source: For expenditure, Appendix table A.9. ing from general tax revenues generally amounts weakened considerably. The rural population in- to less than half of the 2 to 7 percent of GNP allo- creasingly relies on a system of government- cated to health services. There is little or no provided health care financed in part out of gen- insurance. eral revenues, but with substantial cost recovery Until recently, China was an important excep- through user charges, not unlike systems prevail- tion among low-income countries. There, between ing in other low-income countries. 1960 and 1980, state enterprises provided health In the middle-income countries there are two major care directly to their workers or contracted with types of health systems, distinguished by whether government hospitals to do so. Rural communes the government or the private sector provides were required to earmark a portion of their finan- health insurance. Health spending, at $20 to $400 cial resources for health services for all their mem- per capita, is higher than in low-income countries, bers. By the late 1970s insurance covered virtually and both public and private managerial capacity is all the urban population and 85 percent of the rural stronger. In countries with private insurance, such populationa unique achievement for a low- as South Africa and Zimbabwe, the government income developing country. Since the elimination uses general revenues to pay for health care for of communal agriculture and the liberalization of middle- and low-income groups, while upper- industry in the early 1980s, however, these forms income households (less than 20 percent of the of health insurance and service delivery have population) use private insurance to pay for pri- 111 vate physicians and hospitals or for private rooms of essential clinical services. But what should be in government hospitals. the content of this package? Although political In countries with social insurance, mandatory considerations will inevitably affect the decision, contributions from employees and employers, and the most important factors in selecting the essen- sometimes government funds, finance insurance tial package should be the relative cost-effective- for part of the population, including most middle- ness of interventions, the size and distribution of class workers. Health care for the poor is financed the health problems affecting the population, and from general revenues. This is the system that pre- the resources available. vails in Korea, Turkey, and most of Latin America. In the formerly socialist economies of Eastern Eu- Defining the essential package rope and the Soviet Union, general revenue financing with government provision of health A patient's health needs often require several in- services was until recently the only officially recog- terrelated interventions. A child with fever and di- nized form of health care. Public spending on arrhea may require treatment for both acute respi- health now accounts for 3 to 6 percent of GNP in ratory and gastrointestinal infections. A pregnant these countries, or $30 to $200 per capita. Prior to woman needs to receive both prenatal and deliv- the period of political and economic liberalization ery care. For this reason, it makes sense to group in the late 1980s, private payments were fre- certain interventions when analyzing their costs quently made for "public" health services (for ex- and benefits. Five groups, or clusters, of clinical ample, gratuities were given physicians in govern- interventions are likely to be important in every ment hospitals), and drugs often leaked from the country's essential clinical package: prenatal and public sector into private markets. Since the politi- delivery care; family planning services; manage- cal and economic reforms that swept across these ment of the sick child; treatment of tuberculosis; countries in the late 1980s, the health systems and case management of sexually transmitted dis- there have been in crisis. Dwindling public fund- eases (STD5). The first two groups are often dis- ing and deteriorating government services have cussed under the umbrella of "safe motherhood" created strong pressures for new forms of public activities, but for the purposes of cost and benefit and private insurance. estimates they are presented separately here. All The established market economies, with the excep- five sets of interventions are highly cost-effective; tion of the United States, rely on one of the two each costs $50 or less per DALY in low-income types of public financing for more than three-quar- settings and $150 or less per DALY in middle- ters of their health expenditures, which range income settings. Moreover, they deal with wide- from $400 to $2,500 per person per year. Norway, spread health problems that affect the poor. Sweden, and the United Kingdom use general tax Details on the health problems that these five revenues to pay for health services that are pro- groups of clinical interventions help to resolve, on vided directly by the government. In France, Ger- ways of delivering the interventions efficiently to many, and Japan, among others, social insurance patients, and on the cost-effectiveness of the inter- is the dominant mode of financing. The United ventions are given in Boxes 5.1 through 5.4. The States, with annual health spending of about problems addressed are among the largest afflict- $2,800 per capita, has a bewildering combination ing developing countries. Four preventable or eas- of systems, including voluntary private employ- ily treatable infectious diseases of children account ment-based insurance, compulsory insurance for for nearly 7 million child deaths annually. Unsafe federal workers with each employee having a childbirth is responsible for half a million maternal choice of alternative insurers and packages, and deaths each year. Tuberculosis kills more than 2 full public finance and provision for veterans. A million people annually, making it the leading single-payer approach financed from general reve- cause of death among adults. More than 250 mil- nue is used at the federal level for health care for lion new cases of debilitating and potentially fatal the elderly (Medicare) and at the state level for thd STDs occur each year. poor (Medicaid). In addition to these five groups of clinical inter- Selecting and financing the essential clinical ventions, in any realistic setting an essential pack- package age would have to include treatment of minor infection and trauma, as well as advice and allevia- A basic responsibility accepted by governments al- tion of pain for health problems that cannot be most everywhere is to ensure access to a package fully resolved with existing resources and technol- 112 Box 5.1 Making pregnancy and delivery safe Under optimal conditions, about 990 of every 1,000 others to danger signs that may occur during preg- pregnancies that reach the seventh month of gestation nancy and childbirth, and mobilize communities for conclude with a healthy newborn and a healthy transport of women with complications to district mother. For most women in the developing world, hospitals however, childbirth is unsafe. About one in 50 women Community-based obstetrics with trained nurse-mid- in developing countries dies as a consequence of com- wife staff to provide prenatal care, including tetanus plications of pregnancy and childbirth, compared with toxoid immunization, treatment for syphilis, provision only one in 2,700 in the established market economies. of micronutrients (iron, folate, and iodine), and detec- Maternal mortality has profound consequences within tion of complications of pregnancy and delivery; nor- the household; the chances of dying for children under mal delivery, including prophylactic application of an- 5 increase by up to 50 percent when the mother dies. tibiotics against gonorrheal ophthalmia; obstetric first In [987 the international health community, includ- aid, including sedatives for early eclampsia (preg- ing the World Bank, WHO, the United Nations Popula- nancy-related seizures) and manual removal of the pla- tion Fund (UNPF), and agencies in forty-five countries, centa; effective early referral of severe complications; launched the Safe Motherhood Initiative. The prime and safe abortion. goal is to reduce by half the number of maternal deaths District hospital facilities to provide essential obstet- by 2000. The health programs recommended under the ric services (cesarean section, anesthesia, blood re- initiative include family planning and pregnancy- placement, manual procedures, and monitoring of la- related care, prenatal care, and delivery care. (Family bor) and neonatal resuscitation (aspiration of secretions planning and abortion services are discussed more and assisted respiration with oxygen). fully in Chapter 4.) The marginal cost-effectiveness of The emphasis given the different components will pregnancy-related care varies with circumstances, but depend on local conditions. At one extreme are dis- the World Bank has estimated that the average cost per tricts where resources are limited and women are DALY is between $30 and $110, the equivalent of less highly isolated. Here, high priorities would be prenatal than $2,000 per death averted. care aimed mainly at correcting micronutrient deficien- The extension of prenatal, delivery, and postpartum cies and infections such as STDs and malaria. At the care to 80 percent of the world's population would re- other extreme are urban and periurban areas where duce by 40 percent the burden of disease associated referral centers are overwhelmed with normal deliv- with unsafe childbirth, at a cost of between $90 and eries and the quality of care is typically low; here, $255 per birth attended, or $4 to $9 per capita. A rea- health centers should be improved so that they can sonable program of pregnancy-related care would in- deal with normal births, and the quality of hospital care clude three components: should be enhanced to provide better treatment of ob- Information, education, and communications designed stetric complications. to create demand for clinical services, alert women and ogies. Hospital capacity would be sufficient to trained staffto deliver essential services and that handle some emergency care, including most frac- inputs for services outside the nationally defined tures and infrequently needed procedures such as package are not supplied. A district hospital with appendectomies. Local discretion in the provision about one bed per 1,000 population served is of these services would depend on the availability needed to provide inpatient and specialized out- of inputs and on day-to-day capacity. This "lim- patient care, but the hospital would have to per- ited care" and the five groups of interventions to- form only basic surgery. No higher-level hospital gether constitute a minimum package of essential clin- is required for delivery of the minimum package. ical services. Efficient delivery of these essential Although doctors are needed for supervising es- services requires a well-functioning district health sential clinical care and handling more compli- system consisting of health posts and health cn- cated cases, most of the services in the minimum ters as the first point of patient contact and district package can be delivered by nurses and midwives. hospitals as referral facilities, with the two levels A ratio of fully qualified nurses to physicians of linked by emergency transport. between 2 and 4 to 1 (estimates vary to accommo- Governments must ensure that publicly pro- date the availability of physicians and nurses in vided facilities have the necessary inputsdrugs, different regions) and 0.1 to 0.2 physician per supplies, facilities, equipment, and properly 1,000 population would be adequate. Although 113 Box 5.2 Integrated management of the sick child Four groups of infectious diseasediarrhea! diseases, experience with these two disease clusters can be ex- acute respiratory infections (ARIs), measles, and ma- panded to include children with malaria, measles, and lanaaccount for more than half of the 12.7 million malnutrition. Evidence that malaria and pneumonia deaths every year of children under age 5. In the devel- overlap in their clinical presentation and can be treated oping world measles alone causes 860,000 deaths in with the same antibiotic strengthens the case for treat- children under age 5 and accounts for 6 percent of ing several diseases together. DALYs lost in that age group. Malaria causes 4 percent Under the integrated management approach, the of the disease burden in the under-five group. Sick sick child is initially assessed by means of a limited children taken by their mothers to health centers for range of questions and observation of easily recognized diarrhea! disease and for ARTs such as pneumonia of- symptoms. The child's nutritional and immunization ten receive inappropriate diagnosis and treatment, status is measured, and immunization is given if leading to unnecessary complications and deaths. needed. The child's condition is classified according to Whereas preventing diarrhea! diseases and ARIs has disease grouping and severity guidelines, which are proved difficult and is probably not cost-effective, case used as a basis for treatment and possible referrals. The management in community-based programs is feasible final step is to give the mother advice Ofl follow-up and extremely effective. WHO and UNICEF have re- care. cently begun to support national programs on the Inte- The core of the package is to train primary health grated Management of the Sick Child. This initiative care providers to diagnose diseases and prescribe the builds on more than fifteen years of experience with appropriate treatment at the health center level or refer case management of diarrheal diseases, mainly by oral immediately to a district hospital those cases with com- rehydration therapy (ORT), and about seven years of plications. An adequate supply of antibiotics, anti- research on and program implementation of case man- malarial drugs, and other drugs is critical for success. agement of ARIs. The integrated cluster of treatments, including hospital In Nepal a controlled intervention trial that relied services, would cost between $30 and $100 per DALY exclusively on indigenous community health workers saved. Since the walk-in component accounts for ap- (CHWs) to detect and treat pneumonia without hospi- proximately 60 to 70 percent of the reduction in the talization led to a 28 percent reduction in the risk of disease burden, district hospitals are not indispensable death from all causes by the third year of service. Addi- for starting the program, but their presence and proper tional benefits were obtained from the reduction in functioning add substantial health benefits. If high deaths caused by diarrhea and measles. Other research rates of health service use can be achieved, child deaths on similar community-based strategies for children un- in high-mortality communities, according to WHO esti- der age 5 indicates decreases of approximately 50 per- mates, could be reduced by between 50 and 70 percent. cent in infant mortality from ARTs. In Egypt the use of This fact and the relatively low technology involved ORT has in some areas led to a reduction of 50 percent make the management of the sick child a high priority in mortality from diarrhea and 40 percent in overall in countries with child mortality rates of more than mortality among children ages I month to 5 years. The thirty deaths per 1,000 children under age 5. many developing countries can already deliver the drugs) to reduce the risk of cardiovascular disease minimum package, some low-income nations in high-risk individuals, and inexpensive manage- would require additional investments in person- ment of angina and heart attacks. Other treat- nel, equipment, and facilities. ments that might fit into an expanded package in- In developing countries with the financial re- clude hernia repair, meningitis treatment for sources and political will to go beyond the mini- children, management of gastrointestinal ulcers, mum clinical package, a more comprehensive set cataract removal, and treatment of moderately se- of services could cover other interventions with vere injuries and of complications of diabetes. slightly lower cost-effectiveness than those in the Given the large contribution of disability to the minimum package. This set might include a num- burden of disease, inclusion of low-cost rehabilita- ber of interventions for chronic disease, such as tive measures will often be a priority for interven- use of oral hypoglycemics or insulin to control dia- tions beyond the minimum essential package. betes, medical treatment for schizophrenia and These interventionsmost of which respond to manic-depressive illness, screening and treatment conditions that will become increasingly common for breast and cervical cancer, measures (for exam- with the epidemiological transition documented in ple, use of aspirin and of simple antihypertensive Chapter 1can cost as little as $200 to $300 per 114 DALY. Including them and other interventions of essential package of services in their allocation of similar cost-effectiveness would reduce the current public spending on health. In Botswana and Zim- disease burden by 5 to 10 percent. babwe the rapid decline in infant mortality and Many health procedures have such low cost- rise in life expectancy during the 1980s were effectiveness that governments should exclude strongly influenced by government action to ex- them from the essential clinical package. In low- pand the health infrastructure and by the use of income countries these might include heart sur- general tax revenues to finance an array of public gery; treatment (other than pain relief) of highly health and clinical services. Some key services that fatal cancers of the lung, liver, and stomach; ex- were initially left out of the package but are now pensive drug therapies for HIV infection; and in- being incorporated include vitamin A supplemen- tensive care for severely premature babies. It is tation and improved control of STDs. Similarly, difficult to justify using government funds for dramatic health gains in Costa Rica in the 1970s these medical treatments when much more cost- were largely brought about by new basic public effective services that benefit mainly the poor are health and clinical services, financed almost en- not receiving adequate financing. tirely by the Ministry of Health and the national Several developing countries that have been social security agency. The remarkable improve- highly successful at improving the health status of ments in health status in China, Kerala State in their populations have emphasized access to an India, and Sri Lanka are attributable in part to gov- Box 5.3 Treatment of sexually transmitted diseases Sexually transmitted diseases (STDs) are extremely person affected is a member of a "core" group and common infections: according to a 1990 estimate by hence likely to transmit syphilis to others, and Ofl the WHO, there are more than 250 million new cases each case-detection strategy used, curing a case of syphilis year worldwide. These diseases have severe and often can cost as little as $0.10 or as much as $40 per DALY. In irreversible consequences that disproportionately af- Zambia a syphilis treatment demonstration project for fect women, who bear 80 percent of the total DALYs pregnant women achieved a two-thirds reduction in lost to STDs (excluding HIV). Women are more likely stillbirths, low birth weights, and neonatal deaths as- than men to acquire STDs because of the greater effi- sociated with syphilis, even though attendance, ciency of male-to-female transmission for most STD screening, and treatment were not optimal. The cost pathogens, the lack of female-controlled preventive was $12 per adverse outcome prevented, or less than methods, and, in many settings, gender power dy- $1 per DALY. In a similar program in Kenya prevention namics that limit women's ability to determine the con- of one neonatal death cost $50. Unfortunately, in most ditions under which sexual intercourse occurs. Women countries a comprehensive program of perinatal syph- are less likely than men to obtain care for STDs because ilis screening has not been implemented. the majority of infected women are asymptomatic and Effective interventions exist for other STDs, but lack those with symptoms may be deterred by fear of social of simple, appropriate, rapid, and inexpensive diag- stigma. Since STDs also inflict a heavy burden of ill- nostics for use in the field have made these measures ness on men, and since men are responsible for much much more difficult and expensive to implement. Con- STD transmission, disease control measures, to be ef- sequently, when resources are limited, such interven- fective, must be targeted to both sexes. tions must be targeted specifically to core groups to be Because STDs increase the efficacy of transmission of cost-effective. In these groups, case management of HIV, controlling these infections is one of the most im- chlamydia (a bacterial infection of the reproductive portant interventions for containing the spread of tract) and gonorrhea can be highly cost-effective ($10 to AIDS. But even in the absence of AIDS, STDs cause $40 per DALY), as can treatment of chancroid in areas substantial morbidity and mortality. They usually af- where it is common. Much of the cost of intervention fect people in the 15-44 age group, the most economi- for these diseases is related to diagnosis. Calculations cally productive ages. One of the main causes of .neo- suggest that if prevalence of an STD is more than 10 natal morbidity and mortality in some countries is percent, treating everyone in the risk group may be congenital syphilis. Because treatment is simple (one to more cost-effective than screening. This, however, three injections of penicillin) and inexpensive, screen- does not take into account the problems of widespread ing and treatment for syphilis during prenatal care has use of the antibiotics used to treat STDs, including ad- been recommended for the minimum package. De- verse reactions, development of microbial resistance, pending on the prevalence of disease, on whether the and change in microbial flora. 115 Box 5.4 Short-course treatment of tuberculosis Tuberculosis (TB) kills or debilitates more adults ages teen months. Drugs for the short course cost about $50 15-59 than any other disease and is responsible for to $80 per patient. Those for the standard course cost about 2 to 4 percent of the burden of disease. It is the only $10 to $15, but the cost per death averted is higher single leading cause of death in developing countries, because only 30 percent of patients complete treatment accounting for about 2 million deaths a year, or ap- and are cured, as against 60 percent for the short proximately 5 percent of all deaths and 25 percent of course. Other benefits of the short course include a preventable adult deaths in those countries. More smaller number of resistant organisms and less need women of childbearing age die from TB than from for expensive retreatment. (This discussion applies to causes associated with pregnancy and childbirth. More the treatment of sputum-smear-positive TB. Once than half of the world's population is infected with the other forms of TB have been identified, treatment costs TB bacillus. People who are malnourished or have an- should be similar except for serious forms of smear- other severe illness are at particular risk for TB, as are negative TB.) those infected with HIV. The relationship between TB Walk-in treatment is less expensive than hospitaliza- and HIV is highly significant, as each person infected tion, but if this care cannot be closely monitored (as in with HIV and TB could infect twelve other persons many rural areas), hospitalization may be more cost- with TB per year. effective. The program described is modeled on pas- Annual incidence rates of all forms of clinical TB vary sive case investigation, assuming that a person with TB from 50 to 260 per 100,000 in the developing world; will have symptoms such as cough and weight loss and more than half of these cases are infectious (sputum- will seek care and that infected persons discovered by smear-positive). For most forms of TB, 50 to 60 percent active searches will be less likely to continue treatment of those infected will die if untreated. All ages are at than those who seek care. Although the BCG vaccina- risk, but the peak is in young adulthood. In Sub- tion is important in TB control for children, its effec- Saharan Africa the annual risks of infection remain tiveness in adults is still under investigation. high, partly because of poverty and overcrowding The cost of treatment is less than $10 per DALY in all (which are risk factors for TB) and partly because any chemotherapy scenarios. It is estimated that tuber- decrease in the annual risk of infection is offset by the culosis treatment of infectious (smear-positive) individ- HIV epidemic. In India and Sub-Saharan Africa TB is uals prevents one to four new cases by stopping trans- the leading cause of death and the biggest contributor mission. The positive externalities of short-course to the disease burden; it is responsible for about 8 to 11 chemotherapy explain in part the extremely favorable percent of the DALYs lost in the 15-59 age group. cost-effectiveness and justify government intervention. There are two effective approaches to treating TB: Because the cost of drugs, at $50 to $80 per patient, is short-course chemotherapy, which uses three to five probably too high for the poor, public subsidy is espe- drugs over six to eight months, and the "standard" cially warranted for low-income households. course of two to three drugs taken over twelve to eigh- ernment policies that emphasized the financing of tries. This reduction in the burden of disease cost-effective clinical services directed especially at would be equivalent to saving the lives of more the rural poor. than 9 million infants each year. Benefits, costs, and financing of the essential clinical Delivery of the minimum clinical package would package cost an average of about $8 per person each year in low-income countries and about $15 in middle- Widespread adoption of the minimum clinical income countries. Approximately half of these package would have a tremendous positive effect amounts would be for prenatal and delivery care on the health of people in developing countries. If alone. When the cost of selected public health in- 80 percent of the population were reached, 24 per- terventions is added, total costs rise to $12 per cap- cent of the current burden of disease in low- ita in low-income countries and $22 in middle- income countries and 11 percent in middle-income income countries. The differences are the result of countries could be averted (Table 5.3). When the different demographic structures, epidemiological minimum clinical package is combined with the conditions, and labor costs in the two settings. public health package outlined in Chapter 4, the In low-income countries, where governments share of current illness that could be eliminated typically spend about $5.50 per person for health rises to approximately 32 percent in low-income and where total health expenditures are about countries and 15 percent in middle-income coun- $14 per person (Figure 5.2), the affordability of the 116 Table 5.3 Estimated costs and health benefits of selected public health and clinical services in low- and middle-income countries, 1990 Annual cost (dollars)' Per capita cost as share of Disease burden Per case or Per income per averted Country group and package per participant Per capita DALY capita (percent) (percent)" Low-income (per capita income = $350) Public health packager 4.2 - 1.2 8 Minimum essential package of clinical services - 7.8 - 2.2 24 Short-course chemotherapy for tuberculosis 500 0.6 3-5 1 Management of the sick child 9 1.6 30-50 14 Prenatal and delivery care 90 3.8 30-50 4 Family planning 12 0.9 20-30 3 Treatment of STDs' 11 0.2 1-3 1 Limited caree 6 0.7 200-350 1 Total, public health and clinical services 12.0 - 3.4 32 Middle-income (per capita income = $2,500) Public health packager - 6.8 0.3 4 Minimum essential package of clinical services - 14.7 - 0.6 11 Short-course chemotherapy for tuberculosis 275 0.2 5-7 1 Management of the sick child 8 1.1 50-100 4 Prenatal and delivery care 255 8.8 60-110 3 Family planning 20 2.2 100-150 1 Treatment of STDsd 18 0.3 10-15 1 Limited care' 13 2.1 400-600 1 Total, public health and clinical services - 21.5 - 0.9 15 Note: Figures assume coverage of 80 percent of the population. Average costs. Marginal benefits. Includes EPI Plus; school health including deworming, micronutrient supplementation, and health education; information on health, nutrition, and family planning; tobacco and alcohol control programs; monitoring and surveillance; vector control; and programs for prevention of AIDS. Benefits were calculated assuming an AIDS epidemic comparable to that in Sub-Saharan Africa today. Limited care includes assessment, advice, alleviation of pain, treatment of infection and minor trauma, and treatment of more complicated conditions as resources permit. Source: World Bank calculations. Because the poorest countries spend far less on health than do middle-income countries and the public share is smaller, thei have fewer resources available for reallocation. Figure 5.2 Public financing of health services in low and middleincome countries, 1990 El l'ublic and private spending Public spending Health spending per capita (1990 dollars) Percentage of GNP 120 100 80 60 40 20 0 0 1 2 3 4 Lowincome countries Middleincome countries Source: Appendix table A.9. 117 $12 minimum package is problematic. Paying for sential package, some degree of targeting is almost the package would require an increase in public inevitable. In countries where the wealthy do not spending for health (part of which could be fi- use government-financed services because of the nanced by donors in the short term), as well as a greater quality and convenience of privately f i- reorientation of current government expenditure nanced services, targeting may be fairly easy. In for health from discretionary to essential care. In other cases, however, user charges will have to be middle-income countries, where public spending applied selectively, relying on means testing and for health averages about $62 per person, the other targeting devices (see Box 5.5). Special ame- $22 minimum package is financially feasible if nities in teaching hospitals, for example, can be there is adequate political commitment to shift ex- priced at or above cost, since they will be con- isting resources in the direction of public health sumed exclusively by the wealthy. User charges and essential clinical care. Indeed, upper-middle- can help generate additional revenues for the es- income countries can afford public finance of an sential package, amounting to perhaps 10 to 20 essential package that goes well beyond this percent of total government spending for health. minimum. Finally, directing donor funding to public health Who should pay for the essential clinical pack- and essential clinical care can significantly expand age? There are strong efficiency arguments for di- the total resource basket available for the mini- rect financing by developing country governments mum package. of the selected public health interventions dis- User charges for public health services in devel- cussed in Chapter 4. A number of the essential oping countries have sparked much debate since clinical services, including treatment of tuber- the World Bank endorsed the concept in 1987 in a culosis and STDs, have large positive externalities. policy study on health financing. Critics argue that What is more important is that the poor are dispro- fees restrict access to care, especially for the poor. portionately affected by the disease burden of the Yet many developing countries, particularly in conditions listed in Table 5.3 and that, because of Sub-Saharan Africa, have been forced to rely in- larger family sizes, they would benefit dispropor- creasingly on fees to supplement strained budgets. tionately from prenatal, delivery, and childhood Studies on the effect of user fees are inconclusive services. Public finance of a basic package of ser- and contradictory. One reason is that some re- vices is an effective mechanism for reaching the searchers have failed to calculate the true cost to poor. patients of treatment at government clinics. People The main problem with universal government often pay dearly for supposedly "free" health financing of an essential package is that it leads to care. Recent household surveys in India, Indo- public subsidies to the wealthy, who could afford nesia, and Viet Nam indicate that each visit to a to pay for their own services, with the result that government health center actually costs patients fewer government resources go to serve the poor. two to three times the amount of the low official A policy of concentrating public resources on ser- fees. Bribes aside, the indirect costs such as trans- vices for the poor and requiring others to pay all or port and the opportunity cost of time spent seek- part of their own costs makes sense on equity ing care are substantial. grounds but also has some disadvantages. Often, Since patients are already paying for supposedly the administrative costs of targeting are high, and free or low-cost health care, new user fees, when the exclusion of wealthy and middle-income accompanied by a reduction in indirect costs and groups can erode political support for the essential improvement of services, may increase utilization. package, causing a decline in funding and quality Recent studies in four African countriesBenin, of care. Furthermore, problems of cost escalation Cameroon, Guinea, and Sierra Leoneindicate and access to insurance on the part of high-risk that this is the case and, encouragingly, that the groupsdiscussed in Chapter 3can complicate poor benefited most from these changes. Since fa- reliance on private finance. Perhaps for these rea- cilities used fees to fund services not previously sons, most member governments of the OECD fi- available locally, poor patients avoided costly nance (or mandate finance of) comprehensively travel, and the actual cost of care declined. Studies defined essential packages for virtually all their indicate that user fees amounting to less than 1 citizens. percent of annual household income have little im- In low-income countries, where current public pact on the utilization of health services, even by spending for health is less than the cost of an es- the poor. Because higher fees do decrease utiliza- 118 Box 5.5 Targeting public expenditure to the poor When public spending on health is not targeted to the tively well-off. Targeting public finance to those not pooras often happens, according to numerous participating in social insurance will reach the poor, studiesno other source of funds is likely to compen- and administration will be relatively simple. In coun- sate. Which targeting mechanisms work best in prac- tries with established social insurance mechanisms, tice will depend on their impact on demand, their ad- this targeting mechanism will often prove best. ministrative costs, their technical and managerial Let individuals self-target. The essential services requirements, and the level of political support. In are available free of charge to all, but the program is countries where incomes are too low for a minimum designed in such a way as to deter the better-off from essential package to be provided universally, there are using them. Time costs, stigma, and fewer amenities four main mechanisms for targeting the essential pack- associated with services are the usual mechanisms for age of services: encouraging self-targeting. Unfortunately, these same Assess individuals seeking services on the basis of characteristics may discourage the poor as well as the income, nutritional status, or other criteria and, de- better-off. Low-income working mothers, for example, pending on the assessment, provide services from the may find that the time, for themselves and their chil- essential clinical package free of charge or according to dren, involved in using subsidized services is an insur- a sliding scale of fees. In evaluating income, direct mountable obstacle. measurements or proxies (such as housing characteris- Target by type of service. Offer free of charge, or tics) can be used, but this tends to be more adminis- subsidize heavily, services that are needed dispropor- tratively costly than other mechanisms. tionately by the poor. This sort of targeting mechanism Subsidize essential clinical services for easily iden- is inherent in much of the essential package of clinical tified subgroups of the population (for example, all services. Prenatal and delivery services, management those living in certain low-income regions or neighbor- of the sick child, and STD and tuberculosis treatment hoods or all children in public schools). Where social are all services that, if universally available, would es- insurance mechanisms (usually financed through pay- pecially benefit the poor. roll taxes) exist, they generally tend to cover the rela- tion, reducing charges or exempting the poor from ary clinical health careservices outside the essen- the fees may be warranted. tial packageis far less compelling. In fact, gov- In middle-income countries, where a significant ernments can promote both efficiency and equity part of the population may be covered by private by reducingor, when possible, eliminating or social insurance, governments can target public public funding for these services. Doing so re- monies to essential clinical services for the poor by quires recovering the cost of discretionary services legally defining and mandating that the national provided in government health facilities and cut- essential package be covered in all insurance poli- ting subsidies to private and public insurance cies, thereby freeing government resources to tar- schemes that finance discretionary care. By reduc- get the poor. Surprisingly, emerging managed care ing spending on these services, governments can institutions in developing countries often fail to concentrate public expenditure where it will do the cover benefits that would be in such an essential most goodin public health and cost effective clin- package. In the 1980s in Brazil, for example, many ical services. health maintenance organizations failed to include Out-of-pocket payments are the main source of immunizations and family planning in the basic financing for discretionary care in low-income benefits package for their clienteles. countries. They remain substantial in middle- income countries, but insurance becomes increas- Insurance and finance of discretionary clinical ingly important as incomes rise. Because, except services for the very rich, out-of-pocket financing cannot cover expensive care or deal with catastrophic ill- Public financing of a national essential clinical ness, widespread financing of discretionary care is package can be justified because the package cre- possible only through insurance. Countries have ates positive externalities and reduces poverty. two main options for meeting a growing demand The case for government financing of discretion- and need for insurance. One is to move toward the 119 current U.S. system, which relies substantially on the Philippines, a combination of limited private private voluntary insurance. The other is to follow insurance and the ability of upper-income groups the examples of Canada, Japan, and most Eu- to pay makes it feasible for governments to charge ropean countries, where general government reve- for discretionary care delivered in public hospitals. nues or social insurance cover the cost of relatively In Kenya the government is currently attempting comprehensive essential packages, leaving only a to recover the cost of caring for the insured at the small discretionary residual for private insurance. national referral hospital in Nairobi. In Lesotho Government policy can improve the functioning charges in the private ward of the central hospital of insurance markets in three ways. It can strive to in Maseru were increased in 1990 to recover costs eliminate unfair subsidies to insurance. It can from wealthier patients. work to maximize the population covered by in- In middle-income countries insurance becomes surance by preventing selection biasthe ten- more important as a mechanism for financing dis- dency of insurers to discriminate against bad cretionary services. In South Africa private insur- health risks. And it can help to eliminate another ance covers about 15 percent of the population and potential problem with insurance: the explosive accounts for more than a third of total health increases in health care costs that are closely asso- spending. In Brazil, even though everyone is eligi- ciated with fee-for-service payment of health pro- ble for publicly financed health services, about a viders by third-party insurers. fifth of the population is also privately insured. Social insurance, in which payroll deductions are Redirecting public funding from discretionary care earmarked for health care, is widespread in mid- dle-income countries, especially in Latin America. There is substantial scope in the developing world Such payroll taxes account for a quarter or more of for redirecting current public spending away from national health spending in Costa Rica, Korea, and discretionary services. Cost recovery in govern- Panama (Table 5.4). In countries with broader in- ment hospitals, especially from the wealthy and surance coverage there is even greater potential for insured, is one important mechanism. Even in cost recovery than in poorer countries. Public hos- low-income countries, where insurance may ac- pitals in Chile are now being encouraged to count for less than 5 percent of total health spend- charge, particularly for patients who have private ing, as in Ethiopia, Kenya, Lesotho, Pakistan, and insurance. Table 5.4 Social insurance in selected countries, 1990 (percent) Social insurance as share of Share of population covered public sector health Social insurance as share of Health expenditure as Group and country by social insurance expenditure total health expenditure share of GNP Low-income India 5 9 2 6.0 Kenya 10 7 4 4.3 Indonesia 13 17 6 2.0 Middle-income Dominican Rep. 6 9 6 3.7 Ecuador 9 11 7 4.1 Colombia 15 18 8 4.0 Paraguay 18 24 13 2.8 Philippines 38 12 6 2.8 Panama 50 55 43 7.1 Turkey 58 26 14 4.0 Costa Rica 82 85 62 6.5 Korea, Rep. of 90 50 25 6.6 High-income Germany 75 76 63 8.0 Japan 100 64 56 6.5 France 100 95 71 8.9 Netherlands 100 94 73 7.9 Source: World Bank data; Mesa-Lago 1991; de Geyndt 1991; Vogel 1990; Brotowasisto and others 1988; Ikegami 1992; Hurst 1992; and Solon and others 1992. 120 The other way to redirect government spending experience with such arrangements in developing away from discretionary care is to phase out public countries. subsidies to insurance. These subsidies, which are large and widespread, take the form of both direct Eliminating unequal access to clinical care budgetary transfers to insurance institutions and under insurance tax concessions for employers' and employees' in- surance contributions. They benefit the better-off A serious problem with relying on insurance to and are therefore regressive. pay for discretionary care in developing countries In Latin America governments in Guatemala, is that individuals and groups often have unequal Honduras, Mexico, Nicaragua, and Venezuela access to insurance coverage. The problem is espe- contribute a percentage of individual workers' cially acute with voluntary private insurance be- wages to social security sickness and maternity cause of selection bias. When insurers rate individ- funds. In Chile and Uruguay the government cov- ual risks, they often either refuse to insure the sick ers the operating deficits of the funds, and in Co- and elderly or make insurance prohibitively ex- lombia and El Salvador it pays directly for a part of pensive for these clients because their expected the cost of social security health services and ad- costs are so high. They also commonly exclude ministration. Private insurance in South Africa many health conditions that should ideally be cov- and Zimbabwe receives large public subsidies in ered. In Brazil, for example, where government the form of tax deductions for employer and em- regulation of insurance is weak, private insurers ployee contributions. Fees to insured patients frequently refuse to enroll persons who are poor using government hospitals in Zimbabwe are set health risks, and they fail to cover costly risks such at perhaps a third to a quarter of actual costs, pro- as HIV-AIDS. In South Africa and Zimbabwe pri- viding another subsidy to the better-off. vate health insurance schemes known as medical Once subsidies to private and public insurance aid societies have historically covered entire indus- are established, they are extremely difficult to tries and occupational groups, but the recent mar- eliminate. Recipientsgenerally the better-off- keting of individual insurance policies has intro- view them as an important benefit. The current duced discrimination based on risk. Selection bias debate in the United States over reducing the em- is an important reason for the incomplete insur- ployer tax deduction for contributions to medical ance coverage in the United States, where 37 mil- insurance illustrates this strong political resis- lion people, or about 15 percent of the population, tance. To date, very few developing countries have are uninsured. successfully cut subsidies. An exception is Zim- Very few countries, developing or industrial, babwe, where major political shiftsnational inde- have managed to eliminate selection bias under pendence and the advent of a democratically elec- private insurance, even though it could theo- ted government in 1980led to a significant retically be done by prohibiting insurers from rat- scaling back of tax breaks for medical aid pre- ing individuals' health risks and requiring them to miums. The Chilean government is actively con- rate only large groups, or "communities," in sidering eliminating the existing tax concession for which high risks are spread over a large number of employers' contributions to social insurance. people to minimize the cost effect of the risks. Subsidies to insurance systems that cover only Eliminating selection bias is easier with social in- part of the population are invariably regressive, surance because contributions are compulsory and but subsidies can be progressive when insurance are normally a fixed percentage of wages for all has become universal. At that point, public sub- employees regardless of their number of depen- sidies end up benefiting mainly those outside the dents or their individual health risks. The snag is formal labor force, notably the elderly and the that since compulsory insurance generally uses poor. Examples of progressive subsidies are the 30 employment-based contributions, it is able to percent of national social insurance funding that achieve broad coverage only when most of the eco- the Korean government contributes on behalf of nomically active population is in the formal labor low-income households and the 20 percent of Ja- force. pan's social insurance spending that derives from Where only part of the population is covered by government budget transfers for retirees and the insurance, as in most developing countries, access poor. Governments could provide equally pro- to clinical services for the uninsured poor can be gressive, targeted subsidies to buy regulated pri- much more difficult than for insured, better-off vate health insurance for the poor, but there is no people. The quality of care, including essential 121 clinical services, may also be vastly inferior for the successfully upgraded separate clinical services poor. In low-income Africa the wealthy consult targeted to the poor. private doctors for their clinical care while the poor are often forced to use understaffed government Containing the costs of clinical care health centers that lack the most basic drugs and Escalating health spending is perceived as a crisis equipment. In much of Latin America middle-class when it begins to crowd out other sectors of the families receive better care in hospitals and clinics economy or to raise the cost of labor, threatening a belonging to the social security agency than the country's international competitiveness. This is poor are given in the badly deteriorated facilities currently happening in the United States and to a run by the ministry of health. varying extent in other high-income countries (Box One way to eliminate these disparities would be 5.6), and it is about to happen in several middle- to put all health facilities under a single adminis- income developing countries, including Chile and tration and open them to all. Few countries have Korea. taken on this politically and administratively diff i- The sources of excess health costs and growth of cult task. Costa Rica, in which the social security costs are complex and much debated. Health ser- agency manages all government hospitals, is a rare vices have a high labor content, and their produc- exception. Another solution is for the government tivity has grown slowly in comparison with that of to focus spending on the poor by investing heavily other areas of the economy. In the United States in the infrastructure (facilities, equipment, and relatively high levels of underlying morbidity and transport) needed to improve essential clinical care greater amenities in hospitals are part of the rea- for the poor and spending substantially more on son, but inefficiencies are also important. Two the associated personnel and drugs. By focusing types of inefficiency stand out: high administrative investments on peripheral health units (health costs and unnecessary use of an ever-expanding centers, subcenters, and health posts) and on staff array of sophisticated and ever more costly tech- for these facilities, Malaysia and Zimbabwe have nologies for diagnostic tests and surgical pro- Box 5.6 Containing health care costs in industrial countries In 1990 the United States devoted 12.7 percent of its mark, Germany, and Sweden) set overall limits on pay- GNP to health, as against 9.1 percent in Canada, 8.9 ments to both doctors and hospitals. In the latter percent in France, 8.0 percent in Germany, 6.5 percent group, the method of limiting payments to doctors in Japan, and 6.1 percent in the United Kingdom. The varies widely: capitation in Britain, fee-for-service in 2.7 percent annual increase in the health-to-GNP ratio Germany, and salaries in Sweden. In Germany, as a for the United States during the 1980s was the highest means of controlling expenditure, fees are reduced if among the OECD countries. U.S. health expenditures the volume of services exceeds the anticipated level. In of $2,800 per person in 1990 were nearly $1,000 above general, the OECD countries that have contained costs the average for the OECD countries. The price of better have greater government control of health health care services in relation to other goods and ser- spending and a larger public sector share of total health vices also rose much more rapidly in the United States expenditures. This is also the case in eleven developing (2.2 percent per year) than in the other OECD countries with income per capita of more than $6,000. countries. In the poorer countries there is no apparent link be- An important factor in explaining the rapid growth tween the public and private shares of health expendi- in health care costs in the United States is that doctors ture and the proportion of income devoted to health. and hospitals are paid predominantly on a fee-for-ser- In the U.S. health insurance system, with its large vice basis. Countries experiencing moderate spending numbers of insurers reimbursing providers at different growth (Canada and Japan) also use fee-for-service fot rates, administrative costs absorb about 15 percent of outpatient physician care but have devised other ways health expenditures, compared with 5 percent or less of controlling expenditures: a uniform fee structure in the other OECD countries. If the United States re- and aggressive peer review of doctors' spending pat- duced its administrative outlays to the level in these terns in Japan and fixed overall budgets for hospitals in other countries, a total of $80 billion annually could be Canada. Countries with low levels of spending (the saved, equivalent to about a third of total health spend- United Kingdom) or low recent growth of costs (Den- ing for all developing countries combined. 122 cedures. These two kinds of inefficiency appear to outstripping Korea's robust overall economic be closely linked to basic features of the U.S. growth. The share of GNP devoted to health rose health system. Open-ended fee-for-service com- from 3.7 percent in 1980 to 6.6 percent in 1990. By pensation for health providers encourages the de- 1990 Korea was spending $377 per capita for velopment of new equipment, drugs, and pro- health, putting it 50 percent above the expected cedures and leads to exploding costs because level for its income. neither providers nor patients have strong incen- To control health care costs, countries need to tives to hold down utilization or spending. A com- limit payments to health care providers (Table 5.5). plex system of multiple insurance institutions and One approach is to pay a fixed amount for each other payers, each with its own procedures, raises person, as is now done by health maintenance or- administrative overheads substantially. ganizations in Brazil and the United States and by The findings concerning health cost escalation in the British National Health Service. Another the United States and other industrial countries method, used in several OECD countries, is to are especially relevant for middle-income develop- give each hospital or network of physicians a fixed ing countries. Those countries are under pressure total budget (Box 5.7). from medical professionals, manufacturers, and Insurers may jointly negotiate uniform fees for consumers to use new medical technologies, and physicians, as is done by Japan's social insurance they face difficult policy choices related to insur- system and by Zimbabwe's private insurance sys- ance institutions and compensation of providers. tem. They can also set fixed payments for specified Korea's problems with escalating health expendi- medical procedures or standard per diem pay- tures parallel those of the United States and may ments for hospital stays. Brazil's social insurance hold important lessons for other developing system, for example, is paying standard fees for countries. hospital care according to a modified version of the The Korean social insurance system is a nation- system of diagnostic-related groups that was de- wide network of "sickness funds" covering occu- veloped in the United States for Medicare, the pational and regional groups. Worker payroll con- government-funded system for the elderly. Chile's tributions are compulsory. All Koreans are public health system is introducing comparable in- covered, and the government subsidy to extend ternal prices for its hospitals. coverage to the elderly and indigent is highly As worldwide experience amply demonstrates, progressive. As in the United States, Korea's there is no perfect insurance system: every coun- health providers are predominantly private-72 try's insurance institutions have their problems. percent of physicians and 80 percent of hospital But there are two important generalizations: third- beds are in this categoryand they are paid on a party insurance leads to cost escalation, and, in fee-for-service basis. There is little control of the general, social insurance and regulated private in- acquisition of medical technologies. Korea already surance with community rating avoid selection has more sophisticated new medical equipment, bias far better than voluntary private insurance. It such as imaging machines and lithotripters (used is difficult to achieve wide coverage of the popula- to treat kidney stones), per capita than either Can- tion in most developing countries, and without ada or Germany. During 1989-91 spending for wide coverage, public subsidies to insurance will medical devices and diagnostic products grew by inevitably be regressive. Countries, whether they more than 20 percent a year. As in Japan, physi- use social or private insurance, are finding it ex- cians often sell the drugs they prescribe, and in- tremely hard to eliminate these deeply entrenched surance reimburses the cost of nonprescription subsidies. Once widespread insurance coverage is drugs, which creates strong incentives to over- achieved, costs can easily spin out of control un- prescribe and overuse drugs. Prescription phar- less provider compensation is tightly regulated or maceuticals now account for 36 percent of health determined in ways that give incentives for cost spending, one of the highest shares anywhere. Fi- containment. Developing country governments nally, the administrative costs associated with the must be prepared to deal forthrightly with this dif- more than 300 independent insurance funds are 10 ficult set of interlocking issues. to 20 percent of operating expenses, which is simi- lar to the administrative burden of private insur- Delivery of clinical services ance in the United States. Policies related to the delivery of health services in All these causes contributed to a dramatic in- developing countries should have two main objec- crease in health spending in the past decade, far tives. The first is to improve access to essential 123 Table 5.5 Strengths and weaknesses of alternative methods of paying health providers Payment method Strengths Weaknesses Fee for service Provider's reward closely linked to Tends to cause cost inflation level of effort and output Allows for easy analysis of Creates incentives for excessive and provider's practice unnecessary treatment Per case Provider's reward fairly well tied to Technical difficulty of forcing all cases (for example, using output into standard list can lead to mismatch diagnostic-related between output and reward groups) Gives provider incentive to minimize Providers may misrepresent diagnosis resource use per individual treated in order to receive higher payment Capifation Administratively simple; no need to Gives provider incentive to select (per patient under break down physician's work into patients based on risk and to reject continuous care) procedures or cases high-cost patients Facilitates prospective budgeting May create incentives for provider to underservice accepted patients Gives provider incentive to minimize Difficult to analyze provider's practice cost of treatment Allows for consumer clout if patient can select own provider Salary Administratively simplest Loss of patient influence over provider (straight payment per behavior unless patient choice links period of work) provider salary to patient satisfaction Facilitates prospective budgeting Can easily create incentives for provider to underservice patient and to reduce productivity Source: Adapted from Reinhardt 1989. clinical services, especially for the poor. The sec- low efficacy, and drugs and supplies are stolen or ond is to increase the efficiency with which ser- go to waste in government warehouses and hospi- vices are delivered. In the public sector inefficiency tals. Although the problems are many and deep, is widespread. Clinics and outreach programs op- appropriate government policies can do much to erate poorly because of shortages of drugs, trans- reduce or eliminate these kinds of inefficiency. port, and maintenance. Hospitals are poorly orga- Most countries have mixed systems of both pub- nized and managed and keep patients longer than lic and private delivery of clinical services and both necessary. Countries pay too much for drugs of public and private financing. The type of financing 124 a country chooses does not dictate the kind of cient public sector health centers and district hos- health delivery system it should have, or the other pitals (for example, in Chile, China, Sri Lanka, and way around. Some countries, such as Botswana, Zimbabwe), and there are circumstances in which Malaysia, Sweden, and the United Kingdom, have it is impossible or too costly to persuade the pri- mainly public financing and public delivery; some, vate sector to deliver careparticularly in geo- including Brazil and Korea, have public financing graphically remote or extremely poor areas. and private delivery; and still others, such as the The existence of alternative suppliers, both pub- Philippines, the United States, and Zaire, rely pri- lic and private, creates pressure for improved per- marily on private financing and private delivery. formance. In developing countries where the pub- The private health sector typically serves a di- lic system has a near-monopoly on health care verse clientele, and it typically delivers services delivery, a mixed system that exposes the public that are perceived to be high in quality and more services to competition is likely to be more efficient responsive to consumer demand than the govern- and to improve quality of care. Furthermore, in ment's. But there are also examples of highly effi- countries where government health services are -S Box 5.7 Health care reform in the OECD The OECD countries face persistent difficulties in the ing the productivity of rationed resources. There has financing and delivery of health services. Problems in- been some convergence of systems toward contracting clude inequitable access to services, gaps in insurance between public insurers and private providers. Bel- coverage, unacceptably rapid increases in health ex- gium and France have introduced tighter contracts into penditure, inefficiency, and poor quality. These diffi- their reimbursement systems, and the United King- culties are partly the result of circumstances outside the dom has moved away from its integrated National control of governmentsfor example, demographic Health Service and toward more autonomous and and technological change. To some extent, however, competitive physicians and hospitals. Other countries they arise from flaws in the design of the financing, reformed the contractual model itself by emphasizing payment, and regulation systems for health care. Dur- consumer choice, active informed purchasing rather ing the 1980s most OECD countries initiated moderate than passive funding by third parties, and managed or major reforms of their health care systems to correct competition among providers. These reforms are most these flaws. marked in Germany, but elements of such changes Three principal types of reforms can be distin- have also been implemented in Belgium and the guished. Several countries, including Ireland, the Netherlands in the form of mixed payment systems Netherlands, and Spain, have taken steps to extend eli- that combine budgetary caps with work-related pay- gibility for public medical care, bringing the last remain- ment of providers. ing groups of their populations into the public system Finally, the Netherlands and the United Kingdom of coverage for basic medical care. Despite widespread have embarked on differing experiments to introduce calls for privatization of finance, no country has re- competition within their public systems. In the Nether- duced its commitment to public coverage. lands it is envisaged that consumers will be able to Many governments have taken important initiatives choose among quasi-public sickness funds and private to contain costs, through increased cost sharing or insurers, with a central health care fund taking income- through supply-side reforms. Belgium, France, Ger- related premiums and paying out risk-related pre- many, and the Netherlands have set fixed budgets for miums to the competing insurers. This amounts to a hospital expenditure. Virtually all the OECD countries sophisticated health voucher scheme. In the United greatly reduced the rate of growth of health expendi- Kingdom part of the hospital budget is given to large tures during the 1980s. The biggest exception was the (competing) general practices, which will enable gen- United States. Those countries that still relied to some eral practitioners to purchase certain hospital services extent on the reimbursement of patients for medical on behalf of their patients. Since the reforms were in- bills, with no connection between insurers and pro- troduced only recently in the United Kingdom and are viders, were less successful in containing costs han being carried out gradually in the Netherlands, it is not those in which insurers had direct contracts or in which yet possible to evaluate them fully. There are, however, a public agency was both public insurer and health already signs that general practitioner "Iundholders" provider. in the United Kingdom are using their new purchasing Probably the most important reforms of the 1980s power to negotiate a higher quality of hospital services involved the introduction of improved incentives and reg- for their patients. ulations for providers and insurers, with the aim of rais- 125 Table 5.6 Policies to improve delivery of health care Potential impact on / / o Provider and policy Public sector Protect nonsalary recurrent spending Complete the district health delivery infrastructure Retain fees at point of collection Decentralize financial resources and operational authority Subcontract ancillary services to private sector Improve drug selection, procurement, and use Nongovernmental organizations Legalize and simplify registration Provide government subsidies (per case, per diem, or block grants) for essential clinical services Subsidize training for district health workers Private (for-profit) sector Remove legal barriers to practice Promote health maintenance organizations Establish managed competition among suppliers Regulate private hospitals and physicians Provide public subsidies for essential clinical services and selected public health interventions Significant U Moderate 0 Modest 0 No significant impact both overextended and excessively concentrated care facilities to the private sector. At the same on discretionary care at the expense of essential time, governments have to improve the equity and services for the poor, the public system needs to be efficiency of their remaining health programs and scaled back. This means reducing public invest- facilities, through selective and progressive user ment in tertiary care facilities and specialist train- charges, decentralization, managerial incentives, ing and, in some cases, transferring discretionary and better information systems. The key policies 126 for improving delivery of clinical services by gov- operating expenditures for its thirteen predomi- ernment, NGO, and private for-profit providers nantly rural health regions. Countries can also are shown in Table 5.6. protect nonsalary spending by introducing more flexibility into their hiring arrangements. In India Delivering the essential package the Ministry of Health is planning to hire 8,000 In a competitive health system, people seeking workers for a leprosy control project on a per diem health services can choose from a diversity of pro- basis rather than engage them as civil servants viderspublic, private nonprofit, and private for- with virtual lifetime guarantees of employment. profit. As developing countries move toward such Another policy for ensuring adequate spending a competitive system, they face a wide range of on essential clinical services for the poor is to allow policy options that can improve the delivery of the primary care facilities to retain user charges and essential clinical package. spend these revenues on drug supplies and incen- tive bonuses for health workers. In Cameroon THE PUBLIC SECTOR. For many countries an im- when fees were introduced in a group of rural portant step in improving access to government- health centers and the revenues collected were provided essential clinical services is to complete used to replenish drug supplies, utilization of the the basic district health infrastructure by building centersespecially by low-income familiesin- health centers and health posts and training more creased substantially. The percentage of sick peo- nurses, midwives, and other providers of primary ple living near those centers who sought care rose care. Provision of good-quality housing for rural by more than 25 percent compared with the areas health workers as part of the district health infra- where fees were not charged, demonstrating that structure can improve efficiency by encouraging user fees can actually increase services for the staff to move to rural areas. Zimbabwe doubled poor. the number of its rural health centers, from 500 to more than 1,000, during the period 1980-90 with NONGOVERNMENTAL ORGANIZATIONS. Private the goal of making essential clinical services avail- nonprofit institutions, both local and foreign, pro- able within 8 kilometers of home for the whole vide a significant share of health services in devel- population. When housing for rural physicians oping countries, delivering essential clinical ser- and nurses was added to the district health pro- vices to low-income households in the poorest gram in the mid-1980s, the effect on staff recruit- countries. NGOs provide a third or more of all ment and retention was enormous: vacancy rates clinical carein Cameroon, Ghana, Malawi, for physicians and nurses in the areas with new Uganda, and Zambia. They own a quarter of the housing fell to zero, as against 20 to 30 percent health facilities in Bolivia's three largest cities, and elsewhere. they supply more than 10 percent of clinical ser- Government recurrent spending for primary vices in India and Indonesia. Although it is diffi- care inputs other than salaries is particularly vul- cult to compare the performances of NGO and nerable to budget cuts. When revenue shortfalls government health facilities, recent data from Af- occuror tertiary care hospitals overspendser- rica suggest that NGOs are often more efficient vices in peripheral facilities and communities often than the public sector. In Uganda physicians in suffer. In countries such as Chad, Haiti, Mozam- church mission hospitals treated an average of five bique, and Nepal there are widespread reports of times as many patients as physicians in govern- health centers that have staff but few or no drugs ment facilities, and NGO nurses attended more and of mobile health teams with little or no gas- than twice as many patients as their government oline for their vehicles. Under these circumstances counterparts. Governments that have excluded technical efficiency falls to near zerogovern- NGOs or heavily restricted their operations have ments continue to pay staff salaries, but virtually seen essential services deteriorate. When Mozam- no meaningful health services can be delivered. bique decided after independence in 1975 to ban Protecting the nonsalary part of the budget is NGO health activities in favor of government-run extremely difficult but critical. In Senegal the gov- facilities, a wide range of health services in rural ernment is currently committed to increasing its areas suddenly disappeared. Where such bans or budgetary allocation for drugs and medical sup- barriers to NGO activity exist, they should be plies by 10 percent a year in real terms during removed. 1992-96. Mauritania has set targets for annual gov- Beyond this, there are important opportunities ernment spending on drugs and other nonsalary for governments to form constructive partnerships 127 with NGOs to deliver essential clinical services. small-town families are prepared to pay for good One approach being followed in Sub-Saharan Af- care. The average outpatient visit costs $2 and a rica and in some states of India is to incorporate normal delivery $8. NGO health centers into the network of public f a- In Africa and Asia traditional medicine remains cilities by nominating appropriately located NGO an important part of the health care system, some- hospitals as district (first level of referral) hospi- times accounting for more than 10 percent of total tals. The NGOs are expected to provide a range of spending. The number of traditional healersin- public health and clinical services and to perform cluding herbalists, bonesetters, faith healers, and specific districtwide functions such as health plan- traditional birth attendantsis typically many ning, supervision of lower-level clinics and com- times larger than the number of medical physi- munity activities, and maintenance of emergency cians. The ratio of traditional to modern health transport. In return, the government pays some of practitioners has been estimated at nine to one in the NGOs' costs. Sri Lanka, seventeen to one in Indonesia, twenty- This kind of government-NGO collaboration five to one in Ghana, and twenty-eight to one in takes a variety of forms. In Lesotho nine of the Nigeria. Under these circumstances there may be country's eighteen health service areas (districts) opportunities for governments to improve the de- are headed by a church mission hospital that car- livery of essential health services by using tradi- ries out comprehensive health planning and man- tional practitioners. Successful examples include agement for its entire area. In Zimbabwe govern- the use of traditional healers to screen for malaria ment funds for rural health improvement are and to distribute antimalarial drugs in Thailand, to being used to expand mission ("designated dis- promote modern contraceptives in Kenya, and to trict") hospitals and to purchase ambulances for distribute condoms to reduce HIV and STD trans- the NGOs. Ministries of health pay the salaries of mission in Uganda and Zimbabwe. Traditional nursing staff in mission hospitals in Zaire and birth attendants have also been enlisted to im- most of the recurrent costs of NGO facilities in prove pregnancy outcomes in many countries, in- Botswana. Government donation of free vaccines cluding Bangladesh (Box 5.8). and contraceptives to NGO health providers has Making delivery of clinical services more efficient also become a common way to target public sub- sidies to specific health intervention programs. There are many routes by which developing coun- tries can improve the efficiency of clinical services. THE PRIVATE SECTOR. In recognition of the grow- Policies to increase the efficiency of government ing importance of the modern private sector in health services through decentralization and better developing countries' health systems, some gov- hospital management could have an especially ernments have begun to encourage private practi- large positive effect, as could policies for strength- tioners to deliver essential services. In India sev- ening government regulation of a more competi- eral states and nonprofit groups are working with tive private sector. the private sectorincluding the country's 1 mil- lion semiqualified urban and rural medical practi- DECENTRALIZATION. A policy that can improve tionersto improve the quality and effectiveness both efficiency and responsiveness to local needs of basic care. Some innovative approaches are be- is decentralization of the planning and manage- ing tried: private distribution outlets are being ment of government health services. In Africa stocked with condoms and oral rehydration solu- some central ministries of health have given pro- tion to make both more widely available, and the vincial and district offices responsibility for plan- Indian Rural Medical Association is trying to im- ning, day-to-day management of funds, person- prove the skills of rural private practitioners nel, training, maintenance, and other functions. through education on such subjects as family plan- Many problems have arisen: local governments or ning, immunization, and oral rehydration. In local offices of the central health ministry have not Mali, one of the world's poorest countries, the lo had the capacity to plan and manage health activ- cal medical school has started a program to train ities; devolution of responsibility has not always graduates to set up private practices in small been accompanied by allocation of the needed towns of 15,000 to 50,000 inhabitants. Although funds; and local officials have not necessarily been experience is recent, a number of physicians have accountable to their constituents. There have been already established successful private practices in some successes, however, that offer lessons for these towns, and they have demonstrated that other countries. 128 Box 5.8 Traditional medical practitioners and the delivery of essential health services Many simple health activities do not require extensive outreach workers, the volunteers improve case detec- professional training or major facilities and equipment. tion and save the government considerable expense. Health workers based in clinics or in their own commu- A WHO study found that traditional healer volun- nities play an important role in delivering these ser- teers in Thailand were more active in pursuing and vices. The millions of community-based traditional identifying malaria cases than other volunteers and health practitioners have enormous potential as public that they tended to remain in the program longer be- health workers and providers of essential clinical ser- cause their service enhanced their standing in the Com- vices if governments can give them the appropriate munity. Villagers indicated that they felt more confi- training, information, and incentives. dent about having someone they already knew as the Thus far, the experience with modern-traditional col- village traditional healer draw their blood and adminis- laboration has been mixed. A number of projects have ter treatment. failed because of poorly designed training and inade- In western Kenya the African Medical and Re- quate supervision, and many governments need to do search Foundation (AMREF) has trained male and fe- more to curb unnecessary and dangerous practices by male traditional health practitioners who live in remote traditional healers. But there have also been instructive villages to dispense drugs and some types of contra- successes. ceptives. Since the project began, the share of women An evaluation of workers participating in a volun- of reproductive age using modern contraception in six teer program for detection of malaria in northern Thai- pilot sites has risen from less than 10 percent to more land found that the performance of volunteers who than 25 percent. The Kenyan government has asked were traditional healers was superior to that of other AMREF to expand the project. volunteers. The program, which began in 1961, had by In Bangladesh a program to train and support mid- 1988 more than 40,000 malaria-control volunteers dis- wives to work with traditional birth attendants helped tributed across 34,000 Thai villages. Volunteers trained to lower maternal mortality rates by 60 percent over a by the Ministry of Health are expected to examine vil- ten-year period. The results of the program indicate lagers, take blood samples, prepare smears to be sent that, given adequate support systems, community- to the district malaria clinic for analysis, and treat fever. based services could bring about a substantial decline They also provide malaria-related education to the vil- in maternal mortality. lagers. In comparison with alternatives such as paid In Ghana, until recently, decisions on health sponsibility for primary care to local district coun- spending were highly centralized, with inflexible cils. The process began on a pilot basis in 1973 and expenditure levels set by the Ministry of Health for was gradually expanded to cover the entire coun- specific "vertical" programs such as immuniza- try. To support decentralization, the Ministry of tion, control of tuberculosis and leprosy, and fam- Health funded the creation of district health teams ily planning. In the late 1980s the ministry agreed consisting of a medical officer, a public health to delegate financial authority to health teams in nurse, and a health inspector. The central govern- each of the country's 110 districts. Unfortunately, ment continues to finance, through annual block most district health officials did not know the pro- grants, most of the recurrent primary care expen- cedures for obtaining and accounting for funds. To ditures of the district councils, and the councils' remedy this problem, members of the district proposals for capital spending are included in the health management teams were given training ministry's investment budget. But the day-to-day that enabled them to make more timely budgeting management of primary care centers, including and spending decisions, and expenditure levels in- purchases of supplies and hiring of personnel, is creased as much as fivefold. In districts where ear- in the hands of the councils. marked funds from the center were pooled and Those countries that have gone furthest in de- reallocated according to local priorities, technicil centralization have devolved responsibility for efficiency improved because of joint planning of health servicesincluding implementation of gov- work schedules and sharing of transport for out- ernment health programs and management of reach services and supervision. government health facilitiesto subriational levels In Botswana, although secondary and tertiary of government. Such devolution has been going health care is the responsibility of the central Min- on for many years in some large developing coun- istry of Health, the government has devolved re- tries with federal systems of government (for ex- 129 ample, Brazil, India, Mexico, and Nigeria), but it is ficient systems that conform to the expected pat- also becoming increasingly common in a number tern. But in others, such as Indonesia, Jamaica, of other Latin American countries and in the for- and Lesotho, there are no significant differences merly socialist economies. between the different types of hospitals. In Papua The gradual devolution of Chile's health system New Guinea the ideal pattern is reversed: the low over the past decade suggests that success can be turnover rate for district hospitals implies that the achieved through a measured process accom- hospitals either have too many beds or offer such panied by training and institutional development. poor services that patients go directly to higher Chile began the progressive decentralization of its levels. The shorter average stay at the tertiary hos- publicly provided health services in 1979, when pital suggests that many patients who do not re- twenty-six health service areas (HSAs) were estab- quire tertiary care are treated there. lished to cover the country's thirteen administra Disparities in average length of stay across hos- tive regions. Each HSA was given responsibility pitals with a similar mix of cases are another indi- (and, to go with it, additional personnel and a cation of substantial inefficiencies. Data from Latin share of the health budget) for managing all the America for 1980-85 show that average stays in government health facilities in its area. The second hospitals run by ministries of health varied from step in decentralization was taken in 1987, when five days in Colombia to thirteen days in Uruguay, management of the primary care network (consist- and stays in social insurance hospitals varied from ing of more than 2,500 urban and rural clinics, five days in Mexico to twelve days in Peru. In pub- rural health posts, and rural medical stations as lic hospitals in Argentina the average length of well as about 14,000 health personnel) was trans- stay ranged from eight to twenty-seven days. In ferred to the local government or municipalities. Malawi average stays in six government hospitals The HSAs were responsible for monitoring the with a similar case mix varied from five to thirteen municipalities' actions. In a third phase of the de- days. centralization process, scheduled to begin in late To remedy these inefficiencies, major steps to 1993, the central Ministry of Health will withdraw improve the organization and management of completely from service provision, leaving this public hospitals are required. One approach task entirely to the HSAs, which will enter into would be to delegate responsibility for health ser- formal performance contracts with the ministry. vice delivery to individual public institutions, but But not all decentralization has been a success. doing so will require changes in accounting and Countries such as Colombia that have devolved management practices. Government health bud- responsibility in a short period of time, without gets are now often highly aggregated, covering all the requisite financial resources and institutional facilities and programs in a given district, region, capacity at lower levels of government, have or even country. This prevents any detailed analy- found that decentralization can be counterproduc- sis of spending and services. Tracking costs should tive, aggravating existing inefficiencies and ineq- become part of the responsibility of the facility uities in the health system. In Brazil the munici- manager and the district management team. Cost pality of Rio de Janeiro refused to accept analysis would make it easy for these managers to responsibility for all the "decentralized" federal monitor areas of over- or underfunding and for health facilities within its borders on the grounds higher-level officials to compare cost profiles and that federal budget transfers to the municipality unit costs. were inadequate. Some countries have converted public hospitals into semiautonomous foundations or public enter- IMPROVEMENT OF HOSPITAL MANAGEMENT. Low prises in order to improve performance by grant- rates of hospital utilization in developing countries ing greater budgetary and management auton- point to significant inefficiencies in the use of omy. These foundations or parastatals are under buildings and equipment (to which scarce capital fewer restrictions than public facilities in manag- has been devoted). Most common medical condi- ing their budgets and sometimes in hiring and fir- tions can be treated in relatively simple facilities. ing. They can recover costs and collect charitable In a well-run hospital system district hospitals donations. The Tunisian government, for exam- would have the highest turnover rates and the ple, has converted eleven large public hospitals shortest stays, while tertiary hospitals would have into semiautonomous entities over the past two the lowest turnover rates and the longest stays. years and will convert another ten in the next few Some countries, including China and Fiji, have ef- years. Under the new arrangement, each hospital 130 manages its own operational budget and is free to when they are already sick or have a pressing reassign funds across budget categories as needed. health problemand because, in the case of third- To ensure accountability, the hospital is required to party reimbursement, they do not bear the full operate within its annual budget and to provide to cost of the services they consume. the Ministry of Health detailed reports about ser- Governments can encourage efficiency-pro- vices provided and unit costs. Autonomy in per- moting competition among suppliers of health ser- sonnel matters is more limited. All staff except the vices by requiring them to offer a standard pack- general manager are still governed by civil service age of services at a price fixed in advance. regulations. The hospital management cannot fire Consumers can then pick the supplier that offers employees, but it can ask the ministry to reabsorb the most attractive combination of price, service, staff and can use contracted personnel instead. and quality, with competition spurring suppliers The effect of these reforms will be fully assessed to improve quality and reduce costs. This is the over the next few years, but some gains in effi- basic approach taken in "managed competition" ciency are already apparent. One of the first hospi- proposals for health care reform in the United tals converted has fully contracted out all food, States (Box 5.9). Although managed competition cleaning, and security functions and now obtains requires a high degree of government administra- services of much higher quality, at similar or even tive capacity to set the rules and to monitor pro- lower unit cost. vider performance, it may be relevant in some Elsewhere too, an increasing body of evidence middle-income developing countries. suggests that the technical efficiency of govern- As developing countries take steps to encourage ment health facilities can be improved by contract- a diversified system of health service delivery, in- ing out ancillary services. In Venezuela, for exam- cluding use of NGOs and private providers, they ple, Health Ministry hospitals contract out will also need to strengthen their governments' maintenance of large equipment, and social secu- capacity to regulate the private sector. Regulations rity hospitals frequently contract for laundry, gar- are required to ensure that quality standards are dening, food services, and security, as well as for met, that financial fraud and other abuses do not maintenance. This arrangement has several ad- take place, that those entitled to care are not de- vantages. It can be less costly than publicly pro- nied services, and that confidentiality of medical vided services; services can be of better quality; information is respected. Regulation can be carried and they can be more reliable (because less subject out in a number of ways: by inspecting private to strikes and other industrial action). health facilities; by accrediting medical schools; by licensing physicians, nurses, and other health pro- COMPETITION AND REGULATION. Competition fessionals; and by prohibiting certain insurance among health providers in developing countries practices such as exclusion of prior medical condi- can improve the quality of services as perceived by tions. When the government allocates funds to patients and thus increase consumer satisfaction. NGOs, reimburses private providers under public This applies to the poor as well as the rich: compe- insurance schemes, or subcontracts with the pri- tition among private physicians in the slums of vate sector for ancillary services such as catering Bombay, for example, is intense, with private prac- and laundry, it can require an independent audit titioners offering convenient evening hours, short of these private contractors. waiting times, and readily available drugs to win In practice, few developing countries have es- patients from other private practitioners and from tablished such regulatory mechanisms, but the sit- public clinics. uation may be changing. In Brazil, where social It is much less clear, however, whether competi- insurance finances the bulk of health care, and pri- tion among suppliers of health services always vate hospitals and physicians provide 80 percent leads to greater efficiency. In fact, the contrary of hospital services and half of all outpatient care, sometimes happens, especially when competition important regulatory changes are now under dis- among private providers is combined with third- cussion. These include comprehensive accredita- party reimbursement of fees paid for services. Ex- tion of facilities by state governments, standard cessive tests, procedures, and drugs are supplied, licensing examinations for medical school gradu- and costs increase. This supplier-induced ineffi- ates, and the inclusion of representatives of citi- ciency takes place because most patients are un- zens' groups and consumer advocacy organiza- able to judge the value of specific services or to tions in medical ethics boards, which are currently compare prices among suppliersall the more so composed exclusively of physicians. The private 131 Box 5.9 "Managed competition" and health care reform in the United States 'Managed competition," which has attracted wide- zation would enable consumers to choose among com- spread interest in the United States, refers to a health peting packages in an informed and more price- services purchasing strategy designed to promote com- conscious way. One example of "managed competi- petition and to reward those health care providers with tion" is the California Public Employees Retirement the best performance in terms of cost, quality, and pa- System, which operates like the proposed HIPC, ar- tient satisfaction. The strategy is designed to address ranging health coverage and managing competition on the fundamental problems of the current health care behalf of almost I million state employees and retirees financing and delivery system in the United States. and their families. There, health care coverage is mainly employment- Analysts expect that over time competition would based and is far from universal: approximately 37 mil- force third_party insurers to drop out or move in the lion people under age 65 lacked insurance coverage in direction of managed care networks because these net- 1990. Costs are increasing rapidly; if current trends works can use financial incentives and management continue, spending is expected to grow from 12 to 18 tools to achieve efficient care. Indeed, insurers are al- percent of GNP by 2000. Under third-party insurance, ready shifting in this direction: nearly half of all health provider reimbursement methods often create financial insurers now offer some sort of plan involving man- incentives to provide more care. Insurers seek profits aged care. Competition would also give health care by excluding higher-risk individuals rather than by ag- providers clear incentives to become more efficient. gressively pursuing greater efficiency in providing One question about the managed competition model is health services. A major indirect cost of the system is whether it would work in rural areas and areas of low reduced labor mobility as a result of the risks of exclu- population density, and, if not, what alternatives sion from insurance at a new place of employment. would be best. Another concern is the cost of extend- Under managed competition, a health insurance ing universal coverage. purchasing cooperative (HIPC) would be formed to or- Although this proposal was developed to respond to ganize purchasers of health care within a region. The the particular problems of the U.S. system, it has rele- HIPC would establish standards for the region's health vance elsewhere. The Netherlands, which is introduc- plan by, for example, defining a basic benefit package ing choice of insurer under a universal social health of comprehensive health services, and would contract insurance scheme, faces the same challenges regarding with eligible providers for this basic package. During control of risk-selection behavior on the part of in- the annual open-enrollment period the HIPC would surers. Chile has, since 1981, encouraged the growth of provide information about the price of the basic pack- private prepaid health insurance schemes, known as age from different providers and about the quality of ISAPREs. (Box 7.2 provides details on these plans.) care offered. Equity would be improved by requiring Problems have come up with Chile's reforms, how- providers to open their rolls to all consumers, regard- ever: the lack of a standard package has limited effec- less of risk. Universal coverage would be achieved tive competition, and weak regulation has allowed pri- through public subsidies to those not otherwise cov- vate insurers to deny coverage to high-risk individuals. ered so that they could purchase packages. Standardi- Brazilian Association of Hospitals is debating the Reorienting clinical services and beyond establishment of its own accreditation system, and the medical association in São Paulo has begun a It is a fundamental responsibility of governments voluntary pilot effort to certify hospitals in that everywhere to ensure access to a package of essen- state on the basis of adherence to norms, assess- tial clinical health care, with special attention to ment of patient data, and patient satisfaction sur- the poor. Utilization of a minimum package de- veys. Qualifying examinations for licensing physi- fined by its high cost-effectiveness would reduce cians are being tested in another state where, at the total burden of illness in developing countries present, all medical school graduates are automaf- dramatically, by an average of 25 percent. Such a ically licensed for life, without requirements for package is affordable in low-income countries if continuing education or recertification. And the governments reallocate current health expendi- Federal Council of Medicine has proposed legisla- tures and increase public spending and if they im- tion that would expand its authority to monitor the plement policies that encourage selective pay- quality of health care and to discipline poorly per- ments directly from better-off patients and from forming doctors and hospitals. existing insurance schemes. Middle-income coun- 132 tries could easily pay for the minimum package, providers. Key government measures include using the resources currently devoted to health, strengthening legal and financial support to and might wish to enrich the package by adding NGOs that provide health services and creating a services. positive environment for the private sector com- Government efforts to improve health insurance bined with important regulatory safeguards should be aimed at increasing the portion of the against abuse. Gains in technical efficiency can be population covered, reducing subsidies for insur- achieved through a combination of careful decen- ance that benefit primarily the wealthy and the tralization of government health services and im- middle class, and controlling health care spending proved management of public hospitals. financed from insurance. This will require stronger These efforts to reorient government recurrent regulation of private insurance and policies to ex- spending toward essential clinical care and to pro- pand compulsory social insurance based on pay- mqte diversity and competition in the supply of roll taxes. It also means shifting provider payment health services must be accompanied by changes away from open-ended fee-for-service methods to in longer-term investments in health inputsfacil- prepayment through capitation and preset bud- ities and equipment, health personnel, phar- gets for hospitals. maceutical management systems, health informa- Greater diversity and competition in the supply tion, and health research infrastructure. Policies to of health services can do much to improve the de- bring about this reorientation are taken up in livery of an essential clinical package and raise the Chapter 6. technical efficiency of doctors, hospitals, and other 133 Health inputs In recent decades developing countries have in- spending subsidizes high-end facilities, equip- vested heavily in health. Often with help from do- ment, and human resources for private markets. nors, they have constructed hospitals and build- The challenge for public policy is to redress the ings and purchased equipment to fill them. They balance and so permit the efficient delivery of pub- have educated doctors, nurses, and other health lic health and essential clinical services. Where care professionals. And they have set up new sys- cost containment of health spending is a concern, tems to supply drugs, research, and information. public policy can play a useful role in limiting the Worldwide, the number of hospital beds rose be- growth of both public and private investments in tween 1960 and 1980 from 5 million to almost 17 specialist training, equipment, and tertiary facili- million, which more than doubled the per capita ties. For some inputs, such as buildings and hu- supply. The number of physicians increased more man resources, changes will necessarily be slow. than fivefold between 1955 and 1990, from 1.2 mil- For others, such as pharmaceuticals, a new policy lion to 6.2 million. Such investments have created can alter inputs rapidly. This chapter suggests how new opportunities, but they have also led to to set about these tasks. It also assesses how public problems. support for information and research can help Once built, hospitals are extremely difficult to improve health sector performance today and close. Once trained, physicians create pressure to create new health systems and technologies for be employed. In virtually every developing coun- tomorrow. try, facilities, equipment, human resources, and drugs are skewed toward the top of the health Reallocating investments in facilities system pyramid (Figure 6.1). Yet the cost-effective and equipment public health and clinical interventions discussed in preceding chapters of this Report are best deliv- Investments to support delivery of essential clini- ered at the level of the district hospital or below. cal services are best directed at health centers and That they are often delivered through tertiary hos- district hospitals and at improving access in under- pitals simply increases costs without improving served areas. Some public investments in tertiary quality. This problem is found in poor countries in facilities are needed to support research and train- which the principal tertiary teaching hospital in ing, but at levels well below current levels of pub- the capital city consumes a large proportion of the lic financing in most countries. Investments in spe- total resources available for health. It is also found cialized facilities can be left largely to the private in cities such as London, where numerous spe- sector, and public subsidies, where they exist, can cialized teaching hospitals absorb large amounts often be greatly reduced. Redirecting public of resources while failing to address the most spending toward lower-level facilities is difficult common and pressing health problems of city politically, but some countries are moving in this residents. direction. In Papua New Guinea, for example, In many countries public investments are con- public spending on hospitals has for the past dec- centrated unduly on tertiary services, and public ade been limited to 40 percent of the Ministry of 134 Most health care should.take place toward the bottom whole cityan example is Zambia's University of this pyramid. Teaching Hospital of Lusaka, with 1,835 beds. Hospitals absorb the bulk-40 to 80 percentof public spending on health in developing coun- Figure 6.1 The health system pyramid: tries. Industrial countries have much higher health where care is provided expenditures and more chronic disease problems, but the share allocated to hospitals is slightly smaller, 35 to 70 percent. Figure 6.2 shows the marked variations in hospital supply across the hospitals eight demographic regions used in this Report, ASpecialized from about eleven beds per 1,000 population in Central and Eastern Europe to less than one bed District A District per 1,000 population in India. In most developing health countries more than 60 percent of all hospital beds hospitals system are public. The data used in Figure 6.2 unfor- Health centers tunately fail to distinguish between tertiary and and clinics district-level hospitals. The minimum package of essential clinical services described in Chapter 5 requires about one district hospital bed per 1,000 Households population. Given that some of India's and Sub- Saharan Africa's hospital beds are devoted to care outside the essential package, there is likely to be a shortage of district-level hospital beds in parts of those regions. In some countries the underfunding of lower- Health's recurrent budget, which is well below the level facilities has been exacerbated by the creation average in most developing countries. of multiple levels of outpatient facilities (health posts, dispensaries, and rural health centers), Facilities none of which functions well. At the same time, tertiary care hospitals are crowded with patients Tertiary hospitals provide the most specialized and who could be treated in less costly and more acces- sophisticated services and are where most clinical sible district hospitals or health centers. A study in research, education, and training take place. They Chad, for example, revealed that 71 percent of all are usually located in large urban areas. One step central hospital consultations were for problems down the health hierarchy are district hospitals, that could have been treated at lower-level facili- which are typically located in towns or smaller ties. An obvious way to reduce spending without cities serving rural areas but are valuable in large sacrificing any health gains is to make full use of cities too. District hospitals generally have 100 to existing lower-level facilities. Measures for achiev- 400 beds, serve 50,000 to 200,000 inhabitants, and ing this include charging a higher fee to patients include departments of medicine, surgery, pedi- who go straight to tertiary facilities without refer- atrics, obstetrics and gynecology, and dentistry. rals, except in emergencies, and making a referral They also provide basic anesthesia, radiology, and from the primary care provider a mandatory con- clinical laboratory services. The district hospital is dition for specialized services. At the same time, the first level of referral from health centers and however, the quality and responsiveness of ser- provides complementary services such as basic vices at lower-level facilities need to be improved. surgery. It mainly offers inpatient care but also typically provides some outpatient care, day sur- INCENTIVES AND INVESTMENT DECISIONS. Public gery (in which the patient is operated on and dis- sector budgetary procedures often obscure the real charged on the same day), and emergency services costs of investments in health facilities and bias not available at health centers. In many cities that them toward high-profile investments in large have grown rapidly over the past twenty years, hospitals. Major investments, including donor- periurban areas do not have enough health centers financed projects, may be approved by a govern- and district hospitals. In some African capitals one ment body that does not have to face the recurrent extremely large tertiary public hospital serves the costs of operating the facility. Regions can argue 135 Installed hospital ca pacify is lowest in India, Sub-Saha ran Africa, and Asia and highest in the formerly socialist economies of Europe. Figure 6.2 Hospital capacity by demographic region, about 1990 Beds per 1,000 population 0 2 4 6 8 10 12 Sub-Saharan Africa India China Other Asia and islands Latin America and the Caribbean Middle Eastern crescent Formerly socialist economies of Europe Established market economies Source: Organization for Economic Cooperation and Development data; World Health Organization data. for new facilities without having to weigh the bud- zations (HMOs) in the United States have incen- getary consequences if federal-state resource tives for providing care efficiently. As a result, transfers, instead of following predictable and they operate with much less hospital infrastruc- transparent funding formulas (such as population- ture than the health system overall. Large HMOs based schemes), are heavily politicized. (with hospitals) have about 1.5 hospital beds per International assistance has frequently exacer- 1,000 memberswell below the average of 8 beds bated the problem of unsustainable health invest- per 1,000 for established market economies and 5 ments. Donor assistance, particularly for tertiary for the United States overall and slightly below facilities and teaching hospitals, has sometimes regional averages for China and Latin America. been provided even if the incremental recurrent Evidence indicates that quality of care in HMOs is costs from these investments are too high. Too maintained even with significantly lower levels of many donor-financed hospitals have opened only hospitalization and hospital infrastructure. Pri- partially, not at all, or at the expense of existing vately financed health insurance with uncon- facilities. In Rwanda, for example, a 200-bed hos- strained fee-for-service payment, by contrast, pro- pital was completed in 1991 but has not yet opened vides no incentives to rationalize physical capacity. because of the difficulty of financing its high recur- rent costs, which are estimated at about 15 percent DUPLICATION OF HEALTH FACILITIES IN THE PUBLIC of the Ministry of Health's already tightly con- SECTOR. Excess public facilities in urban areas are a strained budget. In Chad, where external assis- problem in many countries. The historical growth tance amounts to about 30 percent of national in- of hospitals, especially in capital cities, has led to a come, a national development plan proved useful proliferation of specialist tertiary services that are for screening out inappropriate donor financing. often linked to medical education and research. Two new hospital construction proposals that Another cause of duplication is the public provi- were found to conflict with the plan are being sion of health services to different subgroups of reconsidered. the population, each with its own hospitals and In the private sector, financial incentives drive health centers. In Poland, for example, parallel investment decisions. Health maintenance organi- health systems exist for workers and their families 136 in the railway, mining, police, and military sectors PRODUCTIVITY. The potential for improving the and for prisoners; another set of facilities serves productivity of installed hospital capacity is large. the general population. This can result in excess In addition to the financing and management re- capacity, with no facilities achieving economies of forms discussed in Chapter 5, efficiency gains can scale. be achieved by taking the following measures, There are two solutions for duplication: creating which will need to be supported by investments in internal markets and instituting central or regional training and infrastructure: planning linked to health budgets. If effective in- Convert some acute care hospital capacity to ternal markets are created within the public sector, less costly extended or chronic care facilities for money will follow patientsand patients, together patients who require less-intensive care for long- with their general practitioners, will have a choice term recovery and for rehabilitation of chronic as to which hospital to use. The availability of conditions. Extended care facilities operate at a good information about quality and price will help lower cost per bed-day than acute care hospitals. efficient providers of specialist services to prosper, In the absence of such lower-level facilities, pa- while less-efficient hospitals will close. The alter- tients occupy high-cost acute care beds. native is rationalization of services by central or Perform outpatient diagnostic tests before ad- regional planning. In the largely publicly financed mitting the patient to the hospital. health systems of the Nordic countries, health re- Support home care as an alternative to long- sources are allocated by region. Each region of term hospitalization for some ailments. about 350,000 inhabitants elects representatives Modify treatment protocolsfor example, re- who make decisions about health care spending. duce unnecessary surgeries, perform low-risk de- These representatives have incentives to avoid du- liveries at maternity centers, and treat tuberculosis plication of services and to capture economies of patients and many surgical cases on an outpatient scale in service delivery. If left to individual hospi- basis. In Cali, Colombia, costs per procedure for tals, decisionmaking for large investments will day surgery are less than 30 percent of the cost tend to reflect the interests of that hospital, not the of traditional treatment in hospital. Outpatient region. Multiple hospitals will want to provide surgery has grown rapidly in many industrial specialized, "prestige" services, leading to countries but is used much less in developing overinvestment. countries. EQuITY CONSIDERATIONS. For the rural poor, lack MAKING THE TRANSITION. The 1985 earthquake of physical infrastructure is the largest obstacle to in Mexico City destroyed about 20 percent of pub- use of health services. Distance to health facilities lic hospital capacity. The Ministry of Health chose limits people's willingness and ability to seek care, to concentrate reconstruction and new construc- particularly when transport is limited. There is a tion in low-income periurban areas that had hith- heavy urban bias in the distribution of health facili- erto been poorly served, and six new 144-bed dis- ties. Large cities are much better served by both trict hospitals were built in these areas. But such public and private health infrastructure than possibilities for rapidly reconfiguring capacity to- would be expected from their roles of serving ur- ward lower-level facilities and underserved areas ban populations and providing referral services for are seldom available. The alternative is to reduce the surrounding population. (Referral hospitals or refrain from public investment in tertiary hospi- are needed for only a small proportionno more tals while simultaneously increasing investment than 10 percentof total hospitalizations.) Wealth- and operating budgets for health centers and dis- ier regions also have better access to infrastruc- trict hospitals. Over time, the tertiary hospitals can ture. In India the richer states of Maharashtra and be operated on a self-financing basis, or they can Gujarat have 1.5 and 1.1 beds, respectively, per be closed, converted to chronic care facilities or 1,000 population; the poorer states of Bihar and district hospitals if these are needed, or even sold Madhya Pradesh have only 0.3 and 0.4 bed, re- to the private sector. But in most countries this spectively, per 1,000 population. Public invest- process will necessarily be slow. ments need to address inequities in the present distribution of health infrastructure. Donors have Equipment an important role in this regard, especially where a significant proportion of investment is donor Developing countries account for about $5 billion, financed. or 7 percent, of the $71 billion spent each year on 137 medical equipment worldwide. This global esti- proving maintenance to increase operating life and mate includes medical and dental supplies, surgi- reduce downtime of equipment is more efficient cal instruments, electromedical and X-ray equip- than buying new equipment. ment, diagnostic tools, and implanted products. Because of the many products on the market The ability of the medical equipment industry to and the speed of change, carrying out technology develop new health care technologies has vastly assessments can be extremely costly. The interna- exceeded the capacity of purchasers to evaluate tional community could help by developing and the clinical value and the cost-effectiveness of such disseminating information on the availability, ef- innovations. At present, approximately 6,000 dis- fectiveness, and prices of equipment and on user tinct types of medical devices (equipment, sup- guidelines. Essential equipment lists could be de- plies, and reagents) and more than 750,000 veloped along the lines of the essential drug lists brands, models, and sizes, produced by perhaps already used by many countries. 12,000 manufacturers worldwide, are on the Equipment procurement would also benefit market. from greater use of competitive buying. Purchas- Efficiency losses from poor selection and main- ing is commonly restricted to local distributor- tenance of medical equipment can be very large. ships, and some countries also heavily protect lo- WHO estimates that less than half of all medical cal industry. These policies reduce competition equipment in developing countries is usable. In and can easily double the purchase price of equip- Brazil an estimated 20 to 40 percent of the $2 billion ment. Developing countries can cut costs by to $3 billion worth of public sector medical equip- adopting competitive purchasing methods or by ment is not working. A study of twelve Kenyan purchasing equipment from international agen- hospitals in 1984 found that sterilizers operated for ciessuch as UNICEF, Equipment for Charity an average of two years instead of the six expected Hospitals Overseas (ECHO), and the International and that incubators lasted only two years rather Dispensary Associationthat offer procurement than eight. Equipment failed prematurely because services for some medical equipment at competi- maintenance budgets were only about 1 percent of tive prices. the value of the capital stock (10 percent might be There are several reasons for government in- considered optimal). In Viet Nam 39 percent of volvement in the development of health infra- urban health centers and 29 percent of urban poly- structure. The government itself, as a provider of clinics surveyed in 1991 lacked a working ster- health services, may finance and use infrastruc- ilizera critical piece of equipment for developing ture. It may also intervene to compensate for mar- countries that have to reuse such supplies as ket failures that can lead to greater investment, syringes. particularly in specialized health inputs, than is Investments in medical equipment can be ratio- socially optimal. Finally, the government has a role nalized by controlling the purchase of expensive, in undertaking technology assessment of medical sophisticated equipment and rejecting most do- equipment, which is a costly public good. nated medical equipment, new or used. To contain To reduce both capital and recurrent costs with- costs, Belgium, France, and Portugal directly con- out sacrificing quality of care, governments can: trol the acquisition of state-of-the-art medical tech- Reallocate public spending toward the facili- nologies by both the public and the private sectors. ties and equipment required for providing public In Canada major capital acquisitions require prior health programs and essential clinical services. approval by the provincial or territorial ministry of Improve the efficiency of installed capacity by health on the basis of a needs assessment and considering alternative uses of facilities, as well as other factors. Alternatively, governments can en- new diagnostic and treatment protocols. (Exam- courage public hospitals to make tough choices by ples are the conversion of some costly acute care limiting their budgets. Even assuming that do- capacity to less costly extended care beds and nated equipment meets local equipment require-. treatment of some surgeries on an outpatient ments, very little of it ever becomes operational, basis.) Such reconfiguration may require modest for a variety of reasons, including missing or dam- new investment. aged parts, lack of disposable inputs and of user When cost containment is a concern, consider and service manuals, and problems with power controls on the purchase of expensive, specialized supply. Standardization of equipment could sim- technologies, whether by public or by private plify management and maintenance and reduce providers. inventory costs. Purchasing decisions could be an- Support and disseminate technology assess- alyzed on a life-cycle cost basis. In many cases im- ments to purchasers. 138 Reduce or eliminate subsidies to private in- too, has long relied on graduates of three-year (in- vestors in facilities and equipment. stead of five-year) medical schools to meet the needs of rural areas. Addressing imbalances in human In some countries tasks traditionally performed resources by physicians have been successfully delegated to lower-level primary care providers as a way of im- Nearly all countries face the same fundamental proving the efficiency of health services. By spe- problems with human resources in the health sec- cializing in certain common procedures (as mid- tor. There are not enough primary care providers wives specialize in deliveries, for example), such and too many specialists. Health workers are con- providers may become better at their tasks than a centrated in urban areas. Training in public health, generalist physician. Surgical technicians in health policy, and health management has been Mozambique perform hysterectomies and ce- relatively neglected. Medical training is subsidized sarean sections and remove ectopic pregnancies. even though physicians may earn high incomes Some nongovernmental organizations (NGOs) in and many work in the private sector. Bangladesh use graduate nurses to do steriliza- There are several ways in which governments tions, and in Thailand public sector nurse-mid- can do something about these problems. Public wives perform this procedure. In these cases, eval- sector pay and employment policies can be im- uations indicate no differences in outcomes proved to be more competitive with the private compared with procedures done by physicians. sector and to relate pay to performance. Career Ophthalmic clinical officers, who are not physi- development paths and in-service training are cians, have performed cataract surgery in Kenya needed to retain staff, especially in managerial po- on a pilot basis, and evaluations indicate accept- sitions. Policies on accreditation and licensing can able results. Africa has only one ophthalmologist be used to limit enrollments in training programs, per 1 million people; without the use of nonphysi- to shape curricula (all physicians might spend time cian services, many patients would not be able to in rural practice during their medical training or be get cataract surgery. required to pass examinations in public health), The distribution of nurses and physicians by re- and to set minimum standards for providers. Edu- gion is shown in Figure 6.3. Appropriate staffing cation finance policies can be used to curtail educa- ratios depend heavily on the organization and f i- tion opportunities for physicians and specialists nancing of care and the specific tasks health per- and to expand them for workers in primary care, sonnel carry out. Health maintenance organiza- public health, health policy, and management. But tions in the United States, for example, operate where oversupply is greatest, as for specialist phy- with about 1.2 physicians per 1,000 enrollees, com- sicians, the only effective solution may be to set pared with about 4.5 in the fee-for-service sector. quotas for training, or at the very least for publicly Evaluations of health outcomes and user satisfac- subsidized training. tion indicate that these savings in resources do not come at the expense of quality. Sub-Saharan Africa Improving the balance between primary care providers has the fewest physicians and nurses of any re- and specialists gion, which is an obstacle to the delivery of the public health interventions and essential clinical A central role in delivery of most cost-effective services described in Chapters 4 and 5 because health interventions belongs to primary care pro- some of the existing personnel are providing other viders, a category that can include physicians, services. The public health and minimum essential nurses, nurse practitioners, or midwives, depend- clinical interventions require about 0.1 physician ing upon how the jobs are defined. Nonphysician per 1,000 population and between 2 and 4 gradu- primary care providers have many advantages. ate nurses per physician. Given resource con- They cost less to train (data from Myanmar, straints, however, the relatively high ratio of Pakistan, and Sri Lanka indicate that between 2.5 nurses to physicians in Sub-Saharan Africa is a and 3 nurses can be trained for the cost of training good sign. There is no optimal level of physicians one physician), and they receive lower salaries. per capita or optimal nurse-to-physician ratio, but They are easier to attract to rural areas and usually a rule of thumb is that nurses should exceed physi- communicate more effectively with their patients. cians by at least two to one. (The ratio is five to one In Sub-Saharan Africa, where the few local physi- in Africa but well under two to one in China, cians are concentrated in urban hospitals, nurses India, Latin America, and the Middle Eastern often function as primary care providers. China, crescent.) 139 The availability and mix of health personnel vary widely across regions. Figure 6.3 Supply of health personnel by demographic region, 1990 or most recent available year Ratio of nurses and midwives Physicians per 1,000 population to physicians 5 6 D Sub-Saharan Africa fl Other Asia and islands D India O Middle Eastern crescent O Latin America and the Caribbean o China I 0 Established market economies Formerly socialist economies Source: See Appendix table A.8. PHYSICIAN OVERSUPPLY. During the 1960s and nonmedical jobs, and 11 percent were in low- 1970s many governments encouraged, primarily income medical jobs or were seeing very few pa- through subsidies to education, rapid expansion in tientswhich is a concern because the physicians physician training to meet the need for primary may not see enough patients to maintain their care providers. In many countries the excess of competence. The quality of medical education also physicians in relation to nurses and of specialists declined with the rapid growth in medical schools. in relation to other physicians has created prob- Furthermore, the expansion of medical training lems. By the early 1980s the established market did nothing to solve the problem of attracting phy- economies, Latin America, and parts of Asia were sicians to rural areas. In 1983 an interinstitutional having trouble absorbing growing numbers of body was created, with representatives from the physicians. These policies have been costly, and it ministries of health and education, health care in- will take many years to correct the imbalances. stitutions, and universities. This group has, by Mexico illustrates the problems. Medical enroll- agreement, reduced enrollments and contained ments in 1970 stood at about 29,000 in twenty- the number of medical schools. More recently, the seven schools. Within ten years there were 93,000 government has begun publishing average exam- in fifty-six schools. Many of the schools offered ination scores of medical school graduates by highly subsidized or free tuition, and some of the school to provide information on educational qual- largest had open enrollment policies. At the same ity for prospective students and employers. time, health services were growing only modestly. Other countries responded to physician over- In 1960 there were 20,600 physicians in Mexico; by supply by restricting medical immigration (Canada 1990 there were 166,000. A survey of physicians in and the United Kingdom), by reducing working major cities in 1986 revealed that 7 percent were hours (Denmark), and by indirectly promoting unemployed, another 11 percent were working in outmigration of medical personnel. The last two 140 Box 6.1 International migration and the global market for health professionals Over the past several decades, large numbers of physi- trained human resources can cause shortages of health cians and nurses have migrated across national bor- workers. In Jamaica vacancy rates of more than 50 per- ders. WHO estimates that 14,000 nurses did so in the cent in nursing positions, in large part because of mas- early 1970s and that in 1972 more than 140,000 physi- sive migration of nurses, have forced the Ministry of cians (or 6 percent of the total) resided outside the Health to close whole wards and to reduce the services countries in which they were born or had been trained. offered in many facilities. In addition, emigrating Over the past half century the main flow of physicians health workers deprive their own countries of the and nurses has been from developing to industrial benefits of (often state-financed) investments in their countries. Developing countries donate a full 56 per- education. For example, the 111 registered nurses who cent of all migrating physicians and receive less than 11 resigned from government service in Jamaica in 1990 percent. The principal donating countries for physi- took with them nearly $1.7 million in government in- cians are India and the Philippines. More than 90 per- vestment in training and education. cent of the nurses who migrate go to North America, Europe, and the high-income countries of the western Policy responses Pacific, while only about 7 percent go to developing countries. Migrating nurses come overwhelmingly In an attempt to alter the patterns of migration, many from the Philippines, which exports each year 2,000 to countries have changed their immigration and licens- 3,000 nurses, many of whom go to North America. In ing laws and regulations. During the 1980s, for exam- 1970 more Filipino nurses were registered in Canada ple, the United States, to address its own nursing and the United States than in the Philippines, and the shortage, changed its policy on immigration of nurses, trend has continued to the present. Other major coun- making it relatively easy for nurses wishing to come to tries of origin for migrating nurses are Australia, Can- the United States to obtain a visa. This had a profound ada, the United Kingdom, and certain West Indian effect on a number of neighboring countries. In the countries. Philippines during the same period it became increas- ingly difficult for registered nurses to obtain travel doc- Consequences uments because of the enormous outflow of nurses from that country. The migration of health professionals has both positive Short-term immigration restrictions, however, may and negative effects. It can help alleviate shortages in have only a limited effect. Other possibilities for en- the receiving countries, and large remittances or tax couraging health professionals to remain in their home revenues from overseas workers can improve the stan- countries include reforming education finance to re- dard of living in the countries of origin. (The Philip- quire that individuals repay some or all of the costs of pines received an estimated $680 million from expatri- state-financed training, through student loans or en- ate workers in all fields in 1986, and an estimated $8 forced service bond requirements. And publicly fi- billion in remittances went to developing countries as a nanced opportunities for overseas training could be re- group in 1975.) On the negative side, the net outflow of stricted because of its tendency to lead to outmigration. L solutions waste valuable resources (see Box 6.1). In cedures. This, in turn, pushes up health care costs some countries the government can limit enroll- and reduces the quality of care. The United States ments in medical schools directly. Egypt has re- has the highest number of cardiologists and car- duced medical enrollments by half since 1982; in diac operating suites per capita in the world and France the Ministry of Health used quotas to cut correspondingly higher rates of surgerya signifi- new enrollments from about 11,000 in 1975 to less cant proportion of which is inappropriate. than 5,000 in 1989. In other countries, such as Ger- While practice guidelines and incentive struc- many and Mexico, universities have autonomy in tures can be important policy tools for curbing determining enrollments, and cuts must be made overuse of procedures, training fewer cardiologists through consensus or through education finance would also help. A classic U.S. study showed that policies. a 10 percent increase in surgeons would bring about a 3 to 4 percent increase in surgical opera- CURTAILING SPECIALIST TRAINING. Most govern- tionsthe phenomenon of "supplier-induced ments will need to limit not only total enrollments demand." but also the training of medical specialists. High Even when specialists function as generalists, numbers of specialists tend to increase the fre- they have more costly styles of practice, ordering quency of unnecessary and often risky pro- more tests and procedures. Analysis of geographic 141 variation in expenditures in the United States indi- slow. The fifty-five member institutions of the in- cates that expenditures on physician services are ternational Network of Community-Oriented Edu- unrelated to the total number of physicians per cational Institutions for Health Sciences have capita but are related to the ratio of primary care adopted curriculum reforms that emphasize com- physicians to specialists. Many OECD countries munity-based and problem-based learning. The limit the number of specialist training oppor- goal is to produce graduates whose competencies tunities. This policy instrument is increasingly and experience correspond closely to community relevant for middle-income countries interested in health needs. Significantly larger proportions of cost containment. Subsidized medical education graduates from these schools have followed ca- has already led to overproduction in some middle- reers in primary care, Similar reform efforts have income developing countries such as Chile, where taken place in nursing education. In Nigeria, Sene- 75 percent of all physicians are specialists, and gal, and Uganda (for basic nurse training) and in Venezuela, where about 55 percent of all physi- Thailand (for public health nurse practitioners) the cians employed in the public sector are specialists nursing curriculum has been oriented more to- In contrast, only 25 to 50 percent of physicians in ward community settings and preventive services. Belgium, France, Germany, and the Scandinavian countries are specialists, and regulatory bodies Attracting primary care providers to underserved areas and committees determine the number to be trained. Health providers are concentrated in urban areas. Few, if any, specialists are needed to deliver the Professional isolation, lack of additional work op- cost-effective clinical interventions discussed in portunities, substandard housing, and other dis- Chapter 5, even with a modest expansion in con- amenities often make staffing rural health facilities tent beyond the minimum essential package. difficult. If public sector wages cannot be in- Some specialists are required for services outside creased, other methods must be found to increase the essential package. The overall proportion of the attractiveness of rural posts. Many countries physician generalists to specialists is an important require a period of rural service following publicly indicator for governments to monitor, but this in- financed medical training. Canadian provinces formation is not at present readily available in have used many incentives, including differential many countries. A reasonable benchmark for the pay scales, settlement allowances, payment of ex- maximum proportion of specialists to physicians penses for continuing education, and provision of in developing countries might be 25 percent, scholarships for later study in return for a certain which is about the lowest proportion found in the number of years of service. established market economies. In many develop- In some settings lack of female health providers ing countries the proportion could be much lower, is an obstacle to utilization of health services. In given the epidemiological characteristics of the Egypt, for example, most physicians are male, but population and the smaller share of the population cultural beliefs constrain women from being seen using clinical services beyond the essential pack- after puberty by men who are not family mem- age. Public regulation and rationing of specialist bers. Even when trained, female primary care pro- training, in addition to the elimination of training viders are hard to attract to underserved areas be- subsidies, may be needed to achieve this. cause of security concerns and the importance of living with their families. The Aga Khan Develop- CONTENT OF TRAINING. Primary care training ment Network in Pakistan, recognizing this prob- should include, at a minimum, the skills necessary lem, has trained women to work in their own com- to provide the essential clinical services discussed munities as lady health visitors. in Chapter 5. In fact, however, basic curricula in Community health workers can complement the medical schools often fail adequately to cover work of primary care providers in rural areas. some of these services, such as family planning Burkina Faso, the Gambia, Ghana, and other services and the proper diagnosis and treatment of countries have trained large numbers of commu- sexually transmitted diseases (STDs). For more nity health workers as part of the national strategy than two decades there have been calls to stop for primary health care; in many other settings training health professionals in high-technology much smaller programs have been set up by NGO tertiary institutions and to expose them thor- groups. Evaluations sometimes show disappoint- oughly to health problems and practice at the ing results: community health workers have often grass-roots level. But progress has been extremely had little impact on health service utilization and 142 Box 6.2 Community health workers Over the past twenty years many countries have exper- to the health system, but their availability to the com- imented with the use of community health workers munity diminished. The program has since been (CHWs) to provide primary health care. Several Afri- greatly reduced. can countries introduced CHW programs in the 1970s Other efforts have been more successful. Perhaps as a way of extending primary health care services at the largest scale NGO-run community health worker low cost nationwide. Health workers' responsibilities program is the Pastoral da Crianca, operated by the typically include providing education on sanitation, Catholic Church in Brazil. This program, initiated in nutrition, family planning, child health, and immuni- 1983, receives strong support from the Ministry of zations, in addition to carrying out some basic health Health and some technical and financial support from interventions. They can also be valuable as a referral UNICEF and from the Bernard Van Leer Foundation point between health centers and the community. Re- and other NGOs. It now has 47,000 CHWs throughout grettably, CHW programs have had mixed results. Brazil. An estimated 1.5 million children were enrolled Studies have shown that in the Gambia and Indonesia in the program in 1992. CHWs provide health educa- traditional birth attendants who were not backed up by tion to low-income mothers regarding the importance skilled services were unable to decrease the risk of ma- of prenatal care, good diet during pregnancy, breast- ternal mortality. feeding, proper weaning, immunizations, and man- A Jamaican program, launched in 1977, that used agement of diarrhea, and they monitor the growth of CHWs in primary health care efforts is an example of a infants and young children. The training process for well-intentioned effort gone awry. Problems emerged CHWs follows a central guideline but is adapted to fit from the beginning, with the selection of personnel. the characteristics of different regions. Special care is CHWs generally demonstrate greater dedication when given to the training programs for illiterate volunteers, they serve the communities in which they live. Unfor- and supervision of CHWs is closely integrated with tunately, too few CHWs were recruited from the target continuing education and motivational support. An communities, and workers who lived elsewhere had to evaluation carried out in 1990 found that health and be enlisted. Inability to recruit male volunteers limited nutritional indicators for young children enrolled in the the success of family planning and STD-prevention program were significantly better than indicators from programs. The CHWsa large groupsought and ob- similar communities in which the Pastoral da Crianca tained civil service benefits, including a set salary struc- had no activities. ture and promotional opportunities. In 1985 salaries for Community health workers are also central to the briefly trained CHWs were to be equivalent to two- successful Aga Khan Health Service primary health thirds those of registered nurses with three years' care programs in remote mountainous areas of rural training. Health center buildings were altered to serve Pakistan. The CHWsvolunteers selected by the vil- as bases for CHW operations. Shortages of higher-level lagerscollect epidemiological information, provide staff prompted many health centers to substitute health education, identify problems, and provide sim- CHWs for nurses, even though the workers lacked the ple treatment and referrals. They are backed up by mo- necessary training. CHWs became increasingly linked bile teams of physicians and nurses. health indicators (Box 6.2). These same evalua- These skills are in short supply in most developing tions point to four necessary (but difficult) condi- countries. Public health often receives little atten- tions for success: community health workers must tion in basic medical curricula, specialty training is be well trained, well supervised, well provided often inadequate, and courses in public health with logistical support, and linked to well-func- schools may be too academic and not relevant to tioning district health systems for referral when local problems and needs. In Sub-Saharan Africa, needed. where public health capacity is weakest, fewer than 100 people receive specialty training in public Increasing training in public health, management, health annually. Some countries are exploring and policy, and planning implementing multidisciplinary training programs that include management and communication Improvements in health systems performance can techniques as well as the traditional public health be facilitated by training adequate numbers of poli- sciences. An innovative example of public health cymaking and management personnel, including training designed to produce future leaders is the public health specialists, policy analysts, hospital Union School of Public Health in Beijing, estab- managers, and drug management specialists. lished in 1989 to stimulate public health training in 143 the entire country. The school offers a master's in a priority sector (such as primary care or public degree in public health and draws students and health) or in an underserved location. Not only teachers both from health disciplines and from would professionals be better distributed and economics, management, and the social and envi- used, but there would be substantial savings of ronmental sciences. Training is based on problem public resources. solving, and more than half of the educational ex- Almost every country today is grappling with perience is in the form of community service. In problems in the mix and quality of its health pro- Zimbabwe, under a new public health training fessionals. Government financial policies can play program, students spend 75 percent of their time a constructive and central role in correcting market in the field. failures that lead to distortions in access to training Health policy and planning and good manage- and in the supply of professionals in different ment are fundamental (albeit insufficient) condi- fields. (For example, if credit is not widely access- tions for better performance of health services. ible, only the better-off may be able to go to medi- Over the past thirty years the role of managers, cal school; if the private rate of return for a certain economists, and planners in health services has specialty greatly exceeds the social rate of return, expanded in the industrial countries. For example, more professionals may choose that field than in many of these countries professional (nonphysi- would be socially optimal.) Government policy cian) hospital managers commonly run hospitals, can: in contrast to developing countries, where hospi- Help meet the need for training primary care tals tend to be run by physicians. As developing providers and other health professionals by im- countries seek to boost efficiency and as they move proving capital marketsusing student loan pro- toward decentralized management of health facili- grams, where feasibleand through national ser- ties, the need for trained managers increases. In vice mechanisms. most developing countries, however, training pro- Increase spending on training of, and im- grams in these areas are poorly developed. prove public sector wages and benefits for, health Distance education can facilitate training in pub- professionals in areas in which social benefits cur- lic health, health economics, and management by rently exceed private returns. These include, in allowing rapid implantation of what are often new particular, nonphysician primary care providers, curricula without the time-consuming task of health care managers, and staff in rural areas. training a new generation of teachers. Distance Limit or eliminate subsidies and financial in- learning has been used, for example, to build centives for specialist training. health research capacity in China. The University of Newcastle in Australia, in collaboration with Improving the selection, acquisition, and use Chinese universities, has set up a postgraduate of drugs distance-learning program in clinical epidemiol- ogy. The printed materials and academic stan- Drugs and vaccines embody much of the power of dards of the distance-learning program are equiva- modern medicine. Governments can enhance lent to those in the Australian program. Chinese their own utilization of drugs and assist the private professors help the students with applied labora- sector in increasing its efficiency through policies tory and research work. that improve selection, rationalize acquisition and production, and encourage better use. Through Reforming the finance of health training drug regulation and the development of a national list of essential drugs of established cost-effective- Many of the problems with human resources in ness, governments can help providers and con- the health sector derive from the fundamental flaw sumers make better choices among the approxi- of public subsidization of medical training. If phy- mately 100,000 different drugscomposed of more sicians paid the full costs of their training, it would than 5,000 different active substancesnow avail- be of no concern if they were later employed in able worldwide. Governments can encourage nonmedical work. Public subsidies could be specif- health systems to buy drugs of assured quality ically targeted to encourage those training and ca- from the lowest-cost supplier, whether domestic reer choices that are in the public interest. or international. They can eliminate the incentives Student loans could replace most of the current that in many countries induce physicians to over- public subsidies for training. Repayment of loans prescribe drugs because of the profits they earn might then be forgone if the trainee agreed to work from directly dispensing them. In China, Japan, 144 and Korea such incentives helped to drive drug therapeutic products. Evidence from the United spending up to 35 to 50 percent of total health Kingdom and other countries shows that the spending. adoption of formularies can contribute to consider- able savings in drug costs if physicians are in- Selecting essential drugs volved in their development and are educated about the results. The Model List of Essential Drugs developed by Governments are also responsible for carrying WHO suggests a basic list of drugs that WHO con- out regulatory functions to ensure that all drugs siders important and effective for dealing with on the market are of acceptable quality, safety, and health problems in developing countries. First efficacy. Building up a national regulatory author- drawn up in 1977 by art expert panel, the original ity requires the creation of a core group of trained list has been revised and updated seven times and staff, enactment of supporting legislation for ad- now includes about 270 products. It is designed to ministrative drug review, and the establishment of serve as a template from which countries can de- quality assurance laboratories. These are impor- velop their own still more specific lists of essential tant areas for donor assistance and perhaps for drugs. internationally shared efforts. Drugs on the national essential list are intended to be available at all times and in the appropriate Acquiring and producing drugs dosage forms in publicly provided health services. At the health center level about thirty to forty In 1990 the public and private sectors in develop- drugs can treat almost all complaints. District hos- ing countries spent an estimated $44 billion, or $10 pitals require no more than 120 drugs. If properly per capita, on pharmaceuticals. Global expendi- purchased, these drugs tend to be relatively inex- tures on pharmaceuticals amounted to about $220 pensive; almost all have multiple suppliers on in- billion, or $40 per capita. Total expenditures on ternational markets. Drugs are listed by interna- human vaccines, excluding those made in devel- tional, nonproprietary (generic) names. Although oping countries, were between $1.6 billion and many countries have created these essential drug $2.0 billion in 1992. Drug expenditures vary lists, only a few have used them to guide purchas- widely, from a low of $2 per capita in parts of Sub- ing and management of public sector (or publicly Saharan Africa and in Bangladesh to a high of $412 financed) drug supplies. And occasionally national in Japan (Table 6.1). drug lists have omitted important products, par- ticularly contraceptives. Table 6.1 Annual drug expenditures per Bangladesh and Sudan use limited lists not only capita, selected countries, 1990 to select drugs for public financing but also to Expenditure guide the national drug registration process, Country (dollars) thereby affecting the mix of drugs available in the Japan 412 private sector as well. Norway has limited the Germany 222 number of drugs registered by incorporating cost- United States 191 effectiveness, among other factors, into the review Canada 124 process. Since 1991 Zimbabwe has used its na- United Kingdom 97 Norway 89 tional list to determine which drugs can be im- Costa Rica 37 ported by the private sector without a permit. Chile 30 The applicability of the essential drug concept is Mexico 28 not limited to developing countries; drug formul- Turkey 21 aries, which are detailed lists of essential drugs, Morocco 17 Brazil 16 are widely used by institutional health providers Philippines 11 (public or private) and insurance companies in in- Ghana 10 dustrial countries. The formulary contains the China 7 names of drugs that are approved or recom- Pakistan 7 mended for health providers and supply systems. Indonesia 5 Kenya 4 It also provides useful information for individual India 3 prescribers. In creating formularies, drugs are as- Bangladesh 2 sessed on the basis of their safety, effectiveness, Mozambique 2 and cost-effectiveness in comparison with other Source; Ballance, Pogany, and Forstner 1992. 145 In most established market economies phar- cedures, and selection of generic drugs on the maceuticals and vaccines account for between 5 basis of its national essential drug list. This, of and 20 percent of health care spending, and, ex- course, is facilitated by Costa Rica's political stabil- cept in Canada and the United States, more than ity. In 1986 several Caribbean islands joined to- half of all drug expenditures are publicly financed. gether to carry out international tenders through In developing countries, households' out-of- the Caribbean Development Bank. In the first year pocket expenditures make up a much larger pro- they saved 44 percent over previous prices. portion of total drug spending. In Côte d'Ivoire The first step toward efficient procurement is and Pakistan, more than 90 percent of household careful quantification of drug and vaccine supply health expenditure is devoted to drugs. In the needs over a given period, using essential drug public sector drugs generally account for between lists or formularies where possible. Large stocks of 10 and 30 percent of total recurrent costs, making low-priority drugs have high opportunity costs: them the second largest category after salaries. they tie up resources and may expire before they Given this high volume of expenditure, achieving can be used. Shortages of high-priority drugs are the substantial improvements in efficiency of also costly; emergency purchases from local sup- procurement that are possible becomes a high pliers are always expensive. Good forecasting per- priority. mits economical purchasing. Some governments and many donors purchase PURCHASING DRUGS AND VACCINES EFFICIENTLY. drugs through international agencies (see Box 6.3). Some countries have achieved savings of 40 to 60 These agencies use international tendering and, percent in pharmaceutical expenditure by improv- because of the scale of their purchases and their ing selection and by competitive purchasing. For low operating margins, pass on very low prices. example, for several years the Costa Rican social (The total amount of drugs procured in this way is, security agency has been able to purchase drugs at however, small in relation to total drug expendi- approximately half the price of its counterpart in- tures in developing countries.) UNICEF purchased stitutions in other Central American countries, about $160 million worth of pharmaceuticals, vac- partly because of its use of centralized purchasing, cines, and related supplies for developing coun- more open and transparent purchasing pro- tries in 1992. Ethiopia, Sudan, Tanzania, and Box 6.3 Buying right: how international agencies save on purchases of pharmaceuticals UNICEF and several nonprofit organizations offer pur- plies on behalf of governments and nonprofit organiza- chasing services that enable countries to obtain favor- tions in more than eighty developing countries. Its cur- able prices for drugs, vaccines, and some medical rent annual turnover amounts to $80 million. IDA also equipment. UNICEF, the biggest in the field, has sup- carries out quality assurance, checking that manufac- plied basic drugs and vaccines since the 1960s. In 1983 turers produce in accordance with internationally ac- it issued its first international invitation to tender for cepted standards. When the drugs are received, IDA the bulk purchase of pharmaceuticals for Tanzania. The tests samples for quality and verifies labels and certifi- prices quoted against the invitation to tender were up cates of analysis. to 50 percent lower than previous price quotations. As Price lists from UNICEF and IDA provide valuable a result of this experience, UNICEF has continued to market information for countries' own procurement. use international tendering for the bulk purchase of Competitive tendering in Mali reduced prices by 40 pharmaceuticals and to pass on these favorable prices percent. In Kenya bulk purchasing of carefully selected to developing countries. UNICEF contracts with the essential drugs was estimated in 1985 to save nearly 40 Danish National Board of Health to provide advice on percent (or $700,000) of the annual drug bill for church quality assurance for pharmaceutical products. In 1992 health institutions. In 1992 the Chinese government UNICEF's purchases of drugs ($61.2 million), vaccines carried out international competitive bidding for drugs ($63.6 million), and refrigeration equipment, syringes, for tuberculosis treatment and_perhaps because of the needles, and sterilizers ($33.4 million) were delivered very large scale of procurement involved, the low-cost to more than 120 countries. packaging requirements, and the desire of manufac- The International Dispensary Association (IDA), es- turers to enter the Chinese marketachieved savings tablished in 1972, is a nonprofit supplier of drugs to of about 70 percent of UNICEF's published prices. developing countries. IDA procures drugs and sup- 146 Zambia have all relied heavily on nonprofit inter- Brazil, Indonesia, and Turkeyhave primary man- national drug suppliers. ufacturing capabilities or the ability to produce But many other developing countries fail to take both therapeutic ingredients and finished prod- advantage of international competition or interna- ucts. But most developing countries either have tional agencies. Purchasing methods, as well as only the capacity to produce finished products import restrictions, tend to restrict competition from imported ingredients or have no manufactur- and thereby raise prices. In addition, price compe- ing capability whatsoever. (Countries in the latter tition is restricted by the industry's extensive drug group are typically very small.) promotion practices and, in the case of patented Except in the largest countries that have primary products, by monopoly power. Some countries, manufacturing capabilities, local pharmaceutical such as Venezuela and Zimbabwe, protect local production in developing countries is likely to pharmaceutical industries from international com- make sense only for intravenous fluids, which petition (imported drugs will not be approved for have relatively high transport costs; for local pack- import and sale if there is a local producer), and aging of bulk imports in finishing plants; and for Belize and other countries impose import tariffs packaging of oral rehydration salts. Even in these even if there is no local production. This results in activities local production may be inefficient and great variation in prices for pharmaceuticals and waste scarce resources. State-run drug and vaccine supplies in developing countries. Cross-country companies, from which the public sector pur- data on the retail price of condoms show remark- chases preferentially, are common in many coun- able variation: condoms cost only $2 to $3 per 100 tries, including Bangladesh, Brazil, India, and in China, Egypt, and Tunisia, $15 to $30 per 100 in Laos. In some countries the local pharmaceutical Costa Rica, Ecuador, and Mexico, and more than industry (public or private) produces drugs that $70 per 100 in Brazil, Burundi, Myanmar, and Ven- could be purchased less expensively elsewhere. ezuela. This price variation is attributable to a com- Such industries survive only because of the protec- bination of factors, including import tariffs, import tion accorded through the prohibition of compet- restrictions, and wholesale and retail marketing ing imports, through import tariffs, or through structures. guaranteed agreements for public purchase re- Some countries purchase directly from a few lo- gardless of price. cal suppliers because of liquidity constraints. Inter- The combination of protection and poor regula- national agencies do not extend credit, and they tion can be particularly damaging. A 1990 study of require payment in hard currency. Local suppliers more than 6,000 infants in Bangladesh revealed often extend credit in exchange for significantly that the mothers' tetanus toxoid vaccinations did higher prices. Changing this practice to take ad- not reduce the risk of tetanus. Subsequent testing vantage of benefits from competitive procurement in reference laboratories of Bangladesh-produced would require the ministries of both health and vaccine indicated no potency in several consecu- finance to make budgetary funds and foreign ex- tive batches, raising questions about the efficacy of change available when needed for large-scale drug the more than 40 million doses already adminis- purchases. Governments can also improve drug tered. Since Bangladesh has no independent na- procurement by passing legislation to facilitate ge- tional control authority for certifying vaccine neric drug prescribing. This can increase the af- safety, all testing had been done by the production fordability of drugs purchased from private facility itself. Evidence suggests that few public outlets. sector pharmaceutical and vaccine producers have been able to operate competitively, in terms of PHARMACEUTICAL PRODUCTION. The cost of de- price and quality, in the highly competitive and veloping a sophisticated pharmaceutical industry rapidly changing pharmaceuticals market. Im- with a significant research base is huge. During proved selection and purchasing practicesrather 1961-90, 90 percent of the approximately 2,000 than protection of drug manufacturingwill usu- "new chemical entities" (new drugs) brought on ally be the best ways to counter the market power the market were discovered in only ten OECD of international suppliers of drugs. countries. Five countries in the developing worldArgentina, China, India, Korea, and Mex- IMPROVING STORAGE AND DISTRIBUTION. Theft, icodiscovered, developed, and marketed at least spoilage, and shortages are major problems facing one new chemical entity between 1961 and 1990. public distribution in many countries. Systems for Several other developing countriesamong them, inventory control, port clearing, storage, and de- 147 livery can address many of these problems. In providers cannot possibly review all the informa- Zimbabwe a standard nationwide system of stock tion available on the quality, safety, and efficacy of control was fundamental to recent reforms in the drugs and vaccines, governmental involvement in drug supply system. Surveys show a gradual im- regulation and in provision of information is nec- provement in drug availability: in 1991 the facili- essary. In addition, the government must manage ties surveyed had 78 percent of the representative drug selection, procurement, and distribution for essential drugs in stock, up from 38 percent in publicly provided health services. To support the 1987. In hospitals, management information sys- rational use of drugs, governments can: tems help to track periods of drug validity and to Develop a national list of essential drugs and analyze rotation rates and drug consumption. direct public finance to those drugs that support the essential package of clinical services and public Influencing prescription and self-medication health interventions. patterns Purchase drugs competitively and reduce or eliminate protection of local pharmaceutical pro- Significant efficiencies can be achieved by improv- duction of vaccines and drugs. These policies work ing prescription and self-medication practices. to consumers' benefit. Efficient local industry is Widespread overprescription and inappropriate best created under competitive conditions. prescription have been documented in most coun- Provide information to public and private pro- tries. For example, recent surveys found that the viders and consumers on drug use and cost-effec- average number of drugs prescribed per single tiveness and establish regulations that discourage consultation in public health centers ranged from overuse or overprescription. 1.3 in Zimbabwe and Ecuador to as high as 3.3 in Indonesia and 3.8 in Nigeria. These surveys also Generating information and strengthening documented that unnecessarily high proportions research of drugs were being administered in the form of injections (which carry the risk of abscesses, nerve In health, as elsewhere, good information facili- injuries, and transmission of infectious disease) tates sound decisionmaking. Although some basic and that extensive overuse of antibiotics was oc- health information is generated by the private sec- curring. A survey of seventy-five pharmacies in tor without government involvement, the govern- three Asian countries found that only sixteen gave ment has a central role in requiring, standardizing, appropriate advice regarding oral rehydration for and financing the collection, analysis, and dissem- treatment of diarrhea in infants. ination of health information, as well as in financ- Public policies for improving prescription and ing health systems research. Governments are al- medication practices include: ready heavily involved in data collection. Distribution to health care providers and Unfortunately, the data are often irrelevant to p01- pharmacists of regularly updated essential drug icy and program design. And too often, the private lists or formularies that include descriptions of sector is ignored when statistics are being gath- use, dose, adverse reactions, and costs; examples ered. Revamping health information systems is an include the British National Formulary and the attractive investment, both because it is relatively Uganda Drug Formulary inexpensive and because poor decisions based on Strengthening of medical and nursing train- inadequate information can be very costly. But the ing regarding pharmacology, appropriate prescrib- impact of information systems depends crucially ing practices, and problems caused by over- on the decisionmaking environment. Even the prescription and unnecessary use of injections best systems may be seen as irrelevant if managers Public education on appropriate drug use, the have no incentive or scope for using information to disadvantages of injections when oral doses are improve efficiency. Information helps guide available, and the importance of compliance with choices among the existing options, and invest- the full course of therapy ments in research and development create new Removal of financial incentives that encour- options, both for households and for providers of age physicians to overprescribe. care. It can be argued that investments in research Unlike facilities, equipment, and human re- have been the source of the enormous improve- sources, pharmaceuticals and vaccines are an area ments in health in this century. This section dis- in which government policies can alter input use cusses ways of ensuring continued benefits from relatively quickly. And good policies could make a research, as well as the role of the international significant contribution. Because consumers and community in this task. 148 Understanding health status and health risks Some countries have established surveillance systems that rely on sentinel districts selected to be An essential step toward improving health is to understand the distribution of disease, death, and roughly representative of the country. To improve disability. This requires the systematic collection, the speed and accuracy of reporting, data collec- analysis, and dissemination of timely and accurate tion systems are upgraded in these districts to a information on mortality, morbidity, and risk fac- greater degree than could be done for the country tors. Such data are a cornerstone of public health as a whole. Cause-specific death rates, vaccine efforts in any country, and the government's role coverage, the effectiveness of vaccines, and the is central in creating them because the private sec- impact of specific health interventions are then tor has little interest in producing such public monitored intensively within the district. National goods. Epidemiological data are used to estimate household surveys can also generate a wealth of the magnitude of health problems, study risk fac- information on health status, risk factors, and the utilization of health services according to age, sex, tors, evaluate health programs and the effective- ness of interventions, detect epidemics, facilitate region, and racial and ethnic group (see Box 6.4). planning, and monitor changes in health practices. Unlike government health service statistics, popu- These data could be used to estimate a national lation-based surveys cover nonusers as well as users of public services. burden of disease similar to the global burden of disease estimated for this Report. The national Monitoring health spending and equity burden of disease would quantify the loss of healthy life from the diseases that are important in Previous chapters have recommended redirecting the specific country. It could be used to monitor public spending to nationally defined essential and track over time improvements in both mortal- clinical services, targeted largely to the poor, and ity and morbidity. to public health interventions, leaving to private Box 6.4 The contribution of standardized survey programs to health information Three internationally supported standardized house- immunizations, health care behavior, and other aspects hold survey programs have contributed immensely to of child health. DHS survey information has been used knowledge of health conditions, particularly those of for purposes as diverse as examining the effects of eco- children, in the developing world over the past three nomic reversals on demographic outcomes and study- decades. The World Fertility Survey (WFS) sponsored ing small area variations in child mortality risks in ur- forty-three surveys between 1974 and 1982, with fund- ban areas. ing from the U.S. Agency for International Develop- Neither survey program has collected detailed eco- ment (USAID) and the United Nations Population nomic information on households and communities. Fund (UNPF) and some country contributions toward The World Bank's Living Standards Measurement Sur- the costs of survey fieldwork. The Demographic and vey (LSMS) was designed to fill this need by studying Health Surveys (DHS) program, started in 1984, has so the determinants and interactions of poverty, health, far implemented thirty-nine surveys in thirty coun- education, nutrition, and labor activities. The survey tries; the third phase of the survey program, with a collects a wealth of information about incomes, pro- planned twenty-five surveys, is about to begin. The duction, and prices. Some LSMS surveys are funded DHS program has received funding from USAID, with through World Bank-financed projects, but many have contributions from countries and other donors. received grant support from a variety of bilateral do- Both the WFS and the DHS program have used a nors, the UNDP, and other agencies. common core questionnaire around which special The experience with these standardized surveys in- topics could be explored. The core WFS questionnaire dicates the great value of using comparable survey pro- was primarily concerned with fertility and fertility- cedures and instruments across countries and the im- related behavior; for each eligible woman it included a portance of rigorous supervision at all stages of the birth history, recording the date of each birth and, if survey operation, from sampling to data processing. the child had died, the age at death. This information The LSMS and DHS programs have been particularly base has provided much of what is known about child successful with respect to turnaround time; prelimi- mortality trends and the relationships between child nary findings from a survey are available within six mortality and birth spacing, maternal education, and weeks of the conclusion of fieldwork, and a final report household characteristics. The DHS questionnaire, in typically becomes available within one year. addition to a birth history, includes questions about .- 149 finance health services outside the essential pack- National research priorities age. Private expenditures are always difficult to estimate, but even in the public sector, spending is Governments have a role in supporting the re- poorly disaggregated by use. By revamping infor- search necessary for understanding specific local mation systems, estimates can be made of spend- health problems and for guiding public policymak- ing on public health interventions and on catego- ing and program design. This "essential national ries of inputs (essential drugs, nonessential drugs, health research," which is also undertaken by the primary care physicians, other primary care pro- private sector, examines health strategy in more viders, specialists, health centers, district hospi- depth than is done with day-to-day budgetary and tals, and tertiary hospitals). Although still imper- management information. The international com- fect, such estimates better capture the nature of munity can help both in gathering data for interna- government spending. In addition, public expen- tional comparisons and in assisting local institu- ditures need to be regularly consolidated across tions to build up capacity in epidemiology, health federal, state, and local levels for analysis. In Bra- economics, health policy, and management. Re- zil, where state and local governments account for search priorities in this area include cost-effective- about half of all public spending on health, expen- ness analysis of health interventions, evaluations diture estimates are available only for federal of medical practice and of variations in practice spending (except for 1984). Much less information (see Box 6.5), and studies of drug utilization, eq- is compiled from state and municipal levels, de- uity, consumer satisfaction, and women's health. spite their importance. Household surveys can Where the national burden of disease is high collect appropriate information for monitoring and cost-effective interventions already exist, re- who benefits from public health spending. In part search can guide program implementation. One because such data are lacking, analyses of equity such example is the problem of intestinal parasitic in health care have been carried out in only a worms. How can local programs be best designed handful of developing countries, among them Co- to reach children? How can involvement of school lombia, Costa Rica, Côte d'Ivoire, Indonesia, Ma- officials be fostered? Another area is tuberculosis, laysia, and Peru. where treatment compliance is a chronic problem; Box 6.5 Evaluating cesarean sections in Brazil Operations research can examine variations in medical systematic variations by region, type of hospital, socio- practice with a view to identifying areas in which economic status of the woman, and reimbursement changes in practice are needed, as well as possible in- patterns. Rates in 1981 were higher in the more pros- struments for modifying provider practice. In the early perous Southeast (38 percent) and lowest in the poor 1980s Brazil was estimated to have the highest overall Northeast (20 percent). In every region the incidence of cesarean section rate in the world-31 percent of all cesarean section increased with family income. A 1986 hospital births in 1981. Although cesarean sections are survey showed that rates were highest for women with a life-saving procedure in certain circumstances, their a university education (61 percent) and for births in unnecessary use raises costs and poses medical risks private hospitals (57 percent). Other studies showed for the mother and the newborn. The financial cost of that rates were lowest among women with no insur- unnecessary publicly financed cesareans in Brazil was ance. Women covered under the social security system estimated at about $60 million annually in the late had higher rates of cesarean section, and women with 1980s. Medical risks stem from incorrect estimation of private insurance had the highest. the length of gestation (leading to premature deliv- The country's social security institute changed its re- eries), infection from surgery, and the use of general imbursement policies in the early 1980s to remove anesthesia. Among the many factors responsible for some of the financial incentives for cesarean sections, the rising rate of cesareans in Brazil are the financial and education campaigns for physicians were initiated. and administrative incentives for hospitals and doctors But it is clear that even stronger policies are needed to to perform cesarean deliveries, the desire to use a Ce- reverse these trends, as cesarean section rates have sarean delivery as a vehicle for obtaining a sterilization, continued at high levels. A large sample of births in the and the widespread view that cesarean section is the state of São Paulo in 1991, for example, indicated a preferred, "modern" way to deliver. cesarean section rate of 47 percent. Brazilian studies of cesarean section rates illustrate 150 patients often stop taking medication once they laundry, food preparation, and laboratory testing. feel better, but before the problem has been effec- Systems that gather information on vaccine utiliza- tively treated. What program approaches work tion, equipment and vehicle inventories, preven- best in different settings to ensure compliance tive maintenance for buildings and equipment, with directions? In nutrition, how can policies and personnel management, and the like are also programs promote dietary change most effec- fundamental. tively? Solutions to these problems are not univer- Ministries of health frequently pay little atten- sal. Research must be local, and often public sup- tion to the activities of private providers, instead port is needed. focusing all data collection efforts on public pro- In its 1990 report the Commission on Health Re- viders. To remedy this, governments can collect search for Development recommended the forma- basic information about private providers and the tion of international partnerships or networks to population covered under private insurance plans. focus on ensuring that national health resources They can require standardized reporting from are used to maximum effect. The International both public and private hospitals through uniform Network for the Rational Use of Drugs (INRUD), hospital discharge data. The information can then established in 1989, is one such network. Another be synthesized to provide consumers, health re- is the International Clinical Epidemiology Net- searchers, and communities with information work (INCLEN), which was started in the early about the quality of care given by providers, both 1980s by the Rockefeller Foundation to build up a public and private, and about variations in medical critical mass of researchers in clinical epidemiol- practice. These systems can generate sophisticated ogy, including epidemiologists, health econo- information; consumers in California, for example, mists, social scientists, and biostatisticians. IN- can obtain risk-adjusted mortality rates by hospital CLEN enrolls midcareer academic physicians who for common procedures. But relatively simple hold positions of influence in the medical systems measures can also be useful; an example is ce- of developing countries. It provides overseas sarean section rates, by hospital, which can help study opportunities, support for research, and the identify overuse of this procedure (see Box 6.5). opportunity to participate in annual scientific Such standardized information about hospital per- meetings. The network concept has permitted formance can help consumers make better choices units to share experiences and teaching materials about health care and can help central authorities and to carry out collaborative research between identify problems to be corrected. clinical epidemiology units, training centers, and If there are incentives for using information in the international health community. Capacity decisionmaking, improvements in data gathering building is a lengthy process, but INCLEN has al- can often be inspired simply by giving those who ready influenced health policy. Research on the need the information more training in how to col- effectiveness and efficiency of hepatitis B immu- lect it and more responsibility for doing so. District nization in the Philippines brought about the addi- medical officers, hospital superintendents, and tion of hepatitis B vaccine to the national EPI pro- health care managers are usually not trained to gram. Studies on the cost-effectiveness of short- make the best use of data. Whenever possible, tab- course chemotherapy for tuberculosis have led to a ulation of data should be decentralized so that lo- change in national treatment policies in Brazil, the cal decisionmakers can immediately use the infor- Philippines, and Thailand. mation instead of relying on feedback from central levels. In Papua New Guinea, for example, when Improving information at the district and facility levels local-level staff began to see the relevance of man- Health organizations also benefit from improve- agement information for their work, they sought ment of the information needed to make everyday to verify data and to eliminate reporting that was management decisions. In publicly provided dis- irrelevant. trict health facilities, simple management informa- To summarize, governments have a twofold role tion systems for measuring costs, inputs, and pro- in health information systems and operational re- duction could be helpful for monitoring program search: generating the information necessary to efforts over time and for making decisions about guide health policies and public spending and pro- how to combine inputs efficiently. Yet many public viding certain types of information about provider facilities operate without such information. With- performance that would be too costly for con- out basic data on costs, it is difficult to decide, for sumers to collect. To this end, governments can: example, whether to contract out services such as Gather and synthesize epidemiological and 151 Table 6.2 Some priorities for research and product development, ranked by the top six contributors to the global burden of disease Associated DALY loss (millions) Demographically developing Disease or injury countries FSE and EME Priority areas Perinatal and 125 4 Methods of lowering costs of intervention and maternal causes improving delivery in rural areas. Respiratory infections 119 4 Impact of indoor air pollution on pneumonia (to guide interventions designed to reduce pneumonia by use of improved stoves); inexpensive or simplified antibiotic regimens; inexpensive, simple, reliable diagnostics; pneumococcal vaccine Diarrheal diseases 99 - Rotavirus and enterotoxigenic E. coli vaccines; improved cholera vaccine; ways of improving hygiene; better case management of persistent diarrhea; prevention of diarrhea by the promotion of breastfeeding and improved weaning practices Ischemic heart and 58 27 Low-cost prevention, diagnosis, and management cerebrovascular disease methods Childhood cluster: 67 Development of new and improved vaccines to reduce diphtheria, polio, patient contacts, permit immunization at younger pertussis, measles, and ages, and improve heat stability of some vaccines tetanus Tuberculosis 46 1 Methods of ensuring compliance; monitoring tools for drug resistance; simpler diagnostics; new and cheaper drugs All conditions" 1,210 152 Note: The demographically developing countries are those in the Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent regions. FSE, formerly socialist economies of Europe. EME, established market economies. DALYs, disability-adjusted life years; see Box 1.3. Less than 0.5 million. Total for all conditions presented in Appendix B. Source: Appendix B. other information necessary to monitor health sta- malaria; oral rehydration therapy; antibiotics and tus, detect disease outbreaks, and guide public other antimicrobials; and synthetic hormonal policy and program design contraceptives. Support research, where needed, to generate Basic research and product development are local solutions to local problems public goods that require support through govern- Facilitate standardization of information ment subsidy or intervention (for example, grants about health production and health outcomes by of patents). In addition, because the poor in devel- district health systems and other major health pro- oping countries lack market power, the system of viders; where necessary, synthesize and publicize patent protection fails to provide incentives to the this information to aid consumers in making in- commercial sector for developments related to dis- formed choices about health care. eases of the poor. Thus, there is a clear argument for government and international assistance to cat- Expanding the range of choice alyze technological development. In the develop- ing world many serious health problems do not A revolution in health care technology has taken present sufficiently attractive commercial markets place in the course of this century. Significant bio- to induce the development by private companies medical breakthroughs that have generated inter- of better methods of prevention, diagnosis, and national benefitsfor developing countries as well treatment. Developing countries account for al- as for the established market economiesinclude most 90 percent of the global burden of disease, the development of measles, pertussis, polio, and and much of that burden is from conditions such tetanus vaccines; chioroquine for the treatment of as malaria or tuberculosis that primarily occur in 152 those countries. Only about 5 percent of the $30 eases: measles, tetanus, pertussis, diphtheria, billion global investment in health research in 1986 polio, and tuberculosis. It requires at least seven went to health problems unique to developing patient contacts (two for the pregnant mother and countries. five for the infant). Possible improvements in vac- cine technology would reduce multidose vaccines Setting priorities to a single dose, improve the heat stability of vac- Where is extra research really likely to pay off? cines, simplify administrative requirements (to Table 6.2 suggests priorities for research on pre- permit greater use of oral vaccines as compared vention, diagnosis, and case management for the with injections, for example), create new combina- six conditions that make the largest contributions tions of vaccines to reduce patient contacts, inte- to the global burden of disease. These conditions grate new vaccines into the immunization sched- account for about 40 percent of the DALYs lost in ule, permit vaccination earlier in life to reduce demographically developing countries and for infant deaths caused by vaccine-preventable dis- about 25 percent of the losses in industrial coun- eases, and add to the menu of interventions new tries (where cardiovascular disease accounts for vaccinesfor example, against diarrhea and pneu- much of the burden). If the global burden caused monia. These innovations would reduce some of by a disease is large, if no cost-effective interven- the costs and improve the effectiveness of vaccina- tions exist, and if experts believe that such inter- tion programs. An important source of support for ventions might be developed, there is a case for this research is the Children's Vaccine Initiative greater investment in research and product devel- (CVI), which is identifying measures for catalyzing opment. One example that meets these criteria is technological development in these areas. The inexpensive, simple, and reliable diagnostics for CVI, which has its secretariat at WHO, is an inter- respiratory infections. For problems that create a national effort to harness new technologies that large burden of disease and for which cost-effec- can advance the immunization of children. tive interventions already exist, there is a need to direct efforts more toward program development TROPICAL DISEASES. It is primarily the rural poor and operational research to guide implementation. who suffer from tropical diseases such as malaria, For example, little is known about low-cost schistosomiasis, lymphatic filariasis, onchocer- methods of managing ischemic heart disease in ciasis (river blindness), trypanosomiasis, and lep- developing country settings. One low-cost ap- rosy. These diseases create a high burden, and ex- proach that is being adopted in many industrial isting interventions are inadequate against many countries is the use of low daily doses of aspirin to of them. The UNDP-World Bank-WHO Special reduce the risk of obstructive blood clots inside the Programme for Research and Training in Tropical arteries. This approach, developed on the basis of Diseases (TDR) is developing partnerships with the results of large-scale assessments of the effi- commercial entities, national governments, scien- cacy of the intervention, illustrates the potential tists, and NGOs to support research and drug de- benefits of research on low-cost case management. velopment for these diseases. One strategy the International agencies and governments can program has adopted is to look for new applica- stimulate research on health and product develop- tions of drugs already in use in human or veteri- ment in several ways. They can provide informa- nary medicine. An example is the use of ivermec- tion on potential markets for new products, in- tin, a drug originally marketed by Merck & Co. for cluding epidemiological data about the disease, treating worms in animals, in the fight against on- the target population, and the technical require- chocerciasis in human populations (see Box 1.1). ments of desirable innovations. They can subsi- The TDR program facilitated the field testing of dize a portion of the development costs. They can this product on a large scale for human use. The facilitate or finance field evaluations in a variety of results showed that the drug was very safe, that it settings and support introduction of the technol- could be distributed by primary health care ogy in the field. Finally, they can provide procure- workers, and that one oral dose per year could ment guarantees for new or improved products at prevent or arrest blindness. As a veterinary prod- an agreed-on price. A few examples illustrate the uct, ivermectin has estimated annual sales of $500 potential. million; Merck & Co. agreed to supply the drug without charge to governments for treatment of Nuw AND IMPROVED CHILDHOOD VACCINES. The human onchocerciasis. The TDR's network of in- EPI currently includes vaccines against six dis- ternationally funded research centers in develop- 153 Box 6.6 An unmet need: inexpensive and simple diagnostics for STDs This Report recommends that concerted efforts be stick, developed by the Program for Appropriate Tech- made to develop or strengthen effective programs for nology for Health with support from Canada's Interna- control of STDs. Such efforts will be hampered by the tional Development Research Centre (IDRC) and from challenges of diagnosing STDs, particularly in women, private funds, uses synthetic peptides and a color for whom the vast majority of infections are asympto- change to provide an easily performed test. The test matic. Current methods are often unreliable and ex- takes twenty minutes, requires only three simple steps, pensive, and their use requires refrigeration, electric- is stable for six months at tropical temperatures, has a ity, and sophisticated equipment and training. In pattern of sensitivity and specificity similar to commer- addition, certain tests require patients to return in one cially available tests, and costs less than $0.20. Thus, or two days, which is not feasible when, as is often the the per patient cost for testing can be brought down to case, the patient must travel a long distance to receive less than $1.00, including a confirmatory second test. health care. Even if patients return, the period of infec- This test is now being commercially produced in India tivity is prolonged by this delay in therapy. Syndromic- and Thailand. The Canadian International Develop- based approaches to treating STDs are currently being ment Agency (CIDA) is funding the establishment of a used to bridge this gap and are effective for men. For production facility in Cameroon, and there is interest women, however, these approaches are less accurate. in production in Brazil, Indonesia, and Zimbabwe. New diagnostics that are inexpensive, simple, and The STD Diagnostics Initiative, which is funded by convenient to use and provide rapid, stable, and accu- multiple donors, was established to facilitate develop- rate results would overcome these problems. An exam- ment of appropriate diagnostics for resource-limited ple of such a tool is a new HIV test. The availability of settings. The initiative, being carried out in collabora- HJV testing has been limited by high cost, complexity, tion with industry, clarifies and validates performance and requirements for reagents that need refrigeration criteria for STD diagnostics, organizes and supports and have a short shelf life. Even when labor costs are field trials, provides seed money for the development excluded, testing and confirmation can cost $25 to $50 of new diagnostics, and brokers bulk purchases to cre- (although this cost is declining rapidly). The HIV dip- ate markets of adequate size. ing countries allowed it to respond quickly and cancer, rapid plasma finger-stick diagnostic tools flexibly to the opportunity to test ivermectin. for syphilis, and new diagnostic tests for malaria Women spend up to half of their reproductive for use at the local level. (Box 6.6 provides another lives pregnant or lactating. Many protocols for example.) Rapid diagnostic tests avoid reliance on treating tropical diseases exclude these women other levels of the health system because the and sometimes even large numbers of women health center, if supplied with the necessary who might be pregnant (such as adolescent girls). drugs, can treat the problem on the spot. Innova- Blanket exclusion of pregnant or lactating women tions in medical equipment to reduce the cost or has been the result not of clear evidence of prob- improve the effectiveness of preventing and treat- lems but of reluctance to carry out appropriate ing problems at the health center level are high drug trials on pregnant women. There is an urgent priorities for research and development. need to evaluate drug treatments for such women so that health services can offer them better treat- International aspects ment. This is part of a much broader problem of the common omission of women from medical Some types of research and product development studies and clinical trials in both developing and are costly; it can cost more than $100 million to industrial countries. bring a new drug to market. But several break- throughs in medical technology have been inex- MEDICAL EQUIPMENT. Another priority area for pensive. (One of them, oral rehydration therapy, research and development is the development of is now widely recognized as an effective way of low-cost and efficient diagnostic technologies for treating acute watery diarrhea, which, untreated, use in health centers in developing countries can weaken or kill young children.) The need for where sophisticated laboratories are unavailable. public support for certain types of research is Examples of potentially important new technolo- widely understood. The international community gies are visual methods of screening for cervical has played an important role in supporting health 154 research, and most governments support some re- dertaken by the private sector at socially optimal search as well. Over the short to medium term, levels. (Even research that is internationally fi- developing countries can use their scarce public nanced will take place principally in developing resources best if: countries and will be done increasingly by scien- Governments reduce or eliminate finance of tists from developing countries.) The total invest- basic biomedical research that generates interna- ment in health technology research relevant to the tional benefits (which is best supported by the in- needs of the developing countries is woefully in- ternational community) and redirect it toward re- adequate in relation to its potential benefits. And search efforts that generate primarily national the level of international coordination and cooper- benefits ation falls well short of what is required. An inter- The international community directs research national mechanism with stable funding over the support toward new and improved technologies medium to long term could more effectively build where the expected social returns are highest and research capacity in developing countries. Donors would benefit many countries. and governments also need to give more support International financing is needed for important to activities for testing new technologies and incor- biomedical research when the benefits transcend porating them into health systems. national borders and the research will not be un- 155 An agenda for action The policy conclusions of this Report can be tai- quire donor backing for major reforms in the al- lored to the widely varying circumstances of de- location of public spending for health and in veloping countries. This chapter highlights the pri- health policy more generally. ority policy issues and actions that are likely to be The effectiveness of donor spending can be im- most relevant for three groups of countries: low- proved through increased investment in basic income countries in Africa and South Asia, mid- public health measures and essential clinical care, dle-income countries in Latin America and East steps to strengthen the policy and regulatory Asia, and the formerly socialist countries of Eu- framework for insurance and for delivery of ser- rope and Central Asia. It describes the reforms vices, and backing for research to expand the needed in the health sector and assesses their fea- range of cost-effective treatments available to the sibility, examines the principal obstacles to reform, poor in developing countries. Aid for lower-prior- and outlines possible strategies for overcoming ity items, including tertiary care hospitals and these obstacles. Although policy reform must deal training of medical specialists, needs to be corre- with difficult underlying problems, the experience spondingly reduced or eliminated. of a number of developing countries with imple- Finally, improved coordination among donors menting significant policy changes shows that suc- could raise the effectiveness of aid. Despite the cess is possible. many serious obstacles, the recent experience of a This chapter also examines the role of the inter- number of African and Asian countries shows that national community in supporting improvements such coordination can be achieved. in health policies and programs in developing countries. Despite widespread calls for more do- Health policy reform in developing countries nor investment in human resources and in poverty reduction programs, aid flows to the health sector The policies that this Report suggests should be at declined from 7 percent of total development assis- the top of the agenda for developing countries and tance in the early 1980s to 6 percent in the latter the donor community are summarized in Table half of the 1980s. Donors need to match their ver- 7.1. This section describes those policies and pro- bal commitments with actions: the share of aid for vides examples of successful policy reforms in var- health should be restored to its previous level im- ious developing countries. mediately and should be increased substantially over the next five years. An additional $2 billion in Low-income countries aid would help to finance the transitional cost of health policy reforms, as well as priority pro- Previous chapters have outlined the main charac- grams, including AIDS prevention. At the same teristics of health systems in low-income coun- time, donors and developing countries need to tries. In general, there is little public or private focus on measures to improve the effectiveness of insurance. Out-of-pocket spending for drugs, tra- external assistance for health. Doing so will re- ditional medicine, and user fees usually accounts 156 Table 7.1 The relevance of policy changes for three country groups Government objectives and policies Foster an enabling environment for households to improve health Pursue economic growth policies that benefit the poor Expand investment in education, particularly for females Promote the rights and status of women through political and economic empowerment and legal protection against abuse Improve government investments in health Reduce government expenditures for tertiary care facilities, specialist training, and discretionary services Finance and ensure delivery of a public health package, including AIDS prevention Finance and ensure delivery of essential clinical services, at least to the poor Improve the management of public health services Facilitate involvement by the private sector Encourage private finance and provision of insurance (with incentives to contain costs) for all discretionary clinical services Encourage private sector delivery of clinical services, including those that are publicly financed Provide information on performance and cost Very relevant 0 Relevant 0 Somewhat relevant Not relevant for more than half of total spending for health. unlicensed practitioners in South Asia) account for Government financing from general tax reve- the remainder of the health facilities and deliver nuesand sometimes substantial donor contribu- most outpatient care. They offer a service that is tionsaccount for the remainder. Government perceived to be of higher quality than that pro- hospitals and clinics provide the bulk of modern vided by the public sector. Large segments of the medical care, but they suffer from highly central- population, especially the rural poor, do not have ized decisionmaking, wide fluctuations in annual access to modern health services. Female literacy budget allocations, and poor motivation of both and enrollment of girls in primary and secondary facility managers and health care workers. Minis- school are low. tries of health and other government agencies of- Five policies for better health are crucial in this ten have only limited capacity to formulate health environment: providing solid primary schooling policy, implement health plans, and regulate the for all children, especially girls; investing more re- private sector. Private providers (mainly religious sources in highly cost-effective public health activ- organizations in Africa and private physicians and ities that can substantially improve the health of 157 the poor; shifting health spending for clinical ser- sary radio and telephone networks; and building vices from tertiary care facilities to district health the capacity to plan and manage health services at infrastructure capable of delivering essential clini- the district level and in individual facilities. In cal care; reducing waste and inefficiency in gov- many low-income countries, focusing on district ernment health programs; and encouraging in- health infrastructure will mean limiting new in- creased community control and financing of vestment in central hospitals and reorienting those essential health care. facilities toward research and teaching activities that are more relevant to key national health prob- INCREASED SCHOOLING. Despite the often formi- lems. At the same time, there is considerable scope dable obstaclesboth in providing access to for improving the efficiency of large government schools and in eliminating cultural barriers that hospitals, especially through performance-linked keep girls outa number of low-income countries incentives for managers and staff and expanded have proved that dramatic change is possible in a cost recovery from the wealthy and insured. short period of time. Between 1970 and 1990 Indo- To deliver essential clinical services, a greater nesia and Kenya, for example, achieved rapid and share of government health budgets needs to be sustained growth of primary school enrollments devoted to the operations of lower-level facilities and raised the proportion of girls to nearly half of and especially to nonsalary recurrent items. Initial all pupils. These gains were brought about by a emphasis needs to be placed on building capacity combination of high-level political commitment to to deliver the services included in the minimum universal primary schooling, information pro- essential package described in Chapter 5. This is grams that created stronger demand on the part of now happening in a number of countries. Senegal parents, and support from the international has set annual targets for increasing its spending community. for drugs, transport, and maintenance. Ghana is INVESTMENT IN PUBLIC HEALTH ACTIVITIES. The trying to reduce the number of civil servants work- public health activities with the largest payoff will ing for the Ministry of Health. In India, where vary from country to country: vitamin A and io- state governments account for more than three- dine supplementation in India and Indonesia, quarters of total public spending for health, the antismoking campaigns in China, and policies to central government is attempting to act as a cata- reduce traffic injuries in urban areas of Sub- lyst for more cost-effective resource allocation by Saharan Africa. Completion of immunization cov- earmarking its funds for immunization, treatment erage should be a high priority in all low-income of leprosy and tuberculosis, and AIDS control. countries, especially in India and in much of Sub- Some low-income countries will need to in- Saharan Africa, where coverage remains low. Sim- crease government outlays for health if they are to ilarly, a greatly intensified effort to reduce trans- finance a package of public health measures and mission of HIV and other sexually transmitted dis- essential clinical services for the poor. In 1990 gov- eases is warranted. In the parts of Africa in which ernment spending for health in low-income coun- the AIDS epidemic is already widespread, behav- tries averaged only $6 per capita-1.5 percent of ioral change through education and condom distri- GNP if foreign assistance is excluded and 1.6 per- bution should be high on the list of public health cent including aid. The analysis in this Report in- actions. And where, as in Bangladesh and Indo- dicates that provision of a minimum package will nesia, the preconditions (widespread commercial cost about $12 per capita in low-income countries, sex and high prevalence of other STD5) exist for or nearly 3 percent of GNP. Effective targeting of rapid spread of HIV, governments urgently need publicly subsidized clinical services to the poor, to take steps to halt the spread of AIDS from high- and corresponding efforts to encourage cost recov- risk groups into the population at large. ery from more affluent groups, would help stretch limited government budgets. Modest fees col- BETrER ALLOCATION OF SPENDING ON CLINICAL lected at health centers could also be retained and SERVICES. Governments should invest in distfict reinvested locally to improve the quality and re- health infrastructure by (as described in Chapter 6) liability of basic services. expanding training programs for primary care pro- But even with these efforts, many governments viders, particularly nurses and midwives; target- in low-income countries will have to increase the ing construction funds to improve health posts, share of the budget allocated to health. (In Sub- health centers, and district hospitals; financing Saharan Africa health spending declined during ambulances and other vehicles needed for effec- the 1980s to an average of less than 4 percent of tive emergency transport, together with the neces- public expenditure and less than 2 percent of 158 Box 7.1 Community financing of health centers: the Bamako Initiative The principal aim of the Bamako Initiative, launched in The initiative is only five years old, but its achieve- 1988, is to "revitalize the public sector health care de- ments are impressive. Eighteen African countries were livery system [byl strengthening district management participating as of late 1991, and nearly 1,800 health [andi capturing some of the resources the people them- centers located in 221 districts were part of the pro- selves are spending on health" (UNICEF 1992). gram. In Benin the first forty-four health centers tar- Both revolving funds for drug purchases and com- geted by the initiative are covering 42 to 46 percent of munity-managed health centers have existed for many their operating costs with user charges, and in the first years in developing countries, but the Bamako Initia- seventeen centers in Guinea's program, user fees cover tive is attempting to implement these schemes on a 38 to 49 percent of expenditures. Utilization of health much larger scale in Africa and other low-income coun- centers has increased. In Benin average monthly visits tries. The initiative is based on two premises: that to pilot health centers rose from 100 in 1987 to 250 in where public institutions are weak, as they are in many 1989. low-income countries, bottom-up action by commu- Despite the initiative's promising accomplishments, nities is badly needed to complement top-down health it is not yet certain that the reforms can be sustained Ofl policy reforms, and that even poor households are will- a large scale. A number of health centers covered by ing to pay for higher-quality and more reliable health the initiative have received both financial and technical services. assistance from UNICEF, WHO, and other donors Under the initiative, members of local communities more than $36 million has come from UNICEF alone. who use a health center or pharmacy agree to pay mod- Problems may emerge when this external assistance est charges for outpatient care, including drugs. The ends, particularly in converting local revenues gener- revenues generated from fees are retained by the ated through user charges into the foreign exchange health centers and managed by local elected commit- needed to purchase imported drugs. In addition, ef- tees. The committees reinvest in additional drugs forts to encourage local private financing of health care (through a revolving fund), in incentive payments for by poor urban and rural households may allow govern- health workers, and in other improvements. The gov- ments to avoid tackling basic reforms of their health ernment and donors assist health centers in purchasing systems, especially the reallocation of public revenues inexpensive generic drugs, thus increasing the cost- from tertiary care hospitals to more basic services. effectiveness of services at the health center. GNP.) Some countries are already moving in this in part through the use of essential national drug direction. Mozambique, for example, is increasing lists, and by purchasing drugs competitively. Nu- government outlays for health in 1992 and 1993 as merous successes have already been recorded. part of a broader program of economic reform, and Bulk procurement of drugs enabled a group of Mauritania is committed to substantial rises in church-run African health associations to save 40 government health spending during 1992-96. percent of their annual drug bill. Similar efforts by Since shifts in domestic budget resources between several Caribbean states led to an average reduc- sectors take several years to implement, donor tion of 44 percent in the price paid for the twenty- funds could play a significant role in increasing five most frequently used drugs. An essential- government health expenditures, including recur- drugs revolving fund for several Central American rent spending, in the early years. The budget re- nations yielded savings of 65 percent of the costs forms in both Mozambique and Mauritania are be- of pharmaceuticals. ing supported by transitional financing from the donor community. COMMUNITY CONTROL AND FINANCING. Com- munity financing, in the form of user charges and REDUCTION OF WASTE AND INEFFICIENCY. There is prepaid insurance schemes, has become a practical substantial scope for reduction of waste and ineffi- necessity in a number of low-income countries. ciency in government health programs, especially But community financing is also a virtuous neces- in drug management. Pharmaceuticals, which ac- sity: it can help to improve the quality and re- count for 10 to 30 percent of public spending for liability of services, in part by making health health in most countries, are the most promising workers more accountable to their clienteles. area for efficiency gains in the short run. Very This is the approach being taken in the Bamako large savings can be achieved by improving the Initiative, sponsored by WHO and UNICEF (Box selection and quantification of drug requirements, 7.1). Recent experience from a number of African 159 countries shows that rural households are pre- of the population receives services financed pared to pay modest charges for drugs in govern- through general tax revenues. Other countries use ment health centers, provided that the quality of social insurance, with part of the population cov- services improves, that fees are retained and uti- ered by mandatory employment-based contribu- lized at the point of service, and that the local tions, usually pooled in a single fund run by a population has a strong voice in the operation of parastatal agency. The share of the population pro- the facility. In Guinea, for example, about half of tected by social insurance varies widely, from less the country's 350 health centers were practicing than 10 percent in the Dominican Republic, Ecua- community financing in 1991. Of these, all the ur- dor, and El Salvador to more than 80 percent in ban-based facilities and a third of the rural clinics Brazil, Costa Rica, and Cuba. Brazil and Chile em- were able to cover their operating expenses with ploy hybrids of private and public insurance. In income from fees. Governments should act cau- Brazil every citizen is legally entitled to services tiously, however. Experience suggests that fees financed from a combination of general revenues substantial enough to cover the full cost of clinical and social security contributions, and social insur- services can discourage utilization by the poor. ance is deducted from the wages of every salaried Under these circumstances, the poor should be worker. Yet more than one-fifth of the country's charged reduced fees or should be exempted from population currently opts for some form of private payment. insurance coverage. Middle-income countries need to focus on at PROBLEMS AND PROSPECTS. Health policy re- least four key areas of policy reform: phasing out forms face formidable obstacles in low-income de- public subsidies to better-off groups; extending in- veloping countries. The health ministry often surance coverage more widely; giving consumers makes only a weak case for a larger share of the a choice of insurer; and encouraging payment (sometimes shrinking) budget. Politicians, doc- methods that control costs. tors, and the urban population exert strong pres- sures for higher spending on tertiary care facilities REDUCTION OF SUBSIDIES TO BETFER-OFF GROUPS. in the major metropolitan areas at the expense of Governments should reduce and eventually elimi- the district health infrastructure. Professional as- nate public subsidies to relatively affluent groups. sociations and trade unions representing doctors This can be done by charging full-cost fees to in- and nurses strongly resist both staff cuts designed sured persons who use government hospitals and to increase nonsalary spending and efforts to re- clinics for services not included in the national es- deploy health workers to rural areas. Despite these sential clinical package and by cutting tax deduc- obstacles, some low-income countries are cur- tions for insurance contributions. In South Africa rently carrying out major health policy reforms. and Zimbabwe privately insured individuals have Malawi, for example, is implementing sweeping been charged less than the full cost of the services changes as part of a World Bank project. It is in- they receive in government health facilities. In ad- creasing the share of the government budget allo- dition, they have been allowed to deduct from tax- cated to health from 7.1 percent in 1991 to 9.1 per- able income part or all of their out-of-pocket pay- cent by 1995, raising the fraction of health ments for health care, as well as their health spending for district health services from 15 to 23 insurance premiums. Employers can also deduct percent, and reducing the share devoted to the their insurance contributions. These measures re- country's three central hospitals from 35 to 25 per- duce the amounts available for financing essential cent. To strengthen the district health system, the services. In South Africa individual tax deductions government is also engaging more than 3,500 new were estimated to be equivalent to 18 percent of lower-level health workers to serve in rural clinics total public sector health expenditures in 1990. In a and communities. Donor funds are being used to recent effort to reverse a similar situation, Zim- help pay for these workers. babwe has sharply limited tax deductions for health care and insurance, raised fees, and inten- Middle-income countries sified efforts to collect fees from privately insured In middle-income developing countries out-of- patients. Government hospitals have learned that pocket payments for health usually account for they can often identify insured patients by offering less than a third of total spending. Some middle- them extra nonmedical amenities, such as private income countries, such as South Africa and Zim- hospital rooms, and can then target them for ag- babwe, have private insurance, even though most gressive cost recovery if they accept. 160 In countries where social insurance covers only government, as in Germany, tend to be more ac- a fraction of the population, governments can in- countable to their members. In a number of Latin crease the extent to which health services are self- American countries monolithic social security "in- financing by eliminating public subsidies to social stitutes" are already heavily discredited because of insurance. These subsidies, which are widespread their past inefficiencies and corruption. Greater in Latin America, mostly benefit the middle classes competition and accountability are two of the main and are therefore regressive. Elimination of the objectives of current proposals for reforming social subsidies would free resources for health services insurance in Argentina. for the poor. Eliminating subsidies also imposes more financial discipline on the social insurance COST CONTAINMENT. Copayment by insured in- agencies, which are often allowed to run deficits dividuals for some services can help to restrain that are later covered by transfers from other social their use of the services but is unlikely to be a very security programs or from the general government powerful cost-containment method. Copayments budget. In Venezuela, for example, the govern- amounting to an average 40 percent of expendi- ment subsidizes contributions to the medical assis- tures in Korea have done little to slow the rate of tance fund within the parastatal social security increase in health spending, which grew from 3.7 agency. Despite this subsidy, in 1990 the fund ran to 6.6 percent of GNP during the 1980s. Similarly, a deficit equivalent to 37 percent of its health the practice, introduced by private U.S. insurers, expenditures. of retrospective reviews of utilization of medical care appears to lead to a modest one-time savings EXTENSION OF INSURANCE. Where the bulk of the in health spending but does not have long-lasting labor force is already employed, government poli- effects on the rate of growth of expenditures. cies that extend insurance coverage to the rest of By contrast, prepayment of health care pro- the populationincluding the self-employed, the viders is a promising approach to containing elderly, and the poorremove the inequities in- health expenditures. Governments could help to herent in multitiered systems of health financing promote such schemes by removing legal barriers and expand the content of the universally available that in many countries prevent the same institu- package of care. When insurance coverage be- tion from acting as both insurer and provider. In comes universal, as in Costa Rica and Korea, sub- South Africa the government recently decided to sidies actually end up targeting the poor and are allow the creation of health maintenance Organiza- thus progressive. But only a few middle-income tions (HMO5), mainly as a way of containing countries that have adequate financial resources, health costs. More than twenty such organizations political resolve, and administrative capacity will have been established in just a few years. They be able to achieve such universal insurance cover- have introduced capitation and negotiated fees, age. Korea's bold initiative to create a national which limit costs more effectively than did the health insurance system from scratch between open-ended fee-for-service payment arrangements 1978 and 1989 and Costa Rica's efforts in the 1980s historically used in South Africa. to universalize a system that had previously cov- Governments can do much to improve the in- ered only the industrial labor force show that this centives created by social insurance. Where the in- is a difficult but achievable goal. Attaining univer- sured use private providers, fee-for-service pay- sal coverage would be more feasible if govern- ment schemes need to be replaced with an ments limited the essential package of insured ser- alternativecapitation or annually negotiated uni- vices to those with high cost-effectiveness. form fees for doctors and hospitals (based on diag- nostic-related groups of procedures, for example) CONSUMER CHOICE. Competition among sup- or preset overall budgets for hospitals. Where so- pliers of a clearly specified prepaid package of cial insurance covers services by government hos- health services would improve quality and encour- pitals, competition with the private sector can im- age efficiency. And even where there is little or no prove performance. Other promising approaches direct competition among insurance funds, as in are to allow government hospitals to compete with Japan and Korea, multiple semi-independent in- one another as semiautonomous enterprises, as in surance institutions may still have advantages the United Kingdom in recent years, and to give over a single large parastatal agency. Local insur- hospital managers financial and career incentives ance funds managed by boards composed of rep- to meet performance targets, as in Chile. resentatives of workers, employers, and local The example of Chile (Box 7.2) illustrates the 161 benefits and perils of health sector reform in a mid- the formerly socialist countries of Eastern Europe dle-income country. Chile has been able to im- and the Soviet Union. Health expenditures were prove efficiency, quality of care, and consumer financed from general revenues. In principle, they choice, but the reforms have also created new were provided free of cost to the population at problems regarding administration, financing, and government clinics and hospitals and at facilities equity. run by state enterprises, but in practice, "infor- Formerly socialist countries mal" payments oiled the wheels of bureaucracy. Today the health systems in these countries are in Historically, the government was responsible for severe crisis. Many doctors and pharmacists are both the finance and the delivery of health care in leaving the government health services to practice Box 7.2 Health sector reforms in Chile Over the past fifteen years Chile has undertaken dra- had few incentives to help supervise municipal matic reforms of its health sector, Its experience shows facilities. that reform is a permanent process, not a one-time ef- Because municipalities were reimbursed for each fort, and that countries undertaking reform must have unit of service delivered, they tended to provide too both the capacity and the political will to review and much high-cost curative care and too few preventive revise health policies continuously. services, which caused costs to explode. The govern- Starting in the late 1970s, Chile (then under a mili- ment then moved to cap allocations to local authorities, tary government) decentralized the government-run using as a basis historical budget shares that favored health system and created private health insurance in- the wealthier municipalities. stitutions. Responsibility for operating primary care The ISAPREs, by targeting the richest segments of services was devolved to the country's 325 munici- Chilean society, impoverished the rest of the social in- palities. The Ministry of Health transferred its primary surance system. Each salaried beneficiary who chose to care budget and about half of its personnel to the mu- shift to an ISAPRE cost the public system 2.5 times the nicipalities, which could also draw for financing on lo- contribution of an average salaried worker. Because the cal tax revenues and on resources from the central gov- ISAPREs are permitted to rate individual health risks, ernment's Municipal Common Fund. More important, they have "skimmed" the population for good risks, the government encouraged the establishment of pri- leaving the public sector to care for the sick and the vately owned and operated health insurance funds, elderly. known as ISAPREs. The roughly 70 percent of the The democratically elected government that came to population covered by social security schemes had the power in 1989 has chosen to maintain the broad thrust option of using their payroll deduction to buy a pre- of the health reforms while seeking ways to overcome paid private health plan. The competing plans were their adverse effects. Municipal elections have been regulated by a new oversight unit (superintendencia) in held to ensure that popularly chosen and accountable the Ministry of Health. By 1990 about 2.5 million peo- officials look after primary health services. Training ple, or 18 percent of the population, were covered by programs have been organized for municipal health thirty-five ISAPREs. officers. Responsibility for hospitals is being decentral- Both decentralization and the creation of the private ized to twenty-seven health service areas that will enter insurers brought about some improvements in the into management contracts with the Ministry of health system. The municipalities expanded primary Health. Finally, under a new proposal, central funds care services. The ISAPREs introduced more competi- would be allocated to the municipalities on a capitation tion and consumer choice into the financing and deliv- basis, with a further adjustment to favor the poorest ery of services and spurred growth in the numbers of localities. private doctors and hospitals. The government is also beginning to look at ways to But the reforms also created new problems. In the reduce inequities in the ISAPRE health financing sys- early years of the reforms, when local officials were tem. The superintendencia that regulates ISAPRE is be- appointed by the military regime, municipal health ser- ing strengthened. It is considering requiring the pri- vices were not responsive to the local population. vate plans to use community risk-rating and to accept Transfers of Ministry of Health staff to the munici- all applicants able to pay the community-rated pre- palities created job insecurity and caused a decline in miums; making it mandatory for all ISAPREs to offer a staff morale. Many municipalities lacked the capacity similar basic medical plan in order to promote direct to plan and manage primary health services. The mu- competition among suppliers (as in the managed care nicipalities tended to overrefer patients to hospitals, systems being developed in the United States); and which were still funded by the ministry. The ministry eliminating the deduction for employer contributions. 162 fee-for-service medicine in the private sector. Since covering two to four provinces, but the provinces real government spending for health has fallen are reluctant to finance such regions. Moreover, dramatically during the recent transition toward a there are political pressures for further decentral- market economy, the government health system is ization to the level of the district governments, also experiencing serious shortages of drugs and where there is now very little capacity for manag- equipment. ing health systems. Largely because they know all too well the prob- At the same time that they decentralize, govern- lems of repressive central government control, ments will have to reduce the size of publicly policymakers, medical professionals, and con- owned health services, which have far too many sumers in the formerly socialist countries are look- hospitals, hospital beds, and physicians. In this ing to systems of public and private insurance in way, governments can free resources for vital pub- industrial countries as possible models for reform. lic health services, including immunization, work- Some countriesfor example, the Czech Republic, place and food safety, environmental regulation, Hungary, and Polandhave much in common measures such as education and higher taxes to with upper-middle-income countries such as Ar- discourage consumption of alcohol and tobacco, gentina, Costa Rica, and Korea. They may be also and quality control of privately delivered clinical able to adapt some features of the systems of the care. The clinical and managerial skills of the re- Nordic countries and the United Kingdom, which maining government health personnel need to be are financed from general revenues, or of the uni- substantially upgraded and reoriented from the versal social insurance approaches of Germany previous system of centralized bureaucratic con- and Japan. Others in this groupincluding the trol toward the emerging system of semi- relatively poor Central Asian republicsface many autonomous health facilities. of the same issues currently confronting lower- middle-income and even low-income countries, NEw MODES OF FINANCING. The examples of such as Pakistan and Yemen. other countries could help the formerly socialist Despite this diversity, the governments of all the countries establish insurance systems that pre- formerly socialist countries need to consider health serve the main virtue of their old systemwide- sector reforms in at least three main areas: improv- spread coverage of the population. It could also ing the efficiency of government health facilities help them to recognize the circumstances under and services, partly by reducing the size of the which general government revenues can play a public system; finding new ways to finance health positive role, as the dominant source of funding care; and encouraging private supply of health (the pattern in the United Kingdom) or as a com- services while strengthening public regulatory plement to insurance (as in Japan). Experience capacity. elsewhere offers important lessons on how to cre- ate financing systems that are sustainable and that EFFICIENCY OF GOVERNMENT SERVICES. De- contain costs by, for example, discouraging fee- centralization of government health services is po- for-service compensation. The formerly socialist tentially the most important force for improving countries will also want to avoid the large and in- efficiency and responding to local health condi- equitable government subsidies commonly pro- tions and demands. It will be successful only when vided to private insurance for the wealthy or to local government health agencies and hospitals social insurance for the middle class. have a sound financial base, solid administrative Most formerly socialist countries are already on capacity, and incentives for improving efficiency the road to reform. The Czech and Slovak repub- and when they are accountable to patients and lo- lics and Hungary are experimenting with forms of cal citizens. Extreme and hasty decentralization social insurance. Because the Czech system in- can create inefficiencies. In Poland, for example, cluded a very comprehensive package of health the government has decentralized health careto benefits and paid private doctors on a fee-for-ser- the level of the country's forty-nine provinces. The vice basis, it encountered serious financial diff i- average provincial population of less than a mil- culties after just a few months of operation. Under lion is proving too small to make efficient use of the recently revised Hungarian health-financing the tertiary care hospitals being built in each pro- system, public sector doctors will be salaried em- vince, and the available medical personnel are be- ployees of the central and local governments, and ing spread too thin. For these reasons, the govern- private general practitioners will be paid on a cap- ment is now experimenting with health regions itation basis. Russia and Ukraine are also prepar- 163 Box 7.3 Reform of the Russian health system Before the political upheavals of 1990-91 that led to the They will sign contracts for care with public and private breakup of the Soviet Union, the 3 to 4 percent of GNP providers. Individuals can then voluntarily purchase that the Russian republic spent on health care for its supplementary private insurance to cover additional nearly 150 million inhabitants was financed from gen- health services. eral government revenues and delivered through a The health insurance legislation has been in effect vast network of public facilities, programs, and em- since late 1991, but progress in implementing it has ployees. This highly centralized and bureaucratic sys- been slow. Some important issues in the design of the tem led to excessive numbers of doctors and hospitals. system still need to be resolved. These include the role It gave few incentives for efficiency or for providing and extent of competition among public and private quality care, and it neglected the preventive measures insurers; whether risks are to be rated on an individual needed to combat the country's most serious environ- basis or across larger pools of individuals; and how the mental and behavioral problems: industrial pollution, insurance funds will pay providerson a fee-for-ser- alcohol and tobacco dependency, and poor nutrition. vice basis, through capitation, or by some other Consequently, the health status of Russians stagnated method or combination of methods. during the 1970s and 1980s. In 1990 life expectancy for The practical obstacles to the implementation of the Russian men was just sixty-four years, a full ten years new system are formidable, partly because of the un- less than in Western Europe, and the infant mortality settled administrative and economic environment. The rate, at twenty-two per 1,000 live births, was twice the regional governments lack the capacity to manage and Western European average. regulate the health system they are inheriting. The The new Russian government has pursued several economy and the government budget are under severe fundamental reforms of the old Soviet health system. strain. Real wages have fallen dramatically in the past Health financing and management are being decentral- few years. The costs of drugs and equipment have in- ized to eighty-eight regions. Much medical practice is creased faster than inflation, leading to serious short- being privatized, and a recent health insurance law ages. Payroll taxes to cover employee benefits already provides for the introduction and regulation of new absorb 38 percent of wages, making it difficult to fi- forms of insurance. Under the law and its proposed nance an affordable package of health services through amendments, each region is to have a social insurance the social insurance system. To help overcome these fund, and a national fund will equalize resources problems, a number of international agencies, includ- across regions. These insurance funds will receive a ing the World Bank, are working closely with Russian combination of compulsory payroll deductions and health officials on designing and carrying out health budget transfers from general government revenues. policy reforms. ing to implement mixed systems of social insur- sion to issue lifetime licenses to doctors without ance and general revenue financing. Box 7.3 establishing strict standards of practice or recer- describes the current efforts in Russia. tification requirements. Since government regula- tory capacity is likely to be weak in the next few COMPETITIVE PROVISION AND PUBLIC REGULATION. years, health system reforms should be designed Although private medical practice is now permit- in ways that minimize the need for direct govern- ted in most of the formerly socialist countries, the ment regulation. Encouraging self-regulation legal and regulatory environment for private doc- through associations of private medical schools, tors, hospitals, and insurance institutions is often doctors, and hospitals would be one such ap- either nonexistent or hostile. With large numbers proach. In the long run, better regulation will re- of private doctors establishing practices and pri- quire both training of government inspectors and vate hospitals and clinics being created, regulation other regulatory personnel and development of of providers will be critical for reducing the inci- government institutions such as medical licensing dence of medical malpractice and financial fraud: boards and national and local medical ethics It is also essential that regulation encourage the committees. development of efficient institutions, such as health maintenance organizations, for financing Directions and prospects for reform and providing clinical care for the bulk of the population. Already there are signs of poorly con- The world's diversity of health care systems is ceived regulations, such as Romania's recent deci- matched by the diversity of reform movements. 164 But several common themes are beginning to Nigeria, and South Africa and in Eastern Europe. emerge. First, governments are increasingly recog- A free press is important, as are consumer advo- nizing the centrality of their own role in public cacy groups, for conveying a diversity of views on healthfor example, in achieving the enormous health reform and for stimulating debate. global gains in immunization coverage. Second, In many countries, maintaining the support of governments are exploring ways to introduce the middle class and of urban groups for health more competition and foster a diversity of public policy reformsincluding the reallocation of pub- and private institutions in the delivery of clinical lic spending from tertiary care to basic public services. Third, governments are examining new health and clinical care for the poorwill require a approaches to finance and insurance, including se- gradual shift in resources rather than wholesale lective user fees in the public sector, systems that changes in just one or two years. For this reason, discourage third-party reimbursement, systems universal government financing (or government- that mix finance from compulsory social insurance mandated financing) for a nationally defined es- and from general tax revenues, and systems that sential package of services will often be more suc- set fixed budgets for each patient or each case. cessful than a highly targeted approach that may Everywhere, health sector reform is a contin- undermine the political base for reform. Similarly, uous and complex struggle. Neither governments continued government ownership of some hospi- nor free markets can by themselves allocate re- tals that offer high-quality tertiary care, with a sources for health efficiently. As policymakers try phased reduction in public subsidies to the to reach compromises, they must deal with power- wealthy for this care, may be more feasible politi- ful interest groups (private doctors, drug com- cally than rapid divestiture to the private sector. panies, medical equipment manufacturers, and External financial assistance can help countries insurers) and strong political constituencies, in- handle these politically difficult tradeoffs and can cluding urban dwellers and industrial workers. ease the process of policy change. Strategies for overcoming these obstacles to health sector reform will vary from country to International assistance for health country, but some common approaches are dis- cernible. Political leadership, beginning with the After growing rapidly in the 1970s, aid for health head of state, is an indispensable element in re- stagnated during the 1980s. As a share of official form programs almost everywhere. The 1990 development assistance, aid for health declined World Summit for Children proved an effective from an average of 7 percent for the period 198 1-85 means for engaging the attention and commitment to 6 percent during 1986-90. Total aid flows to the of heads of state (see Box 2 in the Overview). Se- health sector in 1990 were $4.8 billionalmost $4 nior officials of ministries of health can be strongly billion in official development assistance and $0.8 influenced by the prevailing views of the interna- billion from NGOs and foundations (Figure 7.1). tional health community, particularly those of This amounts to about one dollar per person in WHO and other major donors, and by participa- developing countries. (The figure for official devel- tion in international meetings and seminars on opment assistance is based on reports from donor health policy and management. governments. Only $3.3 billion of the $4 billion can Professional associations may be able to bring be accounted for as receipts by individual coun- about some reorientation of health workers, espe- tries; this is the amount that appears in the total cially physicians. Appeals to the sense of social health expenditure estimated in Chapter 3 and in responsibility of these associations have helped Appendix table A.9.) Bilateral agencies accounted advance agendas for preventive health in the for the largest share (40 percent), followed by United States and elsewhere. Such groups, how- United Nations agencies (33 percent), NGOs (17 ever, are often the sources of the strongest resis- percent), development banks (8 percent), and tance to change. Reshaping the training curricula foundations (2 percent). of medical and nursing schools to include a greater The trend is for donors to provide aid for health emphasis on public health and general practice is through multilateral channels. The share of multi- likely to be a more effective way to enlist the sup- lateral assistance has grown from 25 percent in port of physicians and nurses. 1980 to 40 percent in 1990 and is likely to exceed 50 Public opinion can be a powerful force for health percent by 1995. As a result of the quadrupling of reform, not only in industrial countries but also in World Bank lending for health over the past six developing countries such as Brazil, Chile, years, disbursements of Bank funds are expected 165 External assistance to developing countries for health comes from many sources, public and private. Figure 7.1 Disbursements of external assistance for the health sector, 1990 (millions of dollars) N 1,913 I 3,925 / Bilateral agencies 4,794 / Public agencies 382 242 I 671 / 4,794 Development > banks" N Donor 869 1,601 382 / Developing countries COO 0 tries United 58(1 Private Nations agencies agencies 830 Nongovernmental organizations (1 N 7 68 -J K Foundations a. Includes $84 million in nonconcessional loans. Source: Michaud and Murray, background paper. to grow from about $350 million in 1992 to about $1 Programme (UNDP), in its annual Human Develop- billion in 1995, making the World Bank the largest ment Report, has argued for more donor spending single source of external funding for health. Since on health, and in its recent World Development Re- the portion of aid going to middle-income coun- ports the World Bank has made similar recommen- tries from the World Bank and other development dations. World Development Report 1990 proposed a banks is nonconcessional lending, some of the 3 percent annual increase in aid during the 1990s, projected increase in lending for health will in- to be targeted at poverty-reducing activities, in- volve a hardening of terms. It would be desirable cluding basic health care. The donor community for bilateral grant-funding agencies and conces- needs to review these goals and targets in light of sional arms of the development banks (such as the the actual trends in aid flows for health. World Bank's International Development Associa- The share of aid going to health should be re- tion) to increase their assistance to health as well. stored immediately to its earlier level of 7 percent The amount of health aid has stagnated, and its of total official development assistance and should share in total development assistance has de- rise substantially over the next five years. Such an clined, even as donors continue to express concern increase would have a significant impact on the about health. Over the past ten years the United health status of the poor, particularly if it is di- Nations and other international agencies have rected toward the transitional costs of reallocating called for increased investments in the develop- government spending to public health measures ment of human resources, including health, both and essential clinical care and to seriously under- by developing countries themselves and by the do- funded disease control efforts such as those for nor community. The United Nations Development tuberculosis and AIDS. A rise in donor assistance 166 of $2 billion, for example, could finance a quarter large share of health expenditure. In Africa aid of the estimated additional costs of a basic package makes up an average 10 percent of national health in low-income countries and of strengthened ef- spending (Table 7.2), or 20 percent if South Africa forts to prevent AIDS. Such an increase, which is excluded. Aid covers more than half of all health would boost from 6 to 9 percent the share of total expenditures in countries such as Burkina Faso, official aid going to health, would be feasible if Chad, Guinea-Bissau, Mozambique, and Tan- other donors matched the rise in World Bank dis- zania. In these countries donors finance an impor- bursements for health that is expected to occur in tant share of recurrent costs, as well as investment coming years. It would also be consistent with the items. In Mozambique, for example, aid accounted proposal in the UNDP's Human Development Report for more than half of recurrent spending in 1991 1993 (also endorsed by UNICEF) that 20 percent of and for 90 percent of capital expenditures for aid be spent on health, education, water and sani- health. Even when aid amounts to 2 percent or less tation, and environmental protection for the of total health spending, as in the other develop- world's poor. ing regions, improvements in its use would still be There are a number of ways, in addition to the an important catalyst for reform. traditional annual and multiyear programming of General lessons on improving aid effectiveness aid by individual donors, for the international apply equally to the health sector (Box 7.4). Do- community to mobilize more financial resources nors need to set their priorities carefully and allo- for health. Coordinated sectorwide pledging at cate their resources in accordance with these prior- consultative group meetings and donor round- ities. The productivity of aid would increase tables has been used successfully in countries such substantially if donors were to direct more of their as Tanzania and Zambia. Another approach is pro- assistance to public health measures and essential gram-specific pledging, as illustrated by the clinical services, especially in low-income coun- dozens of national AIDS-control donor meetings tries. They might also usefully focus on capacity chaired by WHO in recent years. The role of debt- building, research, and reform of health policy. for-development swaps as a means of generating Countries that show a willingness to improve ac- extra resources for both government and NGO- cess to health services for the poor and to under- provided health services should be assessed in this take reforms of the health system should be strong context. Ecuador, Sudan, and Zimbabwe have al- candidates for aid. ready carried out swaps, and Nigeria is exploring a The World Bank increasingly stresses policy re- major swap of its debt currently held by donors in form in its lending for health, which has grown return for increased public spending for essential nearly fourfold in recent years (Box 7.5). For some health services. donors, adjustment of priorities would mean spending less on hospitals, sophisticated medical Improving the effectiveness of aid for health equipment, and training for medical specialists It is crucial that the donor community and devel- During 1988-90 Japan spent more than 33 percent oping countries focus on ways to improve the ef- of its bilateral assistance for health on construction fectiveness of existing and future assistance to the of hospitals, France spent 25 percent, and Ger- health sector, particularly in the low-income coun- many and Italy spent nearly 15 percent each. tries where donor assistance already accounts for a Within the domain of public health and essential clinical care, several areas of intervention deserve greater attention from donors, including tuber- Table 7.2 Official development assistance culosis control, the EPI Plus program, micro- for health by demographic region, 1990 nutrient supplementation, AIDS prevention and Health aid control, and programs to reduce tobacco consump- Health aid as a tion. These problems impose a large burden of ill- received Health aid percentage (millions per capita of health ness, in some cases because rapid growth of the Region of dollars) (dollars) expend?ture threat has gone unrecognized. Their control offers Sub-Saharan Africa 1,251 2.45 10.4 large externalities or economies of scale. Often, so- Other Asia and islands 594 0.87 1.4 lutions will require a global effort. Latin America and the The efficiency of aid for health can be greatly Caribbean 591 1.33 1.3 Middle Eastern crescent 453 1.31 1.3 enhanced through better coordination of donor India 286 0.34 1.6 projects and policies. Fragmentation of external China 77 0.07 0.6 support in the health sector is a long-standing Source: Michaud and Murray, background paper. problem in many countries and imposes a heavy 167 r Box 7.4 Health assistance and the effectiveness of aid Recent evaluations of the effectiveness of aid, includ- ronment. When it comes to coordination, both sides ing a classic 1986 study commissioned by the world have been at fault. Donors have pursued their OWfl donor community, point toward the same conclusion: objectives without attempting to ensure that their aid most aid has been successful, but a considerable share, complements that of others. And all too often, aid re- perhaps a third or more, has been much less so, and a cipients have played one donor off against another, small percentage has failed completely or has even while ministers and ministries have focused on their been harmful. These broad-brush averages hide signifi- own concerns rather than looking to the national good. cant regional differences: in Asia and Latin America Aid for health has generally had a good technical performance has been better; in Sub-Saharan Africa it record. It has fit in well with development priorities, has been worse. Aid has been least effective in the especially in recent years, as the concentration on hos- poorest countries, where success is most needed. pitals and high-technology curative medicine has been The reasons for inferior performance lie with both replaced by an emphasis on primary and preventive donors and recipients. Poor countries and those experi- care. There have also been major successes_mainly encing political conflict and instability constitute a diffi- highly focused initiatives such as the program for the cult environment for aid, as they have little administra- eradication of smallpox, the drive against child mortal- tive capacity or infrastructure. But these difficulties ity, and the effort to control river blindness in Africa. have in many cases been compounded by unfortunate What is still lacking is the ability of the aid system to policies. Aid projects have been poorly designed, both help set in place and sustain locally appropriate public technically and because of inadequate understanding health programs and essential clinical services. of the human, social, institutional, and political envi- burden on already overextended government offi- strengthening the public institutions that finance cials. In the extreme, fragmentation can lead to and deliver health services, both through broad conflicting policies being put into effect. Recently civil service reform and through changes within in one West African country, for example, three the health sector. Donors can play an important different cost recovery policies, each sponsored by role in these areas by supporting decentralization a different donor agency, were being applied in and other organizational reforms and by assisting separate regions of the country. The dangers of the groups that formulate national health policies. fragmentation are especially great in poor coun- Additional support is required for initiatives such tries where different donors choose to focus their as the foundation-backed International Health Pol- health sector activities on different provinces or icy Program and for bilateral projects to train districts and either lose sight of or undermine the health planners and managers, economists, and formulation of national policies. sociologists. Much can be done to improve donor coordina- International programs for research and development in tion, globally and regionally, but especially at the health country level. Donors can agree with countries on overall national health and assistance strategies. Investments in health research and development This is especially effective when the government have yielded high returns in better health. For ex- takes the lead in planning and in coordinating the ample, the programs for tropical disease research donors, as has happened recently in Zimbabwe. and human reproduction funded by donors and Another approach is for donors to form large con- executed by WHO have produced a number of sortia to fund national programs, as in the case of new or improved drugs and diagnostic tests and maternal and child health and family planning in have strengthened research capacity in developing Bangladesh. (The experiences of these two coun- countries. Yet according to the 1990 report of the tries are reviewed in Box 7.6.) At a minimum, do- Commission on Health Research for Develop- nors should create informal local groups that meet ment, only 5 percent of global expenditures on periodically to review progress and problems in health research are directed at the health problems the health sector, as in Mozambique and Senegal. unique to developing countries, and less than 10 The efficiency with which aid for health is spent percent of donor assistance for health is devoted to depends critically on building local capacity to research, both biomedical and in the social plan and manage health systems. This requires sciences. 168 The commission identified several serious defi- compounded by donors' limited capacity to stay ciencies in the international health research and abreast of the latest research proposals and to development system. The expertise of the global assess the relative priorities for funding this pharmaceutical industry is not being adequately research. applied to the development of drugs and vaccines To help stabilize funding, to improve the setting that could reduce the toll of early childhood dis- of priorities, and to boost efficiency, developing eases. Technology assessment is weak, as is the countries, donors, and scientists should consider health policy research needed to determine more the development of a global mechanism for better equitable and efficient ways to finance and deliver coordination of international health research. A health services. Most important, the commission number of institutional arrangements are possible, noted, local research capacity in developing coun- including well-defined networks of research cen- tries is woefully inadequate. A number of promis- ters, informal consultative bodies, and large global ing research efforts, including the Children's Vac- funds that pool donor assistance. Examples of cine Initiative and programs to deal with acute these institutional arrangements in other sectors, respiratory infections, tuberculosis, micronutrient such as the Consultative Group for International deficiencies and worm infections, suffer from Agricultural Research and the Global Environment weak and uncertain donor funding. In general, the Facility, may provide models for improving the co- problems of constrained funding for research are ordination of international health research. Box 7.5 World Bank support for reform of the health sector World Bank support for the health sector has grown hensive reforms, including granting greater manage- dramatically over the past six years. The number of ment autonomy to health facilities and decentralizing new World Bank-financed health, population, and nu- resources to the regional level. Doctors, nurses, and trition projects approved each year increased from an other health personnel are being encouraged to work average of eight during fiscal 1987-89 to twenty-one in better-equipped health centers and other basic facili- during fiscal 1990-92, and the value of credits and ties. And health-financing mechanisms are being re- loans committed each year rose from $317 million to vised, with updated fee schedules, new exemption $1,151 million over the same period. As of June 1992, procedures for the poor, and changes to the health eighty-one Bank-financed health projects were being benefits covered by existing insurance schemes. The implemented. As a share of new World Bank lending, Hospital Restructuring Project, supported by the projects for health, population, and nutrition grew World Bank, is assisting the improvement of manage- from less than 1 percent in 1987 to nearly 7 percent in ment systems and the quality of health services in the 1991. largest government hospitals, which were recently Whereas most of these projects continue the Bank's granted autonomous legal status. The project dovetails traditional support for basic health servicesincluding with the concurrent World Bank-financed Population district health infrastructure and personnel, maternal and Family Health Project, designed to improve the and child health, and control of infectious diseases quality and efficiency of public health services and es- World Bank lending for health is increasingly focusing sential clinical care, especially for mothers and chil- on broad policy reforms in the health sector. For exam- dren. It is expected that better basic services at the ple, in connection with a recent Bank project, the gov- health center level will reduce the demand for hospital ernment of Maurifania has developed a financing plan care, thus slowing the expansion of the country's to improve the availability of basic health services for hospitals. its widely dispersed population. The share of the gen- The Rumania Health Rehabilitation Project supports eral recurrent budget going to the Ministry of Public government efforts to diversify sources of health fi- Health will increase from 5.5 percent in 1992 to 7.5 nancing and thus to reduce dependence on the public percent in 1996. The project is introducing commurtity- budget, which is under pressure because of weak and based cost recovery in three of the country's thirteen unstable macroeconomic conditions and rising health regions as a way of improving the efficiency and qual- care costs. The government is pilot testing decentral- ity of services. Revenues are being raised mainly ization of health sector policymaking, planning, man- through the sale of drugs, organized and managed by agement, and evaluation in three subregions. It is also local health communities. discussing how to create a legal and regulatory envi- In Tunisia the government is carrying out compre- ronment to support reform of health financing. 169 Box 7.6 Donor coordination in the health sector in Zimbabwe and Bangladesh The Zimbabwe Second Family Health Project (1992-96) review and the administrative burden on the is the culmination of a long period of interaction be- government. tween the government of Zimbabwe, the World Bank, In Bangladesh the World Bank and ten bilateral agen- and other multilateral and bilateral donors. The $120 cies together are contributing $440 million to the Fourth million project, which supports the government's five- Population and Health Project, and the government is year investment program for population, health, and providing $165 million, for a total of $605 million over nutrition, is designed to benefit directly low-income the five-year project period. The United Nations Popu- households, especially women and children. Zim- lation Fund, WHO, and UNICEF are supplying project babwe has entered a period of economic adjustment management, procurement, and technical assistance. that will necessitate spending Cuts. The project will All the partners in the project belong to the Bangladesh help protect poor and vulnerable households from Population and Health Consortium, which has some adverse effects of adjustment by mobilizing addi- emerged as an important collective force in the health tional resources for human resource development and sector. The Asian Development Bank has joined the by improving the equity and efficiency of spending. consortium with a view to ensuring consistency be- Zimbabwean participation was emphasized from the tween the project and its own $60 million investment in project's beginning. The project preparation committee population and health in Bangladesh. included representatives from various central govern- During project formulation the government and the ment departments, provincial governments, and the donors held several workshops in Dhaka and a special Zimbabwe National Family Planning Council. The conference in Geneva. These workshops were instru- committee set planning guidelines, including a prelimi- mental in forging a consensus on population and nary outline of project components, costs, and financ- health strategies. The consortium approach enables the ing. Using these guidelines, proposals were prepared government and the donor community to agree on an locally and were then reviewed by the committee and overall strategy and to work out a consistent financing by interested donors. plan for the sector. The consortium operates on the Virtually all the major donors to the health sector in basis of strict equality of all the partners, independent Zimbabwe helped with planning the project by assign- of the size of their financial contributions. ing agency officials and technical specialists to the do- In addition to strengthening Bangladesh's popula- nor team that advised the government on design is- tion program and its delivery system for family plan- sues. In the end, the project received financing from ning services, the consortium is attempting to reorient Denmark, the European Community, Norway, the health care system toward public health, including Sweden, the United Kingdom, and the World Bank. maternal health. It is also trying to make basic services Donors monitor project implementation jointly rather more easily accessible to the rural and urban poor. than separately, reducing the time needed for donor Meeting the challenges of health policy reform Policymakers in developing countries and off i- cials of the international donor community face a If policymakers are to accelerate the substantial number of difficult challenges in pursuing this health gains of recent decades, especially for the agenda. The changing demographic profile of the poor in developing countries, the agenda for re- developing world, including the aging of the form is clear. It includes increasing overall rates of population, is creating new patterns of disease. economic growth and expanding basic schooling, Emerging microbial threats, such as AIDS and particularly for girls; reallocating government drug-resistant strains of tuberculosis and malaria, spending for health from tertiary care and special- call for changes in personal behavior, new drugs, ist training to public health measures and essential and new ways of delivering services effectively. clinical services; encouraging more diversity and In virtually every country interest groups will competition in the provision of clinical care and resist health policy reforms of the kind suggested the development of cost-containing approaches to in this Report. Health workers will object to insurance; increasing the efficiency of government changes that threaten their job security, income health services; and fostering greater involvement levels, and degree of professional autonomy. Drug of communities and households in promoting companies, medical equipment manufacturers, healthier behavior on their own part and in man- and other suppliers will try to block policies that aging their local health services. they see as having an adverse effect on their mar- 170 kets, revenues, and profits. Political and economic about reallocating resources, improving access, elites and organized labor groups will seek to pre- and increasing efficiency. To do this, higher and serve existing public subsidies for insurance and sustained rates of macroeconomic growth are re- health services from which they benefit and to quired. In many cases countries will also need to maintain their privileged access to clinical care. enact fundamental political reforms designed to Beyond this, policymakers will have to wrestle increase participation and to improve the account- with the reality that in the area of health there is no ability of governments for their health spending, simple paradigm for policy choice. Free markets service delivery, and regulatory performance. for public health activities and clinical care often The donor community has a major responsibility fail, and when governments intervene in financing to back up with concrete actions its verbal commit- and delivery, as they frequently do, they can fail ment to poverty reduction and to investment in just as badly. Effective government regulation of health and human resources. In particular, donors private suppliers of health services and inputs, should do more to support the formulation of im- combined with public financing of cost-effective proved health policies and more effective health packages of public health and essential clinical ser- sector reform programs in developing countries. vices, is needed to deal with these failures. But this As suggested in this Report, they can do this by in turn requires strong private and public institu- financing some of the transitional costs of real- tionsand institutional capacity is seriously lack- locating government budgets to public health mea- ing in many developing countries. sures and essential clinical care, by building local Despite these obstacles, there have been a num- planning, management, and research capacity, ber of successes in specific intervention programs and by providing sound assessments of the world- such as polio eradication and river blindness con- wide experience with the cost-effectiveness of in- trol and a smaller but still important number of terventions and with reform of systems. successes in broader health sector reform in such If developing country governments and donors countries as Chile, Tunisia, and Zimbabwe, as well accept the challenges and embrace the key health as in many OECD countries. These successes now policy reforms outlined above, improvements in need to be multiplied, especially in the area of sec- human welfare in the coming years will be enor- tor reform, if countries are to address the acute mous. A large share of the current burden of dis- weaknesses in existing institutional structures and easeperhaps as much as one-quarterwill be to lay the foundation for major improvements in prevented. And people around the world, espe- future living standards. cially the more than 1 billion people now living in Developing country governments need to do poverty, will live longer, healthier, and more pro- more to translate into practice today's rhetoric ductive lives. 171 Acknowledgments This Report benefited greatly from ideas, technical Adolfo MartInez-Palomo, David N. Nabarro, Gen inputs, and critical review from a broad range of indi- Ohi, Richard Peto, Kenneth I. Shine, Pravin Visaria, viduals and organizations. Contributions to specific and Richard Zeckhauser chapters are acknowledged in the Bibliographical note. In addition, valuable input was provided Consultations through four other mechanisms: a World Health Or- Ministerial Review of Health Transition Issues ganization Steering Committee, an Advisory Com- June 22-26, 1992, Bellagio Study and Conference mittee, a series of consultations on specific subjects, Center, Villa Serbelloni, Bellagio, Italy. Partial finan- and a series of seminars, mostly held at the World cial support for the consultation was provided by the Bank. Those from outside the World Bank who con- Rockefeller Foundation. tributed to the Report through these mechanisms are listed below. Participants: Alfredo R. A. Bengzon, Demissie Habte (provided written comments), Richard C. A. Fea- World Health Organization Steering Committee chem, Julio Frenk, Mamdouh Cabr, Scott Halstead, Jean-Paul Jardel, Jorge Jiménez de la Jara, Jeffrey R. This committee provided the focal point for the major Koplan, Marthini K. Budi Salyo, Adolfo MartInez- contributions from WHO to the World Development Re- Palomo, Piotr Mierzewski, Rajiv L. Misra, W. Henry port, for WHO's participation in consultations, and Mosley, Samuel Ofosu-Amaah, Raphael Owor, for its critical review of various drafts. Olikoye Ransome-Kuti, and Leonardo Santos Simão Chair: Jean-Paul Jardel Interventions for Nervous System Disorders Members: Andrew L. Creese, Michel Jancloes, Yuji July 6-7, 1992, Pan American Health Organization, Kawaguchi, R. Srinivasan, and Muthu Subramanian Washington, D.C. Chair: Benedetto Saraceno Advisory Committee Participants: Antonio Campino, Vijay Chandra, Brian The Advisory Committee met on October 7-9, 1992, Cooper, Bulent Coskun, Marcelo E. Cruz, Mary Jane at the World Bank in Washington, D.C.; in addition, England, John T. Farrar, Jefferson Fernandes, R. Juan committee members later provided critical review of Ramon de la Fuente, Walter Gulbinat, Itzhak Levav, an early draft of the Report. Thomas McGuire, Kay Redfield Jamison, Norman Sartorius, Carole Siegel, Donald H. Silverberg, T. Chair: Richard C. A. Feachem Takayanagi, and Richard Jed Wyatt Members: Jane C. Baltazar, José Barzellato, Mayra International Aid Flows to the Health Sector Buvini, Lincoln C. Chen, Antoine Degrémont, Nicholas Eberstadt, John Evans, Mahbub ul Haq, Pe- August 27, 1992, Harvard Center for Population and ter Heller, Abraham Horwitz, Jean-Paul Jardel, Rich- Development Studies, Cambridge, Massachusetts ard Jolly, Somkid Kaewsonthi, Pangu Kasa-Asila, Chair: David Bell 172 Participants: Lincoln C. Chen, Nick Drager, Ramesh Participants: Antoine Degremont, Maria Elena Ducci, Govindaraj, Eva Jespersen, Catherine Michaud, Lilia Durán Gonzales, Paul Garner, Greg Goldstein, Christopher J. L. Murray, and David Parker Emile Jeannée, Matthias Kerker, Peter Kilima, Nicolaus Lorenz, Ngudup Paijor, Voahangy Investing in Health Research Ramahatafandry, Allessandro Rossi-Espagnet, John September 16, 1992, World Health Organization, Ge- Seager, Gustavo A. Torres, and Charles Yesudian neva, Switzerland. Financial support was provided Human Resources for Health by the WHO/UNDP/World Bank Special Programme for Research and Training in Tropical Diseases. November 18-19, 1992, McMaster University, Ham- ilton, Ontario, Canada. Financial support was pro- Chair: Carlos Morel vided by the Canadian International Development Participants: Eusebe Alihouno, Barry R. Bloom, David Agency. J. Bradley, Barbro Carlsson, Gelia T. Castillo, Jac- Co-chairs: Julio Frenk and Victor Neufeld queline Cattani, Anthony Cerami, Joseph A. Cook, Henry Danielsson, Ronald W. Davis, Peter de Raadt, Participants: Orvill Adams, Barbara Carpio, Gilles David Evans, Esmat Ezzat, Tore Godal, Melba Dussauld, John Evans, Alfonso MejIa, Hiroshi Na- Gomes, E. Robert Greenberg, Win E. Gutteridge, katani, Kenneth Ojo, Una Reid, Charas Suwanwela, Ralph H. Henderson, H. Robert Horvitz, Nathan K. and Peter Tugwell Kere, Veronique Lawson, David N. Nabarro, Richard Health of the Elderly Peto, Vulimiri Ramalingaswami, Peter Reeve, Hans Remme, Lateef Akinola Salako, Ebrahim M. Samba, November 23-24, 1992, Voksenasen, Norway. The Vladimir Petrovich Sergiev, Carol Viassoff, Gabisiu consultation was organized by the London School of A. Williams, and Richard Wilson Hygiene and Tropical Medicine. Financial support was provided by the Norwegian Ministry of Foreign Resource Flows to the Health Sector Affairs. September 21, 1992, Harvard Center for Population Chair: Alex Kalache and Development Studies, Cambridge, Massachusetts Participants: Jordi Alonso, Nana Apt, Chris Beer, Felix Chair: William Hsiao Bermejo, Ruth Bonita, Carol Brayne, Paul Chen, Participants: Sudhir Anand, Peter Berman, Mirnal Yolande Coombes, Xianglin Du, Denise Eldemire, J. Dutta Choudray, Gnanaraj Chellaraj, Lincoln C. Grimley Evans, Richard G. A. Feachem, Joe Hamp- Chen, Michel Cichon, Andrew Creese, Ramesh son, Hana Hermanova, Benedicte Ingstad, Zhang Govindaraj, Catherine Michaud, Christopher J. L. Kaiti, Roberto Kaplan, Luis Ramos, Melba Sanchez- Murray, Sudhakar Rao, and George Scheiber Ayéndez, Bela Shah, Alberto Spagnoli, Knight Steel, and Renato Veras AIDS Policy District Health Services November 5-6, 1992, Rockefeller Foundation, New York. Financial support was provided by the Rock- November 24-27, 1992, Institute of Health and Devel- efeller Foundation and the Danish International De- opment, University of Dakar, Senegal. Financial sup- velopment Agency. port was provided by Swiss Development Coopera- tion and the Swiss Tropical Institute. Chair: Robert S. Lawrence Co-chairs: Antoine Degremont and Ibrahima Wone Participants: Roy Anderson, José Barzellato, Seth Berkley, Robert Black, Kevin De Cock, Richard G. A. Participants: Abdel Wahed Abassi, Waya Amoula, Feachem, Penelope Hitchcock, King Holmes, Robert Anarfi Asamoa-Baah, Hubert Balique, Wolfgang E. Howells, Jane Hughes, Peter Lamptey, Jonathan Bichmann, Malang Coly, Christian Darras, Pierre Mann, Michael Merson, Daan Mulder, Peter Piot, Daveloose, Annemarie Demazy, Issakha Diallo, Isseu Peer Sieben, Werasit Sittirai, Judith Wasserheit, Fer- Diop-Touré, Gina Etheridge, Georges Fournier, Lucy nando Zacharias, and Richard Zeckhauser Gilson, Kathia Janovsky, Emile Jeannée, Pangu Kasa- Asila, Matthias Kerker, Vincent Litt, Mandiaye Urban Health Loum, Javier Martinez, Sigrun Mogedal, Maty Cissé Samb Ndao, Sène Touré Ngone, Cornelius Oepen, November 9-11, 1992, Basel, Switzerland. Financial Bakary Sambou, Lamine Cissé Sarr, Malick Sarr, support was provided by Swiss Development Coop- Peter Schubarth, Michael Singleton, Thierno Maine eration and the Swiss Tropical Institute. Aby Sy, Al Hadji Ali Tahirou, James Tumwine, Jean- Co-chairs: Marcel Tanner and Trudy Harpham Pierre Unger, Adamou Yada, and Alfredo Zurita 173 Child Health Romer, Elizabeth Sherwin, Peter Smith, Jan Stjern- November 30-December 2, 1992, Baltimore, Mary- swàrd, Rand Stoneburner, Muthu Subramanian, land. The consultation was organized by the Johns Carole Torel, and Godfrey Walker Hopkins School of Hygiene and Public Health. Finan- Health Finance cial support was provided by the U.S. Agency for International Development. Dkember 14-16, 1992, Montebello, Quebec, Canada. Financial support was provided by the Canadian In- Chair: W. Henry Mosley ternational Development Agency. Participants: Fernando Barros, Al Bartlett, Mark Bel- Chair: Stephen Simon sey, Seth Berkley, Robert E. Black, David Boyd, Donald A. P. Bundy, Carlos C. Campbell, Dennis Participants: Nicholas Barr, David Bell, Ricardo Bitran, Carroll, Robert Clay, Felicity Cutts, Steve Esrey, Ake Blomqvist, Joseph Brunet-Jailly, Claude Cas- Ronald Gray, Jerry Gibson, Bill Hausdorf, Jim Heiby, tonguay, Andrew L. Creese, Robert G. Evans, Donald A. Henderson, Terrel Hill, Sandra L. Huff- Claude Forget, William Hsiao, Naoki Ikegami, Daniel man, Jessica Jitta, Pamela Johnson, Charlotte Neu- M. Le Touzé, Mario Taguiwalo, Abdelmajid Tibouti, man, Alok Perti, Phyllis Piotrow, Kenneth F. Schulz, Bokar Touré, and Katarzyna Tymowska Jim Shepperd, William A. Smith, Sally Stansfield, Review of WDR Findings Hope Sukin, Nebiat Tafari, Taha el Tahir Taha, Carl E. Taylor, James L. Tulloch, Roxann Van Dusen, Caby C. January 26, 1993, Institute of Medicine, Washington, Verzosa, Kenneth S. Warren, and Vivian Wong D.C. Women and Health Chair: William H. Foege December 7-9, 1992, Cumberland Lodge, Windsor, Participants: Abdelmonem A. Afifi (provided written England. The consultation was organized by the Lon- comments), Carolyn Asbury, David E. Bell, Richard don School of Hygiene and Tropical Medicine. Finan- Bissell, Barry R. Bloom, Margaret Catley-Carlson, J. cial support was provided by the U.K. Overseas De- Jarrett Clinton, Joseph A. Cook, Richard G. A. Fea- velopment Administration. chem, Harvey V. Fineberg, Julio Frenk, Susan Gibb, Polly F. Harrison, Donald A. Henderson, Jeffrey R. Organizers: Oona Campbell, Wendy Graham, and Veronique Filippi Koplan, Adetokunbo 0. Lucas, Christopher J. L. Murray, June E. Osborn, Adeline Wynante Patterson, Participants: Uche Amazigo, Carmen Barroso, Loretta David P. Rall, Frederick C. Robbins, Timothy Ro- Brabin, Mayra Buvinic, Mirai Chatterjee, Ann Coles, thermel, Kenneth I. Shine, Alfred Sommer, Roxann Richard G. A. Feachem, Zuzana Feachem, Aleya El Van Dusen, Noel S. Weiss, Barbara L. Wolfe, and Bindari Hammad, Sioban Harlow Gillian Holmes, James Wyngaarden Susan Joekes, Marjorie Koblinsky, Joanne Leslie, Environment and Health Claudia Garcia Moreno, Jacky Mundy, Cynthia Myntti, David N. Nabarro, Phoebe Roome, Kasturi February 4-5, 1993, World Bank, Washington, D.C. Sen, Jacqueline Sherris, Godfrey Walker, and Judith Financial support was provided by the Environmen- Wasserheit tal Health Division, World Health Organization. Global Burden of Disease Chair: Wilfried Kreisel December 10-11, 1992, World Health Organization, Participants: Hendrik De Koning, Devra Lee Davis, Geneva, Switzerland. The consultation was orga- Richard G. A. Feachem, Jacobo Finkelman, Gregory nized by the World Health Organization. Financial Goldstein, Tord Kjellstrom, Anthony J. McMichael, support was provided by the Edna McConnell Clark Horst Otterstetter, David P. Rall (provided written Foundation. comments), and Kirk Smith Chair: Jean-Paul Jardel Improving the Effectiveness of International Assistance to Health Participants: Carla Abou-Zahr, David Barmes, Monika Blössner, Luis Lopez Bravo, Anthony Burton, Yan- February 9-10, 1993, World Bank European Office, kum Dadzie, Richard G. A. Feachem, Jacques Ferlay, Paris, France. Partial financial support was provided Tore Godal, Ann Goerdt, Sandra Gove, Walter by the Danish International Development Agency. Gulbinat, Habib Rachmat Hapsara, Joachim Hempel, Mark Kane, Hilary King, Jeffrey R. Koplan, Jacob Chair: Anthony R. Measham Kumaresan, Marie-Hélène Leclerq, Alan Lopez, In- Participants: Marja Antilla, Lynn Bailey, Jose Bar- grid Martin, Alvaro Moncayo, Christopher J. L. Mur- zellato, Alfredo R. A. Bengzon, Luciano Carrino, Ge- ray, Jenny Pronczuk, Jean-Marie Robine, Claude nevieve Chedville-Murray, Zafrullah Chowdhury, 174 Immita Cornaz, Göran Dahigren, Francois Decaillet, Chair: Jeffrey R. Koplan Nicolas de Riviere, Tore Godal, Klaus Gordel, Arm- Participants: Ruth L. Berkelman, Ruth A. Etzel, Fran- elle George-Guiton, Jacques Hallak, Kyo Hanada, çoise F. Hamers, Jeffrey R. Harris, Nancy C. Lee, Anne Kristin Hermansen, Gillian Holmes, Jean-Paul Alan Lopez, Christopher J. L. Murray, Mark L. Jardel, Eva Jesperseri, Jorge Jimenez de la Jara, Mat- Rosenberg, Richard B. Rothenberg, Frank M. Vinicor, thias Kerker, Robert Kestell, Irene Klinger, Roif and Ray Yip Korte, Louise Lassonde, Jean-Marie Laure, Robert S. Lawrence, Rune Andreas Lea, Dominique Maroger, Seminars Catherine Michaud, Rajiv L. Misra, Bernard Mon- taville, W Muchenje, David N. Nabarro, Francois An important source of ideas for this Report was a Orivel, Tom Ortiz, Aagje Papineau Saim, Liu series of seminars. Most were held at the World Bank Peilong, Ines Penn, Martin Pinero, Peter Poore, Vu!- and were cosponsored by the World Bank's Popula- imiri Ramalingaswami, Olikoye Ransome-Kuti, Brett tion, Health, and Nutrition Department. The Har- Ridgeway, Jon Rohde, Yolanda Richardson, Kenneth vard Center for Population and Development Studies Ross, Timothy Rothermel, Philippa Saunders, Chris- held a series of five seminars to assist in developing topher Shaw, Leonardo Santos Simão, Stephen Si- World Development Report themes; these were orga- mon, Margareta Sköld, Guillermo Soberon, Birgit nized by Lincoln C. Chen and Julio Frenk. The Storgaard, Muthu Subramanian, Carl Wahren, Ronald Wilson, Robert Wrin, Carlos Yanez-Barneuvo, George Washington University Center for Interna- tional Health held a seminar, organized by RosalIa and Pat Youri Rodrigues-GarcIa, to critically review the findings of Violence against Women the World Development Report. The World Bank series included presentations by February 12, 1993, Washington, D.C. This follow-up Henry Aaron, John Akin, Kenneth Arrow, Amie Bat- consultation to that on Interventions for Nervous son, Jere Behrman, David Bloom, Michael Cichon, Systems Disorders focused on the health outcomes of Andrew Creese, Anil Deolalikar, Avi Dor and Janet violence against women for the global burden of dis- Hunt-McCool, Alain Enthoven, Michelle Fryer, Paul ease exercise. Gertler, Eric Hanushek, Estelle James, Lawrence J. Chair: Helen Saxenian Lau, Beryl Levinger, Joseph Newhouse, Abdel Omran, Joel Nobel, Francois Orivel, Charles Phelps, Participants: Jacqueline Campbell, Walter Gulbinat, Lori Heise, Dean Kilpatnick, and Christopher 1. L. Samuel Preston, Barry Popkin, Uwe Reinhardt, George Scheiber, T. Paul Schultz, Donald Shepard, Murray John Strauss, Duncan Thomas, Carol Vlassoff, and Review of the Global Burden of Disease Beverly Winikoff. March 15, 1993, Centers for Disease Control and Pre- vention, Atlanta, Georgia 175 Bibliographical note The Report has drawn on a wide range of World Bank Those outside the World Bank who contributed reports and advice and on numerous outside sources. substantially with comments and material include Special thanks go to the World Health Organization Aloysio Achutti, Universidade Federal do Rio Grande (WHO) for providing extensive expert advice, techni- do Sul, who assisted with the preparation of Figure 3; cal materials, and helpful comments. A. A. Afifi, University of California, Los Angeles; The principal sources are noted below and are also Jere Behrman, University of Pennsylvania; Marit listed alphabetically by author or organization in two Berggrav, Einar Heldal, Rune Andreas Lea, Johanne groups: background papers commissioned for this Sundby, and Ann-Karin Valle, Norwegian Agency Report and a selected bibliography. for International Development (NORAD); Barry In addition to the sources listed, many people both Bloom, Albert Einstein College of Medicine; Robert inside and outside the World Bank helped with the H. Cassen, International Development Centre, Ox- Report. In particular, helpful comments were re- ford University; Immita Cornaz, Swiss Development ceived from World Bank staff and consultants, includ- Cooperation; Göran Dahlgren, Swedish International ing Alexandre Abrantes, Masood Ahmed, Michael Development Authority (SIDA); Joe H. Davis, Cen- Azefor, Howard Barnum, Alan Berg, Eduard Bos, Pa- ters for Disease Control and Prevention; Antoine De- tricia Daly, Willy De Geyndt, Janet de Merode, Jean- grémont, Swiss Tropical Institute; David Fraser, Aga Jacques de St. Antoine, Dennis de Tray, Alfred Duda, Khan Institute; Lucy Gilson, London School of Hy- Graham Dukes, Oscar Echeverri, A. Edward Elmen- giene and Tropical Medicine; Ted Greiner, Uppsala dorf, James Green, Charles Griffin, Ann Hamilton, University; Davidson Gwatkin, International Health Jeffrey Hammer, Barbara Herz, Janet Hohnen, Ishrat Policy Program; David J. Halliday, UNICEF; Gillian Z. Husain, Estelle James, Emmanuel Jimenez, Eliz- Holmes and David Nabarro, U.K. Overseas Develop- abeth King, Timothy King, Mubina Kirmani, Kathie ment Administration (ODA); William C. Hsiao, Har- Krumm, Joseph Kutzin, Jean-Louis Lamboray, Kye vard School of Public Health; Valerie Hull, Australian Woo Lee, Danny M. Leipziger, Maureen Lewis, Sam- International Development Assistance Bureau uel Lieberman, Bernhart Liese, James Listorti, Mar- (AIDAB); Pamela Johnson, Richard Seifman, and lame Lockheed, Jack Maas, Jo Martins, Judith Robert Wrin, U.S. Agency for International Develop- McGuire, Mohan Munasinghe, Rieko Niimi, Mead ment (USAID); Joanne Leslie, UCLA School of Public Over, Lisa Pachter, Ok Pannenborg, David Peters, Health; Adetokunbo 0. Lucas, Harvard University; Ian Porter, Juan Prawda, George Psacharopoulos, A. J. McMichael, University of Adelaide; Peter Poore, Sandra Rosenhouse, Anna Sant'Anna, Miguel John Seaman, and David Woodward, Save the Chil- Schloss, Julian Schweitzer, lona Sebastian, Paul dren (U.K.); Barry Popkin, University of North Caro- Shaw, James Socknat, Lyn Squire, Andrew Steer, lina; Vulimiri Ramalingaswami, Task Force on Health Susan Stout, Vinod Thomas, Erik Thulstrup, Anne Research for Development; Patricia L. Rosenfield, Tinker, Vincent Turbat, Jagadish Upadhyay, Denise Carnegie Corporation; Timothy S. Rothermel, United Vaillancourt, Armand Van Nimmen, Herman van der Nations Development Programme (UNDP); A. Pa- Tak, Dominique van de Walle, Claudia Von Monbart, pineau Saim, Ministry of Foreign Affairs, Nether- Marie-Odile Waty, Kin Bing Wu, Guillermo Yepes, lands; Philippa Saunders, OXFAM; Alfred Sommer Mary E. Ming Young, and Shahid Yusuf. and Carl Taylor, Johns Hopkins School of Hygiene 176 and Public Health; Birgit Storgaard, Ministry of For- Bank-UNICEF exercise to be used in UNICEF forth- eign Affairs, Denmark; Noel S. Weiss, University of coming and described in Hill and Yazbeck, back- Washington; and Hans Emblad, Tore Godal, Marcus ground paper. Eduard Bos and My Vu of the World Grant, Fritz Kaferstein, Wilfried Kreisel, Alan D. Bank's Population, Health and Nutrition Department Lopez, and James C. Tulloch, WHO. provided invaluable assistance and advice with the base regional population projections underlying Chapter 1 much of Chapter 1 and Appendix A. Mortality as- sumptions were updated and revised in the light of This chapter draws on technical materials from the discussions with Larry Heligman of the United Na- World Health Organization and the World Bank and tions Population Division, Gareth Jones of UNICEF, on the scientific literature. The smallpox story is and recent data from the Demographic and Health adapted from Fenner and others 1988. The discussion Surveys program provided by Ties Boerma or extrac- of the gains in worker productivity from better health ted from recent reports. Nicholas Eberstadt contrib- draws on studies by Castro and Mokate 1988, Conly uted useful ideas on mortality differentials in adult- 1975, Max and Shepard 1989, Nur and Mahran 1988, hood, and the section further benefited from Pitt, Rosenzweig, and Hassan 1990, Sagan and Afifi Feachem and others 1992. 1979, Schultz and Tansel 1993, and Hill and others, The results in the section on the global burden of background paper, as well as on helpful materials disease are taken from a joint World Bank-World provided by John Caldwell, Gavin Jones, and John Health Organization study (Murray and Lopez back- Anarfi. Anil Deolalikar provided additional material ground paper); many collaborators are listed in Ap- on the economic impact of improved nutrition in In- pendix B. Material on measuring the burden of dis- dia, and John Akin made available unpublished notes ease for Ghana, to establish health care priorities, on the relationship between health and income. The came from Ghana Health Assessment Project Team cost-benefit calculations of malaria eradication in Sri 1981. Feachem 1988 stresses the importance of "mac- Lanka are derived from Barlow and Grobar 1985. roepidemiology" for health planning. Aehyung Kim and Bruce Benton contributed to Box The section on challenges for the future is based on 1.1. Institute of Medicine 1992, Mackay 1993, and WHO The discussion of the education benefits of im- 1992b and 1992c and on information on HIV and proved health and the related economic benefits of AIDS provided by Seth Berkley, Rand Stoneburner, improved education is based on studies by Behrman and WHO staff. D. A. Henderson provided informa- and others 1991, Boissiere, Knight, and Sabot 1985, tion on emerging microbial infections; Tore Godal, Glewwe 1991, Gomes-Neto and Hanushek 1991, Jam- Tekle Haimanot, and Hans Remme on malaria; and ison and Leslie 1990, Jamison and Moock 1984, Nokes Alan D. Lopez and Neil Collishaw on smoking. and others 1992, and Psacharopoulos 1993. Jacobs and others 1993 report on the development of The section on reduced costs of medical care draws a test for drug resistance in tuberculosis. on studies by Ainsworth and Over 1992, notes by The discussion on demographic and epidemiologic David Bloom and Ajay Mahal on the implications of transition draws on studies by Bobadilla and others reducing the rate of HIV transmission among Se- forthcoming, Frenk and others 1989, and Omran ropositive individuals (with additional personal com- 1971. Country-specific discussions of the implications munication from the authors) and Musgrove 1988. of epidemiological transition for health policy may be Martha Ainsworth and Mead Over drafted Box 1.2. found in World Bank 1984a, World Bank 1990a, and Bloom and Lyons 1993 provide analyses pointing to World Bank 1992a. The purchasing power parity per the economic gains associated with AIDS prevention capita incomes used in Figure 1.9 were provided by in a number of Asian countries. Angus Maddison. The discussion of the factors ex- The discussion of the impact of health investments plaining mortality declines is based on Ewbank and on poverty draws on the work of Henry Mosley and Preston 1990, McKeown 1976, and Preston and on World Bank 1980 and 1990a. The record of success Haines 1991. draws heavily on statistical publications of the United The structure and content of the chapter benefited Nations (Demographic Yearbook, various years) and th from presentations made by Abdel Omran and Sam- World Health Organization (Statistics Annual, various uel Preston. Valuable comments on earlier drafts years). Ingram 1992 discusses the greater conver- were made by Joseph Cook and Richard Morrow. gence of social (including health) indicators than of income across countries. Chapter 2 Child mortality estimates are derived in part from United Nations 1988 but were mostly calculated from This chapter draws on academic sources, presenta- data in United Nations 1992 as part of a joint World tions by speakers in the seminar series cosponsored 177 by the World Development Report and the World Bank and 2.5, respectively. Robert Anda, David Bradley, Population, Health and Nutrition Department, and John Briscoe, Mayra Buvinic, Brigitte Duces, Luis Es- on numerous World Bank documents. The discussion cobedo, Paul Gertler, A. K. Shiva Kumar, Joanne on household capacity (income and schooling) was Leslie, Ruth Levine, Jack Molyneaux, Damianos informed by studies that included Anand and Ray- Odeh, Nick Prescott, Luis Serven, John Strauss, and allion 1993, Behrman 1990, Benefo and Schultz 1992, Molly Tees contributed helpful data and resource ma- Fuchs 1979, Grossman 1975, Hill and Palloni 1992, terials. Valuable comments on earlier drafts were re- Jeyaratnam 1985, Lau and others, background paper, ceived from Sue Berryman, Joseph Bredie, Barbara Luft 1978, Natale and others 1992, Oganov 1992, Pal- Bruns, Ishac Diwan, Edward Henevald, Eva Jarawan, loni 1981, Pierce 1989, Pritchett and Summers, back- Himelda Martinez, Kenneth Shine, and David ground paper, Rodgers 1979, Rogot, Sorlie, and John- Woodward. son 1992, Strauss and others 1992, United Nations 1985, Wilkinson 1992, and World Bank 1990a. Chapter 3 The discussion on women's schooling and child health drew on the extensive literature, including This chapter draws on a wide range of published Bhargava and Yu 1992, Bruce and Lloyd 1992, Cald- and unpublished sources, including documentation well 1986, Cleland 1990, Elo 1992, Engle 1991, Hod- and expertise from the World Health Organization dinott and Haddad 1991, Kennedy 1992, King and and the World Bank and on the academic literature. Hill 1993, Leslie 1989b, Lindenbaum, Chakraborty, Discussion of the role of government draws on World and Elias 1985, Louat, Grosh, and van der Gaag 1992, Bank 1991. Jamison and others forthcoming summa- Over and others 1992, Sahn 1990, Summers 1992, rizes the methods and findings of the cost effective- Thomas 1990, Thomas, Strauss, and Henriques 1990, ness analysis that forms the starting point for the and World Bank Water Demand Research Team 1993. analyses used in this report. The data on health ex- The discussion on economic policy reform and ad- penditures in the first section were compiled from a justment lending drew on sources that included background paper by Murray, Govindaraj, and Behrman 1992, Berg and Hunter 1992, Edwards forth- Chellaraj, which used a wide range of government coming, Kakwani, Makonnen, and van der Gaag health budgets, World Bank reports, and other coun- 1990, Serageldin, Elmendorf, and El-Tigani forthcom- try studies of health financing. Heller and Diamond ing, Summers and Pritchett 1993, Thomas, Lavy, and 1990 also treat this issue. Data on equity in health Strauss 1992, Woodward 1992, World Bank 1990b, status, access, and expenditure were drawn from the World Bank 1992e, and World Bank 1993c. Duncan World Bank's Living Standard Measurement Study Thomas contributed materials on protecting nonsal- (LSMS) and were further analyzed by Kalpana ary spending during economic adjustment. Mehra. The analysis of costs and benefits of packages The presentation on education policies was in- of public health measures and essential clinical ser- formed by Alderman and others 1992, Jamison and vices draws on the background paper by Bobadilla Leslie 1990, Jarousse and Mingat 1992, Lockheed, and others. Verspoor, and associates 1991, Minhas 1991, Over Box 3.1 was drafted by Richard Bumgarner. Box 3.2 and Piot forthcoming, Tan and Mingat 1992, and is based on unpublished data provided by the Insti- World Bank 1988. tuto Materno-Infantil de Pernambuco and on The work on policies for empowering women drew UNICEF and IMIP 1992. Box 3.3 is from the chapters on Akin and others 1985, Birdsall and McGreevey on measles and on tuberculosis in Jamison and others 1983, and Leslie 1989a. The discussion of women and forthcoming. The discussion of cost-effective inter- violence benefited from assistance from Jacquelyn ventions also draws on Halstead, Walsh, and Warren Campbell, Rosemary Garner, Lori Heise and Dean 1985, Walsh 1988, and Walsh and Warren 1979. Basic Kilpatrick and drew on Archavanitkui and Pram- economic issues and their application are treated in aualrantan 1990, Bradley 1988, CAMVAC 1985, Coun- Over 1991. The discussion of market failures in health cil on Scientific Affairs 1992, COVAC 1990, Fauveau draws particularly on Arrow 1963. Insurance and reg- and Blanchet 1989, Handwerker 1991, Hosken 1988, ulation are discussed generally in Diamond 1992 and Koop 1989, Koss, Koss, and Woodruff 1991, Plitcha for Brazil in Piola and Vianna 1991. The section on 1992, Shim 1992, Stark 1984, and Stark and Flitcraft government failures in health policy takes examples 1991. from Evans, Barer, and Labelle 1988, Hlady and Lawrence Lau contributed to the drafting of Box others 1992, and 1DB 1988. Equity examples are 2.1. Box 2.2 is based on material provided by Michelle drawn from Black and others 1982, Meerman 1980, Fryer. Carmen Barroso, Lori Heise, and Nahid Toubia Musgrove 1986 and 1993, and President's Commis- contributed to Box 2.3. John Hobcraft and Aloysio sion 1983, as well as from the work of Prescott and Achutti assisted with the preparation of Figures 2.4 others on social spending in Indonesia. The discus- 178 sion of satisfaction with health care uses Bitran and 1990. Joanne Leslie contributed Box 4.1, and Jayshree Mclnnes 1993, Blendon and others 1990, and Gertler Balachander contributed Box 4.2. Harold Alderman, and van der Gaag 1990. Leslie 1989a discusses the George Beaton, Robert Black, Barry Bloom, Leslie time cost of health interventions, an issue that is not Elder, Paul Elliott, Abraham Horwitz, Suraiya Ismail, explicitly addressed in the cost-effectiveness calcula- Francisco Mardones, Reynaldo Martorell, John tions reported here but that deserves further work. Mason, Paul McKeigue, Daan Mulder, Philip Payne, Brook and Lohr 1986 provide evidence pointing to David Pelletier, and Peter Piot provided helpful infor- huge overuse of medical care in the United States mation or comments. beyond what is of value even at zero cost, resulting in The section on fertility drew on Cochrane and Mer- part from third-party financing. rick, background paper, AbouZahr and Royston Nicholas Barr, Peter Diamond, Robert Evans, and 1991, Amadeo, Chernichovsky, and Ojeda 1991, Fernando Figueira provided valuable ideas and Bertrand and Brown 1992, Population Information comments. Program 1992, Population Reference Bureau 1992a Chapter 4 and 1992b, Sanderson and Tan forthcoming, Ste- phenson and others 1992, United Nations forthcom- This chapter draws on documentation and expertise ing, World Bank 1984a, 1992c, and 1993a, and from the World Health Organization and the World Zinanga 1992. Birgitta Bucht, Parker Mauldin, Vin- Bank and from the academic literature, as well as on cent Miller, Richard Osborn, Warren Sanderson, Bev- expert consultations and on papers and discussions erley Winikoff, and the staff at the Rockefeller Foun- in the seminar series sponsored by the World Develop- dation provided helpful materials and advice. John ment Report and the World Bank Population, Health, Hobcraft assisted in the preparation of Figure 4.3. and Nutrition Department. The section on tobacco, alcohol, and drugs bene- The discussion on immunization and other popula- fited from materials and comments from Jerry Husch, tion-based health services draws on ideas and data Judith Mackay, Richard Peto, and Derek Yack. The discussed at the consultation on Child Health held in discussion drew on background materials from James Baltimore in 1992. Berkley and Jamison 1991 discuss Cercone and from the U.S. Surgeon-General's 1992 the cost and effectiveness of school-based programs report on smoking in the Americas, as well as on for mass treatment of worm infections and micro- Gutierrez-Fisac, Regidor, and Ronda 1992, Pierce nutrient deficiencies. Assistance was also provided 1991, Walsh and others forthcoming, Wasserman and by Amie Batson, Donald Bundy, Pamela Johnson, others 1991, and WHO 1991b and 1992e. Marjorie Koblinsky, Jim Shepperd, Jacqueline The section on the environment benefited from the Sherris, and Nebiat Tafari. Other sources were Bour- contributions of participants in a joint WHO-World don, Orivel, and Perrot 1993, Brenzel 1990, Nokes Bank consultation (see Acknowledgments) and from and others 1992, Robertson and others 1992, Shepard additional assistance provided by Carl Bartone, and others 1989, and chapters in Jamison and others David Bates, Sue Binder, Gloria Davis, Roger Detels, forthcoming on measles, polio, hepatitis B, tetanus, John Dixon, Mohamed T. El-Ashry, Gunnar Eske- and helminth infection. land, Ruth Etzel, Philip Graitcer, Peter Kolsky, Tony The section on diet and nutrition drew on Levin McMichael, David Rail, Anand Seth, and Anthony and others forthcoming, Pinstrup-Anderson and Zwi. Data on the health impact of water supply and others forthcoming, and a variety of other sources. sanitation were taken from the extensive literature Valuable summaries of particular topics are given in and from recent reviews by Cairncross 1990, Esrey ACC/SCN 1991, 1992a, and 1992b, Beaton and and others 1991, and Huttly 1990. The material on Ghassemi 1987, Beaton and others 1993, Berg 1987, water and sanitation policy drew on Briscoe 1992, Drèze and Sen 1989, Elliott 1988, Humphrey, West, World Bank 1992f, and World Bank Water Demand and Sommer 1992, Keusch and Scrimshaw 1986, Les- Research Team 1993. Box 4.4 was drafted by Sandy lie 1987, Leslie, Jamison, and Musgrove forthcoming, Cairncross. Box 4.5 relies on Blum and others 1990 McGuire and Popkin 1990, Monteiro 1988, National and on Feachem and others 1978 for time spent col- Research Council 1989, Pelletier 1991, Pollitt 1990, lecting water. Michael Garn, Letitia Obeng, and Popkin 1993, Sen 1981, Tomkins and Watson 1989, Guillermo Yepes contributed data on water and sani- and U.S. Centers for Disease Control and Prevention tation costs. Greg Watters collated the data in Figure 1992. The section also drew on studies by Adair and 4.6. The discussion of indoor air pollution rests on the others 1993, Bhargava 1992, Black 1991, Bouis 1990, reviews by Betty Kirkwood and colleagues and on the Lutter and others 1992, Mardones and Zamora 1989, work of Kirk Smith. Relevant literature included An- Martorell and others 1992, Musgrove 1990, Stamler derson 1979, Chapman and others 1989, Chen and and others 1989, Thomas, Lavy, and Strauss 1992, others 1990, Norboo and others 1991, Pandey and Waaler 1984, and Walter, Olivares, and Hertrampf others 1989, Smith forthcoming, Smith and Liu 1993, 179 and Smith and Rodgers 1992. Christopher Curtis pre- Material on the success of prevention was drawn pared Box 4.6, with assistance from Cohn Leake, from the literature and from a meeting at GPA/WHO making use of data from Alonso and others 1991, in 1992. The costs and benefits of the public health Curtis 1992, and Maxwell and others 1990. Discus- measures in the essential package are presented in sion of housing policy was informed by World Bank Bobadilla and others, background paper. 1993b. The discussion of the wider environment drew on Chapter 5 Doll 1992 and on the comprehensive accounts con- tained in WHO 1992d and World Bank 1992f. The The costs and benefits of the clinical services in the discussion on occupational health drew on Andreoni essential package are described in Bobadilla and 1986, El Batawi and Husbumrer 1987, and Wegman others, background paper. The discussion of the cost 1992. The discussion of the ambient environment of the essential package of clinical services and mech- drew on the extensive literature, including Bellinger anisms for delivering it drew on the work of the and others 1987, Bradley and others 1992, Faiz and World Bank's Africa Technical Department and others forthcoming, Lancet 1992, MRC 1989, Needle- World Bank forthcoming. The components of the es- man and others 1990, Romieu 1992, Romieu, sential package of clinical services for children drew Weitzenfeld, and Finkelman 1990, Schwartz and on analyses from UNICEF 1993 and on priorities pro- Dockery 1992, WHO 1992a, and WHO/UNEP 1992. posed in UNICEF, WHO, and UNESCO 1991. The The material for Box 4.7 is taken from Bobak 1993, analyses of sources of health financing, provider Bobak and Feachem 1992, Bobak and Leon 1992, and compensation, and alternative modes of service de- World Bank 1992d. Box 4.8 is taken from Study livery are based on the work of Arrow 1963, Barr Group for Global Environment and Economics 1991, 1992, Griffin 1992, Hsiao 1992, Hurst 1992, Reinhardt supplied by Tord Kjellstrom. José Carbajo, Paul Gui- 1991, Schneider and others 1992, van Doorslaer, Wag- tink, Zmarak Shalizi, and John Wootton assisted with staff, and Rutten 1993, and World Bank 1992a. It the section on transport risks, which also drew on draws on and is closely linked with a series of papers Barss and others forthcoming, Downing 1991, John- by the World Health Organization: WHO 1991a, on ston 1992, Smith and Barss 1991, TRRL 1991, WHO health care reform in Eastern and Central Europe; 1989a, and Zwi 1992. WHO 1991c, on the public-private mix; and WHO The section on AIDS benefited from the contribu- 1993. tions made by members of the AIDS consultation (see The analysis of user charges and community fi- Acknowledgments) and from additional assistance nancing draws on the work of Abel-Smith and Dua from Richard Hayes, Daan Mulder, Peter Piot, 1988, Gertler and van der Gaag 1990, Hecht, Over- Wendy Roseberry, Allan Rosenfield, Gary Slutkin, holt, and Holmberg 1993, Korte and others 1992, and Peter Smith. Projections of numbers of infections Lewis and Parker 1991, Litvack and Bodart 1993, were generated by Tony Burton, Rand Stoneburner, McPake, Hanson, and Mills 1992, and others. The and other staff of the Global Programme on AIDS of discussion of health insurance in developing coun- the World Health Organization (GPA/WHO). Mate- tries draws on Abel-Smith 1992b, De Geyndt 1991, rial on the core groups is drawn from Moses and Ikegami 1992, Kutzin and Barnum 1992, McGreevey others 1991 and Over and Piot forthcoming. The ac- 1990, Mesa-Lago 1992, Vogel 1989, Yang 1991, and Yu count of community intervention in Zimbabwe is and Anderson 1992. Analysis of the determinants of drawn from material supplied by David Wilson. Ma- health spending in the OECD countries draws on terial on HIV and breastfeeding is drawn from Dunn Gerdtham and others 1992. The review of options for and others 1992 and various WHO materials. Box 4.9 improving public and private delivery of clinical ser- draws on Goodgame 1990, Katabira and Goodgame vices draws on Bennett 1992, Foster 1991, and World 1989, and Muller and others 1992. Information on Bank 1992g. The discussion of managed competition voluntary testing and counseling is drawn from Fos- relies on Enthoven 1988 and Relman 1993. The dis- ter 1990, Muller and others 1992, and WHO Global cussion of decentralization of health services draws Programme on AIDS 1993b. Estimations of the cost of on Mills and others 1990 and World Bank 1992b. worldwide prevention were drawn from WHO Box 5.1 draws on material produced by Marjorie Global Programme on AIDS 1993a and WHO forih- Koblinsky and on Tinker and Koblinsky 1993 and coming, and from work by Doris Schopper. The dis- Walsh and others forthcoming. Box 5.2 was prepared cussion of the cost-effectiveness of treating sexually with information provided by James C. Tulloch and transmitted diseases is drawn from Moses and others Sandra Gove of WHO. Box 5.3 draws on material 1992, and Over and Piot forthcoming. Box 4.10 is provided by Judith Wasserheit. Box 5.4 was based on drawn from Viravaidya, Obremsky, and Myers 1991 Murray, Styblo, and Rouillon forthcoming. Box 5.5 and from materials contributed by Werasit Sittitrai. draws on Grosh 1992. Box 5.6 uses material from 180 Schieber, Poullier, and Greenwald 1992 and the re- coming, Reyes and Picazo 1990, Richards and Fulop suits of analysis by the WDR team of the relationship 1987, Schmidt and others 1991, Schroeder 1984 and between the public share of health spending and 1992, Schwab 1987, Tarlov 1986 as cited in Reinhardt health care costs. Box 5.7 is based on material from 1991, Welch and others 1993, Whitfield 1987, and un- Hurst 1992. Werner 1987 discusses approaches to published material from Ruth Roemer and WHO. In low-cost but effective rehabilitation from disability. Box 6.2 the discussion of community health workers Peter Berman and Louis Vassiliou provided useful in Jamaica draws on Cumper and Vaughan 1985. The material and Alfred Bartlett, Ricardo Bitran, Michael discussion of the Pastoral da Crianca draws on mate- Cichon, Andrew Creese, Jennie Litvack, Kasa Asila rials provided by the Coordenacao Nacional da Pas- Pangu, John Rohde, Abdelmajid Tibouti, Jacques van toral da Crianca and on Victora and Barros 1990. der Gaag, Ronald Wilson, and Zia Yusuf made valu- The drug discussion draws on Andersson 1992, able comments. Caplan 1985, Foster 1990, Hlady and others 1992, Holly and Lee 1992, Kanji and others 1992, Laing Chapter 6 1990, Management Sciences for Health 1992, Nazerali 1992, Office of Technology Assessment 1993, This chapter draws on technical materials from the Thomas, Lavy, and Strauss 1992, Tomson and Sterky World Health Organization and the World Bank and 1986, and WHO 1988a, 1988b, and 1988c. Box 6.5 from the scientific literature. Regional estimates of draws on World Bank material and on Barros and hospital beds, physicians, and nurses are from others 1986 and Fadndes and Cecatti 1993. Informa- OECD, WHO, World Bank, and national statistics. tion on the INCLEN program is from Halstead, Tug- The hospital and district health system discussion well, and Bennet 1991. The information and research draws on Barnum and Kutzin 1993 and World Bank to guide decisionmaking draws on Enthoven 1989. forthcoming. Estimates of global spending on health research are Valuable comments and materials were provided from the Commission on Health Research for Devel- by Orvil Adams, Uche Amazigo, Harvey Bale, Wil- opment 1990. The health research discussion also bert Bannenberg, Pascal Brudon-Jakobowicz, Robert draws on Free 1991, Godal 1993, and WHO 1991d. Cassen, Gilles Dussault, Anibal Faündes, Enrique Feffer, Michael Free, Julio Frenk, John Gil-Martin, Chapter 7 Wendy Graham, Richard Heller, Richard Laing, John Lloyd, Alfonso MejIa, Violaine Mitchell, Hiroshi Na- Information on health policy reform was provided by katani, Vic Neufeld, Joel Nobel, Joao Batista Oliveira, Jonathan Broomberg for South Africa, Louise Fox for Diego Palacio, David Porter, Michael Porter, Jim Romania, Salim Habayeb for India, Evangeline Javier Rankin, Una Reid, and staff of the Aga Khan Devel- for Chile, and Mary E. Ming Young for Poland. The opment Network and the Aga Khan Foundation. section on aid flows is based on the background pa- Box 6.1 was prepared by Tamara Fox and Ruth Le- per on aid by Michaud and Murray, which used a vine. Estimates of medical equipment expenditure wide range of data from Organization for Economic are from Rozynski and Gallivan 1992. Cooperation and Development, United Nations, and Estimates on the efficiency of outpatient surgery in bilateral sources. The discussion of international Colombia are from Shepard and others 1990. Hospital health research draws heavily on Commission on planning experience in the United States is based on Health Research for Development 1990. John Barton Davis and others 1990. The discussion of regional and Selcuk Ozgediz furnished material on the Con- planning is based in part on Jonsson 1989. The rela- sultative Group for International Agricultural Re- tionship between the volume of surgery and health search (CGIAR) and its relevance for health research. outcomes is based on Hughes, Hunt, and Luft 1987. Valuable comments on the draft chapter were re- Papua New Guinea's experience with containing ceived from John Evans. Box 7.1 is based on UNICEF! hospital spending is based on Newbrander 1987. The Bamako Initiative Management Unit 1990 and 1992 medical equipment discussion draws on Bloom 1989, and on the evaluation study by McPake, Hanson, and Bruley 1991, Garber and Fuchs 1991, Gelijns and Mills 1992. Box 7.2 draws on World Bank reports on Halm 1991, Halbwachs 1992, and WHO various Chile and on material provided by Thomas Bossert years. Abel-Smith 1992a and Rublee 1989 provided on evaluations of health-financing reforms and de- information on medical technology policies in indus- centralization. Box 7.3 uses material provided by Dov trial countries. Chernichovsky and George Schieber. Box 7.4 encap- The human resources for health discussion draws sulates the findings of Cassen and others 1986 and on Abel-Smith 1986, Enthoven and Vorhaus 1992, Riddell 1987 in particular. Guy Ellena and Joseph Evans 1981, Foster 1987, Frenk and others 1991, Kutzin assisted with Box 7.5. Box 7.6 draws heavily Fuchs 1978, Institute of Medicine 1988, Javitt forth- on World Bank 1993d. 181 Appendix A Deborah Symmons, B.-I. Thylefors, Ian Timeus, Carol Torel, James C. Tulloch, Ronald Waldman, Appendix A benefited from contributions from many Godfrey Walker, Jay Wenger, William Whang, Erica institutions; particularly valuable were the data re- Wheeler, Russell Wilkins, G. Yang, R. Yip, and ceived from WHO and UNICEF. Richard Bumgarner Anthony Zwi. and Godfrey Walker of WHO provided information Others who contributed include M. Adrian, Ann for many of the health indicators. Gareth Jones, Ashworth-Hill, P. Blake, Uwe Brinkman, C. Broome, UNICEF, contributed sources of data for nutritional Richard Bumgarner, Jacqueline Campbell, P. Car- indicators. Shea Rutstein from the Demographic and levaro, Mary Chamie, Lincoln C. Chen, D. P. J. Health Surveys project provided data on breastfeed- Daumerie, Hans Emblad, R. Etzel, Paul Fine, A. ing. Robert Hartford and Francis Notzon made avail- Galazka, Marito Garcia, S. Gillespie, Marcus Grant, able the database on perinatal and infant mortality R. J. Guidotti, Francoise Hamers, H. R. Hapsara, J. from the National Center of Health Statistics. Roy Harris, Peter Heller, Alan Hill, C. J. Hong, H. Jamai, Miller, USAID, commented on an earlier version of Fritz Kaferstein, Alex Kalache, N. Khaltaeve, Betty the appendix. My Vu of the World Bank's Population, Kirkwood, Arata Kochi, Jacob Kumaresan, N. Lee, Health, and Nutrition Department was responsible Anthony Mann, Ingrid Martin, G. Mayberly, Juan for preparing the statistical appendix for World Bank Menchaca, Michel Mercier, T. R. Mertens, Alvaro forthcoming, which was extensively used in the ap- Moncayo, Richard Morrow, Y. Motarjemi, Shaik Na- pendix, and processed raw data on several of the deen, William Newbrender, M. Noel, Godfrey selected health indicators. James Cercone helped pro- Oakley, D. Peterson, A. Pio, G. R. Quinke, C. P. cess data on mortality by broad causes of death. Ramachandran, M. Rosenberg, Norman Sartorius, Alan M. Schapira, Gordon Smith, Peter Smith, T. Appendix B Studwick, M. Thuriaux, Andrew Tomkins, Patrick Vaughan, S. Vidwans, F. Vinicor, and Diana Weil. The global burden of disease study was directed by The methodology used for this study drew in part Christopher Murray and Alan D. Lopez. The results on the established literature on quality-adjusted life reported here come from Murray and Lopez, back- years (see Torrance 1986). One of the first applica- ground paper; a much expanded discussion will ap- tions to developing countries was Ghana Health As- pear in Murray and Lopez forthcoming (a). Contribu- sessment Project Team 1981. tions from the United Kingdom were coordinated by Jonathan Broomberg. Substantial contributions and comments on spe- Background papers cific diseases and injuries came from the following individuals: Carla AbouZahr, Mike Adams, Paul Ar- Bobadilla, José-Luis, Peter Cowley, Helen Saxenian, and thur, Robert Ashley, Kenneth Bailey, David Barmes, Philip Musgrove. "The Essential Package of Health Ser- L. Barnes, Robert Beaglehole, Mark Belsey, Stephen vices in Developing Countries." Berman, Barry Bloom, M. Blossner, Loretta Brabin, Cochrane, Susan, and Thomas W. Merrick. "Improving Ma- Donald Bundy, A. Burton, P. D. Cattand, Jacqueline ternal and Child Health through Family Planning Cattani, Chen Chunming, Caroline J. Cook, Edward Services." Cooper, P. M. P. Desjeux, Jacques Ferlay, 1. Fomey, Hecht, Robert M., and Vito L. Tanzi. "The Role of NGOs in the Delivery of Health Services in Developing Jean-Claude Funck, Michel Garenne, Tore Godal, Countries." Anne Goerdt, Johnathan Gorstein, Sandra Gove, Hill, Kenneth, and Abdo Yazbeck. "Trends in Child Mortal- Ramesh Govindaraj, Walter Gulbinat, Ivan Gyarfas, ity, 1960-90: Estimates for 84 Developing Countries." Lori Heise, Larry Heligman, Joachim Hempel, Em- Hill, Kenneth, Dean T. Jamison, Lawrence J. Lau, Jee-Peng manuel Jimenez, Mark Kane, Patrick Kenya, Dean Tan, and Abdo Yazbeck. "The Impact of Health Status on Kilpatrick, Hilary King, Jeffrey Koplan, Marie-Helene Economic Growth." Lau, Lawrence, Abdo Yazbeck, Kenneth Hill, Dean T. Jam- Leclerq, Linda Lloyd, Julian Lobb-Levyt, Luis Lopez ison, and Jee-Peng Tan. "Sources of Child Health Gains Bravo, David Mabey, Prasanta Mahapatra, Paul since the 1960s: An International Comparison." McKeigue, Graham Medley, Edwin Michael, Cath- Michaud, Catherine, and Christopher Murray. "Aid Flows erine Michaud, Kenneth Mott, A.-D. Negrel, Mag- to the Health Sector in Developing Countries." daline Orzeszyna, Max Parkin, Richard Peto, P. Pi- Murray, Christopher, and Alan D. Lopez. "The Global Bur- sani, Jenny Pronczuk, E. Pupulin, Xinjian Qiao, Ravi den of Disease in 1990." Murray, Christopher, Ramesh Govindaraj, and G. Rannan-Eliya, Hans Remme, Jean-Marie Robine, Chellaraj. "Global Domestic Expenditures in Health." Claude J. Romer, Richard Rothenberg, Peter Sand- Murray, Christopher, Jay Kreuser, and William Whang. iford, Elizabeth Sherwin, Alan Silman, Buranaj "Cost-Effectiveness Model for Allocating Health Sector Smutharaks, Jan Stjernsward, Rand Stoneburner, Resources." 182 Pritchett, Lant, and Lawrence H. Summers. "Wealthier Is Amadeo, Jesus, Dov Chernichovsky, and Gabriel Ojeda. Healthier." 1991. "The Profamilia Family Planning Program, Colom- Yazbeck, Abdo, Jee-Peng Tan, and Vito L. Tanzi. "Public bia." Policy, Research, and External Affairs Working Pa- Spending on Health in the 1980s: The Impact of Adjust- per Series 759. World Bank, Population and Human Re- ment Lending Programs." sources Department, Washington, D.C. Anand, Sudhir, and Martin Ravallion. 1993. "Human De- Selected bibliography velopment in Poor Countries: On the Role of Private In- comes and Public Services." Journal of Economic Perspec- tives 7(1): 133-50. Aaron, Henry J. 1991. Serious and Unstable Condition: Financ- ing America's Health Care. Washington, D.C.: Brookings Anderson, H. R. 1979. "Chronic Lung Disease in the Papua Institution. New Guinea Highlands." Thorax 34: 647-53. Abel-Smith, Brian. 1986. "The World Economic Crisis, Part Andersson, Fredrik. 1992. "Norway's Pivotal Role in Regu- 2: Health Manpower out of Balance." Health Policy and latory Health Economics and Pharmaceutical Cost Con- Planning 1(4): 309-16. tainment-What Can Other Countries Learn?" Health Pol- icy 21: 17-34. 1992a. "Cost Containment and New Priorities in the European Community." Milbank Memorial Fund Quar- Andreoni, Diego. 1986. "The Cost of Occupational Acci- terly 70(3): 393-416. dents and Diseases." Occupational Safety and Health Se- ries 54. International Labour Office, Geneva. 1992b. Cost Containment and New Priorities in Health Care. Aldershot, U.K.: Ashgate Publishing. Archavanitkui, K., and A. Pramaualrantan. 1990. "Factors Affecting Women's Health in Thailand." Paper presented 1992c. "Health Insurance in Developing Coun- tries: Lessons from Experience." Health Policy and Plan- at the Workshop on Women's Health in Southeast Asia, ning 7(3): 215-26. Population Council, Jakarta, October 29-31. Arrow, Kenneth. 1963. "Uncertainty and the Welfare Eco- Abel-Smith, Brian, and Ajay Dua. 1988. "Community Fi- nomics of Medical Care." American Economic Review 53: nancing in Developing Countries: The Potential for the 941-73. Health Sector." Health Policy and Planning 3(2): 95-108. Ballance, Robert, Janos Pogany, and Helmut Forstner. 1992. AbouZahr, Carla, and Erica Royston, comps. 1991. Maternal The World's Pharmaceutical Industries: An International Per- Mortality: A Global Fact book. Geneva: World Health Organization. Published for spective on Innovation, Competition, and Policy. the United Nations Industrial Development Organization. ACC!SCN (Administrative Committee on Coordination! Aldershot, U.K.: Edward Elgar Publishing. Subcommittee on Nutrition). 1991. Controlling Iron Defi- ciency. Nutrition Policy Discussion Paper 9. Geneva: Barcellos, Tanya, Carmen Barros, Naia Oliveira, Vera Gar- World Health Organization. cia, Laureane Pastore, and Mercedes Rabelo. 1986. "Seg- 1992a. Nutrition and Population Links: Breastfeeding, regacao Urbana e Mortalidade em Porto Alegre." Fun- Family Planning and Child Health. Nutrition Policy Discus- dacao de Economia e Estatistica, Porto Alegre, Brazil. sion Paper 11. Geneva: World Health Organization. Barlow, Robin, and Lisa M. Grobar. 1985. "Costs and Bene- fits of Controlling Parasitic Diseases." World Bank, Popu- 1992b. Second Report on the World Nutrition Situa- tion. Vol. 1: Global and Regional Results. Geneva: World lation, Health, and Nutrition Department, Washington, D.C. Health Organization. Achutti, Aloysio, B. B. Duncan; Luis Ruiz, and M. I. Barnum, Howard. 1987. "Evaluating Health Days of Life Gained from Health Projects." Social Science and Medicine Schmidt. 1988. "Programas de Saüde do Adulto." Paper 24(10): 833-41. prepared for the World Bank Study on Adult Health in Brazil. World Bank, Washington, D.C. Barnum, Howard, and Joseph Kutzin. 1993. Public Hospitals in Developing Countries: Resource Use, Cost, Financing. Balti- Adair, L., Barry M. Popkin, J. VanDerslice, John S. Akin, more, Md.: Johns Hopkins University Press. David K. Guilkey, R. Black, John Briscoe, and W Flieger. 1993. "Growth Dynamics during the First Two Years of Barr, Nicholas. 1992. "Economic Theory and the Welfare State: A Survey and Interpretation." Journal of Economic Life: A Prospective Study in the Philippines." European Literature 30 (June): 741-803. Journal of Clinical Nutrition 47: 42-51. Barros, F. C., and others. 1986. "Why So Many Cesarean Ainsworth, Martha, and Mead Over. 1992. "The Economic Sections: The Need for Further Policy Change in Brazil." Impact of AIDS: Shocks, Responses, and Outcomes." Health Policy and Planning 1(19). Technical Working Paper 1. World Bank, Africa Technical Barss, P., G. S. Smith, D. Mohan, and S. P. Baker. Forth- Department, Washington, D.C. coming. Injuries in Adults in Developing Countries: Epide- Akin, John S., David K. Guilkey, Charles C. Griffin, and miology and Policy. New York: Oxford University Press. Barry M. Popkin. 1985. The Demand for Primary Health Ser- Baumol, William J. 1992. "Private Affluence, Public Squa- vices in the Third World. Totowa, N.J.: Roman and br." C. V. Starr Center for Applied Economics, New York. Allanheld. Beaton, George H., and Hossein Ghassemi. 1987. "Supple- Alderman, Harold, Jere Behrman, Shahrukh Khan, David mentary Feeding Programs for Young Children in Devel- Ross, and Richard Sabot. 1992. "Public Schooling Expen- oping Countries: A Summary of Lessons Learned." In J. ditures in Rural Pakistan: Efficiently Targeting Girls and a Price Gittinger, Joanne Leslie, and Caroline Hoisington, Lagging Region." Paper presented at a World Bank con- eds., Food Policy: Integrating Supply, Distribution, and Con- ference on Public Expenditure on the Poor, July 17-19, sumption. Baltimore, Md.: Johns Hopkins University Washington, D.C. Press. Alonso, P. L., S. W. Lindsay, J. R. M. Armstrong, M. Con- teh, P. H. David, G. Fegan, A. de Francisco, A. J. Hall, F. C. Shenton, K. Cham, and B. M. Greenwood. 1991. "The Effect of Insecticide Treated Bed-Nets on Mortality of Gambian Children." Lancet 337: 1499-1515. 183 Beaton, George H., R. Martorell, K. A. L'Abbe, B. Ed- Birdsall, Nancy, and William McGreevey. 1983. "Women, monston, G. McCabe, A. C. Ross, and B. Harvey. 1993. Poverty and Development." In Mayra Buvinic, Margaret 'Effectiveness of Vitamin A Supplementation in the Con- Lycette, and William Paul McGreevey, eds., Women and trol of Young Child Morbidity and Mortality in Develop- Poverty in the Third World. Baltimore, Md.: Johns Hopkins ing Countries." A project of the International Nutrition University Press. Program, Department of Nutritional Sciences, University Bitran, Ricardo A., and D. Keith Mclnnes. 1993. The Demand of Toronto. for Health Care in Latin America: Lessons from the Dominican Behrman, Jere R. 1990. The Action of Human Resources and Republic and El Salvador. Economic Development Institute Poverty on One Another: What We Have Yet to Learn. Living Seminar Paper 46. Washington, D.C.: World Bank. Standards Measurement Study Working Paper 74. Wash- Black, Douglas, J. N. Morris, Cyril Smith, Peter Townsend, ington, D.C.: World Bank. and Nick Davidson, eds. 1982. The Black Report. New 1992. "The Effect of Structural Adjustment on York: Penguin Books. Food Policy and Nutrition." Paper presented at the Inter- Black, Robert E. 1991. "Would Control of Childhood Infec- national Union for the Scientific Study of Population tious Diseases Reduce Malnutrition?" Acta Paediatrica (IUSSP) seminar on the Demographic Consequences of Scandinavica (Supplement) 374: 133-40. Structural Adjustment in Latin America, Belo Horizonte, Blendon, Robert J., R. Leitman, 1. Morrison, and K. Don- Brazil, September 29.-October 2. elan. 1990. "Satisfaction with Health Systems in 10 Na- Behrman, jere R., David Ross, Richard Sabot, and Matthew tions." Health Affairs 9(2): 185-92. Tropp. 1991. "Improving the Quality versus Increasing Bloom, Barry R., and Christopher J. L. Murray. 1992. "Tu- the Quantity of Schooling." University of Pennsylvania, berculosis: Commentary on a Reemergent Killer." Science Philadelphia, Pa. 257(21 August): 1055-64. Bellinger, D., A. Leviton, C. Waternaux, H. Needleman, Bloom, David E., and Joyce V. Lyons, eds. 1993. Economic and M. Rabinowitz. 1987. "Longitudinal Analyses of Pre- Implications of AIDS in Asia. UNDP Regional Progamme natal and Postnatal Lead Exposure and Early Cognitive Division (Regional Bureau for Asia and the Pacific). New Development." New England Journal of Medicine 316: Delhi: HIV/AIDS Regional Project. 1037-43. Bloom, Gerald. 1989. "The Right Equipment. . . in Working Benefo, Kofi, and T. Paul Schultz. 1992. "Fertility and Child Order." World Health Forum 10(1): 3-10. Mortality in Côte d'Ivoire and Ghana." World Bank, Afri- Blum, D., R. N. Emeh, S. R. A. Huttly, 0. Dosunmu- can Technical Department, Population, Health, and Nu- Ogunbi, N. Okeke, M. Ajala, J. I. Okoro, C. Akujobi, B. trition Division, Washington, D.C. R. Kirkwood, and R. G. Feachem. 1990. "The lmo State Bennett, Sara. 1992. "Promoting the Private Sector: A Re- Nigeria Drinking Water Supply and Sanitation Project. 1. view of Developing Country Trends." Health Policy and Description of the Project, Evaluation Methods, and Im- Planning 7(2): 97-110. pact on Intervening Variables." Transactions of the Royal Berg, Alan. 1987. Malnutrition: What Can Be Done? Lessons Society of Tropical Medicine and Hygiene 84: 309-15. from World Bank Experience. Baltimore, Md.: Johns Hopkins Bobadilla, José-Luis, Julio Frenk, Tomas Freijka, Rafael University Press. Lozano, and Claudio Stern. Forthcoming. "The Epidem- Berg, Elliot, and Graeme Hunter. 1992. "Social Costs of Ad- iological Transition and Health Priorities." In Dean T justment: The Case of Latin America and the Caribbean." Jamison, W. Henry Mosley, Anthony R. Measham, and Prepared for the U.S. Agency for International Develop- José-Luis Bobadilla, eds., Disease Control Priorities in Devel- ment. Development Alternatives, Inc., Bethesda, Md. oping Countries. New York: Oxford University Press. Berkley, Seth, and Dean jamison. 1991. "Summary Re- Bobak, M. 1993. "Air Pollution and Health in Central and port." Conference on the Health of School-Age Children, Eastern Europe: How Bad Is the Problem?" London sponsored by the United Nations Development Pro- School of Hygiene and Tropical Medicine. gramme and the Rockefeller Foundation, Bellagio, Italy, Bobak, M., and R. G. A. Feachem. 1992. "Health Status in August 12-16. the Czech and Slovak Federal Republic." Health Policy and Bertrand, Jane T., and Judith Brown. 1992. "Family Plan- Planning 7: 234-42. ning Success in Two Cities in Zaire." Policy, Research, Bobak, M., and D. A. Leon. 1992. "Air Pollution and Infant and External Affairs Working Paper Series 1042. World Mortality in the Czech Republic, 1986-88." Lancet 340: Bank, Population and Human Resources Department, 1010-14. Washington, D.C. Boissiere, M., J. B. Knight, and R. H. Sabot. 1985. "Earn- Bhargava, Alok. 1992. "Malnutrition and the Role of Indi- ing, Schooling, Ability and Cognitive Skills." American vidual Variation with Evidence from India and the Philip- Economic Review 75: 1026-30. pines." Journal of the Royal Statistical Society 155. Boring, Catherine C., Teresa S. Squires, and Tony Tong. Bhargava, Alok, and hang Yu. 1992. "A Longitudinal Analy- 1993. "Cancer Statistics 1993." CA-Cancer Journal for Cli- sis of Infant and Child Mortality Rate in Africa and Non- nicians 43 (Jan-Feb.): 7-26. African Developing Countries." Background paper pre- Bos, Eduard, My T. Vu, Ann Levin, and Rodolfo A. Bulatao. pared for the Africa Health Study. World Bank, Africa 1992. World Population Projections 1992-93 Edition: Estimates Technical Department, Washington, D.C. and Projections with Related Demographic Statistics. Balti- Birdsall, Nancy. Forthcoming. "Pragmatism, Robin Hood, more, Md.: Johns Hopkins University Press. and Other Themes: Good Government and Social Well- Bouis, Howarth E. 1990. "The Determinants of Household- Being in Developing Countries." In Lincoln C. Chen, Ar- Level Demand for Micronutrients: An Analysis for Philip- thur Kleinman, and Norma C. P. Ware, eds., Health and pine Farm Households." International Food Policy Re- Social Science in International Perspective. Harvard Series on search Institute, Washington, D.C. Population and International Health. Cambridge, Mass.: Harvard University Press. 184 Bourdon, Jean, François Orivel, and Jean Perrot. 1993. Cleland, John. 1990. "Maternal Education and Child Sur- "Peut-on Apprecier un Lien entre les Acquisitions vival: Further Evidence and Explanations." In John Cald- Scolaires et Ia Sante: Le Cas de l'Enseignement Primaire well, Sally Findley, Pat Caldwell, Gigi Santow, Wendy au Tchad." Paper presented at meetings of the Associa- Cosford, Jennifer Braid, and Daphne Broers-Freeman, tion Francais de Sciences Economiques, Dijon. May eds., What We Know about Health Transition: The Cultural, 27-28. Social, and Behavioural Determinants of Health. Health Tran- Bradley, Christine. 1988. "How Can We Help Rural Beaten sition Series, Vol. 1. Canberra: Australian National Wives? Some Suggestions from Papua New Guinea." Pa- University. per presented at International Women's Aid Conference, Coale, Ansley J., and Paul Demeny, with Barbara Vaughan. Cardiff, Wales. 1983. Regional Model Life Tables and Stable Populations. 2d Bradley, David John, Sandy Cairncross, Trudy Harpham, ed. New York: Academic Press. and Carolyn Stephens. 1992. A Reviezt' of Environmental Comanor, William S. 1986. "The Political Economy of the Health Impacts in Developing Country Cities. Urban Manage- Pharmaceutical Industry." Journal of Economic Literature 24 ment Program Paper 6. Washington, D.C.: World Bank. (September): 1178-1217. Brenzel, Logan. 1990. 'The Costs of EPI: Lessons Learned Commission on Health Research for Development. 1990. from Cost and Cost-Effectiveness Studies of Immuniza- Health Research: Essential Link to Equity in Development. tion Programs." Resources for Child Health (REACH) New York: Oxford University Press. Project. John Snow, Inc., Arlington, Va. Conly, G. N. 1975. The Impact of Malaria on Econo,nic Develop- Briscoe, John. 1992. "Poverty and Water Supply: How to ment: A Case Study. Scientific Publication 297. Washing- Move Forward." Finance and Development 29: 16-19. ton, D.C.: Pan American Health Organization. Brook, Robert H., and Kathleen N. Lohr. 1986. "Will We Council on Scientific Affairs, American Medical Association. Need to Ration Effective Health Care?" Issues in Science 1992. "Violence against Women: Relevance for Medical and Technology 3(1): 68-77. Practitioners." JAMA: Journal of the American Medical Asso- Brotowasisto, Oscar Gish, Ridwan Malik, and Paramita ciation 267(23): 3184-89. Sudharto. 1988. "Health Care Financing in Indonesia." COVAC. 1990. "Evaluación de Proyecto para Educación, Health Policy and Planning 3(2): 131-40. Capacitación, y Atención a Mujeres y Menores de Edad en Bruce, Judith, and Cynthia B. Lloyd. 1992. "Finding the Ties Materia de Violencia Sexual, Enero a Diciembre 1990." That Bind: Beyond Headship and Household." Working Asociación Mexicana Contra Ia Violencia a las Mujeres, Paper 41. Population Council, New York. Mexico City. Bruley, Mark E. 1991. "Global Trends in Healthcare Tech- Cumper, George, and Patrick Vaughan. 1985. "Community nology." Keynote address at Health-Tech 91: Health Tech- Health Aides at the Crossroads," World Health Forum 6: nology in Developing Countries: Assessment, Procure- 365-67. ment, and Maintenance, Yaoundé, Cameroon, March 11. Curtis, C. F. 1992. "Spraying Bed-Nets with Deltamethrin in ECRI, Plymouth Meeting, Pa. Sichuan, China." Tropical Diseases Bulletin 89(8): R1-R6. Cairncross, A. M. 1990. "Health Impacts in Developing Davis, Karen, Gerard F. Anderson, Diane Rowland, and Countries: New Evidence and New Prospects." Journal of Earl P. Steinberg. 1990. Health Care Cost Containment. the Institution of Water and Environmental Management 4(6): Johns Hopkins Studies in Health Care Finance and Ad- 571-77. ministration. Baltimore, Md.: Johns Hopkins University Cairncross, Sandy. 1992. "Sanitation and Water Supply: Press. Practical Lessons from the Decade." UNDP-World Bank De Geyndt, Willy. 1991. Managing Health Expenditures under Water and Sanitation Discussion Paper Series 9. World National Health Insurance: The Case of Korea. World Bank Bank, Washington, D.C. Technical Paper 156. Washington, D.C. Caldwell, J. C. 1986. "Routes to Low Mortality in Poor Diamond, Peter. 1992. "Organizing the Health Insurance Countries." Population and Development Review 12(2): Market." Econometrica 60(6): 1233-54. 171-220. Doll, Richard. 1992. "Health and the Environment in the CAMVAC (Centro de Apoyo a Mujeres Violadas). 1985. 1990s." American Journal of Public Health 82: 933-43. "Carpeta de Información Básica para Ia AtenciOn Soli- Downing, A. J. 1991. "Pedestrian Safety in Developing dana y Feminista a Mujeres Violadas." Mexico City. Countries." In Proceedings of the Vulnerable Road User: Inter- Caplan, David. 1985. "Consumption, Production, World national Conference on Traffic Safety, New Delhi, January Trade and Industry Structure." World Bank, Industry and 27-30. New Delhi: Macmillan India. Energy Department, Washington, D.C. Drèze, Jean, and Amartya Sen. 1989. Hunger and Public Ac- Cassen, Robert H., and others. 1986. Does Aid Work? New tion. WIDER Studies in Development Economics. Oxford, York: Oxford University Press. U.K.: Clarendon Press. Castro, E. B., and K. M. Mokate. 1988. "Malaria and Its Dunn, D. 1., M. L. Newell, A. E. Ades, and C. S. Peckham. Socioeconomic Meaning: The Study of Cunday in Co- 1992. "Risk of Human Immunodeficiency Transmission lombia." In A. N. Herrin and P. L. Rosenfield, eds., Eco- through Breastfeeding." Lancet 340: 585-88. nomics, Health and Tropical Diseases. Manila: School of Ea.o- Edwards, Sebastian. Forthcoming. "Latin America a De- nomics, University of the Philippines. cade after the Debt Crisis." World Bank, Latin America Chapman, R. S., J. L. Mumford, X. He, D. B. Harris, R. and the Caribbean Regional Office, Office of the Chief Yang, and W. Jiang. 1989. "Assessing Indoor Air Pollution Economist, Washington, D.C. Exposure and Lung Cancer Risk in Xuan Wei, China." El Batawi, M. A., and C. Husbumrer. 1987. "Epidemiologi- Journal of the American College of Toxicology 8(5): 941-48. cal Approach to Planning and Development of Occupa- Chen, B. H., C. J. Hong, M. R. Pandey, and K. R. Smith. tional I-Iealth Services at a National Level." International 1990. "Indoor Air Pollution in Developing Countries.' Journal of Epidemiology 16: 288-92. World Health Statistics Quarterly 43: 127-38. 185 Elliott, Paul. 1988. "The Intersalt Study: An Addition to the Feachem, Richard G. A., Tord Kjellstrom, Christopher J. L. Evidence on Salt and Blood Pressure, and Some Implica- Murray, Mead Over, and Margaret A. Phillips, eds. 1992. tions." Journal of Human Hypertension 3: 289-98. The Health of Adults in the Developing World. Washington, Elo, T. Irma. 1992. "Utilization of Maternal Health-Care Ser- D.C.: World Bank. vices in Peru: The Role of Women's Education." Health Fenner, Frank, Donald A. Henderson, Isao Arita, Zdenek Transition Review 2(1): 49-69. Jezek, and Ivan D. Ladnyi. 1988. Smallpox and Its Eradica- Engle, Patrice L. 1991. "Mother's Work and Child Care in tion. Geneva: World Health Organization. Pen-Urban Guatemala." Child Development 62. Foster, A. 1987. "Cataract Blindness in Africa." Ophthalmic Enthoven, Alain C. 1988. Theory and Practice of Managed Coin- Surgery 18: 384-88. petition in Health Care Finance. Amsterdam: North-Hol- Foster, Stanley 0., Deborah McFarland, and Meredith John. land. Forthcoming. "Measles." In Dean T. Jamison, W. Henry 1989. "What Can Europeans Learn from Ameri- Mosley, Anthony R. Measham, and José-Luis Bobadilla, cans?" Health Care Financing Review, annual supplement eds., Disease Control Priorities in Developing Countries. New (December) 49-63. York: Oxford University Press. Enthoven, Alain, and Carol Vorhaus. 1992. "Teaching Note: Foster, Susan D. 1990. "Improving the Supply and Use of The Physician Personnel Industry." Stanford University Essential Drugs in Sub-Saharan Africa." Policy, Research, Graduate School of Business, Palo Alto, Calif. and External Affairs Working Paper Series 456. World Esrey, S. A., J. B. Potash, L. Roberts, and C. Shiff. 1991. Bank, Population and Human Resources Department, "Effects of Improved Water Supply and Sanitation on As- Washington, D.C. cariasis, Diarrhoea, Dracunculiasis, Hookworm Infection, 1991. "Supply and Use of Essential Drugs in Sub- Schistosomiasis, and Trachoma." Bulletin of the World Saharan Africa: Some Issues and Possible Solutions." So- Health Organization 69(5): 609-21. cial Science Medicine 32(11): 1201-18. Evans, John R. 1981. Measurement and Management in Medi- Free, Michael. 1991. "Health Technologies for the Develop- cine and Health Services: Training Needs and Opportunities. ing World: Addressing the Unmet Needs." Program for New York: Rockefeller Foundation. Appropriate Technology in Health, Seattle, Washington. Evans, John R., Karen Lashman Hall, and Jeremy Warford. Frenk, Julio, J.-L. Bobadilla, J. Sepülveda, and M. Lopez 1981. "Health Care in the Developing World: Problems of Cervantes. 1989. "Health Transition in Middle-Income Scarcity and Choice." Shattuck Lecture. New England Countries: New Challenges for Health Care." Health Pol- Journal of Medicine 305 (November): 1117-27. icy and Planning 4:29-39. Evans, Robert G., Morris L. Barer, and Roberta J. Labelle. Frenk, Julio, Javier Alagon, Gustavo Nigenda, Alejandro 1988. "Fee Controls as Cost Control: Tales from the Fro- Muñoz-del Rio, Cecilia Robledo, Luis A. Vaquez-Segovia, zen North." Milbank Memorial Fund Quarterly 66(1): 1-64. and Catalin Ramirez-Cuadra. 1991. "Patterns of Medical Ewbank, Douglas C., and Samuel H. Preston. 1990. "Per- Employment: A Survey of Imbalances in Urban Mexico." sonal Health Behaviour and the Decline in Infant and American Journal of Public Health 81 (January): 123-29. Child Mortality: The United States, 1900-1930." In John Fuchs, V R. 1978. "The Supply of Surgeons and the De- Caldwell, Sally Findley, Pat Caldwell, Gigi Santow, mand for Operations." Journal of Human Resources 13 Wendy Cosford, Jennifer Braid, and Daphne Broers-Free- (supp.): 35-56. man, eds., What We Know about Health Transition: The Cul- 1979. "Economics, Health and Post-Industrial So- tural, Social, and Behavioural Determinants of Health. Health ciety." Milbank Memorial Fund Quarterly 57(2): 153-82. Transition Series, Vol. 1. Canberra: Australian National Garber, Alan M., and Victor R. Fuchs. 1991. "The Expand- University. ing Role of Technology Assessment in Health Policy." Faiz, Asif, Christopher Weaver, Kumares Sinha, Michael Stanford Law and Policy Review (Fall): 203-09. Walsh, and Jose Carbajo. Forthcoming. Air Pollution from Gelijns, Annetine C., and Ethan A. HaIm, eds. 1991. Medical Motor Vehicles: Issues and Options for Developing Countries. Innovation at the Crossroads. Vol. 2: The Changing Economics Washington, D.C.: World Bank and UNEP. of Medical Technology. Washington, D.C.: National Acad- Faündes, Anibal, and José Guilherme Cecatti. 1993. "Which emy Press. Policy for Caesarian Sections in Brazil? An Analysis of Gerdtham, Ulf-G., Jes SGgaard, Fredrik Andersson, and Trends and Consequences." Health Policy and Planning Bengt Jonsson. 1992. "An Econometric Analysis of Health 8(1): 33-42. Care Expenditure: A Cross-Section Study of the OECD Fauveau, V., and T. Blanchet. 1989. "Epidemiology and Countries." Journal of Health Economics 11: 63-84. Cause of Deaths among Women in Rural Bangladesh." Gertler, Paul, and Jacques van der Gaag. 1990. The Willing- International Journal of Epidemiology 18(1): 139-45. ness to Pay for Medical Care: Evidence from Two Developing Feachem, R. C. 1988. "Epidemiology and Tropical Public Countries. Baltimore, Md.: Johns Hopkins University Health: Current and Future Contributions with Particular Press. Emphasis on the Role of the London School of Hygiene Ghana Health Assessment Project Team. 1981. "Quantita- and Tropical Medicine." Transactions of the Royal Society of tive Methods of Assessing Health Impact of Different Dis- Tropical Medicine and Hygiene 82: 790-98. eases in Less Developed Countries." International Journal Feachem, Richard G., and Dean T. Jamison. 1991. Disease of Epidemiology 10(1): 73-80. and Mortality in Sub-Saha ran Africa. New York: Oxford Uni- Glewwe, Paul. 1991. Are Rates of Return to Schooling Estimated versity Press. from Wage Data Relevant Guides for Government Investments Feachem, R. C., E. Burns, S. Cairncross, A. Cronin, P. in Education? Evidence from a Developing Coumitry Living Cross, D. Curtis, M. Khalid Kahn, D. Lamb, and H. Standards Measurement Study Working Paper 76. Wash- Southall. 1978. Water, Health and Development: An Inter- ington D.C.: World Bank. disciplinary Evaluation. London: Tn-Med Books. 186 Godal, Tore. 1993. "Introduction." Tropical Disease Research, Hobcraft, J. N., ed., 1991. Proceedings of the Demographic and Progress 1991-92. Eleventh Programme Report of the Health Surveys World Conference, Washington, D.C., 1991. UNDP/World Bank/WHO Special Programme for Re- Vol. 2: 1157-82. search and Training in Tropical Diseases. Geneva: World 1993. "Women's Education, Child Welfare and Health Organization. Child Survival: A Review of the Evidence." Health Transi- Gomes-Neto, J. B., and E. A. Hanushek. 1991. "The Causes tion Review 3(2). and Effects of Grade Repetition: Evidence from Brazil." Hoddinott, John, and Lawrence Haddad. 1991. "Household Working Paper 295. Rochester Center for Economic Re- Expenditures, Child Anthropometric Status and the Intra- search, Rochester, N.Y. household Division of Income: Evidence from the Côte Goodgame, R. W. 1990. "AIDS in Uganda: Clinical and So- d'Ivoire." Oxford University, Unit for the Study of Afri- cial Features." New England Journal of Medicine 323: can Economics, Oxford, U.K. 383-89. Holly, John, and David Lee. 1992. "Costa Rica: Proyecto Griffin, Charles C. 1992. Health Care in Asia: A Comparative Reforma Salud-Estudio de Financiamiento, Adquisición Study of Cost and Financing. World Bank Regional and Sec- y Distribución de Farmacéuticos y Materiales Medicos." toral Studies. Washington, D.C. Management Sciences for Health, Boston, Mass. Grosh, Margaret E., ed. 1992. "From Platitudes to Practice: Hosken, Fran. 1988. "International Seminar: Female Cir- Targeting Social Programs in Latin America." Vol. 1, Syn- cumcision-Strategies to Bring About Change." Women's thesis. LAC Regional Studies Program Report 21. World International Network News 14(3): 24-37. Bank, Washington, D.C. Hsiao, William. 1992. "Health Care Finance and Health In- Grossman, Michael. 1975. "The Correlation between Health surance Mechanisms in Developing Countries." Paper and Schooling." In Nestor E. Terleckyj, ed., Household presented at WHO Workshop on Health Economics, Ge- Production and Consumption. New York: Columbia Univer- neva, June 25. sity Press. Hughes, Robert G., Sandra S. Hunt, and Harold S. Luft. Gutierrez-Fisac, Juan L., Enrique Regidor, and Elena Ronda. 1987. "Effects of Surgeon Volume and Hospital Volume 1992. "Occupational Accidents and Alcohol Consumption on Quality of Care in Hospitals." Medical Care 25(6): in Spain." International Journal of Epidemiology 21(6): 489-503. 1114-19. Humphrey, J. H., K. P. West, Jr., and Alfred Sommer. 1992. Halbwachs, Hans. 1992. "Need for Spares: Excuse or Real- "Vitamin A Deficiency and Attributable Mortality among ity?" Deutsche Gesellschaft fur Technische Zusammenar- Under-5-Year-Olds." Bulletin of the World Health Organiza- belt (GTZ), Division of Health, Population, and Nutrition, tion 70(2): 225-32. Eschborn, Germany. Hurst, Jeremy. 1992. The Reform of Health Care: A Comparative Halstead, Scott, Peter Tugwell, and Kathryn Bennet. 1991. Analysis of Seven OECD Countries. Paris: Organization for "The International Clinical Epidemiology Network Economic Cooperation and Development. (INCLEN): A Progress Report." Journal of Clinical Epide- Huttly, S. R. A. 1990. "The Impact of Inadequate Sanitary miology 44: 579-89. Conditions on Health in Developing Countries." World Halstead, Scott B., Julia A. Walsh, and Kenneth S. Warren, Health Statistics Quarterly 43: 118-26. eds. 1985. Good Health at Low Cost. Proceedings of a con- 1DB (Inter-American Development Bank). 1988. "Summary ference at Bellagio, Italy, sponsored by the Rockefeller of the Ex-Post Evaluations of Public Health Programs." Foundation, April 29-May 3. 1DB, Operations Evaluation Office, Washington, D.C. Handwerker, W. Penn. 1991. "Gender Power Difference Ikegami, Naoki. 1992. "The Economics of Health Care in May Be STD Risk Factors for the Next Generation." Paper Japan." Science 258 (23 October): 614-18. presented at the 90th Annual Meeting of the American Ingram, Gregory. 1992. "Social Indicators and Productivity Anthropological Association, Chicago, Ill. Convergence in Developing Countries." Policy, Research, Hecht, Robert, Catherine Overholt, and R. Hopkins and External Affairs Working Paper Series 894. World Holmberg. 1993. "Improving the Implementation of Cost Bank, Development Economics Research Advisory Staff, Recovery for Health: Lessons from Zimbabwe." Health Washington, D.C. Policy 25(1), July. Institute of Medicine. 1988. The Future of Public Health. Heise, Lori. 1993. "Violence Against Women: The Missing Washington, D.C.: National Academy Press. Agenda." In Marge Koblinsky, Judith Timyan, and Jill 1992. Emerging Infections: Microbial Threats to Health Gay, eds., The Health of Women: A Global Perspective. Boul- in the United States. Washington, D.C.: National Academy der, Cob.: Westview. Press. Heller, Peter S., and Jack Diamond. 1990. International Com- International Monetary Fund. Various years. Government Fi- parisons of Government Expenditure Revisited: The Developing nance Statistics. Washington, D.C. Countries, 1975-86. Occasional Paper 69. Washington, Jacobs, William R., Jr., Raül G. Barletta, Rupa Udani, John D.C.: International Monetary Fund. Chan, Gary Kalkut, Gabriel Sosne, Tobias Kiese, Gary J. Hill, Kenneth, and Alberto Palloni. 1992. "Demographic Re- Sarkis, Graham F. Hatfull, and Barry R. Bloom. 1993. sponses to Economic Shocks: The Case of Latin Amer-. "Rapid Assessment of Drug Susceptibilities of Mycobac- ica." In Proceedings of the Peopling of the Americas Confer- terium tuberculosis by Means of Luciferase Reporter ence, Veracruz, Mexico. Liege: International Union for the Phages." Science 260 (May): 819-22. Scientific Study of Population. Jamison, Dean T Forthcoming. "Disease Control Priorities Hlady, W. Gary, John V. Bennett, Aziz R. Samadi, Jahanara in Developing Countries: An Oveview." In Dean T. Jam- Begum, Abdul Hafez, Aminul I. Tarafdar, and John R. ison, W. Henry Mosley, Anthony R. Measham, and José- Boring. 1992. "Neonatal Tetanus in Rural Bangladesh: Luis Bobadilla, eds., Disease Control Priorities in Developing Risk Factors and Toxoid Efficacy." American Journal of Pub- Countries. New York: Oxford University Press. lic Health 82(10): 1365-69. 187 Jamison, Dean 1., and Joanne Leslie. 1990. "Health and Koss, Mary, Paul Koss, and Joy Woodruff. 1991. "Delete- Nutrition Considerations in Education Planning: The Cost rious Effects of Criminal Victimization on Women's and Effectiveness of School-Based Interventions." Food Health and Medical Utilization." Archives of Internal Medi- and Nutrition Bulletin (International) 12(3): 204-14. cine 151: 342-47. Jamison, Dean T., and Peter Moock. 1984. "Farmer Educa- Kutzin, Joseph, and Howard Barnum. 1992. "Institutional tion and Farm Efficiency in Nepal: The Role of Schooling, Features of Health Insurance Programs and Their Effects Extension Services, and Cognitive Skills." World Develop- on Developing Country Health Systems." International ment 12: 67-86. Journal of Health Planning and Management 7: 51-72. Jamison, Dean T., W. Henry Mosley, Anthony R. Measham, Laing, R. 0. 1990. "Rational Drug Use: An Unsolved Prob- and Jose-Luis Bobadilla, eds. Forthcoming. Disease Control lem." Tropical Doctor 20 (July): 101-03. Priorities in Developing Countries. New York: Oxford Uni- Lancet.1992. "Environmental Pollution: It Kills Trees, But versity Press. Does It Kill People?" 340: 821-22. Jarousse, Jean-Pierre, and Alain Mingat. 1992. "Options for Leslie, Joanne. 1987. "Interactions of Malnutrition and Diar- Accelerated Development of Primary Education in the rhea: A Review of Research." In J. Price Gittinger, Joanne Sahel." World Bank, Sahelian Department, Population Leslie, and Caroline Hoisington, eds., Food Policy: Inte- and Human Resources Operations Division, Washington, grating Supply, Distribution, and Consumption. Baltimore, D.C. Md.: Johns Hopkins University Press. Javitt, Jonathan C. Forthcoming. "Cataract." In Dean 1. 1989a. "Women's Time: A Factor in the Use of Jamison, W. Henry Mosley, Anthony R. Measham, and Child Survival Technologies." Health Policy and Planning José-Luis Bobadilla, eds., Disease Control Priorities in Devel- 4(1): 1-16. oping Countries. New York: Oxford University Press. 1989b. "Women's Work and Child Nutrition in the Jeyaratnam, 1985. "Health Problems of Pesticide Usage in . Third World." In Joanne Leslie and Michael Paolisso, the Third World." British Journal of Industrial Medicine 42: eds., Women, Work, and Child Welfare in the Third World. 505-06. Boulder, Cob.: Westview. Jimenez, Emmanuel. 1987. Pricing Policy in the Social Sectors. Leslie, Joanne, Dean T. Jamison, and Philip Musgrove. Washington, D.C.: World Bank. Forthcoming. "Protein-Energy Balance in the Diet and Johnston, Ian. 1992. "Action to Reduce Road Casualties." Human Growth." World Bank, Washington, D.C. World Health Forum 13: 154-62. Levin, Henry M., Ernesto Pollitt, Rae Galloway, and Judith Jonsson, Bengt. 1989. "What Can Americans Learn from McGuire. Forthcoming. "Micronutrient Deficiency Disor- Europeans?" Health Care Financing Review, annual supple- ders." In Dean 1. Jamison, W. Henry Mosley, Anthony R. ment (December) 79-93. Measham, and José-Luis Bobadilla, eds., Disease Control Kakwani, Nanak, Elene Makonnen, and Jacques van der Priorities in Developing Countries. New York: Oxford Uni- Gaag. 1990. "Structural Adjustment and Living Condi- versity Press. tions in Developing Countries." Policy, Research, and Ex- Lewis, Maureen A., and Clover Parker. 1991. "Policy and ternal Affairs Working Paper Series 467. World Bank, Implementation of User Fees in Jamaican Public Hospi- Population and Human Resources Department, Washing- tals." Health Policy 18: 57-85. ton, D.C. Lindenbaum, S., M. Chakraborty, and M. Elias. 1985. The Kanji, Najmi, Anita Hardon, Jan Willem Harnmeijer, Mas- Influence of Maternal Education on Infant and Child Mortality uma Mamdani, and Gill Walt. 1992. Drags Policy in Devel- in Bangladesh.Special Publication 23. Dhaka: International oping Countries. London: Zed Books Ltd., in association Center for Diarrhoeal Disease Research. with the Danish International Development Agency Litvack, Jennie, and Claude Bodart. 1993. "User Fees and (DANIDA). Improved Quality of Health Care Equals Improved Ac- Katabira, E., and R. W. Goodgame. 1989. "AIDS Care: Diag- cess: Results of a Field Experiment in Cameroon." Social nostic and Treatment Strategies for Health Workers." Science and Medicine 37(3). Uganda Ministry of Health, Health Education Printing Lockheed, Marlaine E., Dean T. Jamison, and Lawrence Unit, Kampala. Lau. 1980. "Farmer Education and Farm Efficiency: A Kennedy, Eileen. 1992. "Effects of Gender of Head of Survey." Economic Development and Cultural Change 29(1): Household on Women's and Children's Nutritional Sta- 37-76. tus." Paper presented at workshop on Effects of Policy Lockheed, Marlaine E., Adriaan M. Verspoor, and associ- and Programs on Women, January 16. International Food ates. 1991. Improving Primary Education in Developing Coun- Policy Research Institute, Washington, D.C. tries. New York: Oxford University Press. Keusch, Gerald T., and Nevin Scrimshaw. 1986. "Control of Louat, Frederic, Margaret E. Grosh, and Jacques van der Infection to Reduce Malnutrition." Reviews of Infectious Gaag. 1992. Welfare lmplications of Female Headship in Jamai- Diseases 8(2): 298-312. can Households. Living Standards Measurement Study King, Elizabeth M., and M. Anne Hill. 1993. "Overview." Working Paper 96. Washington, D.C.: World Bank. In Elizabeth M. King and M. Anne Hill, eds., Women's Luft, Harold S. 1978. Poverty and Health. Economic Causes and Education in Developing Countries: Barriers, Benefits, and Pol- Consequences of Health Problems. Cambridge, Mass.: icies. Baltimore, Md.: Johns Hopkins University Press. Ballinger. Koop, C. Everett. 1989. "Violence against Women: A Global Lutter, Chessa K., Jean-Pierre Habicht, Juan A. Rivera, and Problem." Address to the Pan American Health Organi- Reynaldo Martorell. 1992. "The Relationship between En- zation, Washington, D.C., May 22. ergy Intake and Diarrhoeal Disease in Their Effects on Korte, R., Heide Richter, F. Merkle, and H. Gorgen. 1992. Child Growth: Biological Model, Evidence, and Implica- "Financing Health Services in Sub-Saharan Africa: Op- tions for Public Health Policy." Food and Nut rition Bulletin tions for Decisionmakers during Adjustment." Social Sci- 14(1): 36-42. ence and Medicine 34(1): 1-9. 188 Mackay, Judith. 1993. The State of Health Atlas. New York: Mosley, W. Henry, and Lincoln Chen, eds. 1984. Child Sur- Simon and Schuster. vival: Strategies for Research. New York: Population Management Sciences for Health. 1992. International Drug Council. Price Indicator Guide 1991. Boston, Mass. MRC (Medical Research Council). 1989. "The Neuro- Mardones, Francisco, and Rafael Zamora. 1989. "Evalua- psychological Effects of Lead in Children: A Review of the ciones Socio-econOmicas del Programa Nacional de Ali- Research 1984-1988." London. mentación Complementaria: Que es lo que se ha Muller, 0., L. Barugahare, B. Schwartlander, E. By- hecho?" Revista Chilena de Nut ricidn 17(3): 169-74. aruhanga, P. Kataaha, D. Kyeyune, W. Heckmann, and Martorell, Reynaldo, Juan Rivera, Haley Kaplowitz, and Er- M. Ankrah. 1992. "HIV Prevalence, Attitudes and Behav- nesto Pollitt. 1992. "Long-Term Consequences of Growth iour in Clients of a Confidential HIV Testing and Counsel- Retardation during Early Childhood." Paper presented at ling Centre in Uganda." AIDS 6: 869-74. the Sixth International Congress of Auxology, Madrid. Murray, Christopher, and Alan Lopez, eds. Forthcoming Amsterdam: Elsevier Science Publishers. (a). "The Global Burden of Disease." Harvard Center for Max, Emmanuel, and Donald S. Shepard. 1989. "Produc- Population and Development Studies, Cambridge, Mass. tivity Loss Due to Deformity from Leprosy in India." In- Forthcoming (b). "Global Causes of Death Pattern ternational Journal of Leprosy 57(2). in 1990." Lancet. Maxwell, C. A., C. F Curtis, H. Haji, S. Kisumku, A. I. Murray, Christopher, Karel Styblo, and Annik Rouillon. Thalib, and S. A. Yahya. 1990. "Control of Bancroftian Forthcoming. "Tuberculosis." In Dean T. Jamison, W. Filariasis by Integrating Therapy with Vector Control Henry Mosley, Anthony R. Measham, and José-Luis Bo- Using Polystyrene Beads in Wet Pit Latrines." Transactions badilla, eds., Disease Control Priorities in Developing Coun- of the Royal Society of Tropical Medicine and Hygiene 84: tries. New York: Oxford University Press. 709-14. Musgrove, Philip. 1986. "Measurement of Equity in McGreevey, William. 1990. Social Security in Latin America: Health." World Health Statistics Quarterly 39: 325-35. Issues and Options for the World Bank. World Bank Discus- 1988. "Is the Eradication of Polio in the Americas sion Paper 110. Washington, D.C. Economically Justified?" Bulletin of the Pan American Sani- McGuire, Judith, and Barry M. Popkin. 1990. "Beating the tary Bureau 22(1). Zero-Sum Game: Women and Nutrition in the Third 1990. Fighting Malnutrition: An Evaluation of Bra- World: Part 2." Food and Nutrition Bulletin (International) zilian Food and Nutrition Programs. World Bank Discussion 12: 3-11. Paper 60. Washington, D.C. McKeown, Thomas. 1976. The Modern Rise of Population. 1993. "Relaciones entre Ia Salud y el Desarrollo." London: Edward Arnold. BoletIn de la Oficina Sanitaria Panamericana 114(2): 115-29. McPake, Barbara, Kara Hanson, and Anne Mills. 1992. "Ex- Natale, Joanne E., Jill G. Joseph, Randall Bergen, Ravilla D. perience to Date of Implementing the Bamako Initiative: Thulasiraj, and Laxmi Rahmathullah. 1992. "Prevalence A Review and Five Country Case Studies." London of Childhood Disability in a Southern Indian City: Inde- School of Hygiene and Tropical Medicine, Department of pendent Effect of Small Differences in Social Status." In- Public Health and Policy. ternational Journal of Epidemiology 21(2): 367-72. Meerman, Jacob. 1980. "An Analysis in Development Per- National Research Council. 1989. Diet and Health: Implica- spective of Bolivia's Largest Public Enterprises: COM- tions for Reducing Chronic Disease Risk. Washington, D.C.: IBOL, YPFB, and ENAF." World Bank, Development Eco- National Academy Press. nomics Department, Washington, D.C. Nazerali, Hanif. 1992. "ZEDAP: A Retrospective Mesa-Lago, Carmelo. 1991. Social Security and Prospects for 1987-1992." Essential Drugs Monitor (Newsletter of WHO Equity in Latin America. World Bank Discussion Paper 140. Action Programme on Essential Drugs and Vaccines) 13: Washington, D.C. 6-8. 1992. "Health Care for the Poor in Latin America Needleman, H. L., A. Schell, D. Bellinger, A. Leviton, and and the Caribbean." PAHO Scientific Publication 539. E. N. Allred. 1990. "The Long-Term Effects of Exposure to Pan American Health Organization, Washington, D.C. Low Doses of Lead in Childhood: An 11-Year Follow-up Mills, Anne, Patrick J. Vaughan, Duane L. Smith, and Iraj Report." New England Journal of Medicine 322(2): 83-88. Tabibzadeh, eds. 1990. Health System Decentralization: Con- Newbrander, William. 1987. "Papua New Guinea's Expen- cepts, Issues and Country Experience. Geneva: World Health diture on Hospitals: Policy and Practice since Indepen- Organization. dence." Health Policy and Planning 2(3): 227-35. Minhas, B. 5. 1991. Educational Deprivation and Its Role as a Nokes, C., S. M. Grantham-McGregor, A. W. Sawyer, E. S. Spoiler of Access to Better Life in India. New Delhi: Indian Cooper, B. A. Robinson, and D. A. P. Bundy. 1992. "Mod- Statistical Institute. erate to Heavy Infections of Trichuris trichiura Affect Cog- Monteiro, Carlos Augusto. 1988. Sadde e Nutrição das Cri- nitive Function in Jamaican School Children." Parasitology anças de São Paulo. São Paulo: Editora Hucitec. 104: 539-47. Moses, S., F. A. Plummer, E. N. Ngugi, N. J. Nagelkerke, Norboo, T., M. Yahya, N. C. Bruce, J. A. Heady, and K. P. A. 0. Anzala, and J. 0. Ndinya-Achola. 1991. "Contyol- Ball. 1991. "Domestic Pollution and Respiratory Illness in ling HIV in Africa: Effectiveness and Cost of an Interven- a Himalayan Village." International Journal of Environmen- tion in a High-Frequency STD Transmitter Core Group." tal Health 20: 749-57. AIDS 5: 407-11. Nur, E. T. M., and H. A. Mahran. 1988. "The Effect of Moses, S., F. Manji, J. E. Bradley, N. J. Nagelkerke, M. A. Health on Agricultural Labor Supply: A Theoretical and Malisa, and F. A. Plummer. 1992. "Impact of User Fees on Empirical Investigation." In A. N. Herrin and P. L. Rosen- Attendance at a Referral Centre for Sexually Transmitted field, eds., Economics, Health and Tropical Diseases. Manila: Diseases in Kenya." Lancet 340: 463-66. School of Economics, University of the Philippines. 189 Office of Technology Assessment, U.S. Congress. 1993. Pollitt, Ernesto. 1990. Malnutrition and Infection in the Class- Phar?naceutical R & D: Costs, Risks and Rewards. Report room. Paris: UNESCO. OTA-H-522, February. Washington, D.C.: U.S. Govern- Popkin, Barry M. 1993. "Nutritional Patterns and Modes of ment Printing Office. Change." Population and Development Review. Oganov, Raphael G. 1992. "Cardiovascular Diseases in Population Information Program. 1992. Population Reports Russia." Paper presented at Institute of Medicine meeting Series M (11): December. in Washington, D.C., October. National Research for Pre- Population Reference Bureau. 1992a. "Adolescent Sexual ventive Medicine, Moscow. Activity and Childbearing in Latin America and the Carib- Omran, Abdel R. 1971. "The Epidemiological Transition: A bean: Risks and Consequences." Washington, D.C. Theory of the Epidemiology of Population Change." Mu- 1992b. "Adolescent Women in Sub-Saharan Af- bank Memorial Fund Quarterly 49(4): Part 1, 509-38. rica: A Chartbook on Marriage and Childbearing." Wash- Over, Mead. 1991. Economics for Health Sector Analysis: Con- ington, DC cepts and Cases. Economic Development Institute Technical Potts, Malcolm, Roy Anderson, and Marie-Claude Boily. Materials. Washington, D.C.: World Bank. 1991. "Slowing the Spread of Human Immunodeficiency Over, Mead, and Peter Piot. Forthcoming. "HIV Infection Virus in Developing Countries." Lancet 338 (Septem- and Sexually Transmitted Diseases." In Dean T. Jamison, ber 7). W. Henry Mosley, Anthony R. Measham, and José-Luis President's Commission for the Study of Ethical Problems in Bobadilla, eds., Disease Control Priorities in Developing Medicine and Biomedical and Behavioral Research. 1983. Countries. New York: Oxford University Press. "Securing Access to Health Care." Washington, D.C. Over, Mead, Randal P. Ellis, Joyce Huber, and Orville Solon. Preston, Samuel H., and Michael R. Haines. 1991. Fatal 1992. "The Consequences of Adult Ill-Health." In Richard Years: Child Mortality in Late Nineteenth-Century America. Feachem and others, eds., The Health of Adults in the Devel- Princeton, N.J.: Princeton University Press. oping World. New York: Oxford University Press. Preston, Samuel H., Nathan Keyfitz, and Robert Schoen. Palloni, Alberto. 1981. "Mortality in Latin America: Emerg- 1972. Causes of Death: Life Tables for National Populations. ing Patterns." Population and Development Review 7(4): New York: Seminar Press. 623-51. Psacharopoulos, George. 1993. "Returns to Education: A Pandey, M. R., J. S. M. Boleij, K. R. Smith, and E. M. Global Update." Policy, Research, and External Affairs Wafula. 1989. "Indoor Air Pollution in Developing Coun- Working Paper Series 1067. World Bank, Latin America tries and Acute Respiratory Infection in Children." Lancet and the Caribbean Technical Department, Washington, 1: 427-29. D.C. Pauley, Mark V. 1986. "Taxation, Health Insurance, and Reinhardt, Uwe E. 1989. "The U.S. Health Care Financing Market Failure in the Medical Economy." Journal of Eco- and Delivery System: Its Experience and Lessons for nomic Literature 24 (June): 629-75. Other Nations." International Symposium on Health Pelletier, David L. 1991. "Relationships between Child An- Care Systems, Taiwan, China. thropometry and Mortality in Developing Countries: Im- 1991. "Health Manpower Forecasting: The Case of plications for Policy, Programs and Future Research." Physician Supply." In Eli Ginzberg, ed., Health Services Cornell Food and Nutrition Policy Program Monograph Research: Key to Health Policy. Cambridge, Mass.: Harvard 12. Cornell University, Ithaca, N.Y. University Press. Phelps, Charles E. 1992. Health Economics. New York: Har- Relman, Arnold S. 1993. "The Marketplace in Health Care per-Collins. Reform: The Demographic Limitations of Managed Com- Pierce, John P. 1989. "International Comparisons of Trends petition." New England Journal of Medicine 328 (January): in Cigarette Smoking Prevalence." American Journal of 148-52. Public Health 79(2): 152-57. Reyes, Edna A., and Oscar F. Picazo. 1990. "Health Man- 1991. "Progress and Problems in International power Employment and Productivity in the Philippines." Public Health Efforts to Reduce Tobacco Usage." Annual Working Paper Series 90-19. Philippine Institute for De- Review of Public Health 12: 383-400. velopment Studies, Manila. Pinstrup-Anderson, Per, Susan Burger, Jean-Pierre Habicht, Richards, R., and T. Fulop in collaboration with J. Banner- and Karen Peterson. Forthcoming. "Protein Energy Mal- man, G. Greenholm, J.-J. Guilbert, and M. Wunderlich. nutrition." In Dean T. Jamison, W. Henry Mosley, An- 1987. Innovative Schools for Health Personnel: Report on Ten thony R. Measham, and José-Luis Bobabdilla, eds., Dis- Schools Belonging to the Network of Community-Oriented Edu- ease Control Priorities in Developing Countries. New York: cational Institutions for Health Sciences. Geneva: World Oxford University Press. Health Organization. Piola, Sérgio Francisco, and Solon Magalhaes Vianna. 1991. Riddell, Roger C. 1987. Foreign Aid Reconsidered. Baltimore, "PolIticas e Prioridades do Sistema CJnico de Saüde- Md.: Johns Hopkins University Press. SUS." World Bank, Latin America Country Department Robertson, Robert L., Andrew J. Hall, Paolo E. Crivelli, Ya- 1, Washington, D.C. moundow Lowe, Hazel M. Inskip, and Sharon K. Snow. Pitt, Mark M., Mark R. Rosenzweig, and Nazmul Hassan. 1992. "Cost Effectiveness of Immunizations: The Gambia 1990. "Productivity, Health and Inequality in the Intra- Revisited." Health Policy and Planning 7(2): 111-22. houshold Distribution of Food in Low-Income Coun- Rodgers, G. B. 1979. "Income and Inequality as Determi- tries." American Economic Review 80: 1139-56. nants of Mortality: An International Cross-Section Analy- Plitcha, Stacey. 1992. "The Effects of Woman Abuse on sis." Population Studies 33(2): 343-51. Health Care Utilization and Health Status: A Literature Roemer, Milton I. 1991. National Health Systems of the World. Review." Women's Health (Jacobs Institute of Women's Vol. 1: The Countries. New York: Oxford University Press. Health) 2(3): 154-61. 190 Rogot, E., P. D. Sortie, and N. J. Johnson. 1992. "Life Ex- Serageldin, Ismail, A. Edward Elmendorf, and El-Tigani E. pectancy by Employment Status, Income, and Education El-Tigani. Forthcoming. "Structural Adjustment and in the National Longitudinal Mortality Study." Public Health in Africa in the 1980s." World Bank, Africa Techni- Health Reports 107(4): 457-61. cal Department, Washington, D.C. Romieu, Isabell. 1992. "Air Pollution and Pneumonia in Shepard, Donald S., R. L. Robertson, C. S. M. Cameron III, Young Children." Centro Panamericano de EcologIa Hu- P. Saturno, M. Pollack, J. Manceau, P. Martinez, P. mana y Salud, Mexico City. Meissner, and J. Perrone. 1989. "Cost Effectiveness of Romieu, Isabel!, Henyk Weitzenfeld, and Jacobo Finkelman. Routine and Campaign Vaccination Strategies in Ecua- 1990. "Urban Air Pollution in Latin America and the Ca- dor." Bulletin of the World Health Organization 67(6): 649-62. ribbean: Health Perspectives." World Health Statistics Shepard, Donald, Julia Walsh, Wolfgang Munar, Laura Quarterly 43(3): 153-67. Rose, Rodrigo Guerrero, Luis F. Cruz, Guillermo Reyes, Rozynski, Edward M., and Matthew S. Gallivan. 1992. Gail P. Orsolani, and Carlos Solarte. 1990. "Cost-Effec- "The Global Medical Device Market Report." Health In- tiveness of Ambulatory Surgery in Cali, Colombia." Pa- dustry Manufacturers Association, Washington, D.C. per presented at conference on Outpatient Hospitals: Rublee, Dale A. 1989. "Medical Technology in Canada, Ger- Their Role in Health Care Systems in Developing Coun- many, and the United States." Health Affairs (Fall):178-81. tries, Harvard School of Public Health, Boston, Sagan, L. A., and A. A. Afifi. 1979. "Health and Economic November. Development Factors Affecting Mortality." In G. T Good- Shim, Young-Hee. 1992. "Sexual Violence against Women man and W. D. Rowe, eds., Energy Risk Management. Lon- in Korea: A Victimization Survey of Seoul Women." Pa- don: Academic Press. per presented at conference on International Perspectives: Sahn, David E. 1990. Malnutrition in Côte d'Ivoire: Prevalence Crime, Justice and Public Order, St. Petersburg, Russia, and Determinants. Social Dimensions of Adjustment in June 21-27. Sub-Saharan Africa Working Paper 4. Washington, D.C.: Smith, G. S., and P. Barss. 1991. "Unintentional Injuries in World Bank. Developing Countries: The Epidemiology of a Neglected Sanderson, Warren C., and Jee-Peng Tan. Forthcoming. Problem." Epidemiologic Reviews 13: 228-65. "Population Issues in Asia: Context, Policies, and Pros- Smith, Kirk R. Forthcoming. "Combustion, Air Pollution, pects." World Bank, Asia Technical Department, Wash- and Health in Developing Countries." Annual Review of ington, D.C. Energy and Environment 18(20). Schieber, George J., Jean-Pierre Poullier, and Leslie M. Smith, Kirk R., and Liu Youcheng. 1993. "Indoor Air Pollu- Greenwald. 1992. "U.S. Health Expenditure Perfor- tion in Developing Countries." In Jonathan Samet, ed., mance: An International Comparison and Data Update." The Epidemiology of Lung Cancer. New York: Marcel Health Care Financing Review 11(4): 9-15. Dekker. Schmidt, Henk G., Victor R. Neufeld, Zohair M. Nooman, Smith, K. R., and S. Rodgers. 1992. "Interventions to Re- and Toye Ogunbode. 1991. "Network of Community-Ori- duce Morbidity and Mortality from Pneumonia in Chil- ented Educational Institutions for the Health Services." dren: Reducing Exposure to Household Biomass Smoke." Academic Medicine 66: 259-63. London School of Hygiene and Tropical Medicine. Schneider, Markus, Rudolf K-H. Dennerlein, Aynur Kose, Solon, Orville, Rhais M. Gamboa, J. Brad Schwartz, and and Lea Scholtes. 1992. Health Care in the EC Member Alejandro N. Herrin. 1992. "Health Sector Financing in States. Health Policy Monographs, Vol. 1. Amsterdam: Philippines." Health Finance Development Project. Elsevier. HFDP Monograph 2. Government of the Philippines De- Schroeder, Steven A. 1984. "Western European Responses partment of Health, Manila. to Physician Oversupply-Lessons for the United States." Stamler, Jeremiah, Geoffrey Rose, Rose Stamler, Paul Elliot, JAMA: Journal of the American Medical Association 252 Alan Dyer, and Michael Marmot. 1989. "Intersalt Study (July): 373-84. Findings: Public Health and Medical Care Implications." 1992. "Physician Supply and the U.S. Medical Hypertension 14: 570-77. Marketplace." Health Affairs (Spring): 234-43. Stark, Evan. 1984. "The Battering Syndrome: Social Knowl- Schultz, T. Paul, and Aysit Tansel. 1993. "Estimates of Wage edge, Social Therapy and the Abuse of Women." Ph.D. Returns to Adult Health in Côte d'Ivoire and Ghana." dissertation. State University of New York, Binghamton, Yale University, New Haven, Conn. Department of Sociology. Schwab, L. 1987. "Cost Effective Cataract Surgery in Devel- Stark, Evan, and Anne Flitcraft. 1991. "Spouse Abuse." In oping Nations." Ophthalmic Surgery 12: 307-09. M. Rosenberg and Anne Finley, eds., Violence in America: Schwartz, H., and D. W. Dockery. 1992. "Increased Mortal- A Public Health Approach. New York: Oxford University ity in Philadelphia Associated with Daily Air Pollution Press. Concentrations." American Reviews of Respiratory Disease Stephenson, Patricia, Marsden Wagner, Mihaela Badea, and 145: 600-04. Florina Servanescu. 1992. "Commentary: The Public Scrimshaw, Nevin S. 1986. "Consequences of Hunger for Health Consequences of Restricted Induced Abortion- Individuals and Society." In Barry R. Bloom, ed., Proceed- Lessons from Romania." American Journal of Public Health ings from the Symposium on the Biomedical Aspects of World 82(10): 1328-31. Famine. Federation Proceedings (Federation of American So- Strauss, John, Paul Gertler, Omar Rahman, and Kristin Fox. cieties for Experimental Biology) 45(10): 2421-26. 1992. "Gender and Life-Cycle Differentials in the Patterns Sen, Amartya R. 1981. Poverty and Famines: An Essay on Enti- and Determinants of Adult Health." Prepared for the tlement and Deprivation. Oxford, U.K.: Clarendon Press. Government of Jamaica. RAND, Santa Monica, Calif. 191 Study Group for Global Environment and Economics. 1991. UNICEF and IMIP (lnstituto Materno-Infantil de Pernam- Pollution in Japan: Our Tragic Experience." Environ- buco). 1992. Cuidados Basicos de Sadde em Comunidndes de ment Agency, Office of Policy, Planning and Research, Baixa Renda. Recife. Tokyo. UNICEF, WHO, and UNESCO. 1991. Facts for Life. New Summers, Lawrence H. 1989. "What Can Economics Con- York. tribute to Social Policy? Some Simple Economics of Man- United Nations. 1985. Socio-Economic Differentials in Child dated Benefits." AEA Papers and Proceedings 79(2): 177-83. Mortality in Developing Countries. Population Study 97. 1992. "Investing in All the People." Policy, Re- New York. search, and External Affairs Working Paper 905. World 1988. Mortality of Children under Age Five. New Bank, Washington, D.C. York. Summers, Lawrence H., and Lant H. Pritchett. 1993. "The 1990. World Population Prospects. New York. Structural Adjustment Debate." American Economic Re- 1991. Demographic Yearbook. New York. view Conference Proceedings, American Economic Asso- 1992. Child Mortality since the 1960s. New York. ciation meeting, Anaheim, Calif. Forthcoming. The Health Rationale for Family Plan- Tan, Jee-Peng, and Alain Mingat. 1992. Education in Asia: A ning: Timing of Births and Child Survival. New York. Comparative Study of Cost and Financing. World Bank Re- U.S. Centers for Disease Control and Prevention. 1992. gional and Sectoral Studies, Washington, D.C. "Famine-Affected, Refugee, and Displaced Populations: Tarlov, Alvin R. 1986. "HMO Enrollment Growth and Phy- Recommendations for Public Health Issues." Morbidity sicians: The Third Compartment." Health Affairs Spring: and Mortality Weekly Report 41 (June 24-RR-13). 23-35. Van Doorslaer, Eddy, Adam Wagstaff, and Frans Rutten, Thomas, Duncan. 1990. "Intra-Household Resource Alloca- eds. 1993. Equity in the Finance and Delivery of Health Care: tion: An Inferential Approach." Journal of Human Re- An International Perspective. CEC Health Services Research sources 25(4): 635-64. Series 8. Oxford, U.K.: Oxford University Press. Thomas, Duncan, Victor Lavy, and John Strauss. 1992. Pub- Victora, Cesar, and Fernando Barros. 1990. "Avaliacao Pre- lic Policy and Anthropometric Outcomes in Côte d'Ivoire. Liv- liminar do Impacto da Pastoral da Crianca sobre Alguns ing Standards Measurement Study Working Paper 89. Indicadores de Saüde e de Utilizacão de Servicos." Washington, D.C.: World Bank. UNICEF, Brasilia. Thomas, Duncan, John Strauss, and Maria-Helena Henri- Viravaidya, M., S. Obremskey, and C. H. Myers. 1991. ques. 1990. "How Does Mother's Education Affect Child "The Economic Impact of AIDS on Thailand." Population Height?" Journal of Human Resources 26(2): 183-211. and Community Development Association, Bangkok. Tinker, Anne, and Marjorie Koblinsky. 1993. Making Mother- Vogel, R. J. 1989. "Trends in Health Expenditures and Reve- hood Safe. World Bank Discussion Paper 202. Washington, nue Sources in Sub-Saharan Africa." Paper prepared in D.C. conjunction with the World Bank Sub-Saharan African Tomkins, Andrew, and Fiona Watson. 1989. Malnutrition and Health Policy Study, Washington, D.C. Infection. Administrative Committee on Coordination! 1990. "Health Insurance in Sub-Saharan Africa: A Subcommittee on Nutrition. Nutrition Policy Discussion Survey and Analysis." Policy, Research, and External Af- Paper 5. New York: United Nations. fairs Working Paper Series 476. World Bank, Africa Tech- Tomson, Gäran, and Göran Sterky. 1986. "Self-Prescribing nical Department, Washington, D.C. by Way of Pharmacies in Three Asian Developing Coun- Waaler, Hans T. 1984. "Height, Weight, and Mortality: The tries." Lancet 2(13 September): 620-21. Norwegian Experience." Acta Medica Scandinavica 679: Torrance, G. W. 1986. "Measurement of Health State Util- 3-56. ities for Economic Appraisal." Journal of Health Economics Walsh, Julia A. 1988. Establishing Health Priorities in the Devel- 5: 1-30. oping World. UNDP Division for Global and Interregional Toubia, Nahid. 1993. "A Call for Global Action against Fe- Programmes. Boston, Mass.: Adams Publishing Group. male Genital Mutilation." Women, Ink., New York. Walsh, Julia A., and Kenneth Warren. 1979. "Selective Pri- TRRL (Transport and Road Research Laboratory). 1991. "To- mary Health Care-An Interim Strategy for Disease Con- wards Safer Roads in Developing Countries: A Guide for trol in Developing Countries." New England Journal of Planners and Engineers." Crowthorne, U.K. Medicine 301 (1 November): 967-74. UNDP (United Nations Development Programme). 1993. Walsh, J. A., C. M. Feifer, A. R. Measham, and P. J. Gertler. Human Development Report 1993. New York: Oxford Uni- Forthcoming. "Maternal and Perinatal Health Problems." versity Press. In Dean T. Jamison, W. Henry Mosley, Anthony R. Mea- UNICEF (United Nations Children's Fund). 1993. The State sham, and José-Luis Bobadilla, eds., Disease Control Priori- of the World's Children 1993. Oxford: Oxford University ties in Developing Countries. New York: Oxford University Press. Press. Forthcoming. The Progress of Nations 1993. New Walter, Tomas, Manuel Olivares, and Eva Hertrampf. 1990. York. "Field Trials of Food Fortification with Iron: The Experi- UNICEF/Bamako Initiative Management Unit. 1990. The - ence in Chile." In Bo Lonnerdal, ed., Iron Metabolism in Bamako Initiative: Reaching Health Goals through Strength- Infants. Boca Raton, Fla.: CRC Press. ened Services Delivery. New York. Wasserman, Jeffrey, Willard G. Manning, Joseph P. New- 1992. "Progress Report and Recommendation on house, and John D. Winkler. 1991. "The Effects of Excise the Bamako Initiative," presented to the UNICEF Execu- Taxes and Regulations on Cigarette Smoking." Journal of tive Board, 1992 session. New York. Health Economics 19: 43-64. Wegman, David H. 1992. "The Potential Impact of Epidem- iology on the Prevention of Occupational Disease." A,ner- ican Journal of Public Health 82: 944-54. 192 Weisbrod, Burton A. 1991. "The Health Care Quadrilemma: Various years. "Health Equipment Management: An Essay on Technological Change, Insurance, Quality of A World Health Organization Newsletter." (English Care, and Cost Containment." Journal of Economic Lit era- only.) Division of Strengthening of Health Services ture 29 (june): 523-52. (SHS), Geneva. Welch, W. Pete, Mark E. Miller, H. Gilbert Welch, Elliott WHO Global Programme on AIDS. 1993a. "The Costs of Fisher, and John E. Wennberg. 1993. "Geographic Varia- HIV/AIDS Prevention Strategies in Developing Coun- tion in Expenditures for Physicians' Services in the United tries." GPA/Dir/93.2. Geneva. States." New England Journal of Medicine 328:9: 621-27. 1993b. "Statement from the Consultation on Test- Werner, David. 1987. Disabled Village Children: A Guide for ing and Counseling for HIV Infection 1992." WHO/CPA! Community Health Workers, Rehabilitation Workers, and Fain,- Inf/93.2. Geneva. lies. Palo Alto, Calif.: Hesperian Foundation. WHO/UNEP (United Nations Environment Programme). Whitfield, R., Jr. 1987. "Dealing with Cataract Blindness- 1992. Urban Air Pollution in Megacities of the World. Oxford, Part III: Paramedical Cataract Surgery in Africa." Ophthal- U.K.: Blackwell Reference. mic Surgery 18: 765-67. Wilkinson, R. G. 1992. "Income Distribution and Life Ex- WHO (World Health Organization). 1977. International Clas- pectancy." British Medical Journal 304 (January): 165-68. sification of Diseases. Ninth Revision. Geneva. Woodward, David. 1992. Debt, Adjustment and Poverty in De- 1988a. Estimating Drug Requirements: A Practical veloping Countries. Vol. 1: National and International Dimen- Manual. Action Programme on Essential Drugs and Vac- sions of Debt and Adjustment in Developing Countries. Vol. 2: cines. Geneva. The Impact of Debt and Adjustment at the Household Level in 1988b. Guidelines for Developing National Drug Poli- Developing Countries. London: Pinter Publishers in associa- cies. Geneva. tion with Save the Children. 1988c. The World Drug Situation. Geneva. World Bank. 1980. World Development Report 1980. New York: 1989a. New Approaches to Improve Road Safety. Tech- Oxford University Press. nical Report Series 781. Geneva. 1984a. China: The Health Sector. A World Bank 1989b. World Health Statistical Annual. Geneva. Country Study. Washington, D.C. 1991a. "Organization and Financing of Health 1984b. World Development Report 1984. New York: Care Reform in Countries of Central and Eastern Eu- Oxford University Press. rope." Report by WHO Task Force on Health Develop- 1987. Financing Health Services in Developing Coun- ment in Countries of Central and Eastern Europe for a tries: An Agenda for Reform. A World Bank Policy Study. meeting at the World Health Organization, April 22-26, Washington, D.C. Geneva. 1988. Education in Sub-Saharan Africa: Policies for 1991b. "Programme on Substance Abuse: Inter- Adjustment, Revitalization, and Expansion. A World Bank Regional Meeting on Alcohol-Related Problems, April Policy Study. Washington, D.C. 2-8, Tokyo." Division of Health Services, Geneva. 1990a. Brazil: The New Challenge of Adult Health. A 1991c. "The Public/Private Mix in National Health World Bank Country Study. Washington, D.C. Systems and the Role of Ministries of Health." Interregio- 1990b. World Development Report 1990. New York: nal Meeting, Hacienda Cocoyoc, State of Morelos, Mex- Oxford University Press. ico. Division of Health Services, Geneva. 1991. World Development Report 1991. New York: 1991d. Tropical Diseases: Progress in Research 1989- Oxford University Press. 1990. Tenth Programme Report of the UNDP/ World Bank/ 1992a. China: Long-Term Issues and Options in the WHO Special Programme for Research and Training in Health Transition. A World Bank Country Study. Washing- Tropical Diseases (TDR). Geneva. ton, D.C. 1991e. World Health Statistics Annual. Geneva. 1992b. Poland: Health System Reform. A World Bank 1992a. Acute Effects on Health of Smog Episodes. Country Study. Washington, D.C. WHO Regional Publications, European Series 43. 1992c. Romania: Human Resources and the Transition Copenhagen. to a Market Economy. A World Bank Country Study. Wash- 1992b. "Global Health Situation and Projections- ington, D.C. Estimates." Division of Epidemiological Surveillance and 1992d. "Setting Environmental Priorities in Cen- Health Situation and Trend Assessment, Geneva. tral and Eastern Europe." World Bank, Washington D.C. 1992c. "Implementation of the Global Strategy for 1992e. Adjustment Lending and Mobilization of Private Health for All by the Year 2000, Second Evaluation and and Public Resources for Growth. Policy and Research Series Eighth Report on the World Health Situation." March. 22. Washington, D.C. Geneva. 1992f. World Development Report 1992. New York: 1992d. Our Planet, Our Health. Report of the WHO Oxford University Press. Commission on Health and Environment. Geneva. 1992g. Zimbabwe: Financing Health Services. A 1992e. "Programme on Substance Abuse: Atlas World Bank Country Study. Washington, D.C. Report." Geneva. - 1993a. Effective Family Planning Programs. Washing- 1993. "Evaluation of Recent Changes in the Fi- ton, D.C. nancing of Health Services." Report of a WHO Study 1993b. Housing: Enabling Markets to Work. A World Group. Technical Report Series 829. Geneva. 'Bank Policy Paper. Washington, D.C. Forthcoming. "Investing in Care for Persons with 1993c. Implementing the World Bank's Strategy to Re- AIDS." Draft. Geneva. duce Poverty: Progress and Challenges. Washington, D.C. 193 1993d. Poverty Reduction Handbook. Washington, Yu, Seung-Hum, and Gerard F. Anderson. 1992. "Achiev- D.C. ing Universal Health Insurance in Korea: A Model for Forthcoming. Better Health in Africa. Washington, Other Developing Countries?" Health Policy 20: 289-99. D.C. Zeckhauser, R., and Donald Shepard. 1976. "Where Now World Bank Water Demand Research Team. 1993. "The De- for Saving Lives?" Law and Contemporary Problems 40: mand for Water in Rural Areas: Determinants and Policy 5-45. Implications." World Bank Research Observer 8(1): 47-70. Zinanga, Alex. 1992. "Development of the Zimbabwe Fam- Yang, Bong-Mm. 1991. "Health Insurance in Korea: Oppor- ily Planning Program." Policy, Research, and External Af- tunities and Challenges." Health Policy and Planning 6(2): fairs Working Paper Series 1053. World Bank, Population 119-29. and Human Resources Department, Washington, D.C. Yoshikawa, Aki, Norihiko Shirouzu, and Matthew Holt. Zwi, A. 1992. "Injury in Developing Countries: A Review of 1991. "How Does Japan Do It? Doctors and Hospitals in a the Literature." London School of Hygiene and Tropical Universal Health Care System." Stanford Law and Policy Medicine, Department of Public Health and Policy. Review 3 (FaIl): 111-37. 194 Appendix A. Population and health data Tables A.1 and A.2 present summary data on popula- vital registration data and official life tables were used tion and GNP. Tables A.3 through A.9 provide popu- in combination with child mortality estimates aver- lation and health data as a supplement to the main aged across the countries of each region. For China text and to the data presented in the World Develop- mortality in 1990 was obtained by adjusting upward ment Indicators. All the tables except A.1 and A.2 are the deaths by age recorded in the 1990 census, using organized by demographic region as presented in Ta- an adjustment factor of 1.20 for males and 1.25 for ble A.10. Economies are listed in ascending order of females; these adjustment factors were derived from GNP per capita as estimated for mid-1991. Regional a comparison of the deaths by age and sex with the totals include all the economies listed in Table A.10, population distribution in 1982 and 1990. For India but country data are presented only for those econ- the 1988 Sample Registration System life table was omies with estimated populations of 3 million or taken for 1990 without adjustment. For Latin America more in mid-1990. Taiwan (China) is not presented and the Caribbean 1990 mortality was based on separately in this appendix but is included in the esti- deaths and population by age as available in the Pan mates for Other Asia and islands. Countries for American Health Organization (PAHO) data base for which GNP per capita is uncertain are listed at the 1990. For Sub-Saharan Africa, Other Asia and is- end of the regional groups and italicized. lands, and the Middle Eastern crescent mortality esti- Although the data reported here are drawn from mates were based on country-specific estimates of the most authoritative sources available, compara- child mortality, combined with indicators from a bility is limited because of variation in data collection, small number of accurate life tables of the relation- statistical methods, and definitions. Differences in ship between child and adult mortality. Fertility esti- the reliability of the data are indicated by presenting mates by region for the period 1950 to 1990 were de- in italics the figures that are deemed to be less termined by the 1990 age distributions and the precise. mortality assumptions. Fertility estimates for the pe- riod 1990 to 2050 were taken as weighted averages of Table A.3 Population structure and dynamics the country-specific values used by Bos and others 1992. Estimates of migration were obtained indirectly Population in 1990 by country and the percentages from United Nations, World Population Prospects 1990. for under age 15 and for age 60 and over were taken Regional summaries for various columns of Table from Bos and others 1992. Regional totals were ob- A.3total fertility rate, total live births, life expectancy, tained by addition. The regional population totals median age at death, child mortality rate, and adult mor- provide the basis for the regional projections carried tality rate by sexare taken directly from the regional out for this Report for the period 1950 to 2050. The projections. basis for the mortality assumptions for these projec- Country-specific values for total fertility rate and to- tions varies by region. For the established market tal live births in 1990 are taken from projection data economies and the formerly socialist economies of bases in Bos and others 1992. Mortality indicators are Europe vital registration data from about 1990 were based on the child mortality estimates for 1960, 1975, used; future mortality trends were then adapted to and 1990, which, for developing countries, are largely agree with Bos and others 1992 for 2050. For the past, based on the special exercise described below. 195 Life expectancy at birth, e(0), is the number of years able used is the logarithm of the observed child mor- that a person born in a given year could expect to live, tality rates. For estimates beyond the range of the given the age-specific mortality rates for that year. Life observations, extrapolation is used; all estimates expectancy in 1960 and 1990 and male and female adult based on extrapolation are shown in italics. For coun- mortality rates for 1990 were derived from the child tries not included in Child Mortality since the 1960s, mortality estimates for that year, combined with as- estimates of child mortality by period have been sumptions about the relationship between child and taken from United Nations, Mortality of Children under adult mortality based on the country-specific projec- Age Five (1988). For these countries, point estimates tions in Bos and others 1992. The adult mortality rate for calendar years have been obtained by averaging for a given sex is the probability of dying between estimates for adjacent five-year periods; thus, for ex- ages 15 and 60, expressed per 1,000. Median age at ample, child mortality for 1960 for Ethiopia is ob- death is the age below which half of all deaths occur in tained as the average of the estimated values for a year. 1955-60 and 1960-65. The perinatal mortality rate is the number per 1,000 births of perinatal deaths (late fetal deaths, occurring Table A.4 Population and deaths by age group at twenty-eight weeks of gestation or thereafter, and Population projections for the eight demographic re- early neonatal deaths, occurring within the first gions were made for the period 1990 to 2030, and seven days of life). Estimates of perinatal mortality reverse projections were made back to 1950. The were derived from various data sources. Vital regis- starting point for these forward and reverse projec- tration data were used for most of the established tions is the age and sex distribution of each region's market economies and for Argentina, Chile, China, population, as reported in Bos and others 1992. For Singapore, and Uruguay. Vital registration data for reverse projections, mortality assumptions are based the republics of the former U.S.S.R. were corrected on child mortality estimates derived for this Report as for underreporting of perinatal deaths by using a re- described above for child mortality rates, with addi- gression model of perinatal mortality on postneonatal tional information derived from United Nations, Mor- mortality based on a data series extending over about tality of Children under Age Five (1988); migration as- forty-five years (1945-91) from forty countries with sumptions were obtained indirectly from United complete vital registration. The remaining estimates Nations, World Population Prospects (1990). For the for- drew on community- and hospital-based studies at ward projections, assumptions on fertility, mortality, the district or other subnational level that were ex- and migration were chosen to agree with Bos and panded to the national level using either percentage others 1992, starting from the levels used in the re- of the population living in urban areas in 1990 or verse projection for 1985-90. The population projec- percentage of births attended by trained health staff. tion follows the component projection model used by Vital statistics for the established market economies the World Bank. and the historical data base were obtained from the U.S. National Center for Health Statistics (NCHS). Table A.5 Mortality risk and life expectancy across the life cycle The child mortality rate is defined as the probability of dying by exact age 5. Estimates for the period 1960 The figures in Table A.5 are derived from the regional to 1990 were obtained from a special exercise carried population projections described in the note to Table out jointly for the World Development Report and the A.4. United Nations Children's Fund (UNICEF), the re- sults of which will be published as UNICEF, The Prog- Table A.6 Nutrition and health behavior ress of Nations 1993. (The methodology is described in Stunting is defined as low height-for-age; data are for Hill and Yazbeck, background paper.) The sources of children ages 24-59 months. Wasting is defined as low information are those given in United Nations, Child weight-for-height; data are for children ages 12-23 Mortality since the 1960s (1992), augmented by re- months. Moderate to severe stunting or wasting cently available census and survey data. For each ob- means, respectively, height-for-age or weight-for- servation of child mortality, there is a corresponding height more than 2 standard deviations below the observation of the date to which the measure refers: median of the NCHS reference population. What Point estimates of child mortality were obtained by matters is the excess over the prevalence in the refer- fitting a line to the observations using weighted least ence population; the latter includes genetic effects as squares, the independent variables being years (to well as malnutrition. Stunting is interpreted as mea- account for trends) and the weights being based on suring chronic malnutrition and wasting as measur- consensus judgment about the relative robustness of ing acute or short-term malnutrition, whether the estimates derived from different types of data. In or- cause is inadequate food intake or infectious disease der to focus on rates of change, the dependent van- or both. Mild or moderate malnutrition is not consid- 196 ered disease, but all degrees of malnutrition increase The incidence rate of tuberculosis has been esti- the risk of death in children. Fully breastfed babies are mated using the most recent available information defined as those given breast milk with or without recorded by governments and corrected for many de- water, juice, or other liquids but no food or non- veloping countries with additional information from breast milk before age 4 months. Women are classi- epidemiological studies. The data source is WHO's fied as anemic when the blood hemoglobin level is Tuberculosis Programme. below the WHO norm of 110 grams per liter. Tobacco consumption per year is an estimate of kilo- Table A.8 Health infrastructure and services grams of consumption of dry-weight tobacco per Doctor is defined to include only individuals with the adult (age 15 and older). Where consumption in raw- professional degree of medical doctor. The definition leaf equivalent is not available, data are derived by of nurse includes only registered nurses and regis- converting data on consumption or sales of products. tered midwives. Hospital bed is defined as beds in In some cases consumption is calculated from pro- clinics and hospitals; beds in long-term care facilities duction of and net trade in leaf and products. Con- and nursing homes are excluded. Data sources are sumption of tobacco for 1990 and 2000 was projected the World Bank, the Organization for Economic Co- through a model that used assumptions on the operation and Development (OECD), PAHO, and growth of private consumption expenditure to derive WHO. per capita demand for tobacco. The demand func- Immunization data refer to DPT3three completed tions and elasticities were based on analysis of recent doses of vaccine against diphtheria, pertussis national family budget surveys and past time series of (whooping cough), and tetanusand to measles. The consumption. Antismoking campaigns and other denominator for estimating coverage is the number of preventive activities that have influenced the level of surviving infants age 1 year. The source of data is tobacco consumption were considered for some WHO's Expanded Programme on Immunization. countries through a trend factor, independent of in- come and price. Table A.9 Health expenditure and total flows from Sources for data on child nutrition are World external assistance Health Organization (WHO), Demographic and Health Surveys, Institute for Resource Development Health expenditure includes outlays for prevention, of Macro Systems (IRD), and UNICEF; for anemia, promotion, rehabilitation, and care; population activ- WHO; and for tobacco consumption, the Commodi- ities; nutrition activities; program food aid; and emer- ties and Trade Division of the Food and Agriculture gency aid specifically for health. It does not include Organization of the United Nations (1990). water and sanitation. Per capita expenditures and per capita aid flows are based on World Bank midyear Table A.7 Mortality by broad cause, and incidence of population estimates. tuberculosis Total health expenditure is expressed in official ex- Mortality rates are standardized for age by the direct change rate U.S. dollars. Data on public and private method, using world population as the standard health expenditure for the established market econ- population. Infectious diseases and reproductive health omies and Turkey are from the OECD. For other problems include all deaths from infectious diseases countries, information on government health expen- listed in the International Classification of Diseases, ditures is from national sources, supplemented by Ninth Revision (1977), plus influenza and pneumonia, Government Finance Statistics (published by the Inter- nutritional disorders and anemia, maternal causes of national Monetary Fund), World Bank sector studies, death (including abortion), and perinatal causes of and other studies. Data on parastatal expenditures death. Injuries include all violent causes, whether in- (for health-related social security and social insurance tentional, unintentional, or unknown. Noncommuni- programs) are from the Social Security Division of the cable diseases include all other causes of death. Deaths International Labour Office (ILO) and the World of which the cause was coded as ill-defined are dis- Bank. Data are drawn from Murray, Govindara;, and tributed among the three groups in proportion to the Chellaraj, background paper. number of deaths in each group. The source is WHO Public sector expenditures include government data derived from national vital statistics. Estimates health expenditures, parastatal expenditures, and for economies with incomplete death registration foreign aid, making the figures comparable with (less than 90 percent of deaths), high levels of non- those for OECD countries. Private sector expenditures medical certification of causes of death (more than 15 for countries other than OECD members are based on percent), or high proportions of deaths from ill-de- household surveys carried out by the ILO and other fined causes (more than 20 percent) are considered sources, supplemented by information from United unreliable and are shown in italics. Nations National Income Accounts, World Bank 197 studies, and other studies published in the scientific Countries (OPEC). Major international NGOs in- literature. clude the International Committee for the Red Cross Estimates for countries with incomplete data were (ICRC) and the International Planned Parenthood calculated in three steps. First, where data on either Federation (IPPF). National NGOs were not included private or public expenditures were lacking, the miss- because the available information was not separated ing figures were imputed from data from countries by recipient country. for which information was available. The imputation Information on ODA from bilateral and multilateral followed regressions relating public or private expen- organizations was completed by data from the diture to GDP per capita. Second, for a country with OECD's Development Assistance Committee (DAC) no health expenditure data, it was assumed that the and Creditor Reporting System (CRS) and from the share of GDP spent on health was the same as the Advisory Committee for the Coordination of Infor- average for the corresponding demographic region. mation Systems (ACCIS). DAC has compiled annual Third, if GDP was also unknown but population was aggregate ODA statistics, by sector, since 1960. The known, it was assumed that per capita health spend- OECD's CRS, established in 1970, complements the ing was the same as the regional average. DAC statistics by identifying contributions allocated Estimates for development assistance for health are ex- by sector. The CRS data base is the most complete pressed in official exchange rate U.S. dollars. Total aid source of information for bilateral ODA, but its com- flows represent the sum of all health assistance for pleteness varies among OECD countries and from health to each country by bilateral and multilateral year to year. ACCIS has kept, since 1987, a Register of agencies and by international nongovernmental or- Development Activities of the United Nations that ganizations (NGOs). Direct bilateral official develop- lists sources of funds and executing agencies for all ment assistance (ODA) comes from the OECD coun- United Nations projects by sector. tries. Sources of multilateral development assistance The estimates of development assistance in this ta- include United Nations agencies, development banks ble were prepared by the Harvard Center for Popula- (including the World Bank), the European Commu- tion and Development Studies as a background paper nity, and the Organization of Petroleum Exporting for this Report. 198 Table A.1 Population (midyear) and average annual growth Population (millions) Average annual growth (percent) Count ry group 1965 1973 1980 1990 1991 2000 2030 1965-73 1973-80 1980-90 1990-2000 2000-2030 Low- and middle-income economies 2,602 3,166 3,662 4,445 4,528 5,294 7,736 2.5 2.1 2.0 1.8 1.3 Low-income economies 1,776 2,169 2,507 3,066 3,127 3,686 5,459 2.5 2.1 2.0 1.9 1.3 Middle-income economies 826 997 1,155 1,379 1,401 1,608 2,273 2.3 2.2 1.8 1.5 1.2 Severely indebted 274 332 389 477 486 569 841 2.4 2.3 2.1 1.8 1.3 Sub-Saharan Africa" 233 288 351 474 489 635 1,313 2.7 2.9 3.1 3.0 2.4 EastAsiaandthePacific 1,009 1,240 1,399 1,641 1,667 1,891 2,442 2.6 1.7 1.6 1.4 0.9 South Asia 632 765 903 1,128 1,152 1,368 2,004 2.4 2.4 2.2 1.9 1.3 Europe and Central Asia . . 448 489 492 517 566 0.9 0.6 0.5 Latin America and the Caribbean 249 305 358 438 455 516 721 2.6 2.4 2.0 1.7 1.1 MiddleEastandNorthAfrica 114 141 173 236 244 315 600 2.8 3.0 3.2 2.9 2.2 High-income economies 671 725 766 817 822 864 920 1.0 0.8 0.6 0.6 0.2 OECD members 649 698 733 777 783 820 871 0.9 0.7 0.6 0.5 0.2 World 3,281 3,895 4,428 5,262 5,351 6,157 8,664 2.2 1.8 1.7 1.6 1.2 Note: Because of incomplete coverage, discrepancies between summed subgroup figures may occur. Projections. For the assumptions used in the projections, see technical notes for Table 26 in the World Development Indicators. Excludes South Africa. Table A.2 GNP, population, GNP per capita, and growth of GNP per capita 1991 GNP 1991 1991 GNP per capita Average annual growth of GNP per capita (percent) (billions of population Country group dollars) (millions) (dollars) 1965-73 1973 -80 1980-90 1989 1990 1991 Low- and middle-income economies 4,571 4,528 1,010 4.3 2.7 1.2 1.1 -0.1 -2.1 Low-income economies 1,097 3,127 350 2.5 2.6 4.0 2.9 2.9 2.1 Middle-income economies 3,474 1,401 2,480 . . . . 0.5 0.7 -1.0 -3.4 Severely indebted 1,130 486 2,320 5.2 3.4 -0.8 -0.6 -5.4 -2.5 Sub-Saharan Africa" 173 489 350 1.7 0.9 -1.3 0.5 -1.4 -0.6 East Asia and the Pacific 1,081 1,667 650 5.0 4.8 6.2 4.5 5.3 5.0 South Asia 372 1,152 320 1.2 1.7 3.2 2.9 3.3 -0.7 Europe 1,314 492 2,670 1.4 1.4 -2.7 -9.8 Latin America and the Caribbean 1,065 445 2,390 4.6 2.2 -0.4 -1.1 -1.4 1.7 Middle East and North Africa 474 244 1,940 6.0 1.7 -2.5 -0.2 -0.2 -1.3 High-income economies 16,920 822 20,570 3.7 2.1 2.3 2.7 1.6 0.3 OECD members 16,463 783 21,020 3.8 2.1 2.3 2.7 1.6 0.1 World 21,464 5,351 4,010 2.8 1.3 1.2 1.6 0.5 -0.1 Note: Because of incomplete coverage, discrepancies between summed subgroup figures may occur. Projections. For the assumptions used in the projections, see technical notes for Table 26 in the World Development Indicators. Excludes South Africa. 199 Table A.3 Population structure and dynamics Population and fertility General mortality Age.specific mortality rates Total Under 15 Total live births per Median Pen natal Adult mortality rate, 1990 Population, years old, 60 years and fert it it5 year, 1990 age at mortality Life expectancy at birth Child mortality rate (ages 15-59) 1990 1990 over, 1990 rate, (hundreds of death, rate, Demographic region and economy (millions) (percent) (percent) 1990 thousands) 1960 1990 1990 1990 1961) 1975 1990 Male Female Sub-Saharan Africa 510 1 46 w 5w 6.4 w 251.8 1 43 w 52 w 5w 68w 251 w 212 w 175 w 381 w 322 w Mozambique 16 44 5 6.4 7.2 39 43 2 75 280 280 280 490 421 Tanzania 25 47 5 6.6 11.7 42 49 5 71 242 202 165 379 335 Ethiopia 51 47 5 7.5 26.5 37 48 4 87 294 262 197 404 329 Uganda 16 49 5 7.3 8.5 44 47 4 85 224 173 185 424 367 Burundi 5 46 5 6.8 2.7 40 47 11 87 255 209 180 424 367 Chad 6 42 6 6.0 2.5 35 47 7 74 326 271 212 445 358 Madagascar 12 46 5 6.3 5.3 42 51 Il 76 250 200 170 389 333 Sierra Leone 4 43 5 6.5 1.9 34 38 2 72 391 375 360 503 436 Malawi 9 47 4 7.6 4.6 35 47 4 83 361 313 201 426 369 Rwanda 7 48 4 8.3 3.9 45 44 3 86 210 223 222 453 395 Mali 8 47 5 7.0 4.3 33 48 4 80 413 321 200 417 361 Burkina Faso 9 46 5 6.5 4.2 35 49 4 85 318 254 159 429 352 Niger 8 47 4 7.1 3.9 35 38 3 79 320 320 320 513 454 Nigeria 06 47 4 6.0 42.5 47 49 7 71 204 198 191 406 354 Kenya 24 50 4 6.6 11.1 46 59 15 77 203 139 83 315 259 Benin 5 48 5 6.4 2.2 35 50 6 69 307 228 170 387 316 Central Africa Rep. 3 42 5 5.8 1.3 35 55 15 64 332 209 132 346 288 Ghana 15 47 5 6.3 6.6 45 52 7 71 213 169 170 344 282 Togo 4 48 5 6.7 1.8 39 54 7 75 26.4 193 143 325 268 Guinea 6 46 4 6.5 2.7 35 44 2 76 347 297 268 452 395 Zimbabwe 10 45 4 5.0 3.6 52 62 26 55 159 120 58 269 216 Cdte d'lvoire 12 47 4 6.7 5.4 40 57 10 68 260 194 90 332 277 Senegal 7 47 4 6.5 3.4 35 50 15 73 303 265 156 397 340 Cameroon 12 46 6 5.9 4.8 40 57 16 68 265 194 125 316 256 South Africa 36 38 6 4.3 12.1 48 62 41 50 192 141 91 278 209 Somalia 8 46 5 6.8 3.8 36 45 4 70 294 262 214 443 390 Zaire 37 46 4 6.3 17.0 37 49 6 68 286 223 190 387 319 Sudan 25 46 5 6.3 11.2 46 57 13 78 203 152 104 267 234 Zambia 8 49 4 6.7 4.0 45 47 11 63 213 167 190 422 354 Angola 10 45 5 6.5 4.7 . 35 46 3 75 346 281 214 434 381 India 850 37 7 4.0 258.1 47 58 37 64 235 195 127 272 229 China 1,134 27 9 2.5 251.3 43 69 64 25 210 85 43 201 150 Other Asia and islands 6831 37w 6w 3.3w 188.71 SOw 62w 42w 49w 182w 135w 97w 243w 177w Nepal 19 42 5 5.7 7.6 44 56 12 90 279 202 135 312 243 Cambodia 8 35 5 4.6 3.3 45 50 30 85 218 239 174 347 274 Bangladesh 107 43 5 4.6 37.2 46 56 12 75 251 236 137 295 244 Lao PDR 4 44 5 6.7 2.0 44 50 8 85 232 209 171 345 280 Sri Lanka 17 32 8 2.4 3.5 58 72 73 19 140 69 22 158 92 Indonesia 178 36 6 3.1 45.9 46 59 47 40 214 151 111 278 212 Philippines 61 40 5 3.6 17.8 59 64 49 27 103 75 62 234 172 Papua New Guinea 4 41 5 5.1 1.4 47 52 22 43 204 185 169 374 327 Thailand 56 33 6 2.4 12.0 52 68 71 25 149 85 36 242 163 Malaysia 18 38 6 3.8 5.5 58 71 63 25 106 54 20 177 120 Korea, Rep. 43 26 8 1.8 6.9 53 72 74 10 133 29 10 149 67 Hong Kong 6 21 13 1.5 0.7 64 78 77 8 53 17 7 91 44 Singapore 3 24 8 1.9 0.5 65 74 76 8 48 16 8 135 64 Myanmar 42 37 6 3.9 12.7 43 61 41 50 234 153 101 256 187 VietNam 66 40 7 3.9 20.4 57 67 50 40 105 68 46 180 118 Korea, Dem. People's Rep. 22 28 7 2.4 4.7 53 70 71 20 133 55 31 179 84 Latin America and the Caribbean 4441 36w 7w 3.3w 124.6 I 54 w 70w 55w 33w 161w 104w 60w 228w 163w Nicaragua 4 46 4 5.4 1,5 50 62 13 35 191 149 106 283 264 Haiti 6 40 6 4.8 2.3 47 54 18 43 221 208 156 413 406 Honduras 5 45 5 5.3 2.0 49 67 23 39 203 126 62 220 162 Bolivia 7 43 5 4.9 2.6 43 60 13 37 251 205 125 330 269 Guatemala 9 45 5 5.5 3.6 49 64 23 40 205 152 84 287 227 Dominican Rep. 7 37 6 3.3 2.0 56 68 49 35 149 114 56 212 147 Ecuador 10 39 6 3.8 3.1 53 70 57 37 174 120 42 218 157 Peru 22 38 6 3.8 6.6 45 65 47 40 233 157 73 272 221 El Salvador 5 44 6 4.3 1.7 51 69 32 39 188 146 52 318 217 Colombia 32 35 6 2.7 7.9 58 73 66 33 132 88 21 200 109 Paraguay 4 41 5 4.7 1.5 64 70 42 37 92 70 37 261 210 Chile 13 31 9 2.6 2.9 55 73 69 14 155 68 20 214 112 Venezuela 20 38 6 3.6 5.7 67 72 62 26 78 59 26 196 105 Argentina 32 30 13 2.8 6.5 67 72 72 28 73 56 26 168 90 Uruguay 3 26 16 2.3 0.5 71 74 73 17 55 58 23 194 101 Brazil 150 35 7 3.3 40.4 52 66 57 35 179 110 69 250 182 Mexico 86 37 6 3.3 23.8 56 70 60 30 148 95 38 212 164 Puerto Rico 4 26 14 2.3 0.6 67 76 75 18 70 27 15 155 77 Cuba 11 23 12 1.9 1.9 71 76 77 17 49 34 12 134 95 200 Population and fertility General mortality Ages pecif1 mortality rates Total Adult mortality Under 15 Total live t,irths per Median Pen natal rate, 1990 Population, years old, 60 years and fertility year, 1990 age at mortality Lifeexpectancyal birth Child mortality rate (ages 15-59) 1990 1990 over, 1990 rate, (hundreds of death, rate, Demographic region and economy (millions) (percent) (percent) 1990 thousands) 1960 1990 1990 1990 1960 1975 1990 Male Female Middle Eastern crescent 5031 41w 6w 5.0 w 195.8 1 44 w 61 w 24w 46w 242w 174w 111 w 228w 174 w Pakistan 112 44 5 5.9 47.1 49 56 7 65 222 163 139 296 263 Yemen, Rep. 11 49 5 7.7 6.1 33 49 4 60 378 270 183 334 327 Egypt 52 39 6 5.6 16.2 40 64 38 58 256 212 56 214 158 Morocco 25 41 6 4.6 8.7 45 62 41 45 215 174 71 214 183 Tajikistan 5 45 6 5.0 2.0 . . 65 22 37 . . 115 75 190 133 Jordan 3 44 4 5.5 1.3 5.4 69 35 40 145 85 34 138 93 Syrian Arab Rep. 12 48 4 6.5 5.5 47 66 23 45 199 98 44 157 121 Uzbekistan 21 42 6 4.0 6.6 . . 67 37 33 . . 67 60 225 135 Tunisia 8 38 6 3.7 2.3 41 67 58 40 245 140 45 166 136 Kyrgyzstan 4 38 9 3.7 1.3 68 64 31 63 53 268 131 Georgia 5 24 16 2.1 0.9 72 71 24 39 28 218 94 Azerbaijan 7 33 9 2.7 1.7 . . 69 66 30 . . 69 52 239 106 Turkmenistan 4 41 6 4.1 1.2 . . 64 48 41 . . 101 93 270 155 Turkey 56 35 7 3.5 15.9 47 65 52 45 217 172 94 175 107 Algeria 25 44 5 5.2 9.1 43 65 47 40 242 174 82 135 105 Armenia 3 30 11 2.4 0.6 . . 72 67 25 . . 43 32 195 100 Iran 56 44 5 6.3 25.1 42 63 18 56 234 164 64 174 124 Kazakhstan 17 32 10 2.8 3.7 . . 68 65 27 . . 48 39 291 131 Saudi Arabia 15 46 4 7.0 6.4 38 64 18 40 292 166 81 175 138 Israel 5 31 12 2.9 1.1 72 76 76 11 38 26 10 110 72 Afghanistan 20 45 4 6.9 9.8 34 40 2 75 358 314 307 421 421 Iraq 19 47 4 6.2 8.0 52 63 24 52 163 106 72 194 129 Libya 5 46 4 6.7 2.0 39 62 16 35 269 146 82 191 144 Formerly socialist economies of Eumpe (FSE) 3461 23w 17w 2.2 w 52.9 1 66w 72w 72w 19w 68w 36w 22w 281w 112w Romania 23 24 16 2.2 3.7 63 70 71 12 82 43 31 233 119 Poland 38 25 15 2.2 5.9 65 71 72 15 70 29 20 263 102 Bulgaria 9 20 20 1.9 1.1 67 73 73 11 62 29 21 217 97 Moldova 4 32 11 2.9 0.8 69 68 24 . . 51 32 271 153 Ukraine 52 21 19 2.1 7.5 72 73 22 . . 25 22 270 107 Czechoslovakia 16 23 17 2.0 2.2 70 72 73 10 32 23 13 243 98 Lithuania 4 30 16 2.0 0.6 72 73 20 . . 23 18 276 108 Hungary 11 20 19 1.8 1.3 68 71 73 15 57 33 20 305 133 Belarus 10 23 18 2.2 1.6 73 73 21 22 18 272 64 Russian Federation 148 24 17 2.3 23.5 71 71 22 . . 33 27 304 110 Albania 3 33 8 3.0 0.8 51 70 67 45 164 71 36 250 110 Yugoslavia" 22 23 15 2.1 2.8 59 71 71 16 113 47 28 195 94 Established market economies (EME) 7981 19w 18w 1.7w 104.01 70w 76w 75w 9w 36w 21w 11w 147w 73w Portugal 10 21 18 1.6 1.3 60 75 75 13 108 49 13 169 82 Greece 10 19 20 1.5 1.1 68 76 76 13 50 29 13 133 71 Ireland 4 27 15 2.2 0.6 70 74 75 10 35 20 10 186 98 New Zealand 3 23 15 2.0 0.6 71 75 75 7 26 18 11 159 86 Spain 39 20 19 1.5 4.4 68 76 75 10 56 22 10 148 79 United Kingdom 57 19 21 1.9 7.9 71 76 77 8 27 18 9 156 87 Australia 17 22 15 1.9 2.5 71 77 76 10 24 16 9 148 74 Italy 58 16 16 1.3 5.7 68 77 77 12 56 25 11 128 72 Netherlands 15 18 18 1.6 1.9 73 77 77 10 21 12 9 141 72 Belgium 10 18 21 1.7 1.2 70 76 77 10 38 19 11 156 75 Austria 8 18 20 1.5 0.9 69 76 77 8 47 24 10 162 76 France 56 20 19 1.8 7.6 70 77 78 9 33 16 9 159 66 Canada 27 21 16 1.8 3.9 71 77 76 8 33 16 9 146 65 United States 250 22 17 1.9 38.6 70 76 76 10 31 19 11 157 75 Germany 79 16 20 1.6 9.2 69 76 78 7 43 22 9 159 76 Denmark 5 17 20 1.6 0.6 72 75 77 9 25 12 10 162 90 Finland 5 20 18 1.8 0.7 63 75 76 8 27 11 8 168 86 Norway 4 19 21 1.9 0.6 73 77 78 8 22 12 10 140 68 Sweden 9 17 23 2.0 1.2 73 78 78 7 19 10 8 135 71 Japan 124 18 17 1.6 13.5 68 79 78 6 37 11 6 120 63 Switzerland 7 17 20 1.7 0.8 71 78 78 8 25 12 9 136 63 FSE and EME 1,1441 20w 18w 1.9w 156.8 I 69w 75w 74w 12w 46w 25w 15w 188w 86w Demographically developing group 4,1231 36w 7w 3.8 w 1,270.3 1 46 w 63w 39w 45w 226w 152w 106 w 250w 199 w World 5,2671 32w 9w 3.4 w 1,427.11 53 w 65 w 55w 40w 195w 135w 96 w 234w 169 w Note: In this appendix the demographically developing group includes the Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent regions. Regional totals and averages include relevant information for less populous countries as listed in Table AiD, except for perinatal mortality. Refers to former Czechoslovakia because disaggregated data are not yet available. Refers to former Socialist Federal Republic of Yugoslavia because disaggregated data are not yet available. 201 Table A.4 Population and deaths by age group Population (millions) Deaths (millions) Demographic region and age group 1950 1980 1990 2000 2030 1950 1980 1990 2000 2030 Sub-Saharan Africa 179 376 510 724 1,628 4.4 6.2 7.9 9.3 117 0-4 32 69 95 139 206 2.3 3.3 4.0 4.8 3.4 5-14 46 111 140 205 392 0.4 0.5 0.7 0.7 0.6 15-59 93 179 252 348 945 1.1 1.5 1.9 2.1 3.7 60+ 9 17 23 32 85 0.6 0.9 1.3 1.7 4.0 India 358 684 850 1,003 1,357 8.1 9.1 9.3 9.7 11.8 0-4 55 97 117 111 102 4.4 3.9 3.2 2.5 1.0 5-14 84 168 197 223 202 0.6 0.6 0.6 0.6 0.2 15-59 198 378 477 593 870 1.6 1.9 2.3 2.6 2.9 60+ 20 42 59 76 182 1.5 2.7 3.3 4.0 7.7 China 547 988 1,134 1,296 1,610 15.2 7.7 8.9 9.1 13.9 0-4 76 97 118 116 105 6.0 1.4 1.1 0.8 0.3 5-14 108 253 187 240 232 1.3 0.3 0.1 0.2 0.1 15-59 322 566 728 808 927 4.5 2.3 2.3 2.3 2.2 60+ 41 73 101 132 346 3.4 3.7 5.4 5.8 11.3 Other Asia and islands 281 552 683 808 1,108 5.7 5.7 5.5 6.5 9.2 0-4 42 82 86 87 83 2.5 2.2 1.6 1.5 0.6 5-14 68 141 164 169 165 0.4 0.4 0.4 0.3 0.1 15-59 155 298 390 494 698 1.6 1.5 1.5 1.9 2.3 60+ 17 31 43 58 161 1.2 1.6 2.0 2.8 6.2 Latin America and the Caribbean 166 355 444 538 765 2.9 2.7 3.0 3.1 6.0 0-4 27 52 56 59 56 1.3 0.9 0.7 0.6 0.3 5-14 40 89 103 113 114 0.2 0.1 0.1 0.1 0.1 15-59 90 194 254 323 474 0.7 0.9 0.9 0.7 1.5 60+ 9 21 31 43 121 0.6 0.9 1.2 1.7 4.1 Middle Eastern crescent 148 382 503 667 1,240 3.9 4.7 4.4 5.6 7.6 0-4 24 62 81 102 130 2.0 2.1 1.8 2.1 1.2 5-14 35 99 127 173 252 0.3 0.3 0.2 0.3 0.2 15-59 79 201 266 353 748 1.0 1.2 0.9 1.4 1.9 60+ 10 20 29 39 110 0.7 1.1 1.4 1.8 4.3 Formerly socialist economies of Europe (FSE) 269 324 346 361 395 3.1 3.2 3.8 3.7 4.3 0-4 27 27 27 25 26 0.7 0.3 0.1 0.1 0.0 5-14 50 49 54 51 50 0.1 0.1 0.0 0.0 0.0 15-59 165 204 208 217 225 0.8 0.5 0.9 0.7 0.5 60+ 26 45 57 67 94 1.5 2.3 2.7 2.9 3.7 Established market economies (EME) 564 757 798 832 869 6.5 7.5 7.1 8.0 10.1 0-4 58 52 51 50 49 0.6 0.2 0.1 0.1 0.1 5-14 96 117 104 104 99 0.1 0.1 0.0 0.0 0.0 15-59 342 461 497 516 460 1.5 1.0 1.1 1.1 0.7 60+ 69 127 145 163 260 4.3 6.2 5.9 6.8 9.3 FSE and EME 832 1,077 1,144 1,194 1,267 9.6 10.7 10.9 11.7 143 0-4 85 79 78 75 75 1.3 0.5 0.2 0.2 0.1 5-14 146 166 158 156 149 0.2 0.1 0.0 0.0 0.0 15-59 507 665 705 732 685 2.2 1.5 2.0 1.8 1.2 60+ 95 167 203 231 358 5.8 8.5 8.6 9.7 13.0 Demographically developing group 1,678 3,337 4,123 5,034 7,708 40.2 36.2 39.1 43.3 60.2 0-4 255 458 552 613 682 18.5 13.9 12.4 12.3 6.8 5-14 381 851 919 1,123 1,348 3.2 2.2 2.1 2.2 1.3 15-59 937 1,902 2,367 2,918 4,672 10.5 9.2 9.8 11.0 14.5 60+ 106 126 286 380 1,005 8.0 10.9 14.6 17.8 37.6 World 2,511 4,414 5,267 6,228 8,975 49.8 46.9 50.0 55.0 74.5 340 536 631 687 757 19.8 14.4 12.7 12.5 6.9 5-14 527 1,017 1,077 1,279 1,497 3.4 2.3 2.2 2.2 1.3 15-59 1,443 2,489 3,072 3,600 5,358 12.7 10.7 11.8 12.8 15.7 60+ 200 371 488 662 1,363 13.8 19.4 23.2 27.5 50.6 202 Table A.5 Mortality risk and life expectancy across the life cycle Life expectancy at different Probability of dying (percent) ages (years) Demographicregionandagegroup 1950 1980 1990 2000 2030 Demographic region and age 1950 1980 1990 2000 2030 Sub-Saharan Africa Sub-Saharan Africa 0-4 28.6 19.9 17.5 15.2 7.9 0 39 49 52 55 64 5-14 6.2 3.7 3.1 2.7 1.3 5 50 56 58 59 64 15-59 47.9 36.6 34.5 30.6 21.5 15 43 48 50 51 55 60-75 58.0 51.5 49.4 47.7 42.1 60 13 15 15 15 17 India India 0-4 30.4 17.2 12.4 10.5 4.5 0 42 55 58 61 70 5-14 5.0 3.0 2.7 2.3 1.0 5 54 61 61 63 68 15-59 38.5 26.3 25.0 23.3 15.1 15 47 52 53 54 59 60-75 61.1 51.2 48.9 44.6 33.8 I 60 14 15 16 16 18 China China 0-4 31.5 6.7 4.3 3.6 1.61 0 38 63 69 71 77 5-14 6.3 1.2 0.8 0.6 0.3 5 47 64 67 69 73 15-59 53.4 21.6 17.5 15.3 9.7 15 40 55 58 59 63 60-75 65.2 43.5 41.5 34.8 23.0 60 12 17 18 18 20 Other Asia and islands Asia and islands 0-4 23.4 12.0 9.7 8.2 3.8 0 44 58 62 64 72 5-14 4.9 1.9 1.7 1.3 0.6 5 52 61 63 65 70 15-59 45.0 24.5 21.2 19.7 13.9 15 44 51 54 56 60 60-75 59.3 47.8 44.9 41.2 31.4 60 13 15 16 17 19 Latin America and the Latin America and the Caribbean Caribbean 0-4 19.9 8.6 6.0 5.1 2.3 0 49 65 70 71 76 5-14 3.7 1.5 1.2 0.9 0.3 5 56 66 69 70 73 15-59 36.6 19.1 19.1 13.4 9.6 15 48 57 60 60 63 60-75 51.5 37.9 33.0 30.7 23.4 60 15 17 19 19 20 Middle Eastern crescent Middle Eastern crescent 0-4 5-14 30.1 6.8 15.0 2.5 11.1 1.9 9.4 1.5 0.7 r 0 5 38 49 56 60 61 63 63 65 71 69 15-59 50.1 26.5 20.1 19.3 14.0 15 42 52 55 56 60 60-75 59.2 46.7 42.0 39.4 31.0 60 13 15 17 17 19 Formerly socialist economies of Formerly socialist economies of Europe (FSE) Europe (FSE) 0-4 12.8 3.0 2.2 1.9 0.9 59 71 72 74 78 5-14 2.4 0.5 0.4 0.3 0.1 5 63 68 69 70 74 15-59 22.3 20.2 19.7 13.3 8.9 15 54 58 59 60 64 60-75 54.8 38.9 36.3 32.8 22.1 60 16 18 18 19 21 Established market economies Established market economies (EME) (EME) 0-4 6.0 1.7 1.1 1.0 0.6 0 65 74 76 77 81 5-14 1.0 0.3 0.3 0.2 0.1 5 65 71 72 73 77 15-59 20.7 11.4 10.7 9.3 6.4 15 55 61 62 63 67 60-75 43.1 31.1 27.6 24.9 16.5 60 17 19 20 20 22 FSE and EME FSE and EME 0-4 8.4 2.1 1.5 1.3 0.7 0 64 73 75 76 81 5-14 1.5 0.4 0.3 0.2 0.1 5 65 70 71 72 76 15-59 60-75 Demographically developing group 0-4 21.2 47.0 28.6 15.0 33.7 . 13.5 13.7 30.4 10.6 - 10.6 27.5 9.5 7.2 18.3 4.9 15 60 Demographically developing 0 group 55 17 40 60 19 59 61 19 63 62 20 65 66 22 71 5-14 5.7 2.6 2.2 1.6 0.8 5 50 62 64 66 69 15-59 46.7 26.7 23.5 21.9 15.3 15 43 53 55 56 60 60-75 60.3 46.7 43.8 40.4 32.6 60 13 16 17 17 19 World World 0-4 24.8 12.0 9.6 8.7 4.5 0 48 62 65 67 73 5-14 4.7 2.1 1.9 1.5 0.8 5 55 64 66 67 70 15-59 40.6 24.7 20.7 19.7 11.7 15 47 55 56 57 61 60-75 56.0 40.1 40.1 38.8 31.0 14 17 17 17 19 203 Table A.6 Nutrition and health behavior Prevalence of Percentage of children affected by: anemia in pregnant Tobacco consumption per year Percentage of women, 1970s and Stunting, 1980-90 Wasting, 1980-90 (kilograms per capita in adult children fully 1980s (percentage (ages 24-59 (ages 12-23 population) breastfed, 1985-90 below the norm for Region and economy months)a months)a (ages 0-3 months)a hemoglobin) a 1974-76 1990 2000 Sub-Saharan Africa 39w lOw 63w 41 w Mozambique 58 0.5 0.4 0.4 Tanzania 80 0.8 0.6 0.6 Ethiopia 43 19 6 Uganda 45 2 76 Burundi 48 6 98 68 Chad 13 37 Madagascar 56 17 Sierra Leone 43 14 45 Malawi 61 8 49 0.4 6.4 Rwanda 34 1 Mali 24 11 82 65 Burkina Faso 28 11 24 Niger 38 23 47 Nigeria 43 9 61 43 o.4 o4 o.i Kenya 32 5 48 57 Benin 55 Central Africa Rep. 67 Ghana 30 81 64 Togo 29 6 60 47 Guinea Zimbabwe 56 Côte d'lvoire 20 17 34 1.3 1.0 1.0 Senegal 25 6 77 55 Cameroon 43 2 70 8 South Africa 53 10 28 2.3 1.4 1.1 Somalia 30 40 73 Zaire 27 3 4 42 0.8 Sudan 32 13 84 36 Zambia 59 10 72 34 Angola 29 India 65 27 88 0.8 0.8 0.9 China 41 8 25 1.6 2.6 2.9 Other Asia and islands 53w 11w 58w 1.6w 1.7w 1.8w Nepal 69 14 33 Cambodia Bangladesh 0.9 1.0 Lao PDR 44 20 62 Sri Lanka 27 13 79 62 . . . . Indonesia 67 9 45 74 1.0 1.4 1.6 Philippines 43 13 48 1.5 1.5 1.7 Papua New Guinea 47 10 Thailand 22 6 33 i.6 2.0 Malaysia 32 6 34 1.5 1.8 2.1 Korea, Rep. 18 2 2.4 2.9 3.5 Hong Kong 2.0 1.6 1.8 Singapore 16 18 6.3 3.4 3.2 Myanmar 50 11 58 3.0 3.0 3.1 Viet Nam 49 12 0.6 1.0 1.1 Korea, Dem. Peovle's Rev. 4.3 4.0 3.9 Latin America and the Caribbean 26w 5w 41w 35w 2.0w 1.6w 1.7w Nicaragua 22 0 Haiti 51 17 64 Honduras 34 2 Bolivia 38 2 75 36 Guatemala 57 13 Dominican Rep. 19 1 42 0.8 1.1 1.1 Ecuador 39 4 54 46 Peru 37 2 58 53 El Salvador 36 6 14 1.2 6.9 Colombia 23 1 40 24 1.9 2.0 2.3 Paraguay 17 0 63 1.9 0.9 0.9 Chile 10 1 20 1.3 0.9 1.0 Venezuela 7 4 29 1.8 1.5 1.7 Argentina 3.1 1.9 1.9 Uruguay 16 Brazil 29 6 34 i.i i. Mexico 22 6 44 41 1.4 1.0 1.1 Puerto Rico Cuba 1 4.8 4.5 5.0 204 Prevalence of anemia in pregnant Percentage of children affected by: Tobacco consumption per year Percentage of women, 1970s and (kilograms per capita in adult Stunting, 1980-90 Wasting, 1980-90 children fully 1980s (percentage population) (ages 24-59 (ages 12-23 breastfed, 1985-90 below the norm for Region and economy months)a monfhs)a (ages 0-3 mont hS)a hemoglobin)a 1974-76 1990 2000 Middle Eastern crescent Pakistan 50 9 25 57 1.4 1.7 1.9 Yemen, Rep. 15 15 Egypt 31 1 66 1.8 1.8 Morocco 25 4 66 46 1.3 1.7 1.9 Tajikistan Jordan 20 3 32 50 Syrian Arab Rep. 52 Uzbekistan Tunisia 18 3 60 38 Kyrgyzstan Georgia Azerbaijan Turkmenistan Turkey 74 2:. 2i 2.3 Algeria 13 4 42 1.8 1.9 2.1 Armenia Iran 55 23 28 o. Kazakhstan Saudi Arabia 24 1.5 20 21 Israel 25 2.1 2.4 2.4 Afghanistan Iraq 3.0 29 Libya Formerly socialist economies of Europe (FSE) Romania 2.0 2.0 2.0 Poland 16 3.4 3.5 3.7 Bulgaria 3.6 4.1 4.3 Moldova Ukraine Czechoslovakiab 23 20 2.5 2.6 Lithuania Hungary 29 16 Belarus Russian Federation Albania Yugoslavia' 4 2.5 2.5 2.8 Established market economies (EME) 15w 3.2w 2.4w 2.0w Portugal 1.3 1.9 2.0 Greece 24 3.2 3.0 3.3 Ireland 3.2 2.4 2.1 New Zealand 3 1 22 3.2 2.1 1.9 Spain 9 2.5 2.4 2.6 United Kingdom 2 19 2.6 1.9 1.6 Australia 8 2.9 2.0 1.7 Italy 2 1 10 2.2 1.9 2.0 Netherlands 18 3.8 3.0 2.7 Belgium 6 3.5 2.9 2.7 Austria 2.3 2.1 1.9 France 6 0 18 2.8 2.3 2.1 Canada 5 1 3.8 2.6 2.2 United States 2 2 17 3.8 2.6 2.2 Germany 12 3.2 2.3 2.1 Denmark 3.5 2.6 2.3 Finland 2.2 1.6 1.4 Norway 2.3 2.0 1.9 Sweden 1.9 1.5 1.3 Japan 4 3.5 2.4 1.9 Switzerland 3.7 2.9 2.3 FSE and EME 4w 3w 4w 2.9w 2.2w 1.8w Demographically developing group 46w 13w 47w 49w 1.4w 1.7w 1.9w World 42w 12w 42w 1.7w 1.9w 1.9w Each value refers to one particular but not specified year within the time period denoted. Refers to former Czechoslovakia because disaggregated data are not yet available. Refers to former Socialist Federal Republic of Yugoslavia because disaggregated data are not yet available. 205 Table A.7 Mortality, by broad cause, and tuberculosis incidence Mortality rates by major cause of death, 1985-90 (deaths per 100,000 population, standardized for age) Annual incidence rate Communicable diseases of tuberculosis, 1990 and maternal and Noncommunicable (per 100,000 Demographic region and economy perinatal causes diseases Injuries population) Sub-Saharan Africa . 220 w Mozambique 189 Tanzania 140 Ethiopia 155 Uganda 300 Burundi 367 Chad 167 Madagascar 310 Sierra Leone 167 Malawi 173 Rwanda 260 Mali 289 Burkina Faso 289 Niger 144 Nigeria 222 Kenya 140 Benin 135 Central Africa Rep. 139 Ghana 222 Togo 244 Guinea 166 Zimbabwe 207 Côte d'Ivoire 196 Senegal 166 Cameroon 194 South Africa 250 Somalia 222 Zaire 333 Sudan 211 Zambia 345 Angola 225 India 470 761 97 220 China 117 6% 88 166 Other Asia and islands 201 w Nepal 167 Cambodia 235 Bangladesh 220 Lao PDR . . . 235 Sri Lanka 232 459 194 167 Indonesia . 220 Philippines 280 Papua New Guinea 275 Thailand . 173 Malaysia . 67 Korea, Rep. 113 454 194 162 Hong Kong 71 354 28 140 Singapore 114 498 39 82 Myanmar 189 Viet Nam . . 166 Korea, Dem. People's Rep. . . . 162 Latin America and the Caribbean 193 w 494 w 95 w 92 w Nicaragua 110 Haiti . . 333 Honduras . 133 Bolivia .. . . . . 335 Guatemala 595 523 113 110 Dominican Rep. 206 443 88 110 Ecuador 210 448 119 166 Peru 327 392 53 250 El Salvador 202 385 201 110 Colombia 67 Paraguay . . . . . 166 Chile 131 444 88 67 Venezuela 151 449 110 44 Argentina 107 530 59 50 Uruguay 98 519 67 15 Brazil . . . . . . 56 Mexico 168 490 102 110 Puerto Rico 78 447 59 8 Cuba 73 472 82 10 206 Mortality rates by major cause of death, 1985-90 (deaths per 100,000 population, standardized for age) Annual incidence rate Communicable diseases of tuberculosis, 1990 and maternal and Noncommunicable (per 100,000 Demographic region and economy perinatal causes diseases Injuries population) Middle Eastern crescent ll6w 619w 72w 99w Pakistan 150 Yemen, Rep. 96 Egypt 78 Morocco . . . . . . 125 Tajikistan 182 558 53 133 Jordan 14 Syrian Arab Rep. . . . . . . 58 Uzbekistan 137 601 65 55 Tunisia . . . . . . 55 Kyrgyzstan 124 651 95 68 Georgia 69 591 56 36 Azerbaijan 110 595 46 47 Turkmenistan 216 737 68 72 Turkey 57 Algeria . . . . . . 53 Armenia 60 580 66 127 Iran . . . . . . 83 Kazakhstan 86 700 103 77 Saudi Arabia . . . . . . 22 Israel 64 444 53 12 Afghanistan 278 Iraq 111 Libya 12 Formerly socialist economies of Europe (FSE) 52w 658w 94w 52w Romania 93 685 65 70 Poland 73 603 80 43 Bulgaria 73 619 64 30 Moldova 54 704 104 54 Ukraine 32 673 93 50 Czechoslovakiab 51 646 62 22 Lithuania 25 598 107 82 Hungary 55 690 90 38 Belarus 28 625 90 50 Russian Federation 47 704 115 56 Albania . 40 Yugoslaviac 87 559 68 30 Established market economies (EME) 47 w 416 w 49 w 20w Portugal 70 429 78 57 Greece 51 393 48 12 Ireland 57 526 39 18 New Zealand 50 487 58 10 Spain 45 410 42 49 United Kingdom 49 478 31 10 Australia 31 424 48 6 Italy 38 425 39 25 Netherlands 40 416 36 9 Belgium 52 459 68 16 Austria 30 437 55 20 France 40 362 70 16 Canada 39 395 48 8 United States 54 447 58 10 Germany 35 468 45 18 Denmark . . . . . . 7 finland 43 450 76 15 Norway 52 399 53 8 Sweden 41 397 46 7 Japan 51 306 41 42 Switzerland 0 . . 18 FSE and EME 49 w 488 w 63 w 29 w Demographically developing group 253 w 692 w 94w 173 w World 187 w 626 w 84 w 142 w Each value refers to one particular but not specified year within the time period denoted. Refers to former Czechoslovakia because disaggregated data are not yet available. Refers to former Socialist Federal Republic of Yugoslavia because disaggregated data are not yet available. 207 Table A.8 Health infrastructure and services Percentage of children immunized, age less than 1 year Doctors per Hospital beds 1,000 Nurse-to- per 1,000 Third dose of Demographic region and population doctor ratio, population, DPT Measles economy 1988_92a 1988_92a 1985_90a l990_91a 1990_91a Sub-Saharan Africa 0.12 w 5.1 w 1.4 w 52 w 52 w Mozambique 0.02 13.1 0.9 19 23 Tanzania 0.03 7.3 1.1 79 75 Ethiopia 0.03 2.4 0.3 44 37 Uganda 0.04 8.4 0.8 77 74 Burundi 0.06 4.3 1.3 83 75 Chad 0.03 0.9 . . 18 28 Madagascar 0.12 3.5 0.9 46 33 Sierra Leone 0.07 5.0 1.0 75 74 Malawi 0.02 2.8 16 81 78 Rwanda 0.02 1.7 1.7 89 89 Mali 0.05 2.5 . . 35 40 Burkina Faso 0.03 8.2 0.3 37 42 Niger 0.03 11.3 . . 18 24 Nigeria 0.15 6.0 1.4 65 70 Kenya 0.14 3.2 1.7 36 36 Benin 0.07 5.8 . 67 70 Central Africa Rep. 0.04 4.5 0.9 82 82 Ghana 0.04 9.1 1.5 39 39 Togo 0.08 6.2 1.6 73 61 Guinea 0.02 4.3 0.6 41 39 Zimbabwe 0.16 6.1 2.1 89 87 Côte d'Ivoire 0.06 4.8 0.8 48 42 Senegal 0.05 2.6 0.8 60 59 Cameroon 0.08 6.4 2.7 56 56 South Africa 0.61 4.5 4.1 67 63 Somalia 0.07 7.1 0.8 18 30 Zaire 0.07 2.1 1.6 32 31 Sudan 0.09 2.7 0.9 63 58 Zambia 0.09 6.0 . . 79 76 Angola 0.07 16.4 1.2 26 39 India 0.41 1.1 0.7 83 77 China 1.37 0.5 2.6 95 96 Other Asia and islands 0.31 w 3.0 w 1.8 w 81 w 78 w Nepal 0.06 2.7 0.3 74 63 Cambodia 0.04 8.0 2.2 38 38 Bangladesh 0.15 0.8 0.3 87 83 LaoPDR 0.23 5.9 2.5 22 47 Sri Lanka 0.14 5.1 2.8 86 79 Indonesia 0i4 2.8 0.7 86 80 Philippines 0.12 3.1 1.3 88 85 PapuaNew Guinea 0.08 8.1 3.4 64 63 Thailand 0.20 5.5 1.6 69 60 Malaysia 0.37 3.9 2.4 90 79 Korea, Rep. 0.73 1.0 3.0 74 93 Hong Kong 0.93 4.5 4.2 83 42 Singapore 1.09 3.8 3.3 91 92 Myanmar 0.08 4.0 0.6 69 73 VietNam 0.35 4.9 3.3 85 85 Korea, Dem. People's Rep. 2.72 13.5 90 96 Latin America and the Caribbean 1.25w 0.5w 2.7w 71w 75w Nicaragua 0.60 0.5 1.8 71 54 Haiti 0.14 0.8 0.8 41 31 Honduras 0.32 1.0 1.1 94 86 Bolivia 0.48 0.7 1.3 58 73 Guatemala 0.44 2.5 1.7 63 48 Dominican Rep. 1.08 0.7 2.0 47 69 Ecuador 1.04 0.3 1.7 89 54 Peru 1.03 0.9 1.5 71 59 El Salvador 0.64 1.5 1.5 60 53 Colombia 0.87 0.6 1.5 84 75 Paraguay 0.62 1.7 1.0 79 74 Chile 0.46 0.8 3.3 91 93 Venezuela 1.55 0.5 2.9 54 54 Argentina 2.99 0.2 4.8 84 99 Uruguay 2.90 0.2 4.6 88 82 Brazil 1.46 0.1 3.5 75 83 Mexico 0.54 0.8 1.3 64 78 Puerto Rico 2.55 . . 4.0 0 . . Cuba 3.75 1.7 5.0 99 99 208 Percentage of children immunized, age less than 1 year Doctors per Hospital beds 1,000 Nurse-to- per 1,000 Third dose of Demographic region and population doctor ratio, population, DPT Measles economy 1988_92a 1988_92a 1985_90a 1990_91a 1990_91a Middle Eastern crescent 1.04w 1.5w 2.9w 75w 74w Pakistan 0.34 0.8 0.6 81 77 Yemen, Rep. 0.18 2.9 0.9 62 57 Egypt 0.77 1.2 1.9 86 89 Morocco 0.21 4.5 1.2 79 76 Tajikistan 2.71 2.8 10.6 89 89 Jordan 1.54 0.3 1.9 92 85 Syrian Arab Rep. 0.85 1.2 1.1 89 84 Uzbekistan 3.58 2.9 12.4 57 81 Tunisia 0.53 2.7 2.0 90 80 Kyrgyzstan 3.67 2.8 12.0 78 94 Georgia 5.92 2.2 11.1 65 74 Azerbaijan 3.93 2.4 10.2 89 91 Turkmenistan 3.57 2.8 11.3 78 68 Turkey 0.74 1.5 2.1 72 66 Algeria 0.26 4.7 2.6 89 83 Armenia 4.28 2.5 9.0 88 92 Iran 0.32 1.1 1.5 88 84 Kazakhstan 4.12 3.0 13.6 84 94 Saudi Arabia 1.52 1.5 2.7 94 90 Israel 2.90 2.3 6.3 88 88 Afghanistan 0.11 0.8 0.3 . Iraq 0.58 1.2 1.6 69 73 Libia 1.04 2.9 4.1 62 59 Formerly socialist economies of Europe (FSE) 4.07w 2.2w 11.4w 77w 86w Romania 1.79 8.9 97 92 Poland 2.06 . . 6.6 98 94 Bulgaria 3.19 2.1 9.8 99 97 Moldova 4.00 3.0 7.8 87 95 Ukraine 4.40 2.7 13.6 78 88 Czechoslovakia' 3.23 2.4 7.9 99 98 Lithuania . . . . . . 80 92 Hungary 2.98 1.1 10.1 100 100 Belarus 4.05 13.2 90 97 Russian Federation 4.69 . 13.8 65 83 Albania 1.39 2.5 4.1 94 87 Yugoslaviac 2.63 1.9 6.0 79 75 Established market economies (EME) 2.52 w 2.1 w 8.3 w 80 w 77 w Portugal 2.57 0.8 4.2 95 96 Greece 1.73 1.6 5.1 54 76 Ireland 1.58 4.7 3.9 65 78 New Zealand 1.74 0.1 6.6 81 82 Spain 3.60 1.1 4.8 73 84 United Kingdom 1.40 2.0 6.3 85 89 Australia 2.29 3.8 5.6 90 68 Italy 4.69 0.6 7.5 95 50 Netherlands 2.43 3.4 5.9 97 94 Belgium 3.21 0.1 8.3 94 75 Austria 4.34 2.4 10.8 90 60 France 2.89 1.6 9.3 95 69 Canada 2.22 4.7 16.1 85 85 United States 2.38 2.8 5.3 67 80 Germany 2.73 1.7 8.7 80 90 Denmark 2.56 5.6 5.7 95 86 Finland 2.47 4.3 10.8 95 97 Norway 2.43 4.4 4.8 89 90 Sweden 2.73 3.4 6.2 99 95 Japan 1.64 1.8 15.9 87 66 Switzerland 1.59 2.6 11.0 90 90 FSE and EME 3.09w 2.1 w 9.3 w 79 w 80 w Demographically developing group 0.78w 0.9w 2.0w 80w 79w World 1.34w 1.4 w 3.6 w 80 w 79w Note: Regional totals and averages include relevant information for less populous countries, as listed in Table A. 10, except for the indicator "percentage of children immunized." Each value refers to one particular but not specified year within the time period denoted. Refers to former Czechoslovakia because disaggregated data are not yet available. Refers to former Socialist Federal Republic of Yugoslavia because disaggregated data are not yet available. 209 Table A.9 Health expenditure and total flows from external assistance - - - Development assistance for health Total health expenditure Health expenditures as a (official exchange rate percentage of GDP Aid flows as a Total aid flows percentage of dollars) Public Private in dollars, Aid flows total health Millions, Per capita, Total, sector, sector, 1990 per ca pita, expenditure, Demographic region and economy 1990 1990 1990 1990 1990 (millions)a 1990 1990 Sub-Saharan Africa 12,080 t 24 w 4.5 w 2.5 w 2.0 w 1,251 t 2.5 w 10.4 w Mozambique 85 5 5.9 4.4 1.5 45 2.9 52.9 Tanzania 109 4 4.7 3.2 1.5 53 2.1 48.3 Ethiopia 229 4 3.8 2.3 1.5 43 0.8 18.8 Uganda 95 6 3.4 1.6 1.8 46 2.8 48.4 Burundi 36 7 3.3 1.7 1.6 15 2.8 42.7 Chad 76 13 6.3 4.7 1.6 33 5.8 43.0 Madagascar 79 7 2.6 1.3 1.3 17 1.5 21.5 Sierra Leone 22 5 2.4 1.7 0.8 7 1.7 33.0 Malawi 93 11 5.0 2.9 2.1 22 2.5 23.3 Rwanda 74 10 3.5 1.9 1.6 29 4.1 39.5 Mali 130 15 5.2 2.8 2.4 36 4.3 27.7 BurkinaFaso 219 24 8.5 7.0 1.5 42 4.7 19.4 Niger 126 16 5.0 3.4 1.6 43 5.6 34.0 Nigeria 906 9 2.7 1.2 1.6 58 0.6 6.4 Kenya 375 16 4.3 2.7 1.6 84 3.5 22.3 Benin 79 17 4.3 2.8 1.6 33 7.0 41.8 Central Africa Rep. 55 18 4.2 2.6 1.6 20 6.5 35.8 Ghana 204 14 3.5 1.7 1.8 29 1.9 14.2 Togo 67 18 4.1 2.5 1.6 14 3.9 21.0 Guinea 106 19 3.9 2.3 1.6 20 3.5 23.8 Zimbabwe 416 42 6.2 3.2 3.0 42 4.2 10.0 Côte d'Ivoire 332 28 3.3 1.7 1.6 11 0.9 3.4 Senegal 214 29 3.7 2.3 1.4 36 4.9 16.9 Cameroon 286 24 2.6 1.0 1.6 38 3.3 13.4 South Africa 5,671 158 5.6 3.2 2.4 2 . Somalia 60 8 1.5 0.9 0.6 27 3.5 45.6 Zaire 179 5 2.4 0.8 1.5 48 1.3 26.7 Sudan 300 12 3.3 0.5 2.8 39 1.5 13.0 Zambia 117 14 3.2 2.2 1.0 6 0.7 4.9 Angola . . 28 2.8 India 17,740 21 6.0 1.3 4.7 286 0.3 1.6 China 12,969 11 3.5 2.1 1.4 77 0.1 0.6 Other Asia and islands 41,752 t 61 w 4.5 w 1.8 w 2.7 w 594 0.9w 1.4w Nepal 141 7 4.5 2.2 2.3 33 1.8 23.6 Cambodia . . . 0 . . . . . . . Bangladesh 715 7 3.2 1.4 1.8 128 1.2 17.9 LaoPDR 22 5 2.5 1.0 1.5 5 1.2 22.7 Sri Lanka 305 18 3.7 1.8 1.9 26 1.5 7.4 Indonesia 2,148 12 2.0 0.7 1.3 159 0.9 7.4 Philippines 883 14 2.0 1.0 1.0 69 1.1 7.8 Papua New Guinea 142 36 4.4 2.8 1.6 7 1.8 4.9 Thailand 4,061 73 5.0 1.1 3.9 36 0.7 0.9 Malaysia 1,259 67 3.0 1.3 1.7 3 0.1 0.2 Korea, Rep. 16,130 377 6.6 2.7 3.9 32 0.2 HongKong 4,060 699 5.7 1.1 4.6 . . . Singapore 658 219 1.9 1.1 0.8 1 0.2 0.1 Myanmar . . . . . . . . . . 12 0.3 Viet Nam 157 2 2.1 1.1 1.0 25 0.4 15.9 Korea, Dem. People's Rep Latin America and the Caribbean 46,660 105 w 4.0 w 2.4 w 1.6 w 591 1.3w 1.3w Nicaragua 133 35 8.6 6.7 1.9 27 6.6 20.0 Haiti 193 30 7.0 3.2 3.8 33 5.1 17.0 Honduras 134 26 4.5 2.9 1.6 20 4.0 15.1 Bolivia 181 25 4.0 2.4 1.6 37 5.1 20.3 Guatemala 283 31 3.7 2.1 1.6 32 3.4 11.1 Dominican Rep. 263 37 3.7 2.1 1.6 11 1.5 4.1 Ecuador 441 43 4.1 2.6 1.6 31 3.0 7.0 Peru 1,065 49 3.2 1.9 1.3 29 1.4 2.7 El Salvador 317 61 5.9 2.6 3.3 44 8.5 13.9 Colombia 1,604 50 4.0 1.8 2.2 26 0.8 1.6 Paraguay 160 37 2.8 1.2 1.6 10 2.4 6.4 Chile 1,315 100 4.7 3.4 1.4 10 0.7 0.7 Venezuela 1,747 89 3.6 2.0 1.6 2 0.1 0.1 Argentina 4,441 138 4.2 2.5 1.7 11 0.3 0.2 Uruguay 383 124 4.6 2.5 2.1 5 1.7 1.4 Brazil 19,871 132 4.2 2.8 1.4 84 0.6 0.4 Mexico 7,648 89 3.2 1.6 1.6 65 0.8 0.9 Puerto Rico Cuba 0.3 210 Development assistance for health Total health expenditure Health expenditures as a (official exchange rate percentage of GDP Aid flows as a dollars) Total aid flows percentage of Public Private in dollars, Aid flows total health Millions, Per capita, Total, sector, sector, 1990 per capita, expenditure, Demographic region and economy 1990 1990 1990 1990 1990 (mihions)a 1990 1990 Middle Eastern crescent 38,%1 t 77 w 4.1 w 2.4 w 1.7 w 453 t 0.9 w 1.2w Pakistan 1,394 12 3.4 1.8 1.6 76 0.7 5.4 Yemen, Rep. 217 19 3.2 1.5 1.7 25 2.2 11.6 Egypt 921 18 2.6 1.0 1.6 111 2.1 12.1 Morocco 661 26 2.6 0.9 1.6 20 0.8 3.0 Tajikistan 532 100 6.0 4.4 1.6 . . . Jordan 149 48 3.8 1.8 2.0 18 5.9 12.4 Syrian Arab Rep. 283 23 2.1 0.4 1.6 20 1.6 7.1 Uzbekistan 2,388 116 5.9 4.3 1.6 . . . Tunisia 614 76 4.9 3.3 1.6 18 2.3 3.0 Kyrgyzstan 517 118 5.0 3.3 1.6 Georgia 830 152 4.5 2.8 1.7 Azerbaijan 785 98 4.3 2.6 1.7 . . . Turkmenistan 459 125 5.0 3.3 1.7 2 0.5 0.4 Turkey 4,281 76 4.0 1.5 2.5 23 0.4 0.5 Algeria 4,159 166 7.0 5.4 1.6 2 0.1 0.1 Armenia 506 152 4.2 2.5 1.7 Iran 3,024 54 2.6 1.5 1.1 2 Kazakhstan 2,572 154 4.4 2.8 1.7 . Saudi Arabia 4,784 322 4.8 3.1 1.7 1 0.1 Israel 2,301 494 4.2 2.1 2.1 3 0.6 0.1 Afghanistan 53 2.6 Iraq 4 0.2 Libya . Formerly socialist economies of Europe (FSE) 49,143 t 142 w 3.6 w 2.5 w 1.0 w Romania 1,455 63 3.9 2.4 1.5 Poland 3,157 83 5.1 4.1 1.0 Bulgaria 1,154 131 5.4 4.4 1.0 Moldova 623 143 3.9 2.9 1.0 Ukraine 6,803 131 3.3 2.3 1.0 Czechoslovakia1' 2,711 173 5.9 5.0 0.9 Lithuania 594 159 3.6 2.6 1.0 Hungary 1,958 185 6.0 5.0 0.9 Belarus 1,613 157 3.2 2.2 1.0 Russian Federation 23,527 157 3.0 2.0 1.0 Albania 84 26 4.0 3.4 0.6 Yugoslavia' 4,512 205 3.0 4.0 1.0 Established market economies (EME) 1,483,1% t 1,860 w 9.2 w 5.6 w 3.5 w Portugal 3,970 383 7.0 4.3 2.7 Greece 3,609 358 5.5 4.2 1.3 Ireland 3,068 876 7.1 5.8 1.4 New Zealand 3,150 925 7.2 5.9 1.3 Spain 32,375 831 6.6 5.2 1.4 United Kingdom 59,623 1,039 6.1 5.2 0.9 Australia 22,736 1,331 7.7 5.4 2.3 Italy 82,214 1,426 7.5 5.8 1.7 Netherlands 22,423 1,500 7.9 5.7 2.2 Belgium 14,428 1,449 7.5 6.2 1.3 Austria 13,193 1,711 8.3 5.5 2.8 France 105,467 1,869 8.9 6.6 2.3 Canada 51,594 1,945 9.1 6.8 2.4 United States 690,667 2,763 12.7 5.6 7.0 Germany 120,072 1,511 8.0 5.8 2.2 Denmark 8,160 1,588 6.3 5.3 1.0 Finland 10,200 2,046 7.4 6.2 1.2 Norway 7,782 1,835 7.4 7.0 0.3 Sweden 20,055 2,343 8.8 7.9 0.9 Japan 189,930 1,538 6.5 4.8 1.6 Switzerland 16,916 2,520 7.5 5.1 2.4 FSE and EME 1,532,340 t 1,340 w 8.7 w 5.4 w 3.4 w Demographically developing group 170,115 t 41 w 4.7w 2.3 w 2.5 w 3,252 0.8 w 1.9 w World 1,702,455 t 323 w 8.0 w 4.9 w 3.2 w Note: Regional totals and averages include relevant information for less populous countries, as listed in Table A.10. Aid flows are official development assistance and include only a small portion of private flows, that is NGO assistance. Refers to former Czechoslovakia because disaggregated data are not yet available. Refers to former Socialist Federal Republic of Yugoslavia because disaggregated data are not yet available. 211 Table A.10 Economies and populations by demographic region, mid-1990 (population in thousands) Sub-Saharan Africa (49 economies) 510,271 Latin America and the Caribbean (continued) Nigeria 96,203 Burkina Faso 9,016 Less than 3 million population Ethiopia 51,180 Malawi 8,507 Zaire 37,320 Mali 8,460 Costa Rica 2,807 St. Vincent 107 South Africa 35,919 Zambia 8,111 Jamaica 2,420 French Guiana 92 Sudan 25,188 Somalia 7,805 Panama 2,418 Grenada 91 Tanzania 24,517 Niger 7,666 Trinidad and Tobago 1,236 Antigua and Barbuda 79 Kenya 24,160 Senegal 7,404 Guyana 798 Dominica 72 Uganda 16,330 Rwanda 7,118 Suriname 447 Aruba 66 Mozambique 15,707 Guinea 5,717 Guadeloupe 387 St. Kitts and Nevis 40 Ghana 14,870 Chad 5,680 Martinique 360 Cayman Islands 24 Côte d'Ivoire 11,902 Burundi 5,427 Barbados 257 British Virgin Islands 13 Cameroon 11,739 Benin 4,740 Bahamas, The 255 Montserrat 12 Madagascar 11,673 Sierra Leone 4,136 Netherlands Antilles 189 Turks and Caicos Islands 9 Angola 10,012 logo 3,638 Belize 188 Anguilla 8 Zimbabwe 9,805 Central Africa Republic 3,035 St. Lucia 150 Falkland/Malvinas Islands 2 Virgin Islands (U. S.) 110 Less than 3 million population Liberia 2,561 Comoros 475 Middle Eastern crescent (34 economies) 503,075 Congo 2,276 Djibouti 427 Mauritania 1,969 Equatorial Guinea 417 Pakistan 112,351 Yemen, Rep. 11,282 Namibia 1,780 Cape Verde 371 Turkey 56,098 Tunisia 8,060 Lesotho 1,768 São Tome and Principe 117 Iran 55,779 Azerbaijan 7,149 Botswana 1,254 Mayotte 73 Egypt 52,061 Georgia 5,462 Gabon 1,136 St. Helena Morocco 25,091 Tajikistan 5,302 Guinea-Bissau 980 Ascencion Algeria 25,056 Israel 4,659 Gambia, The 875 Tristan da Cunha 0.3 Uzbekistan 20,532 Libya 4,546 Swaziland 797 Afghanistan 20,445 Kyrgyzstan 4,395 Iraq 18,914 Turkmenistan 3,669 Kazakhstan 16,746 Armenia 3,325 India 849,515 Saudi Arabia [4,870 Jordan 3,098 Syrian Arab Rep. 12,360 China 1,133,698 Less than 3 million population Other Asia and islands (49 economies) 682,533 Lebanon 2,681 Gaza Strip 588 Kuwait 2,143 Bahrain 503 Indonesia 178,232 Nepal 18,916 United Arab Emirates 1,592 Qatar 439 Bangladesh 106,656 Malaysia 17,857 Oman 1,554 Malta 354 Viet Nam 66,312 Sri Lanka 17,002 West Bank 1,088 Western Sahara 179 Philippines 61,480 Cambodia 8,469 Cyprus 702 Thailand 55,853 Hong Kong 5,806 Korea, Rep. 42,797 Lao PDR 4,186 Myanmar 41,609 Papua New Guinea 3,915 Demographically developing group (180 economies) 4,123,389 Korea, Dem. People's Rep. 21,576 Singapore 3,003 Taiwan, China 20,313 Formerly socialist economies of Europe (14 economies) 346,237 Less than 3 million population Russian Federation 148,255 Hungary 10,553 Ukraine 51,860 Belarus 10,278 Mongolia 2,124 Kiribati 70 Poland 38,180 Bulgaria 8,823 Bhutan 1,433 Seychelles 68 Romania 23,199 Moldova 4,367 Mauritius 1,074 Marshall Islands 41 Yugoslavia 23,808 Lithuania 3,731 Fiji 744 American Samoa 39 Czechoslovakiab 15,662 Albania 3,250 Reunion 593 Northern Mariana Islands 23 Macao 459 Cook Islands 19 Less than 3 million population Solomon Islands 316 Trust Territory of the Pacific 18 Latvia 2,686 Estonia 1,583 Brunei 256 WaIlis and Futuna Islands 14 Maldives 214 Nauru 9 French Polynesia 197 Tuvalu 9 Established market economies (35 economies) 797,788 New Caledonia 165 Niue 3 United States 249,975 Greece 10,067 Western Samoa 165 Midway Island 2 Japan 123,519 Belgium 9,956 Vanuatu 151 Tokelau Island 2 Germany 79,484 Sweden 8,559 Guam 137 Wake Island 2 Italy 57,663 Austria 7,712 Fed. States of Micronesia 103 Johnston Island United Kingdom 57,395 Switzerland 6,712 Tonga 99 Pitcairn Island 0.1 France 56,440 Denmark 5,140 Spain 38,959 Finland 4,986 Canada 26,522 Norway 4,242 Latin America and the Caribbean (46 economies) 444,297 Australia 17,085 Ireland 3,503 Brazil 150,368 Bolivia 7,171 Netherlands 14,943 New Zealand 3,405 Mexico 86,154 Dominican Rep. 7,074 Portugal 10,354 Colombia 32,345 Haiti 6,472 Argentina 32,293 El Salvador 5,213 Less than 3 million population Peru 21,663 Honduras 5,105 Luxembourg 378 Andorra 47 Venezuela 19,738 Paraguay 4,314 Iceland 255 Gibraltar 30 Chile 13,173 Nicaragua 3,853 Channel Islands 144 Liechtenstein 28 Cuba 10,617 Puerto Rico 3,530 Isle of Man 66 Monaco 28 Ecuador 10,284 Uruguay 3,094 Bermuda 59 San Marino 23 Guatemala 9,197 (Continued in next column) Greenland 56 St. Pierre and Miquelon 6 Faeroe Islands 48 Holy See FSE and EME (49 economies) 1,144,025 World (229 economies) 5,267,414 Refers to former Socialist Federal Republic of Yugoslavia because disaggregated data are not yet available. Refers to former Czechoslovakia because disaggregated data are not yet available. 212 Appendix B. The global burden of disease, 1990 The World Bank and the World Health Organization age 40 is assumed to represent a stream of lost life have undertaken a joint exercise for this Report that that is equal to the female expectation of life at age 40, quantifies the impact in loss of healthy life from about or 43 years. Longer streams of life lost as a result of a 100 diseases and injuries in 1990 (Murray and Lopez, female death as compared with a male death at any background paper). The global burden of disease given age seem to be justified by data for high-income (GBD) combines the loss of life from premature death groups in low-mortality populations. These data in 1990 with the loss of healthy life from disability. show that women's expectations of life at birth are The GBD is measured in units of disability-adjusted still two to three years higher than males'. life years (DALYs) except in Table B.8, which presents Value of a healthy year of life lived at each age. Most only deaths. Disease and injury categories are based societies attach more importance to a year of life lived on the International Classification of Diseases, Ninth Re- by a young or middle-aged adult than to a year of life vision (1977). The criterion for selecting the diseases lived by a child or an elderly person. The relative and injuries studied was the expected magnitude of value of a year of life at each age has been modeled the burden within a specific age group. The selected for this exercise as an exponential function of the diseases and injuries account for more than 90 per- form 1w exp(-Ba), where a is age and B is equal to 0.04. cent of premature deaths and probably for a similar This function rises quickly from zero at birth to a peak proportion of the burden attributable to disability. at age 25 and then declines asymptotically toward The tables presented in this appendix include sub- zero. The constant k is chosen so that the total num- totals for different disease groups; the numbers pre- ber of DALYs is the same as though uniform age sented in the subtotals include DALYs lost as a result weights had been used (see Box figure 1.3). It is im- of the specified diseases and, in some cases, several portant to note that while the first year of life receives residual conditions. a very low weight, the life of a newborn is valued Calculation of the disease burden is based on sev- according to the weights of all the years he or she is eral assumptions, some of which involve decisions expected to live, that is, according to the sum of the about ethical values or social preferences. The key function over future years. In the absence of dis- choices are the potential years of life lost as a result of counting, therefore, the greatest loss of DALYs from a death at a given age; the relative value of a year of premature death occurs from infant deaths. The in- healthy life lived at different ages; the discount rate, troduction of discounting means that the greatest loss or extent of time preference for human life and from premature death occurs in early adulthood, but health; and the disability weights used to convert life that loss is only slightly greater than the loss from an lived with a disability to a common measure with infant death. Largely because loss of life is valued premature death. The choices are described below. A according to the future stream of age-specific age full presentation of the methodology will appear in weights and not just the weight for one year, the Murray and Lopez (forthcoming). results of the analysis are not very sensitive to the Duration of life lost due to a death at each age. The introduction of nonuniform age weights. number of years of life lost as a result of a death at Time preference. Since the stream of damage to each age is obtained from a standard schedule of ex- health from current illness and injury can extend pectations of life at that age. A Coale and Demeny years or even decades into the future, it must be de- (1983) "West" family model life table (level 26 with cided how to value the future relative to the present. an expectation of life of 82.5 years) has been used as This is a controversial and unresolved issue because the standard for females. A comparable model with there are two different arguments concerning how to an expectation of life of 80 years has been used as the regard the future. First, societies typically prefer to standard for males. For example, a female death at have a given amount of consumption today rather 213 than tomorrow. This "pure social rate of time prefer- with mortality, and it is easy to apply to nonfatal ence" is usually assumed to be quite low, of the order disabilities. Disability-adjusted life years attributable of 0-3 percent per year, meaning that the future is to premature mortality are calculated on the basis of valued exactly or almost as much as the present. Sec- 1990 deaths by cause, as presented in Table B.8. ond, there is a reason for discounting the future Estimates of mortality by cause were based on much more, if resources are not consumed today but three types of source: vital registration data, model- are invested so as to generate higher consumption based estimates, and epidemiological estimates for tomorrow. The expected rate of growth of consump- particular diseases. Vital registration data for all tion times the elasticity of utility with respect to con- countries with good registration systems that attrib- sumption yields a term typically assumed to be about ute a cause of death were used with only minor mod- 8-10 percent, comparable to the rate of return on in- ifications; such countries include most of the estab- vestments involving risk and taxation. In evaluating lished market economies, the formerly socialist the global burden of disease, it is assumed that this economies of Europe, and a large proportion of the argument does not apply to human lives, which are countries of Latin America and the Caribbean. For incommensurable with consumption; adding to China high-quality sample registration data from the healthy years does not necessarily raise consumption Disease Surveillance Points System were used with per person. This Report therefore uses a discount rate some modifications. Vital registration data were also of 3 percent per year, which could be entirely attrib- used for some countries in the Middle Eastern cres- uted to pure time preference. cent and Other Asia and islands regions. Higher discount rates would reduce the total bur- A variety of models relating cause-specific mortal- den of disease because future health damage from ity to total mortality by age have been developed on health losses in 1990 would count for less. More im- the basis of the patterns of causes of death recorded portant, higher discount rates would also alter the in nations with good registration systems. For the relative importance of different diseases. Because the large groups of causescommunicable plus maternal stream of life lost as a result of mortality is, on aver- and perinatal, noncommunicable, and injuriesthese age, longer than that caused by disability, higher dis- models have been used to allocate deaths to cause count rates raise the importance of disability com- groups. Whenever possible, these estimates have pared with that of premature mortality. For the same been validated by examining the results of small-scale reason, higher discount rates reduce the importance longitudinal population surveillance systems. Esti- of premature deaths at young ages in relation to those mates for more detailed causes were built up from at older ages. assessments by disease experts of incidence, remis- Disability weights. Disabilities were assigned sever- sion, and case-fatality rates. These epidemiological ity weights ranging from zero, representing perfect estimates by cause have been constructed so as to add health, to one, representing death. These disability up to total mortality. weights were determined at a meeting of experts in A different approach was used to estimate the international health who had not participated in the DALYs lost through life lived with a disability. A study. In order to reduce the number of weights to be group of experts estimated the incidence, age of on- assigned and to emphasize large differences in the set, and duration of disability for each specific disease severity of disability, each disability condition was on the basis of community-based epidemiological assigned to one of six classes of severity. The disabil- data, routine health facilities information, and, where ities in a particular class differ in kind (for example, necessary, expert judgment. Separate estimates were blindness versus paralysis) but were considered to be made for the five age groups, two sexes, and eight of equal severity. Each participant then voted on the regions in the study. When prevalence was used to weight to be assigned to the entire class, not to indi- estimate incidence, the GBD incidence prevalence vidual disabilities, and the class was weighted model was used to check for consistency between the according to the average vote. It is important to note estimated incidence, remission rate, case-fatality rate, that many disabling conditions lead to two or more and general background mortality rate. The same distinct disabilities, which may be classified in more model was also used to check estimated duration for than one class of severity. each disability. A completed first-round set of esti- The burden of disease could be computed using a mates was reviewed at a conference hosted by the prevalence perspective (the extent of burden during a World Health Organization, and comments were given year, no matter when a disease condition be- taken into account. A second round of estimates was gan) or an incidence perspective (the future burden of undertaken and was subjected to the same review. that year's new cases or incident diseases). An inci- The third round of estimates is presented here. A dence perspective was chosen: the burden of disease fourth and final round of revisions will appear in is the future stream of disability caused by incident Murray and Lopez (forthcoming). cases in 1990. This is the more logical way of dealing 214 Table B.1 Burden of disease by age and sex, 1990 Millions of disability-adjusted Percentage DALYs per 1,000 population life years (DALYs) lost of DALYs Male/female Demographic region and age group MAle Female Total lost Male Female Total ratio Sub-Saharan Africa 152.8 139.9 292.7 100.0 606 542 574 1.12 0-5 83.8 73.3 157.1 53.7 1,765 1,559 1,662 1.13 5-14 19.7 17.5 37.2 12.7 280 251 266 1.12 15-44 36.2 35.6 71.9 24.6 349 335 342 1.04 45-59 7.9 7.1 15.1 5.1 391 323 355 1.21 60+ 5.2 6.2 11.4 3.9 494 491 492 1.01 India 145.3 147.1 292.4 100.0 331 359 344 0.92 0-5 67.3 70.0 137.2 46.9 1,125 1,234 1,178 0.91 5-14 15.8 16.5 32.3 11.1 155 173 164 0.90 15-44 30.8 35.0 65.8 22.5 154 191 172 0.81 45-59 16.5 12.3 28.9 9.9 348 268 308 1.30 60+ 14.9 13,3 28.2 9:6 500 460 480 1.09 China 103.5 97.8 201.3 100.0 177 178 178 0.99 0-5 23.7 26.2 49.9 24.8 394 452 422 0.87 5-14 8.8 7.4 16.3 8.1 91 82 87 1.11 15-44 31.0 31.4 62.4 31.0 101 111 106 0.91 45-59 17.3 12.2 29.6 14.7 239 190 216 1.26 60+ 22.6 20.5 43.1 21.4 462 397 429 1.16 Other Asia and islands 95.1 81.6 176.7 100.0 277 240 259 1.15 0-5 36.8 30.4 67.1 38.0 840 724 783 1.16 5-14 15.0 11.7 26.7 15.1 179 146 163 1.22 15-44 24.3 22.4 46.8 26.5 151 140 146 1.08 45-59 10.1 8.1 18.2 10.3 296 230 263 1.29 60+ 8.9 9.0 17.9 10.1 441 397 418 1.11 Latin America and the Caribbean 57.2 45.7 102.9 iEio.o 258 205 232 1.26 0-5 18.1 14.4 32.5 31.5 629 520 575 1.21 5-14 6.5 5.5 12.0 11.6 125 108 116 1.16 15-44 20.8 15.6 36.4 35.4 199 150 175 1.33 45-59 6.4 4.9 11.3 10.9 287 209 247 1.37 60+ 5.5 5.3 10.8 10.5 383 316 347 1.21 Middle Eastern crescent 73.9 70.3 144.2 100.0 288 285 287 1.01 0-5 38.1 36.2 74.3 51.5 925 912 919 1.01 5-14 9.3 8.2 17.5 12.1 142 133 137 1.07 15-44 14.4 15.4 29.8 20.6 127 143 135 0.88 45-59 6.3 4.7 11.1 7.7 284 211 248 1.35 60+ 5.8 5.8 11.6 8.0 427 372 398 1.15 Formerly socialist economies of Europe (FSE) 33.2 25.0 58.2 100.0 201 138 168 1.45 0-5 3.0 2.4 5.4 9.2 216 181 199 - 1.19 5-14 1.0 1.0 2.0 3.4 37 37 37 1.00 15-44 11.3 6.3 17.6 30.2 148 84 116 1.77 45-59 9.1 5.0 14.1 24.3 337 168 248 2.00 60+ 8.8 10.3 19.1 32.9 420 284 334 1.48 Established market economies (EME) 52.0 41.6 93.6 100.0 133 102 117 1.30 0-5 3.5 2.9 6.4 6.8 132 115 124 1.15 5-14 1.3 0.9 2.2 2.4 24 19 21 1.30 15-44 16.1 11.4 27.4 29.3 87 64 76 1.37 45-59 11.6 7.2 18.7 20.0 175 106 140 1.65 60+ 19.6 19.2 38.8 41.5 324 227 267 1.43 FSE and EME 85.2 66.6 151.8 100.0 153 113 133 1.35 0-5 6.5 5.3 11.7 7.7 161 138 150 1.17 5-14 2.3 1.9 4.2 2.8 29 25 27 1.15 15-44 27.3 17.7 45.0 29.7 105 70 87 1.51 45-59 20.6 12.2 32.8 21.6 222 125 172 1.77 60+ 28.4 29.5 58.0 38.2 349 244 286 1.43 Demographically developing group 628.0 582.3 1,210.3 100.0 299 288 294 1.04 0-5 267.7 250.5 518.1 42.8 952 924 938 1.03 5-14 75.1 66.9 142.0 11.7 160 149 155 1.07 15-44 157.6 155.5 313.1 25.9 159 165 162 0.97 45-59 64.6 49.4 114.0 9.4 295 231 264 1.27 60+ 62.9 60.2 123.0 10.2 458 406 431 1.13 World 713.1 648.9 1,362.1 100.0 269 248 259 1.08 0-5 274.1 255.7 529.9 38.9 853 822 840 1.03 5-14 77.4 68.8 146.2 10.7 140 131 136 1.07 15-44 185.0 173.1 358.1 26.3 148 144 146 1.02 45-59 85.3 61.6 146.9 10.8 273 198 236 1.38 60+ 91.3 89.7 181.0 13.3 417 333 371 1.25 Note: In this appendix the demographically developing group includes the Sub-Saharan Africa, India, China, Other Asia and islands, Latin America and the Caribbean, and Middle Eastern crescent regions. 215 Table B .2 Burden of disease in females by cause, 1990 (hundreds of thousands of DALYs lost) Formerly Latin socialist Established Sub- Other America Middle economies market Saha ran Asia and and the Eastern of Europe economies FSE and Demographically Disease or injury Africa India China islands Caribbean crescent (FSE) (EME) EME developing group World Communicable, maternal, and perinatal 1,038.7 772.9 281.4 419.8 207.6 387.8 25.7 48.8 74.5 3,108.2 3,182.7 Infectiousandparasitic 716.7 409.9 133.0 228.8 120.6 189.6 8.5 22.8 31.4 1,769.6 1,830.0 Tuberculosis 62.1 45.2 24.4 37.7 10.6 18.8 0.5 0.5 1.0 198.8 199.8 STDs excluding H1V 45.8 32.0 33.3 12.6 21.6 6.3 5.0 15.4 20.5 151.7 172.2 Syphilis 23.9 3.4 0.0 0.1 1.6 0.0 0.0 0.0 0.0 29.0 29.1 Chlamydia 1.5 2.7 3.1 2.3 1.5 0.6 0.3 0.8 1.1 11.7 12.8 Gonorrhea 1.6 0.1 0.2 0.2 0.1 0.1 0.0 0.0 0.1 2.2 2.3 Pelvic inflammatory disease 18.7 25.8 30.0 10.1 18.3 5.7 4.8 14.5 19.3 108.7 128.0 Humanimmunodeficiencyvirus 89.9 13.6 0.0 4.9 10.2 0.6 0.2 3.4 3.6 119.1 122.8 Diarrheal diseases 146.2 143.9 21.7 68.8 27.6 78.9 1.1 1.2 2.2 487.2 489.4 Acute watery 78.8 78.9 14.1 37.6 15.9 41.7 1.0 1.0 1.9 267.1 269.0 Persistent 44.6 42.6 3.6 20.5 7.1 24.6 0.0 0.1 0.1 142.9 143.0 Dysentery 22.8 22.4 4.0 10.8 4.6 12.6 0.1 0.1 0.2 77.2 77.4 Childhood cluster 132.8 98.7 8.1 36.5 7.5 43.4 0.2 0.4 0.6 327.1 327.7 Pertussis 22.0 15.2 2.7 5.6 3.4 7.9 0.2 0.3 0.5 56.8 57.3 Polio 6.0 7.5 0.9 1.7 1.0 2.8 0.0 0.0 0.0 19.9 19.9 Diphtheria 0.2 0.6 0.0 0.2 0.1 0.1 0.0 0.0 0.0 1.2 1.2 Measles 77.5 48.5 1.5 19.3 1.9 19.9 0.0 0.0 0.1 168.5 168.5 Tetanus 27.2 27.0 3.0 9.8 1.1 12.7 0.0 0.0 0.0 80.7 80.7 Meningitis 6.4 8.2 2.8 4.1 3.3 5.3 0.6 0.5 1.0 30.1 31.1 Hepatitis 1.3 1.7 2.2 1.4 0.9 1.0 0.2 0.3 0.4 8.4 8.9 Malaria 154.1 4.7 0.0 12.5 2.2 1.5 0.0 0.0 0.0 175.0 175.0 Tropical cluster 25.8 7.5 2.3 1.0 13.4 1.0 0.0 0.0 0.0 51.0 51.0 Trypanosomiasis 8.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.8 8.8 Chagas' disease 0.0 0.0 0.0 0.0 12.6 0.0 0.0 0.0 0.0 12.6 12.6 Schistosomiasis 11.8 0.9 1.5 0.3 0.6 0.3 0.0 0.0 0.0 15.4 15.4 Leishmaniasis 2.0 5.0 0.4 0.4 0.1 0.6 0.0 0.0 0.0 8.6 8.6 Lymphatic filariasis 0.5 1.6 0.4 0.3 0.0 0.1 0.0 0.0 0.0 2.9 2.9 Onchocerciasis 2.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.7 2.7 Leprosy 1.1 2.6 0.0 0.8 0.3 0.2 0.0 0.0 0.0 5.1 5.1 Trachoma 6.9 2.0 3.6 7.0 0.7 3.6 0.0 0.0 0.0 23.7 23.7 Intestinal helminths 4.3 10.0 30.6 28.5 11.9 2.6 0.0 0.0 0.0 87.9 87.9 Ascaris 2.2 5.7 18.7 15.7 6.7 2.4 0.0 0.0 0.0 51.4 51.4 Trichuris 1.5 2.4 10.9 11.6 4.5 0.0 0.0 0.0 0.0 30.9 30.9 Hookworm 0.5 1.9 0.9 1.2 0.7 0.2 0.0 0.0 0.0 5.6 5.6 Respiratory infections 153.9 161.9 69.0 93.0 29.9 84.9 6.6 11.7 18.3 592.5 610.8 Lowerrespiratoryinfections 148.5 154.0 61.0 87.3 26.5 80.9 4.8 8.0 12.8 558.3 571.1 Upper respiratory infections 1.7 2.8 2.7 2.3 1.3 0.4 1.0 2.3 3.3 11.1 14.4 Otitis media 3.7 5.1 5.3 3.3 2.2 3.5 0.8 1.4 2.2 23.1 25.3 Maternal 79.9 78.2 25.0 43.6 18.0 42.1 4.8 5.5 10.4 286.8 297.2 Hemorrhage 14.3 13.7 6.4 7.5 2.5 4.7 0.5 0.9 1.4 49.1 50.4 Sepsis 27.5 27.5 6.3 15.8 4.1 17.5 1.5 2.0 3.5 98.7 102.2 Eclampsia 4.2 3.9 0.7 2.0 1.6 1.3 0.0 0.1 0.1 13.7 13,8 Hypertension 2.0 1.9 0.3 2.5 0.7 1.8 0.1 0.0 0.1 9.3 9.4 Obstructed labor 19.0 19.4 7.9 10.2 5.7 13.3 2.0 2.3 4.3 75.4 79.8 Abortion 7.9 9.5 0.9 2.6 2.2 2.0 0.4 0.1 0.6 25.1 25.6 Perinatal 88.2 122.9 54.4 54.5 39.1 71.2 5.7 8.8 14.5 430.3 444.7 Noncommunicable 280.5 578.9 558.1 349.8 212.0 255.2 203.5 334.8 538.3 2,234.5 2,772.8 Malignant neoplasms 22.3 53.6 72.0 36.7 27.8 22.3 36.0 79.5 115.5 234.7 350.1 Mouth and oropharynx 1.0 6.3 1.8 2.5 0.4 1.1 0.4 0.9 1.3 13.2 14.5 Esophagus 0.7 3.5 5.3 0.7 0.2 0.7 0.3 0.7 1.0 11.2 12.2 Stomach 1.6 2.4 11.1 2.1 1.4 1.5 3.9 4.3 8.2 20.0 28.2 Colon and rectum 0.8 1.9 4.4 1.9 1.3 1.1 3.5 9.4 12.9 11.5 24.4 Liver 2.1 0.6 9.0 1.8 0.2 0.6 0.3 0.8 1.1 14.3 15.4 Pancreas 0.4 0.5 1.1 0.3 0.3 0.3 0.4 2.6 3.0 2.8 5.8 Trachea, bronchus, and lung 0.5 0.9 6.0 2.0 0.7 1.0 2.2 8.3 10.5 11.1 21.6 Melanoma 0.5 0.1 0.1 0.1 0.3 0.1 0.5 1.2 1.6 1.2 2.8 Breast 2.5 6.1 4.0 4.0 4.7 3.1 5.2 15.7 20.8 24.3 45.1 Cervix 4.6 9.6 3.1 4.7 4.3 [.9 2.0 2.3 4.4 28.1 32.5 Uterus 0.4 0.5 0.9 0.4 0.8 0.4 0.6 2.3 2.8 3.4 6.2 Ovary 1.0 2.0 1.6 1.5 0.7 0.8 1.4 4.0 5.4 7.6 13.0 216 Formerly Latin socialist Established Sub- Other America Middle economies market Saha ran Asia and and the Eastern of Europe economies FSE and Demographically Disease or injury Africa India China islands Caribbean crescent (FSE) (EME) EME developing group World Noncommunicable diseases, malignant neoplasms (continued) Prostate 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Bladder 0.6 0.2 0.4 0.3 0.2 0.4 0.6 1.1 1.7 2.0 3.7 Lymphoma 1.3 1.4 1.2 1.1 1.1 0.7 1.0 3.5 4.6 6.8 11.4 Leukemia 0.6 2.3 7.1 1.6 1.1 1.3 1.3 2.6 3.9 14.0 18.0 Diabetes mellitus 2.1 10.3 4.1 6.9 5.8 6.3 2.3 7.1 9.5 35.5 45.0 Nutritional and endocrine 39.9 90.6 38.6 41.6 23.9 27.4 5.1 8.6 13.7 262.1 275.8 Protein-energy malnutrition 10.4 29.2 10.1 4.4 4.6 5.1 1.1 1.2 2.4 63.9 66.3 Iodine deficiency 8.5 6.9 4.9 6.4 2.6 6.7 0.0 0.0 0.0 35.8 35.9 Vitamin A deficiency 10.7 20.2 4.9 12.2 6.9 2.6 0.0 0.0 0.0 57.7 57.7 Anemia 6.0 25.0 16.6 13.9 5.9 8.9 3.0 3.9 6.9 76.2 83.0 Neuropsychiatric 41.3 83.6 78.5 57.4 35.7 38.9 29.3 62.0 91.3 335.3 426.6 Depressive disorders 11.2 20.2 31.9 17.7 11.8 11.9 7.9 14.3 22.1 104.8 127.0 Bipolar affective disorders 0.6 1.0 1.6 0.9 0.6 0.6 0.4 0.6 1.0 5.2 6.2 Psychoses 2.7 11.4 7.8 4.2 2.7 2.8 1.2 2.9 4.1 31.6 35.7 Epilepsy 5.5 9.5 6.9 7.0 3.5 5.1 2.6 2.8 5.4 37.4 42.9 Alcohol dependence 2.2 2.4 2.4 1.4 2.0 0.5 2.4 3.7 6.1 10.9 17.0 Alzheimers and other dementias 4.3 9.7 13.6 7.3 4.7 4.9 8.7 23.2 31.9 445 76.4 Parkinson's disease 0.3 0.7 1.0 0.6 0.6 0.3 0.8 1.9 2.7 3.4 6.2 Multiple sclerosis 0.7 1.3 1.7 1.1 0.5 0.7 0.6 1.1 1.7 5.9 7.6 Drug dependence 0.9 1.7 0.7 2.3 2.1 0.9 0.6 3.9 4.5 8.7 13.3 Posttraumatic Stress disorder 3.2 5.1 7.0 4.4 2.8 3.1 2.7 4.5 7.2 25.5 32.7 Sense organ 7.9 11.4 9.0 7.5 3.3 3.1 0.3 0.6 0.9 42.2 43.1 Glaucoma 1.7 1.7 3.4 2.9 0.4 0.1 0.1 0.4 0.5 10.2 10.7 Cataract 5.8 8.8 3.6 4.2 2.6 2.7 0.2 0.1 0.3 27.7 28.0 Cardiovascular 66.3 138.0 133.4 86.5 45.9 62.8 82.2 98.1 180.3 532.9 713.2 Rheumatic 6.1 12.7 12.2 3.9 1.8 3.1 1.8 0.9 2.7 39.8 42.6 lschemic heart disease 5.4 31.9 17.6 26.7 11.3 10.6 34.3 37.2 71.5 103.4 175.0 Cerebrovascular 25.3 35.0 58.1 21.8 14.0 18.5 29.7 26.6 56.3 172.6 229.0 Pen-, endo-, and myocarditis and cardiomyopathy 16.0 31.7 4.4 9.4 7.2 8.9 1.3 2.5 3.8 77.5 81.3 Respiratory 21.2 39.5 85.4 18.4 15.0 16.8 7.6 14.9 22.6 1%.3 218.9 Chronic obstructive pulmonary disease 3.0 7.3 50.9 3.7 2.9 3.0 3.1 5.7 8.7 70.7 79.4 Asthma 10.0 8.7 17.1 7.0 5.6 4.9 2.7 5.8 8.4 53.2 61.7 Digestive 25.4 55.3 37.8 24.6 12.8 21.9 8.5 14.7 23.2 177.8 201.0 Peptic ulcer disease 1.5 3.6 3.9 1.9 0.7 1.2 0.8 1.8 2.6 12.8 15.4 Cirrhosis 4.8 8.4 10.0 5.3 3.0 2.7 1.4 5.1 6.5 34.3 40.8 Genitourinary 10.1 20.3 12.8 11.4 6.6 8.2 3.8 6.2 10.0 69.5 79.5 Nephritis and nephrosis 5.6 10.6 11.1 6.9 4.5 4.5 1.9 4.3 6.2 43.2 49.4 Benign prostatic hypertrophy 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Musculoskeletal 4.6 8.4 40.0 17.0 13.0 7.2 11.3 24.5 35.8 90.2 126.1 Rheumatoid arthritis 1.1 1.1 5.2 1.6 4.6 3.6 4.1 11.3 15.4 17.3 32.7 Osteoarthritis 1.7 3.1 31.1 12.9 6.6 1.9 6.4 11.3 17.7 57.2 74.9 Congenital abnormalities 29.6 45.9 35.4 25.5 14.9 27.1 7.3 11.2 18.6 178.4 197.0 Oral health 5.2 14.1 8.9 11.7 5.9 10.2 9.0 5.7 14.6 56.0 70.6 Dental caries 0.7 1.6 1.1 1.3 2.2 1.6 0.8 0.4 1.2 8.5 9.6 Periodontal disease 4.2 10.6 4.9 5.2 2.9 2.9 0.4 0.5 0.9 30.8 31.7 Edentulism 0.2 1.9 2.9 5.2 0.8 5.6 7.8 4.8 12.6 16.7 29.3 Injuries 79.6 119.4 138.5 46.2 37.2 59.9 21.1 32.1 53.1 480.8 533.9 Unintentional 42.6 104.8 84.7 34.7 31.1 37.0 15.3 22.9 38.2 335.1 373.3 Motor vehicle 8.1 9.4 12.3 8.5 17.9 16.4 4.5 9.2 13.7 72.6 86.3 Poisoning 1.4 0.8 6.6 1.3 0.3 1.0 1.8 0.7 2.5 11.4 13.9 Falls 9.5 21.1 19.2 7.3 2.7 4.6 2.8 7.4 10.2 64.4 74.6 Fires 4.2 8.5 5.6 2.9 1.9 2.6 1.1 1.9 3.1 25.7 28.8 Drowning 4.2 8.3 16.6 3.0 1.1 2.6 0.8 0.5 1.4 35.7 37.1 Occupational 0.5 3.9 1.1 0.7 0.5 0.4 0.4 0.3 0.7 7.0 7.7 Intentional 36.9 14.6 53.8 11.5 6.1 22.8 5.8 9.2 14.9 145.7 160.6 Self-inflicted 3.0 10.8 40.0 4.4 1.0 2.7 2.6 4.9 7.4 61.9 69.3 Homicide and violence 4.7 2.8 13.8 5.8 3.2 3.6 3.2 4.3 7.5 34.0 41.4 War 29.2 0.9 0.0 1.3 1.9 16.5 0.0 0.0 0.0 49.8 49.8 Total 1,398.8 1,471.1 978.0 815.7 456.9 702.9 250.2 415.7 665.9 5,823.4 6,489.4 217 Table B.3 Burden of disease in males by cause, 1990 (hundreds of thousands of DALYs lost) Formerly Latin socialist Established Sub- Other America Middle economies market Saha ran Asia and and the Eastern of Europe economies FSE and Demographically Disease or injury Africa India China islands Caribbean crescent (FSE) (EME) EME developing group World Communicable, maternal, and perinatal 1,046.8 704.4 228.0 438.0 226.3 347.2 24.5 42.3 66.8 2,990.9 3,057.7 Infectious and parasitic 763.7 404.9 117.6 258.4 137.7 180.2 8.0 18.2 26.3 1,862.5 1,888.8 Tuberculosis 74.6 62.8 34.7 51.6 15.1 21.6 3.1 1.1 4.1 260.5 264.7 STDs excluding HIV 28.9 5.3 0.8 0.6 2.4 0.3 0.1 0.2 0.3 38.3 38.6 Syphilis 27.2 4.7 0.1 0.1 2.1 0.0 0.0 0.0 0.0 34.1 34.1 Chlamydia 0.3 0.6 0.6 0.5 0.3 0.3 0.1 0.1 0.2 2.5 2.7 Gonorrhea 1.4 0.0 0.1 0.1 0.1 0.0 0.0 0.0 0.0 1.8 1.8 Pelvic inflammatory disease 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Human immunodeficiency virus 93.7 27.1 0.0 8.0 34.1 2.6 1.4 12.4 13.7 165.5 179.3 Diarrheal diseases 157.3 136.4 20.7 78.5 31.3 75.1 1.1 1.2 2.3 499.4 501.7 Acute watery 84.1 75.0 14.1 42.6 17.5 39.5 1.0 1.0 2.0 272.8 274.8 Persistent 48.8 40.2 2.7 23.7 8.7 23.7 0.1 0.1 0.1 147.7 147.8 Dysentery 24.4 21.3 4.0 12.2 5.1 12.0 0.1 0.1 0.2 79.0 79.1 Childhood cluster 148.1 95.8 9.3 43.3 8.6 42.5 0.2 0.4 0.7 347.7 348.4 Pertussis 26.1 14.3 2.8 7.1 3.9 7.5 0.2 0.4 0.6 61.7 62.2 Polio 8.3 10.9 1.4 2.4 1.3 3.9 0.0 0.0 0.0 28.1 28.2 Diphtheria 0.1 0.5 0.0 0.2 0.1 0.1 0.0 0.0 0.0 1.1 1.1 Measles 83.0 44.9 1.5 22.3 1.9 18.8 0.0 0.1 0.1 172.5 172.6 Tetanus 30.6 25.3 3.6 11.3 1.3 12.2 0.0 0.0 0.0 84.2 84.2 Meningitis 11.5 11.9 4.0 9.7 3.8 7.4 0.8 0.6 1.4 48.4 49.8 Hepatitis 1.1 1.4 4.5 1.4 0.7 0.8 0.2 0.4 0.6 9.8 10.4 Malaria 161.0 4.8 0.1 12.9 2.2 1.3 0.0 0.0 0.0 182.3 182.3 Tropical cluster 39.0 11.3 3.8 3.1 16.3 1.5 0.0 0.0 0.0 75.0 75.0 Trypanosomiasis 9.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.0 9.0 Chagas'disease 0.0 0.0 0.0 0.0 14.8 0.0 0.0 0.0 0.0 14.8 14.8 Schistosomiasis 23.1 1.7 2.8 0.7 1.2 0.5 0.0 0.0 0.0 29.9 29.9 Leishmaniasis 1.9 6.8 0.6 1.5 0.3 0.9 0.0 0.0 0.0 12.0 12.0 Lymphatic filariasis 1.3 2.8 0.5 0.9 0.0 0.1 0.0 0.0 0.0 5.6 5.6 Onchocerciasis 3.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.7 3.7 Leprosy 1.2 2.6 0.0 0.8 0.3 0.2 0.0 0.0 0.0 5.1 5.1 Trachoma 2.1 1.1 1.1 2.4 0.4 2.2 0.0 0.0 0.0 9.3 9.3 Intestinal helminths 4.2 10.6 32.6 29.6 12.0 2.8 0.0 0.0 0.0 91.8 91.8 Ascaris 2.2 6.0 19.9 16.3 6.8 2.6 0.0 0.0 0.0 53.8 53.8 Trichuris 1.5 2.5 11.6 12.0 4.5 0.0 0.0 0.0 0.0 32.2 32.2 Hookworm 0.5 2.1 1.0 1.2 0.7 0.2 0.0 0.0 0.0 5.8 5.8 Respiratory infections 162.5 155.7 60.0 103.4 33.9 80.8 8.2 12.5 20.7 596.3 617.1 Lower respiratory infections 157.1 147.3 52.1 97.4 30.2 76.9 6.5 8.9 15.4 560.9 576.4 Upper respiratory infections 1.6 3.2 2.8 2.5 1.3 0.5 0.9 2.1 2.9 11.9 14.8 Otitis media 3.8 5.1 5.0 3.6 2.4 3.5 0.9 1.5 2.3 23.6 25.9 Maternal 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Hemorrhage 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sepsis 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Eclampsia 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Hypertension 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Obstructed labor 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Abortion 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Perinatal 120.6 143.8 50.4 76.2 54.8 86.2 8.3 11.6 19.8 532.0 551.8 Noncommunicable 287.9 601.3 609.6 359.7 228.6 264.0 231.6 399.0 630.6 2,351.0 2,981.6 Malignant neoplasms 22.5 65.7 113.1 41.4 25.3 26.5 49.9 99.5 149.4 294.6 444.0 Mouth and oropharynx 1.2 12.8 3.8 3.8 1.3 1.9 2.1 3.3 5.4 24.7 30.0 Esophagus 1.5 5.0 11.7 1.3 0.7 1.0 1.3 2.8 4.1 21.2 25.3 Stomach 1.6 4.7 19.9 3.3 2.4 2.1 6.4 7.0 13.4 34.0 47.4 Colon and rectum 0.6 2.5 5.3 1.9 1.2 1.1 3.4 10.2 13.7 12.7 26.4 Liver 3.9 1.6 27.4 4.8 0.3 0.9 0.4 2.5 2.9 38.8 41.7 Pancreas 0.3 0.8 1.8 0.5 0.3 0.5 0.6 3.3 3.9 4.3 8.2 Trachea, bronchus, and lung 1.3 4.7 13.6 5.5 2.2 4.0 13.2 22.4 35.6 31.3 66.9 Melanoma 0.3 0.1 0.1 0.1 0.2 0.1 0.5 1.5 2.0 1.0 2.9 Breast 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Cervix 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Uterus 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Ovary 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 218 Formerly Latin socialist Established Sub- Other America Middle economies market Saharan Asia and and the Eastern of Europe economies FSEand Demographically Disease or injury Africa India China islands Caribbean crescent (FSE) (EME) EME developing group World Noncommunicable diseases, malignant neoplasms (continued) Prostate 2.2 1.9 0.3 0.9 1.6 0.7 1.1 6.7 7.8 7.7 15.6 Bladder 0.8 0.8 1.4 0.7 0.7 1.3 1.8 3.3 5.1 5.7 10.9 Lymphoma 3.2 3.1 2.6 1.8 1.7 1.6 1.6 4.8 6.4 13.9 20.3 Ieukemia 0.7 3.2 8.1 2.7 1.5 1.9 1.7 3.5 5.2 18.1 23.3 Diabetes mellitus 1.2 8.4 3.6 4.6 4.3 4.8 1.7 6.2 7.9 26.8 34.7 Nutritional and endocrine 42.5 91.4 27.7 40.5 23.0 25.4 2.9 6.9 9.8 250.6 260.3 Protein-energy malnutrition 11.2 26.3 6.5 4.9 5.2 5.2 0.6 1.2 1.8 59.3 61.1 Iodine deficiency 8.2 7.1 5.0 6.5 2.6 6.9 0.0 0.0 0.0 36.3 36.3 Vitamin A deficiency 11.1 20.9 4.9 12.9 7.2 2.9 0.0 0.0 0.0 59.9 59.9 Anemia 4.1 19.7 10.2 9.5 3.9 6.3 1.3 2.2 3.5 53.6 57.2 Neuropsychiatric 55.8 93.7 81.7 66.7 46.8 41.7 35.1 78.2 113.3 386.4 499.8 Depressive disorders 5.3 10.7 16.6 8.6 5.7 6.0 3.5 6.8 10.3 52.8 63.1 Bipolar affective disorders 0.5 1.1 1.6 0.9 0.6 0.6 0.3 0.6 0.9 5.3 6.1 Psychoses 3.1 10.0 10.3 4.7 3.1 3.3 1.4 3.3 4.8 34.5 39.2 Epilepsy 8.0 13.3 10.2 9.9 4.7 7.3 2.3 4.1 6.4 53.4 59.8 Alcohol dependence 14.9 16.6 16.6 9.4 14.1 3.7 16.4 24.5 40.9 75.3 116.2 Alzheimer's and other dementias 4.1 9.8 12.6 6.8 4.0 4.6 5.1 16.7 21.8 41.9 63.7 Parkinson's disease 0.3 0.8 0.9 0.6 0.6 0.3 0.5 1.6 2.1 3.6 5.7 Multiple sclerosis 0.6 1.4 1.5 1.0 0.4 0.6 0.5 1.0 1.5 5.5 7.0 Drug dependence 2.8 4.8 2.2 6.8 6.3 2.7 1.8 11.8 13.6 25.7 39.3 Posttraumatic stress disorder 1.9 3.3 4.4 2.6 1.7 1.9 1.2 2.7 3.9 15.8 19.6 Sense organ 6.0 12.4 7.5 5.8 3.1 3.9 0.2 0.3 0.5 38.6 39.1 Glaucoma 0.0 2.3 1.5 1.3 0.2 0.3 0.1 0.1 0.2 5.6 5.7 Cataract 5.5 9.3 4.0 4.2 2.5 3.3 0.1 0.1 0.2 28.8 29.1 Cardiovascular 55.6 146.7 148.4 85.0 48.9 64.0 89.8 120.9 210.7 548.6 759.3 Rheumatic 2.4 6.0 9.1 2.0 1.4 1.6 1.6 0.5 2.1 22.5 24.6 lschemic heart disease 6.7 49.5 24.8 35.4 16.0 15.5 45.2 56.4 101.7 147.9 249.6 Cerebrovascular 17.4 27.5 68.4 16.1 13.2 15.5 21.8 23.1 449 158.2 203.1 Peri-, endo-, and myocarditis and cardiomyopathy 17.1 36.3 5.2 9.6 8.7 9.5 2.3 4.1 6.4 86.2 92.6 Respiratory 22.6 38.5 96.0 21.5 17.3 21.1 13.0 21.2 34.2 217.1 251.3 Chronic obstructive pulmonary disease 3.5 9.7 60.8 4.8 4.1 4.2 6.4 10.3 16.7 87.1 103.8 Asthma 7.8 8.5 18.2 7.5 5.4 6.6 2.7 5.6 8.3 54.0 62.3 Digestive 28.8 55.3 49.9 30.3 20.5 23.0 14.3 23.7 37.9 207.8 245.7 Peptic ulcer disease 2.0 6.4 7.3 2.9 1.2 1.7 1.8 2.7 4.5 21.4 26.0 Cirrhosis 8.7 18.5 22.0 10.9 8.7 4.5 3.3 11.7 15.1 73.4 88.5 Genitourinary 10.9 18.7 21.8 12.3 7.7 10.4 5.3 8.5 13.8 81.8 95.6 Nephritis and nephrosis 6.2 10.4 14.8 7.3 4.2 4.6 2.4 4.5 6.9 47.6 54.5 Benign prostatic hypertrophy 2.3 3.7 5.5 2.7 2.2 2.3 1.3 2.5 3.9 18.6 22.5 Musculoskeletal 1.9 4.0 13.3 10.9 8.8 2.0 2.9 14.4 17.3 40.9 58.2 Rheumatoid arthritis 0.5 1.0 2.3 1.0 2.5 0.5 1.6 2.4 4.0 7.9 11.9 Osteoarthritis 0.6 1.2 8.5 8.9 5.6 0.7 0.8 11.1 11.9 25.4 37.3 Congenital abnormalities 33.8 48.4 34.8 27.8 16.2 29.7 8.4 12.3 20.7 190.7 211.4 Oral health 5.0 15.1 9.2 11.3 5.6 10.4 7.5 5.0 12.5 56.5 69.0 Dental caries 0.7 1.7 1.2 1.3 2.2 1.7 0.7 0.4 1.0 8.7 9.8 Periodontal disease 4.1 11.5 5.3 5.2 2.8 3.1 0.4 0.5 0.9 32.0 32.8 Edentulism 0.2 1.9 2.7 4.8 0.7 5.5 6.4 4.1 10.5 15.8 26.4 Injuries 193.6 147.6 197.6 153.8 117.2 127.9 75.5 78.8 154.3 937.8 1,092.1 Unintentional 107.9 126.3 148.3 109.2 79.4 76.2 53.6 50.8 104.4 647.3 751.7 Motor vehicle 29.0 23.1 33.0 32.6 41.2 31.2 17.2 24.0 41.2 190.1 231.3 Poisoning 3.9 2.1 8.0 3.9 0.5 2.3 6.7 2.0 8.7 20.7 29.5 Falls 20.3 28.9 23.5 21.0 7.2 9.7 5.7 7.9 13.6 110.6 124.2 Fires 5.9 7.1 8.1 5.1 2.7 3.9 2.6 3.2 5.8 32.7 38.6 Drowning 11.3 9.0 27.6 10.5 4.8 7.0 4.8 2.2 7.0 70.2 77.3 Occupational 3.6 5.1 5.0 3.9 2.8 1.9 2.7 1.7 4.4 22.3 26.7 Intentional 85.7 21.4 49.3 44.6 37.8 51.7 21.9 28.0 49.9 290.5 340.4 Self-inflicted 13.9 11.1 29.8 14.9 2.9 7.2 11.6 14.6 26.3 79.7 106.0 Homicide and violence 21.2 8.2 19.5 27.1 30.8 11.5 10.2 13.4 23.7 118.3 142.0 War 50.6 2.0 0.0 2.6 4.1 33.1 0.0 0.0 0.0 92.5 92.5 Total 1,528.3 1,453.3 1,035.2 951.4 572.1 739.2 331.6 520.1 851.7 6,279.7 7,131.4 219 Table B.4 Burden of disease by age and the three main groups of causes, 1990 Communicable diseases and maternal and perinatal causes Noncommunicable diseases Injuries Rate Rate Rate DALYs lost (per 1000 DALYs lost (per 1,000 DALYs lost (per 1,000 Demographic region and age group (millions) Percentage population) (millions) Percentage population) (millions) Percentage population) Sub-Saharan Africa 208.6 100.0 408.7 56.8 100.0 111.4 27.3 100.0 53.5 0-4 132 63.5 1,4AJ1.5 18.2 32.1 193.0 6.4 23.4 67.7 5-14 25.5 12.2 181.8 7.2 12.7 51.5 4.6 16.7 32.5 15-44 44.8 21.5 213.2 12.3 21.6 58.4 14.8 54.3 70.6 45-59 3.7 1.8 87.7 10.2 18.0 240.5 1.1 4.2 27.1 60+ 92.4 8.9 ia0-4 2.1 147.7 97.9 1.0 100.0 66.3 173.9 840.7 118.0 33.4 15.7 100.0 28.3 383.0 138.9 286.6 0.4 26.7 5.9 1.4 100.0 22.2 16.7 31.4 50.9 5-14 15.4 10.4 78.3 9.8 8.3 49.6 7.1 26.7 36.1 15-44 27.1 18.3 70.5 27.3 23.1 71.1 11.5 43.1 30.0 45-59 4.7 3.2 50.2 22.6 19.2 242.0 1.5 5.7 16.3 60+ 2.6 1.8 44.8 25.0 21.1 425.2 0.6. 2.3 10.4 China 50.9 100.0 44.9 116.8 100.0 103.0 33.6 100.0 29.7 0-4 25.5 50.0 215.5 16.4 14.0 138.4 8.1 24.0 68.3 5-14 7.9 15.5 42.1 4.7 4.1 25.3 3.6 10.8 19.4 15-44 12.3 24.2 20.9 33.0 28.2 55.8 17.1 50.9 29.0 45-59 2.7 5.4 20.0 24.3 20.8 177.1 2.6 7.6 18.6 60+ 2.5 4.9 24.6 38.4 32.9 381.7 2.2 6.7 22.2 Other Asia and islands 85.8 100.0 125.7 70.9 100.0 103.9 20.0 100.0 29.3 0-4 49.7 58.0 580.1 14.4 20.3 167.8 3.0 15.0 35.0 5-14 15.2 17.8 92.7 8.0 11.2 48.5 3.5 17.6 21.5 15-44 16.0 18.7 50.1 19.2 27.1 59.9 11.5 57.6 36.0 45-59 2.7 3.1 38.5 14.2 20.0 204.8 1.3 6.7 19.3 60+ 2.1 2.4 48.5 15.2 21.5 355.0 0.6 3.1 14.4 Latin America and the Caribbean 43.4 100.0 97.7 44.1 100.0 99.2 15.4 100.0 34.8 22.3 51.5 396.1 8.6 19.5 152.7 1.5 9.7 26.7 5-14 6.2 14.4 60.7 3.6 8.1 34.5 2.2 14.1 21.2 15-44 13.0 29.9 62.2 13.2 30.0 63.5 10.2 66.3 49.2 45-59 1.2 2.8 26.7 9.0 20.5 198.2 1.0 6.6 22.2 60+ 0.6 1.5 20.8 9.6 21.8 310.0 0.5 3.2 16.1 Middle Eastern crescent 73.5 100.0 146.1 51.9 100.0 103.2 18.8 100.0 37.3 0-4 55.1 75.0 681.5 15.2 29.3 188.3 4.0 21.1 48.9 5-14 7.0 9.5 54.9 6.2 12.0 48.7 4.3 22.8 33.6 15-44 9.1 12.4 41.1 11.5 22.1 51.8 9.2 49.1 41.7 45-59 1.2 1.7 27.2 8.9 17.2 200.1 0.9 4.8 20.4 60+ 1.1 1.5 36.7 10.1 19.5 347.2 0.4 2.2 14.0 Formerly socialist economies of Europe (FSE) 5.0 100.0 14.5 43.5 100.0 125.7 9.7 100.0 27.9 0-4 2.5 50.0 93.4 2.3 5.2 84.9 0.6 5.8 20.9 5-14 0.1 2.7 2.6 1.2 2.7 22.1 0.7 7.0 12.5 15-44 1.8 35.0 11.6 9.7 22.4 64.5 6.1 62.8 40.1 45-59 0.3 6.8 6.0 12.2 28.0 213.7 1.6 16.6 28.0 60+ 0.3 5.4 4.7 18.1 41.6 315.9 0.8 7.9 13.3 Established market economies (EME) 9.1 100.0 11.4 73.4 100.0 92.0 11.1 100.0 13.9 0-4 2.7 29.3 51.8 3.2 4.3 61.5 0.5 4.9 10.6 5-14 0.2 2.1 1.9 1.4 2.0 13.8 0.6 5.4 5.8 15-44 4.4 48.8 12.2 16.0 21.8 44.1 7.0 62.8 19.2 45-59 0.5 5.8 3.9 16.9 23.0 126.2 1.3 11.8 9.7 60+ 1.3 14.0 8.8 35.9 48.9 247.1 1.7 15.1 11.5 FSE and EME 14.1 100.0 124 116.9 100.0 102.2 20.7 100.0 18.1 0-4 5.2 36.6 66.1 5.4 4.7 69.5 1.1 5.3 14.1 5-14 0.3 2.4 2.1 2.6 2.2 16.6 1.3 6.1 8.1 15-44 6.2 43.9 12.1 25.8 22.1 50.1 13.0 62.8 25.3 45-59 0.9 6.2 4.6 29.1 24.9 152.3 2.9 14.0 15.2 60+ 1.5 10.9 7.6 54.0 46.2 266.6 2.4 11.7 12.0 Demographically developing group 609.9 100.0 .147.9 458.5 100.0 111.2 141.9 100,0 34.4 0-4 383.1 62.8 693.7 106.2 23.2 192.3 28.9 20.3 52.3 5-14 77.3 12.7 84.1 39.5 8.6 42.9 25.3 17.8 27.5 15-44 122.3 20.0 63.2 116.4 25.4 60.2 74.5 52.5 38.5 45-59 16.3 2.7 37.6 89.3 19.5 206.4 8.5 6.0 19.6 60+ 11.0 1.8 38.7 107.2 23.4 375.4 4.8 3.4 16.7 World 624.0 100.0 118.5 575.4 100.0 109.2 162.6 100.0 30.9 0-4 388.2 62.2 615.7 111.7 19.4 177.1 30.0 18.4 47.5 5-14 77.6 12.4 72.1 42.1 7.3 39.1 26.6 16.3 24.7 15-44 128.5 20.6 52.5 142.2 24.7 58.1 87.5 53.8 35.7 45-59 17.1 2.7 27.5 118.3 20.6 189.8 11.4 7.0 18.3 60+ 12.6 2.0 25.8 161.2 28.0 330.3 7.2 4.4 14.8 220 Table B.5 Burden of disease by consequence, sex, and age, 1990 (millions of DALYs lost) As result of premature death As result of disability Demographic region and age group Males Females Total Males Females Total Sub-Saharan Africa 119.5 105.7 225.2 33.3 3.4.2 67.5 0-4 71.6 61.8 133.4 12.2 11.6 23.7 5-14 13.4 12.7 26.1 6.3 4.8 11.1 15-44 26.1 2.2.7 48.8 10.1 12.9 23.1 45-59 4.9 4.4 9.3 3.0 2.8 5.8 60+ 3.5 4.1 7.6 1.7 2.1 3.8 India 100.8 99.7 200.6 44.5 47.4 91.9 0-4 53.0 55.1 108.1 14.2 14.9 29.1 5-14 9.4 10.8 20.2 6.4 5.7 12.1 15-44 18.9 17.7 36.6 12.0 17.2 29.2 45-59 9.8 7.3 17.2 6.7 5.0 11.7 60+ 9.7 8.7 18.4 5.2 4.6 9.8 China 62.2 53.3 115.5 41.3 44.5 85.8 0-4 16.6 18.7 35.3 7.1 7.5 14.6 5-14 3.2 2.3 5.5 5.7 5.1 10.8 15-44 17.4 13.2 30.6 13.7 18.2 31.9 45-59 10.8 7.0 17.8 6.6 5.2 11.8 60+ 14.3 12.1 26.4 8.3 8.5 16.7 OtherAsiaandislands 62.9 50.1 113.0 32.2 31.5 63.7 0-4 29.7 23.9 53.6 7.0 6.5 13.5 5-14 8.4 6.3 14.7 6.6 5.4 12.0 15-44 13.3 10.1 23.4 11.1 12.3 23.4 45-59 5.8 4.3 10.1 4.3 3.8 8.1 60+ 5.7 5.5 11.2 3.3 3.5 6.7 Latin America and the Caribbean 33.8 24.3 58.0 23.5 21.4 44.9 0-4 13.2 10.1 23.3 4.8 4.3 9.1 5-14 2.6 2.0 4.6 3.9 3.5 7.3 15-44 11.4 6.6 18.0 9.4 9.0 18.5 45-59 3.4 2.6 6.0 3.0 2.3 5.3 60+ 3.1 3.0 6.1 2.3 2.3 4.7 Middle Eastern crescent 51.5 46.9 98.4 22.4 23.4 48 0-4 31.1 29.6 60.7 7.0 6.6 13.6 5-14 5.4 4.9 10.4 3.8 3.3 7.1 15-44 7.7 6.2 13.9 6.8 9.1 15.9 45-59 3.6 2.6 6.2 2.7 2.1 4.9 60+ 3.7 3.6 7.3 2.1 2.2 4.3 Formerly socialist economies of Europe (FSE) 21.5 13.0 345 11.7 12.1 23.7 0-4 2.1 1.5 3.5 0.9 0.9 1.8 5-14 0.5 0.3 0.9 0.5 0.7 1.1 15-44 6.9 2.1 9.0 4.4 4.2 8.6 45-59 6.1 2.7 8.8 3.0 2.3 5.3 60+ 59 64 12.3 2.9 3.9 6.9 Established market economies (EME) 29.5 19.6 49.2 22.5 22.0 0-4 2.0 1.5 3.5 1.5 1.4 2.9 5-14 0.5 0.3 0.8 0.8 0.6 1.4 15-44 8.5 3.5 12.0 7.6 7.9 15.4 45-59 6.7 3.7 10.3 4.9 3.5 8.4 60+ 11.9 10.6 22.6 7.7 8.5 16.2 FSE and EME 51.0 32.6 83.6 34.1 34.0 68.2 0-4 4.0 3.0 7.0 2.4 2.3 4.7 5-14 1.1 0.6 1.7 1.3 1.3 2.6 15-44 15.4 5.6 21.0 11.9 12.1 24.0 45-59 12.7 6.4 19.1 7.9 5.9 13.8 60+ 17.8 17.0 3.4.9 10.6 12.5 23.1 Demographically developing group 430.7 380.0 810.7 197.2 202.4 399.6 0-4 215.3 199.1 414.4 52.4 51.3 103.7 5-14 42.4 39.1 81.5 32.7 27.8 60.5 15-44 94.6 76.6 171.3 63.0 78.8 141.9 45-59 38.3 28.2 66.5 26.3 21.2 47.5 60+ 40.1 37.0 77.0 22.8 23.2 46.0 World 481.8 412.6 894.3 231.4 236.4 467.8 0-4 219.3 202.1 421.4 54.8 53.6 108.5 5-14 43.5 39.7 83.2 33.9 29.1 63.0 15-44 110.0 82.2 192.3 74.9 90.9 165.9 45-59 51.1 34.5 85.6 34.2 27.1 61.3 60+ 57.9 54.0 111.9 33.4 35.7 69.1 221 Table B.6 Distribution of the disease burden in children in demographically developing economies, showing the ten main causes, 1990 Children under 5 Children ages 5-14 Female Male Female Male Total DALYs lost (millions) 250 268 67 75 Diseases and injuries Rank Percent Rank Percent Rank Percent Rank Percent Communicable and perinatal 73.2 74.6 57.1 52.0 Infectious and parasitic 37.5 37.2 48.5 45.2 Tuberculosis 0.5 0.5 WA 5.7 WA 4.1 STDs and HIV 1.0 WI 1.0 WI 2.4 1.9 Syphilis 0.5 0.5 0.0 0.0 Human immunodeficiency virus 0.5 0.5 0.3 0.1 Diarrhea! diseases WI 16.2 I 31 15.7 13 7.1 El 6.1 Childhood cluster 41 10.7 13 10.6 WI 8.6 WI 8.1 Pertussis 1.8 1.9 1.6 1.5 Polio 0.3 0.4 2.0 2.5 Measles 5.6 5.5 4.1 3.4 Tetanus 2.9 2.9 0.8 0.7 Malaria 61 4.7 El 4.7 El 4.9 El 4.3 Intestinal helminths 1 0.0 0.0 0.0 0.0 1 12.3 7.6 'I 11.4 7.1 Ascans Trichuris 0.0 0.0 4.6 4.2 Respiratory infections lii 18.5 WI 17.6 I 31 7.9 El 6.9 Perinata! 1 21 17.2 19.9 0.0 0.0 Noncommunicable 21.1 19.9 28.4 27.2 Nutritional and endocrine 6.4 6.1 3.7 2.4 Protein-energy malnutrition WA 2.4 13 2.1 0.3 0.3 Iodine deficiency WI 1.3 WI 1.2 0.2 0.2 Vitamin A deficiency '3 2.3 WA 2.2 0.0 0.0 Anemias 0.3 0.3 13 3.0 1.9 Neuropsychiatric 1.1 1.0 7.5 9.1 Epilepsy 0.2 0.3 I 91 2.6 I 81 3.5 Respiratory 2.0 1.8 3.2 4.0 Asthma 0.2 0.2 2.3 2.6 Congenital I 51 6.5 WA 6.6 1.2 1.0 Injuries 5.7 5.5 14.5 20.7 Unintentional 4.6 4.6 12.6 18.1 Motor vehicle 0.4 0.4 I 71 3.7 I 51 4.4 Falls 110 1.2 1.0 1.9 110 3.1 Drowning 0.6 0.7 1.7 WI 3.2 Intentional 1.0 0.8 2.0 2.6 Notes The rankings refer to health intervention priorities; disease groups are ranked only when there is a single intervention or accepted cluster of interventions for controlling the diseases included in the group. Can be substantially controlled with cost-effective intervention; less than $100 per DALY saved. Can be partially controlled with moderately cost-effective interventions; $250 to $999 per DALY saved. (There are few or no interventions in the range of $100 to $250 per DALY saved.) I I Cannot be controlled in a cost-effective manner; $1,000 or more per DALY saved. I I Preventive and therapeutic interventions have not been evaluated for cost-effectiveness. Table B.7 Distribution of the disease burden in the adult and elderly populations in demographically developing economies, showing the ten main causes, 1990 Young adults (ages 15-44) Mature adults (ages 45-59) Elderly (60+ years old) Female Male Female Male Female Male Total DALYs lost (millions) 155 158 49 65 60 63 Diseases and injuries Rank Percent Rank Percent Rank Percent Rank Percent Rank Percent Rank Percent Communicable and maternal 50.0 28.2 13.2 15.1 8.4 9.5 Infectious and parasitic 29.5 25.8 10.6 13.6 3.8 5.5 Tuberculosis 7.0 8.4 5.6 9.3 J 1.9 4.0 Sexually transmitted diseases 8.9 1.5 0.3 0.2 0.0 0.0 Human immunodeficiency virus 6.6 9.3 0.3 0.6 0.0 0.0 Respiratory infections 2.5 2.4 2.0 1.5 4.6 4.0 Maternal I 1] 18.0 0.5 0.0 Noncommunicable 37.6 36.8 81.6 75.8 87.8 86.5 Malignant neoplasms 4.5 4.3 16.7 15.9 10.4 14.5 Stomach 0.3 0.3 1.4 2.2 1.3 I 8 2.3 Liver 0.3 0.8 1.1 1101 2.5 0.7 1.4 Trachea, bronchus, and lung 0.1 0.2 0.9 2.0 0.8 2.3 Cervix 0.6 IIJ 2.6 1.0 Diabetes mellitus 0.4 0.4 I 81 2.8 1.6 I 7] 2.4 1.5 Nutritional and endocrine 3.6 3.5 2.4 1.4 1.4 0.9 Anemia I 8] 2.5 1.5 1.3 0.9 0.7 0.5 Neuropsychiatric 11.9 12.1 6.9 8.5 6.7 7.1 Depressive disorders I 51 5.8 I 71 2.9 2.2 0.9 0.5 0.2 Alcohol dependence syndrome 0.4 9] 2.7 0.7 3.6 0.2 1.6 Alzheimer's disease and other dementias 0.1 0.1 2.2 1.7 41 4.8 I 4! 4.1 Sense organ 0.2 0.2 4.6 3.0 2.1 2.1 Cataracts 0.2 0.1 3.1 2.3 IN 1.6 1.7 Cardiovascular 6.0 6.5 25.2 23.7 44.3 39.3 Ischemic heart disease 0.6 1.5 I 31 4.7 21 7.6 I 21 11.6 I 2I 11.7 Cerebrovascular 1.5 1.4 I ii 8.7 31 6.7 I ii 16.5 III 13.8 Pen-, endo-, and myocarditis 1.1 1.6 I 41 3.2 61 3.4 I 61 3.6 1 71 3.6 Respiratory 2.3 2.3 5.4 4.5 10.5 11.7 Chronic obstructive pulmonary 0.3 0.3 I 71 2.8 71 2.7 I 31 8.1 I 31 9.6 Digestive 2.7 4.0 5.8 7.2 3.8 4.8 Cirrhosis 0.8 1.9 2.4 4] 4.2 1.2 1101 2.1 Genitounnary system 1.4 1.1 3.1 4.2 2.5 2.4 Benign prostatic hypertrophy 0.0 8] 2.5 0.4 Musculoskeletal 3.2 1.1 3.9 2.1 2.3 1.2 Osteoarthritis I 91 2.2 0.7 2.7 1.5 1.5 0.6 Oral health 0.6 0.6 4.4 3.4 1.1 0.9 Periodontal disease 0.1 0.1 I 5] 3.1 I 9] 2.5 0.0 0.0 Injuries 12.4 35,0 5.2 9.1 3.8 4.0 Unintentional 6.4 20.7 3.4 6.4 3.0 3.1 Motor vehicle injuries 1101 2.1 3] 8.2 0.9 1.8 0.3 0.6 Falls 0.4 8] 2.8 0.8 1.5 I 9] 1.8 1.2 Intentional 6.0 14.3 1.9 2.7 0.8 0,9 Self-inflicted I 61 3.2 61 4.0 1.1 1.3 0.5 0.6 Homicide and violence 1.0 4] 6.1 0.3 1.0 0.1 0.2 War 1.8 51 4.2 0.4 0.5 0.1 0.1 Nofe: The rankings refer to health intervention priorities; disease groups are ranked only when there is a single intervention or accepted cluster of interventions for controlling the diseases included in the group. Can be substantially controlled with cost-effective intervention; less than $100 per DALY saved. I I Can be partially controlled with moderately cost-effective interventions; $250 to $999 per DALY saved. (There are few or no interventions in the range of $100 to $250 per DALY saved.) I I Cannot be controlled in a cost-effective manner; $1,000 or more per DALY saved. I I Preventive and therapeutic interventions have not been evaluated for cost-effectiveness. 223 Table B .8 Deaths by cause and demographic group, 1990 (thousands of deaths) Demographically developing group FSE and EME Males Females Males Females Ages Age 5 Ages Age 5 Ages Age 5 Age 5 Disease or injury 0-4 and older 0-4 and older Total 0-4 and older Ages 0-4 and older Total World Communicable, maternal and perinatal 5,539 2,801 5,038 2,738 16,115 73 243 52 207 575 16,690 Infectious and parasitic 2,814 2,179 2,651 1,658 9,301 9 92 7 46 153 9,454 Tuberculosis 34 1,187 37 720 1,978 0 29 0 9 38 2,016 STDs excluding HIV 41 62 37 53 192 0 0 0 1 1 193 Syphilis 39 62 35 50 186 0 0 0 0 0 186 Chlamydia 0 1 0 1 1 0 0 0 0 0 Gonorrhea 2 0 1 0 3 0 0 0 0 0 3 Pelvic inflammatory disease 0 0 0 2 2 0 0 0 0 0 2 Human immunodeficiency virus 29 101 27 92 248 1 35 1 6 43 291 Diarrheal diseases 1,263 191 1,211 201 2,866 2 1 2 2 7 2,873 Acute watery 635 148 609 155 1,547 2 1 2 2 7 1,553 Persistent 439 5 421 6 871 0 0 0 0 0 872 Dysentery 189 38 181 40 448 0 0 0 0 0 448 Childhood cluster 824 129 774 133 1,860 0 0 0 0 1 1,861 Pertussis 146 22 131 22 321 0 0 0 0 0 321 Polio 4 9 3 7 24 0 0 0 0 0 24 Diphtheria 1 1 1 2 4 0 0 0 0 0 4 Measles 442 69 421 74 1,006 0 0 0 0 0 1,006 Tetanus 232 27 219 28 505 0 0 o o 0 505 Meningitis 71 73 50 39 232 3 3 2 3 10 242 Hepatitis 6 33 6 27 72 0 3 0 2 6 77 Malaria 332 143 301 151 926 0 0 0 0 0 926 Tropical cluster 6 105 6 82 199 0 0 0 0 0 199 Trypanosomiasis 2 27 3 24 55 0 0 0 0 0 55 Chagas' disease 0 11 0 12 23 0 0 0 0 0 23 Schistosomiasis 1 23 0 14 38 0 0 0 0 0 38 Leishmaniasis 4 26 3 20 54 0 0 0 0 0 54 Lymphatic filariasis 0 0 0 0 0 0 0 0 0 0 0 Onchocerciasis 0 17 0 12 30 0 0 0 0 0 30 Leprosy 0 1 0 1 3 0 0 0 0 0 3 Trachoma 0 0 0 0 0 0 0 0 0 0 0 Intestinal helminths 0 15 0 14 29 0 0 0 0 0 29 Acaris 0 7 0 6 13 0 0 0 0 0 13 Trichuris 0 5 0 5 9 0 0 0 0 0 9 Hookworm 0 3 0 3 6 0 0 0 0 0 6 Respiratory infections 1,371 622 1,339 652 3,984 12 151 9 158 330 4,314 Lower respiratory infections 1,343 620 1,311 650 3,924 12 150 9 157 328 4,251 Upper respiratory infections 3 3 3 2 10 0 1 0 1 2 12 Otitis media 25 0 25 0 51 1 0 0 0 1 52 Maternal 0 0 0 428 428 0 0 o 3 3 431 Hemorrhage 0 0 0 130 130 0 0 0 0 0 130 Sepsis 0 0 0 79 79 0 0 0 0 0 79 Eclampsia 0 0 0 45 45 0 0 0 0 0 45 Hypertension 0 0 0 31 31 0 0 0 0 0 32 Obstructed labor 0 0 0 40 40 0 0 0 0 0 40 Abortion 0 0 0 60 60 0 0 0 1 1 61 Perinatal 1,353 0 1,048 0 2,402 52 0 36 89 0 2,491 Noncommunicable 693 9,562 687 8,612 19,553 37 4,626 30 4,808 9,502 29,055 Malignant neoplasms 16 2,150 27 1,504 3,698 2 1,352 2 1,075 2,431 6,129 Mouth and oropharynx 0 211 1 103 315 0 39 0 11 51 366 Esophagus 0 220 0 112 332 0 43 0 14 58 389 Stomach 0 337 0 185 522 0 143 0 101 244 766 Colon and rectum 0 116 0 99 215 0 136 0 144 280 496 Liver 0 299 1 119 420 0 30 0 13 44 463 Pancreas 0 44 0 29 73 0 49 0 47 96 169 Trachea, bronchus, and lung 0 329 0 111 441 0 402 0 124 526 967 Melanoma 0 8 0 9 17 0 17 0 14 31 48 Breast 0 0 0 158 158 0 0 0 175 175 332 Cervix 0 0 0 183 183 0 0 0 32 32 215 Uterus 0 0 0 27 27 0 0 0 29 29 56 Ovary 0 0 1 50 51 0 0 0 56 56 106 224 Demographically developing group FSE and EME Males Females Males Females Ages Age 5 Ages Age 5 Ages Age 5 Ages Disease or injury 0-4 and older 0-4 and older Total 0-4 and older Ages 0-4 and older Total World Noncommunicable diseases, malignant neoplasms (continued) Prostate o 105 0 0 105 0 108 0 0 108 213 Bladder 0 61 0 18 80 0 51 0 20 71 151 Lymphoma 3 74 3 42 121 0 51 0 45 96 218 Leukemia 6 72 10 55 143 1 41 1 34 77 219 Diabetes mellitus 0 198 0 284 483 0 69 0 108 177 660 Nutritional and endocrine 119 148 127 185 578 1 30 1 41 74 651 Protein-energy malnutrition 66 26 82 33 207 0 2 0 4 6 213 Iodine deficiency 4 7 3 6 19 0 0 0 0 0 19 Vitamin A deficiency 19 0 16 0 35 0 0 0 0 0 35 Anemia 20 31 16 79 147 0 7 0 10 17 163 Neuropsychiatric 36 309 38 218 600 2 111 2 117 232 832 Depressive disorders 0 0 0 0 0 0 0 0 0 0 0 Bipolar affective disorders 0 1 0 1 1 0 0 0 2 2 3 Psychoses 0 26 0 17 42 0 7 0 9 16 58 Epilepsy 4 56 4 39 103 0 8 0 5 13 115 Alcohol dependence 0 27 0 4 31 0 18 0 4 22 53 Alzheimer's and other dementias 5 45 7 43 100 1 36 1 58 95 195 Parkinson's disease 0 19 0 12 32 0 16 0 15 30 62 Multiple sclerosis 0 11 0 13 25 0 3 0 5 8 33 Drug dependence 0 10 0 4 14 0 3 0 1 3 18 Posttraumatic stress disorder 0 0 0 0 0 0 0 0 0 0 0 Senseorgan 3 10 3 9 25 0 0 0 0 1 26 Glaucoma 0 1 0 1 2 0 0 0 0 0 2 Cataract 0 1 0 1 1 0 0 0 0 0 1 Cardiovascular 64 4,436 59 4,459 9,017 2 2,399 2 2,925 5,328 14,345 Rheumatic 1 146 2 292 440 0 17 0 29 46 486 Ischemic heart disease 2 1,348 1 1,118 2,469 0 1,283 0 1,395 2,678 5,147 Cerebrovascular 8 1,516 6 1,652 3,181 0 565 0 882 1,448 4,629 Pen-, endo-, and myocarditis and cardiomyopathy 32 621 36 540 1,229 0 70 1 64 136 1,364 Respiratory 78 1,172 79 1,008 2,336 1 308 1 199 509 2,845 Chronic obstructive pulmonary disease 11 938 9 756 1,714 0 228 0 130 358 2,072 Asthma 5 61 5 77 147 0 17 0 18 34 181 Digestive 87 738 106 484 1,416 2 231 1 194 427 1,843 Peptic ulcer disease 1 121 1 71 194 0 26 0 20 46 241 Cirrhosis 4 380 3 176 563 0 97 0 49 146 709 Genitourinary 12 264 8 251 535 0 83 0 85 169 704 Nephritis and nephrosis 6 171 4 146 327 0 49 0 50 100 427 Benign prostatic hypertrophy 0 30 0 0 31 0 10 0 0 10 41 Musculoskeletal 0 41 1 63 106 0 10 0 27 37 143 Rheumatoid arthritis 0 2 0 2 4 0 2 0 8 10 14 Osteoarthritis 0 0 0 0 0 0 0 0 0 0 0 Congenital abnormalities 271 40 232 52 595 25 8 21 7 61 656 Oral health 0 0 0 0 0 0 0 0 0 0 0 Injuries 254 2,010 233 923 3,420 13 559 8 227 807 4,227 Unintentional 217 1,297 192 531 2,237 12 375 8 163 558 2,794 Motor vehicle 28 457 22 130 637 2 160 2 55 219 856 Poisoning 13 75 7 38 132 1 41 1 13 56 188 Falls 10 87 12 59 168 1 45 0 51 97 265 Fires 18 29 12 23 83 1 10 1 6 18 100 Drowning 59 172 48 71 349 2 27 1 6 36 384 Occupational 0 88 0 28 115 0 19 0 3 22 137 Intentional 37 712 41 393 1,183 1 184 1 64 249 1,432 Self-inflicted 0 359 0 269 629 0 140 0 49 190 818 Homicide and violence 11 171 15 37 233 1 43 1 15 59 292 War 27 182 27 86 322 0 0 0 0 0 322 Total 6,485 14,372 5,958 12,273 39,088 123 5,428 91 5,242 10,883 49,971 Note: FSE, Formerly socialist economies of Europe; EME, established market economies. 225 World Development Indicators Contents Introduction 233 Key 230 Tables 1 Basic indicators 238 Production 2 Growth of production 240 3 Structure of production 242 4 Agriculture and food 244 5 Commercial energy 246 6 Structure of manufacturing 248 7 Manufacturing earnings and output 250 Domestic absorption 8 Growth of consumption and investment 252 9 Structure of demand 254 10 Structure of consumption 256 Fiscal and monetary accounts 11 Central government expenditure 258 12 Central government current revenue 260 13 Money and interest rates 262 Core international transactions 14 Growth of merchandise trade 264 15 Structure of merchandise imports 266 16 Structure of merchandise exports 268 17 OECD imports of manufactured goods: origin and composition 270 18 Balance of payments and reserves 272 External finance 19 Official development assistance from OECD and OPEC members 274 20 Official development assistance: receipts 276 21 Total external debt 278 22 Flow of public and private external capital 280 23 Aggregate net resource flows and net transfers 282 24 Total external debt ratios 284 25 Terms of external public borrowing 286 Human and natural resources 26 Population growth and projections 288 27 Demography and fertility 290 28 Health and nutrition 292 29 Education 294 30 Income distribution and PPC estimates pf GDP 296 31 Urbanization 298 32 Women in development 300 33 Forests, protected areas, and water resources 302 Table la Basic indicators for other economies 304 Technical notes 305 Data sources 325 Classification of economies 326 229 Key In each table, economies are listed within their are usually derived from the same sources. Expla- groups in ascending order of GNP per capita ex- nations of how World Bank estimates and projec- cept those for which no GNP per capita can be tions are derived from the sources, as well as more calculated. These are italicized, in alphabetical or- information on the sources, are given in World der, at the end of their group. The ranking below Population Projections, 1992-93 Edition. refers to the order in the tables. Note that two Figures in colored bands in the tables are sum- economies, Zimbabwe and Botswana, with re- mary measures for groups of economies. cently revised population data, have not been The letter w means weighted average; m, me- moved in the list to reflect their new ranking. dian value; t, total. The key shows the years of the most recent All growth rates are in real terms. census and the years of the latest demographic Data cutoff date is April 30, 1993. survey or vital registration-based estimates. This The symbol. . means not available information is included to show the currentness of The figures 0 and 0.0 mean zero or less than half the sources of demographic indicators, which can the unit shown. be a reflection of the overall quality of a country's A blank means not applicable. indicators. Beyond these years, demographic esti- Figures in italics indicate data that are for years mates may be generated by projection models, or periods other than those specified. extrapolation routines, or other methods. Other The symbol t indicates economies classified by demographic indicators, such as life expectancy, the United Nations or otherwise regarded by their birth and death rates, and under-5 mortality rates, authorities as developing. Country ranking Population Infant Economy Total in tables census mortality fertility Algeria 73 1987 1985 1984 Argentina 89 1991 1988 1990 Armenia 75 1989 1991 1991 Australia 113 1991 1990 1990 Austria 117 1981 1991 1991 Azerbaijan 67 1989 1991 1991 Bangladesh 12 1991 1989 1989 Belarus 93 1989 1991 1991 Belgium 116 1991 1991 1991 Benin 24 1979 1981-82 1981-82 Bhutan 5 1969 Bolivia 1984 41 *1992 1989 1989 Botswana 84 1991 1988 1988 Brazil 91 1991 1986 1986 Bulgaria 71 1985 1991 1991 Burkina Faso 17 1985 1976 1961 Burundi 8 1979 1987 1987 Cameroon 46 1987 1991 1991 Canada 119 1991 1990 1991 Central African Rep. 25 1975 1975 1959 230 Country ranking Population In[ant Total Economy in tables census mortality fertility Chad 9 1964 1964 1964 Chile 76 1982 1990 1990 China 22 1990 1990 1987 Colombia 57 1985 1990 1990 Congo 55 1984 1974 1974 Costa Rica 72 1984 1990 1990 Côte d'Ivoire 42 1988 1979 1988 Czechoslovakiaa 81 1991 1991 1991 Denmark 122 1981 1991 1991 Dominican Rep. 48 1990 1991 1991 Ecuador 49 1990 1989 1989 Egypt, Arab Rep. 37 1986 1988 1988 El Salvador 54 1971 1988 1988 Estonia 98 1989 1990 1990 Ethiopia 3 1984 1988 Finland 123 1990 1991 1991 France 118 1990 1991 1991 Gabon 97 1981 1960-61 1960-61 Georgia 66 1989 1991 1991 Germany" 121 1987 1991 1991 Ghana 26 1984 1988 1988 Greece 103 1991 1991 1991 Guatemala 47 1981 1987 1987 Guinea 29 1983 1954-55 1954-55 Guinea-Bissau 6 1989 1991 1991 Haiti 23 1982 1987 1987 Honduras 34 1988 1987-88 1987-88 tHong Kong 110 1991 1990 1990 Hungary 87 1990 1991 1991 India 19 1991 1986 1985 Indonesia 36 1990 1991 1991 Iran, Islamic Rep. 77 1986 1986 1975 Ireland 106 1986 1990 1991 tlsrael 107 1983 1990 1990 Italy 114 1991 1991 1991 Jamaica 60 1991 1989 1990 Japan 126 1990 1991 1991 Jordan 51 1979 1990-91 1990-91 Kazakhstan 82 1989 1991 1991 Kenya 20 1989 1989 1989 Korea, Rep. 102 1990 1985 1985 Kyrgyzstan 64 1989 1991 1991 Lao PDR 8 1985 1988 1988 Latvia 95 1989 1990 1990 Lesotho 35 1986 1977 1986 Lithuania 86 1989 1991 1991 Madagascar 10 1974-75 1992 1992 Malawi 14 1987 1982 1977 Malaysia 83 1991 1988 1984 Mali 16 1987 1987 1987 Mauritania 32 1988 1975 1987-88 Mauritius 80 1990 1991 1987 Mexico 92 1990 1987 1987 Moldova 78 1989 1991 1991 Morocco 50 1982 1992 1992 Mozambique 1 1980 1980 1980 231 Country ranking Population Infant Total Economy in tables census mortality fertility Namibia 62 1991 1992 Nepal 7 1991 1987 1987 Netherlands 115 1991 1991 1991 New Zealand 108 1991 1991 1991 Nicaragua 30 1971 1985 1985 Niger 18 1988 1992 1992 Nigeria 21 *1991 1990 1990 Norway 124 1980 1991 1991 Oman 100 . . 1986 1986 Pakistan 27 1981 1990-91 1990-91 Panama 74 1990 1985-87 1990 Papua New Guinea 45 1990 1980 1980 Paraguay 58 1982 1990 1990 Peru 53 1981 1991-92 1991-92 Philippines 44 1990 1986 1988 Poland 70 1988 1991 1991 Portugal 99 1991 1991 1991 Puerto Rico 101 . . Romania 61 1992 1990 1990 Russian Federation 94 1989 1991 1991 Rwanda 15 1991 1983 1992 tSaudi Arabia 104 *1992 Senegal 43 1988 1986 1986 Sierra Leone 11 1985 1971 1975 tSingapore 111 1990 1991 1991 South Africa 85 1985 1980 1981 Spain 109 1992 1990 1990 Sri Lanka 31 1981 1988 1989 Sudan 39 1983 1989-90 1989-90 Sweden 125 1990 1991 1991 Switzerland 127 1980 1991 1991 Syrian Arab Rep. 56 1981 1981 1981 Tajikistan 52 1989 1991 1991 Tanzania 2 1988 1991-92 1991-92 Thailand 65 1990 1989 1987 Togo 28 1981 1988 1988 Trinidad and Tobago 96 1990 1989 1989 Tunisia 63 1984 1988 1990 Turkey 69 1990 1988 1988 Turkmenistan 68 1989 1991 1991 Uganda 4 1991 1988-89 1988-89 Ukraine 79 1989 1991 1991 United Kingdom 112 1992 1991 1991 United States 120 1990 1991 1991 Uruguay 90 1985 1990 1990 Uzbekistan 59 1989 1991 1991 Venezuela 88 1990 1989 1989 Yemen Rep. 33 1986/1988 1991-92 1991-92 Yugoslavia 105 1991 1990 1990 Zambia 40 1990 1992 1992 Zimbabwe 38 1982 1988-89 1988-89 * Census data are not yet incorporated in the population estimates. Note: Economies with sparse data or with populations of more than 30,000 and fewer than I million are included only as part of the country groups in the main tables but are shown in greater detail in Table la. For data comparability and coverage throughout the tables, see the technical notes. a. In all tables data refer to the former Czechoslovakia; disaggragated data are not yet available. b. In all tables, data refer to the unified Germany, unless otherwise stated. c. In all tables, data refer to the former Socialist Federal Republic of Yugoslavia; disaggregated data are not yet available. 232 Introduction This sixteenth edition of the World Development In- concessionality of a country's external obligations. dicators provides economic, social, and natural re- Table 24, Total external debt ratios, includes two new source indicators for selected periods or years for 200 indicators: concessional debt as a percentage of total economies and various analytical and geographic external debt, and multilateral debt as a percentage of groups of economies. Although most of the data col- total external debt. These two sources of external fi- lected by the World Bank are on low- and middle- nancing are shown separately to reflect their impor- income economies, comparable data for high-income tance for many developing economies. economies, where readily available, are also included In Table 28, Health and nutrition, the prevalence of in the tables. Additional information may be found in malnutrition in children under 5 is used; it is consid- the World Bank Atlas, World Tables, World Debt Tables, ered a better measure of nutritional status than the and Social Indicators of Development. These data are previously used indicator, daily calorie supply. now also available on diskette, in the World Bank's Table 30, on income estimates, has been refined to *STARS* retrieval system. make use of the purchasing power of currencies Changes in this edition (PPCs). It contains country-specific observations as well as data derived from a regression equation. With the independence of several new economies during the past year, coupled with space limitations Classification of economies in the main tables, a new criteriondata availability The main criterion used to classify economies and has been introduced. To be included in the main ta- broadly distinguish different stages of economic de- bles, an economy must have reasonable coverage of velopment is GNP per capita. This year the per capita key socio-economic indicators. Basic indicators for income groups are low-income, $635 or less in 1991 economies excluded for lack of data are presented, (40 economies); middle-income, $636 to $7,909 (65 along with countries with fewer than 1 million popu- economies); and high-income, $7,910 or more (22 lation, in Table la, following Table 33. economies). Economies with populations of fewer To preserve a 20-year interval between the two than 1 million and those with sparse data are not years shown for most indicators, the earliest year pre- shown separately in the main tables but are included sented has been changed from 1965 to 1970 or 1975. in the aggregates. Basic indicators for these econ- Readers wanting data for earlier periods can refer to omies may also be found in Table la. previous editions or to the publications noted above, Further classification of economies is by geographic which present data in time series. location. "Europe and Central Asia" now includes The following changes have also been made. (They the newly independent economies of the former So- are described more fully in the technical notes.) viet Union. Economies formerly grouped in "Other Estimates of fish products as a percentage of daily economies" are now included in the appropriate in- protein supply have been added to Table 4, Agricul- come and geographic groupings. Other classifications ture and food, because fish is an important source of include severely indebted middle-income economies protein for some countries and is not included in the and fuel exporters. (See Definitions and data notes, food production per capita estimates in the table. and the tables on classification of economies at the Table 21, Total external debt, includes two new indi- back of this book.) cators: total arrears on long-term debt outstanding and disbursed, and ratio of present value to nominal Methodology value of debt. Total arrears denotes principal and in- The Bank continually reviews methodology in an ef- terest due but not paid, and the present value to fort to improve the international comparability and nominal value of debt is a measure of the degree of analytical significance of the indicators. Differences 233 between data in this year's and last year's edition third or more of the overall estimate, however, the reflect not only updates but also revisions to histori- group measure is reported as not available. The cal series and changes in methodology. weightings used for computing the summary mea- All dollar figures are U.S. dollars unless otherwise sures are stated in each technical note. stated. The various methods used for converting from national currency figures are described in the Terminology and data coverage technical notes. In these notes the term "country" does not imply Summary measures political independence but may refer to any territory whose authorities present for it separate social or eco- The summary measures in the colored bands are to- nomic statistics. tals (indicated by t), weighted averages (w), or me- The unified Germany does not yet have a fully dian values (m) calculated for groups of economies. merged statistical system. Throughout the tables, Countries for which individual estimates are not data for Germany are footnoted to explain coverage; shown, because of size, nonreporting, or insufficient most economic data refer to the former Federal Re- history, have been implicitly included by assuming public, but demographic and social data generally re- that they follow the trend of reporting countries dur- fer to the unified Germany. The data for China do not ing such periods. This gives a more consistent aggre- include Taiwan, China, but footnotes to Tables 14, 15, gate measure by standardizing country coverage for 16, and 18 provide estimates of the international each period shown. Group aggregates include coun- transactions for Taiwan, China. In all tables, tries for which country-specific data do not appear in Czechoslovakia refers to the former Czechoslovakia the tables. Where missing information accounts for a and Yugoslavia to the former Socialist Federal Repub- Groups of economies For this map, economies are classified by income group, as they are for the tables that follow. Low-income economies are those with a GNP per capita of $635 or less in 1991; middle-income, $636-7,910; high-income, $7,911 or more. I Low-income economies Middle-income economies High-income economies Data not available Ki Bryal) Peru (fn) Sam,, (US. Bolivia Tonrga on I) A,5vt and Barbud, Paraguay US V,ngrn 55055 (US) St KiSs Svad,)oupe (Fr) and Nears Dominic, Chile 5,155,5 055 MartThrqu, (Fr) AnUSes )N,(U Ancb, S Ut Laci, (trUth) St. Vincent S B,tbados .Sdrsn,da Thn)d,d and 1,0550 234 lic of Yugoslavia because disaggregated data are not those classified by the United Nations or otherwise yet available. regarded by their authorities as developing. Table content Technical notes The indicators in Table 1 give a summary profile of The technical notes and the footnotes should be re- economies. Data in the other tables fall into six broad ferred to in any use of the data. The notes outline the areas: production, domestic absorption, fiscal and methods, concepts, definitions, and data sources monetary accounts, core international transactions, used in compiling the tables. A bibliography at the external finance, and human and natural resources. end of the notes lists the data sources, which contain The table format of this edition follows that of pre- comprehensive definitions and descriptions of con- vious years. In each group, economies are listed in cepts used. Country notes to the World Tables provide ascending order of GNP per capita, except those for additional explanations of sources used, breaks in which no such figure can be calculated. These are comparability, and other exceptions to standard sta- italicized and in alphabetical order at the end of the tistical practices that Bank staff have identified in na- group deemed appropriate. This order is used in all tional accounts and international transactions. tables except Table 19, which covers only high-in- come OPEC and OECD countries. The alphabetical Comments and questions relating to the World list in the key shows the reference number for each Development Indicators should be addressed to economy; here, too, italics indicate economies with Socio-Economic Data Division, International Eco- no current estimates of GNP per capita. Economies in nomics Department, The World Bank, 1818 H Street, the high-income group marked by the symbol are N.W., Washington, D.C. 20433. FPaoc Is Ida Iceland NelharIae Russian Federation ale of Murr (UIr lrsiarrd Clrarrre&lslaoda (UK) Kaaalrhntan Luxembourg Mangaira LrOsfrtefl stern Uabekrstaa KFrgyaatao Ondorra Dell Fr0p Arrnonur Aaeor Par toga erstan - Rnp. or Korea ReP. ci SiSraItae (UK) Knre&, Arab 'i-I Chrea Rep. Iraq )( Afghan rSlae Pakistan -V Nepal Sitar UdiArair Lax Peapl,'s Unite Sangledexlr Dent, Rep IndIa Myanmar\ Hong Kaln5 (UK) Mauritania g Maoaa (Foot) C ape Uerdn Sret Na Sudan She Gambia il/p pie en Canr bad Uu,m (US)' \ Guinea-Oissau Go flea Sierra Leone Ceethal Atrrcan Loeka I. r1 Federated Staten at Micronesia rM1rahall stands Stun Libel Repub lie ToOl Equalorral Gureea Uganda Kenya San TarE and PrrncrpeN Naueu0 Kiribalr 0 Rwaeda Zaire Suruedi Tanoanra bras Seychelles Zaanbln May05, Malaw (Fe) Mnaaorbi, ieebnbWe Mauridua IIeuoiao (Fe) alec South A man New Basera ar°'°1 BulQanio Zealand Sal Mar 110 Heezagovina' Fed Rep at \ V - 0050sIaala Italy sun. an Mar,daCl Albania 'OraaggreaawaauuaarrnrntrwarlaaarerrMraoaerCaeaaaalaaaua 'a aareaareu Mu are oar nor ear/Jew ror roe rower s raaar 0608/Jr RoMe/Ic an nuqaauea 235 Population density Population per square kilometer 0-19 20-49 For this map, population density is calculated by 50-199 dividing a country's population by its total surface area 200 or more (square kilometers of land and inland water area). See Table 1 for the population and area of the 127 economies in the main tables, and Table la for an additional 73 Data not available economies. Fertility and mortality Total fertility Infant mortality Life expectancy Births per woman Deaths per 1,000 live births Years 8 150 80 100 n 50 0 1 1970 1991 2000 1970 1980 1991 1970 1980 1991 - Lowincome economies - Highincome economies Middleincome economies Note: For explanation of terms or methods, see the technical notes for Tables 27, 28, and 32. 236 Share of agriculture in GDP Less than 6 percent I 6-9percent 10-19 percent 20-29 percent 30 percent or more Share of agriculture in GDP is calculated by taking the value added of an economy's agriculture sector and dividing it by gross domestic product. The shares say nothing about absolute values of production. For economies with high levels of subsistence farming, the share of agriculture in GDP is difficult to measure because of difficulties in assigning subsistence farming its appropriate value. For more details, see the technical Data not available note for Table 3. Median age atdeath, 1990 Years of age 10-29 30-59 60-74 75 or more :0-9 The median age at death measures the age below which half of all deaths occur in a year. For more information, [ 1 Data not available see Table A.3 in the Population and Health Data Appendix. 237 Table 1. Basic indicators GNPper capitaa Average annual Lf Area Average annual expectancy Adult Iltreracy Population (thousands growth rate rate of nflaft ona (percent) at birth (millions) ofsquare Dollars (percent) (percent) (years) Female Total mid-1991 kilometers) 1991 1980-91 1970-80 1980-91 1991 1990 1990 Low-income economies 3,127.3: 38,828: 350 w 39 w 8.2 w 12.6 w 62 w 52w 40w China and India 2,016.0 12,849 350 5.6 w 4.3 w 6.9 66w SOw 37w Other low-income 1,111.21 25,980: 350 1.0w 15.7 w 23.4 w 55 56w 45w I Mozamb,que 16.1 802 80 -1.! . . 37.6 47 79 67 2 Tanzania 25.2 945 100 -0.8 14.1 25.7 51 3 Ethiopia 52.8 1,222 120 -1.6 4.3 2.4 48 . 4 Uganda 16.9 236 170 . . 46 65 52 5 Bhulan 1.5 47 180 8.4 48 75 62 6 Guinea-Bissau 1.0 36 180 1.1 5.7 56.2 39 76 64 7 Nepal 19.4 141 180 2.1 8.5 9.1 53 87 74 8 Bumndi 5.7 28 210 1.3 10.7 4.3 48 60 50 9 Chad 5.8 1,284 210 3.8 7.7 1.! 47 82 70 10 Madagascar 12.0 587 210 -2.5 9.9 16.8 51 27 20 II Sierra Leone 4.2 72 210 -1.6 12.5 59.3 42 89 79 12 Bangladesh 110.6 144 220 1.9 20.8 9.3 51 78 65 13 Lao PDR 4.3 237 220 . . . . . . 50 14 Malawi 8.8 118 230 0.1 8.8 14.9 45 . 15 Rwanda 7.1 26 270 -2.4 15.1 4.1 46 63 50 16 Mali 8.7 1,240 280 -0.1 9.7 4.4 48 76 68 17 BurkinaFaso 9.3 274 290 1.2 8.6 3.8 48 91 82 18 Niger 7.9 1,267 300 -4.1 9.7 2.3 46 83 72 19 India 866.5 3,288 330 3.2 8.4 8.2 60 66 52 20 Kenya 25.0 580 340 0.3 10.1 9.2 59 42 31 21 Nigeria 99.0 924 340 -2.3 15.2 18.1 52 61 49 22 China 1,149.5 9,561 370 7.8 0.9 5.8 69 38 27 23 Haiti 6.6 28 370 -2.4 9.3 7.1 55 53 47 24 Benin 4.9 113 380 -0.9 10.3 1.6 51 84 77 25 CentralAfricanRep. 3.1 623 390 -1.4 12.1 5.1 47 75 62 26 Ghana 15.3 239 400 -0.3 35.2 40.0 55 49 40 27 Pakistan 115.8 796 400 3.2 13.4 7.0 59 79 65 28 Togo 3.8 57 410 -1.3 8.9 4.4 54 69 57 29 Guinea 5.9 246 460 . . . . . . 44 87 76 30 Nicaragua 3.8 130 460 -4.4 12.8 583.7 66 31 Sri Lanka 17.2 66 500 2.5 12.3 11.2 71 17 12 32 Mauritania 2.0 1,026 510 -1.8 9.9 8.7 47 79 66 33 Yemen,Rep. 12.5 528 520 . . . . . . 52 74 62 34 Honduras 5.3 112 580 -0.5 8.1 6.8 65 29 27 35 Lesotho 1.8 30 580 -0.5 9.7 13.6 56 36 Indonesia 181.3 1,905 610 3.9 21.5 8.5 60 32 23 37 Egypt, Arab Rep. 53.6 1,001 610 1.9 9.6 12.5 61 66 52 38 Zimbabwe 10.1 391 650a -0.2 9.4 12.5 60 40 33 39 Sudan 25.8 2,506 14.5 51 88 73 40 Zambia 8.3 753 . . 7.6 . . 49 35 27 Middle-income economies 1,401.0 t 40,7961 2,480 w 0.3 w 28A w 67J 68w 26w 21w Lower-middle-income 773.8: 19,3091 1,590 w -0.1 w 22.8 w 23.1 w 67 w 32w 26w 41 Bolivia 7.3 1,099 650 -2.0 21.0 263.4 59 29 23 42 Côte d'Ivoire 12.4 322 690 -4.6 13.0 3.8 52 46 60 43 Senegal 7.6 197 720 0.1 8.5 6.0 48 75 62 44 Philippines 62.9 300 730 -1.2 13.3 14.6 65 II 10 45 Papua New Guinea 4.0 463 830 -0.6 9.1 5.2 56 62 48 46 Cameroon 11.9 475 850 -1.0 9.8 4.5 55 57 46 47 Guatemala 9.5 109 930 -1.8 10.5 15.9 64 53 45 48 Dominican Rep. 7.2 49 940 -0.2 9.1 24.5 67 18 17 49 Ecuador 10.8 284 1,000 -0.6 13.8 38.0 66 16 14 50 Morocco 25.7 447 1,030 1.6 8.3 7.1 63 62 5! 51 Jordan' 3.7 89 1,050 -1.7 1.6 69 30 20 52 Tajikistandl 5.5 143 1,050 . . . . . . 69 . 53 Peru 21.9 1,285 1,070 2.4 30.1 287.3 64 21 15 54 El Salvador 5.3 21 1,080 -0.3 10.7 17.4 66 30 27 55 Congo 2.4 342 1,120 -0.2 8.4 0.4 52 56 43 56 SyrianArabRep. 12.5 185 1,160 -1.4 11.8 14.3 67 49 36 57 Colombia 32.8 1,139 1,260 1.2 22.3 25.0 69 14 13 58 Paraguay 4.4 407 1,270 -0.8 12.7 25.1 67 12 10 59 Uzbekistan'1 20.9 447 1,350 . . . . . . 69 . 60 Jamaica 2.4 II 1,380 0.0 17.3 19.6 73 1 2 61 Romania 23.0 238 1,390 0.0 . . 6.2 70 62 Namibia 1.5 824 1,460 -1.2 . . 12.6 58 . 63 Thnisia 8.2 164 1,500 1.1 8.7 7.3 67 44 35 64 Kyrgyzstan" 4.5 199 1,550 . . . . . . 66 . 65 Thailand 57.2 513 1,570 5.9 9.2 3.7 69 10 7 66 Georgiad 5.5 70 1,640 . . . 73 Note. For other economies see Table Ia. For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 238 GNPpercapitau Life Area Average annual Average annual Adult illiteracy expectancy rate of inflationa (percent) Population (thousands growth rate at birth (millions) of square Dollars (percent) (pe,re,tt) (years) Female Total mid-1991 kilometers) 1991 1980-91 1970-80 1980-91 1991 1990 1990 67 Azerbaijan' 7.1 87 1,670 71 68 Turkmenistand 3.8 488 1,700 . . . . 66 . 69 Turkey 57.3 779 1,780 2.9 29.4 44.7 67 29 19 70 Poland 38.2 313 1,790 0.6 . 63.1 71 . 71 Bulgaria 9.0 111 1,840 1.7 . . 7.8 72 . 72 CostaRicu 3.1 51 1,850 0.7 15.3 22.9 76 7 7 73 Algeria 25.7 2,382 1,980 -0.7 14.5 10.1 66 55 43 74 Panama 2.5 77 2,130 -1.8 7.5 2.4 73 12 12 75 Armenia' 3.4 30 2,150 . . . . . . 72 . 76 Chile 13.4 757 2,160 1.6 188.1 20.5 72 7 7 77 Iran, Islamic Rep. 57.7 1648 2,170 -1.3 22.4 13.8 65 57 46 78 Moldovad 4.4 34 2,170 . . 69 79 Ukraincd 52.0 604 2,340 . . . . 70 80 Mauritius 1.1 2 2,410 6.1 15.3 8.1 70 81 CzechoslovakiaC 15.7 128 2,470 0.5 . . 3.5 72 82 Kuzakhstan' 16.8 2,717 2,470 . . . . . . 69 . 83 Malaysia 18.2 330 2,520 2.9 7.3 1.7 71 30 22 Upper-middle-income 627.0 1 21,486 1 3,530 w 0.6 w 31.7 w 95.4 w 69 w 17 w 14 w 84 Botswana 1.3 582 2,530a 5.6 11.6 13.2 68 35 26 85 South Africa 38.9 1,221 2,560 0.7 13.0 14.4 63 86 Lithuania'' 3.7 65 2,710 . . . . . . 71 87 Hungary 10.3 93 2,720 0.7 2.8 10.3 70 . 88 Venezuela 19.8 912 2,730 -1.3 14.0 21.2 70 10 12 89 Argentina 32.7 2,767 2,790 -1.5 133.9 416.9 71 5 5 90 Uruguay 3.1 177 2,840 -0.4 65.1 64.4 73 4 4 91 Brazil 151.4 8,512 2,940 0.5 38.6 327.6 66 20 19 92 Mexico 83.3 1,958 3,030 -0.5 18.1 66.5 70 15 13 93 Belams1 10.3 208 3,110 . . . . . . 71 94 Russian Federationd 148.7 17,075 3,220 . . . . 69 95 Latvia 2.6 65 3,410 . . . . . 69 96 TrinidadandTobago 1.3 5 3,670 -5.2 18.5 6.5 71 97 Gabon 1.2 268 3,780 -4.2 17.5 1.5 54 52 39 98 Estonia' 1.6 45 3,830 . . . . . . 70 99 Portugal 9.9 92 5,930 3.1 16.7 17.4 74 19 15 100 Oman 1.6 212 6,120 4,4 28.0 -3.1 69 101 Puerto Rico 3.6 9 6,320 0.9 6.5 3.4 76 102 Korea, Rep. 43.3 99 6,330 8.7 20.1 5.6 70 7 4 103 Greece 10.3 132 6,340 1.1 14.5 17.7 77 II 7 104 Saudi Arabia 15.4 2,150 7,820 -3.4 24.9 -2.4 69 52 38 105 YugoslaviaC 23.9 256 18.4 123.0 73 12 7 Low- and middle-income 4,528.0 79,624 1,010w 1.0 w 21.8 w 53.9 w 64w 46w 35 '4 Sub-Saharan Africa 488.9 23,066 350 w -1.2 w 13.9 w 18.4 w 51 w 62 w 50 a East Asia & Pacific 1,666.51 16,369 650 w 6.1 w 9.1 w 6.3 w 68 w 34 w 24 a South Asia 1,152.2 5,133 320 w 3.1 w 9.7 w 8.3 w 59 w 69 w 54 '4 Europe and Central Asia 492.0 2,3141 2,670 w 0.9 w 18.7 w 18.2 w 70 w 22 w 16 w Middle East & N. Africa 244.1 11,015 1 1,940 w -2.4 w 18.8 w 8.6 w 64w 57 w 45 '4 Latin America & Caribbean 445.3 20,507 2,390 w -0.3 w 43.1 w 208.2 w 68 w 17w 16 w Severely indebted 486.2 1 23,5741 2,350 w -1.0 w 39.1 w 189.6 w 67w 27w 22w High-income economies 822.3 31,682 21,050 w 2.3 w 9.1w 4.5w 77 w 5w 4w OECD members 783.1 t 31,1351 21,530 w 2.3 w 9.0w 4.3w 77 w 5w 4w 106 Ireland 3.5 70 11,120 3.3 14.2 5.8 75 107 '(Israel 4.9 21 11,950 1.7 39.5 89.0 76 108 New Zealand 3.4 271 12,350 0.7 12.5 10.3 76 109 Spain 39.0 505 12,450 2.8 16.1 8.9 77 110 tHong Kong 5.8 l3,430 5.6 9.2 7.5 78 Ill '(Singapore 2.8 1 14,210 5.3 5.9 1.9 74 . 112 United Kingdom 57.6 245 16,550 2.6 14.5 5.8 75 f 113 Australia 17.3 7,687 17,050 1.6 11.8 7.0 77 114 Italy 57.8 301 18,520 2.2 15.6 9.5 77 115 Netherlands 15.1 37 18,780 1.6 7.9 1.8 77 116 Belgium 10.0 31 18,950 2.0 7.8 4.2 76 117 Austria 7.8 84 20,140 2.1 6.5 3.6 76 f f 118 France 57.0 552 20,380 1.8 10.2 5.7 77 f f 119 Canada 27.3 9,976 20,440 2.0 8.7 4.3 77 120 United States 252.7 9,373 22,240 1.7 7.5 4.2 76 121 Germanyc 80.1 357 23,65oh 22h 51h 2.8 76 122 Denmark 5.2 43 23,700 2.2 10.1 5.2 75 f 123 Finland 5.0 338 23,980 2.5 12.3 6.6 76 124 Norway 4.3 324 24,220 2.3 8.4 5.2 77 f 125 Sweden 8.6 450 25,110 1.7 10.0 7.4 78 f 126 Japan 123.9 378 26,930 3.6 8.5 1.5 79 127 Switzerland 6.8 41 33,610 1.6 5.0 3.8 78 World 5,351.01 111,3061 4,010 w 1.2 w 11.2w 15.4w 66w 45w 35w Fuel exporters 262.81 12,3871 1,990 w -3.1 w 19.6w 9.6w 60w 54w 44w t Economies classified by the United Nations or otherwise regarded by their authorities as developing, a. See the technical notes. b. In all tables GDP and GNP data cover mainland Tanzania only. c. In all tables, data for Jordan cover the East Bank only. d. Estimates for economies of the former Soviet Union are subject to more than the usual range of uncertainty and should be regarded as very preliminary. e. See the footnotes to the Key for data coverage. f. According to UNESCO, illiteracy is less than 5 percent. g. Data refer to ODE h. Data refer to the Federal Republic of Germany before unification. 239 Table 2. Growth of production Average annual growth rate (percent) GDP Agriculture Industry Manufacturinga Services, etc.b 1970-80 1980-91 1970-80 1980-9) 1970-80 1980-91 1970-80 /980-91 1970-80 1980-9/ Low-income economies 4.5 w 6.0 w 2.1 w 3.7 w 6.3 w 7.5 w 7.2 w 9.3 w 5.7 w 7.0 w China and India 4.3 w 7.5 w 2.2 w 4.4 w 6.3 w 9.3 w 7.6 w 9.9 w 5.2 w 8.8 w Other low-income 4.8w 3.7w 2.1 w 2.5 w 6.4 w 4.0w 5.7w 6.8 w 6.5 w 4.8 w I Mozambique . . -0.! . . 1.6 . . -3.6 . . . . . . -1.7 2 Tanzania 3.0 2.9 0.7 4.4 2.6 -2.4 3.7 -0.8 8.4 2,0 3 Ethiopia 1.9 1.6 0.7 0.3 1.6 1.8 2.5 1.9 3.9 3.! 4 Uganda . . . . . . . . . . . . . . . . . 5 Bhutan . . 7.6 . . 4.8 . . 14.8 . . /5.2 . . 7.3 6 Guinea-Bissau 2.4 3.7 -1.2 5.0 2.1 2.6 /1.0 2.8 7 Nepal . . 0.5 4.9 8 Bunindi 4.2 4.0 3.2 3.1 11.6 4.6 3.8 5.8 3.5 5.4 9 ChadC 0. I 5.5 -0.4 3.4 -2./ 7.1 2.2 7.3 10 Madagascar 0.5 1.! 0.4 2.4 0.6 0.9 0.6 0.2 II Sierra Leone 1.6 1,1 6.0 2.7 -3.2 -0.8 -2.1 -1.4 2.3 0.9 12 Bangladeshc 2.3 4.3 0.6 2.7 5.2 4.9 5.1 2.9 3.8 5.6 13 Lao PDRC 14 Malawi 5.8 3.! 4.4 2.4 6.3 3.3 . . 3.9 7.0 3.7 IS Rwandac 4,7 0.6 7.1 -1.5 0.5 . . 4.9 0.5 . . 3.1 16 Malic 4.9 2.5 4.2 2.4 2.0 4.0 6.9 2.4 17 BurkinaFaso 4.4 4.0 1.0 3.2 2.5 3.8 4. 1 2.6 /9.9 5.4 18 Niger 1.7 -1.0 -3.7 11.3 2.9 19 India 3.4 5.4 1.8 3.2 4.5 6.3 4.6 4.6 6.7 20 Kenya 6.4 4.2 4.8 3.2 8.6 4.0 9.9 4.9 6.8 4.9 21 Nigeria 4.6 1.9 -0.1 3.5 7.3 -0.4 5.2 . 9.6 3.! 78d . 22 ChinaC 5.2 9.4 2.6 57 II ,od 9.5 /1.1 6.1 11.2 23 Haiti 3.7 -0.7 . . . . . . . . . . . . . 24 BeninC 2.2 2.4 1.8 4.9 1.4 3.6 . . 4.8 2.7 0.5 25 Central African Rep. 2.4 1.4 I.9 2.4 4.1 3.2 . . . . 2.3 -0.2 26 Ghana -0.1 3.2 -0.3 1.2 -1.0 3.7 -0.5 4.1 1.1 6.6 27 Pakistan 4.9 6.1 2.3 4.2 6. I 7.5 5.4 7.8 6.3 6.6 28 Togo 4.0 1.8 1.9 5.3 7.7 1.5 . . 2.5 3.6 -0.2 29 Guineac 30 NicaraguaC 1.1 -1.9 1.9 -2.2 -1.7 2.8 1.1 -3.1 0.4 -1.7 31 Sri Lanka 4.1 4.0 2.8 2.3 3.4 4.7 1.9 6.3 5.7 4.6 32 Mauritania 1.3 1.4 -1.0 0.7 0.5 4.9 3.7 0.5 33 Yemen, Rep.' . . . . . . . . . , . . . . . . , 34 Honduras 5.8 2.7 2.2 2.9 6.7 3.3 6.9 3.7 7.1 2.5 35 Lesotho 8.6 5.5 0.2 1.8 27.8 8.2 /8.0 12.8 13.6 5.3 36 lndonesia 7.2 5.6 4.1 3.1 9.6 5.9 14.0 12.3 7.7 6.8 37 Egypt, Arab Rep. 9.5 4.8 2.8 2.4 9.4 4.2 . . . . 17.5 6.2 38 Zimbabwe L6 3.1 0.6 2.2 1.1 2.1 2.8 3.1 2.4 4.0 39 Sudan 5.6 3.3 4.5 3.9 . . 8.1 40 Zambiac 1.4 0.8 2.1 3.3 1.5 0.9 2.4 3.7 1.2 0.0 Middle-income economies 2.3 w Lower-middle-income 2.7 w 41 Boliviac 4.5 0.3 3.9 1.8 2.6 -0.8 6.0 -0.! 6.8 -0.1 42 Côte d'Ivoire 6.6 -0.5 2.7 -1.2 9.1 -1.6 . . . . 10.9 0.8 43 Senegal' 2.3 3.1 1.3 2.7 5.3 3.8 2.4 5.1 2.0 PhilippinesC 3.0 44 6.0 1.1 4.0 1,1 8.2 -0.5 6.1 0.4 5.1 2.8 45 PapuaNewGuineac 2.2 2.0 2.8 1.6 2.4 0.3 . . . . 1.8 46 Cameroonc 7.2 1.4 4.0 1.1 10.9 2.2 7.0 . . 7.8 1,1 47 Guatemalac 5.8 4.6 -0.2 1.1 1.2 7.7 6.2 -7.0 5.6 1.0 48 Dominican Rep.c 6.5 1.7 3.1 0.3 8.3 1.6 6.5 0.6 7.2 2.3 49 Ecuador 9.6 2.1 2.8 4.4 13.9 1.1 10.5 -0.1 9.5 2.1 50 Moroccoc 5.6 4.2 6.8 1.1 6.5 3.0 . . 4.2 7.0 4.2 51 Jordan . . -1.5 8.1 . . -0.2 . . 1.4 -2.7 52 Tajikistan . . . . . . . . . . . . . . . . . 53 Peruc 3.5 -0.4 0.0 2.2 4.4 -1.1 3.1 -0.7 4.6 -0.9 54 El SaIvador 4.2 1.0 3.4 -0.4 5.2 1.6 4.1 1.4 4.0 1.1 55 Congoc 58 3.3 2.5 3.3 10.3 4.7 . . 6.3 4.5 2.4 56 Syrian Arab Rep.c 9.9 2.6 8.6 -0.6 9.0 6.8 . . . . II.! 1.6 57 Colombia 5.4 3.7 4.6 3.2 5.1 4.8 5.8 3.4 5.9 3.1 58 Paraguayc 8.5 2.7 6.2 3.6 11.2 0.2 7.9 2.2 8.6 3.5 59 Uzbekistan . . . . . . . . . . . . . . . . . 60 Jamaica' -1.2 1.6 0.3 1.0 -3.4 2.6 -2.1 2.5 0.0 1.1 6! Romania 0.1 0.1 -0.8 1.9 62 Namibia 1.0 0.3 -2.0 1.7 3.1 63 Tunisia 6.3 3.7 4. 3.1 6.8 2.9 10.3 6.2 4.3 64 Kyrgyzstan 65 Thailand' 7.l 7.9 4.4 3.8 9.5 9.6 10.5 9.4 7.2 8.0 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 240 Average annual growth rate (percent) GDP Agriculture industry Manufaczuringa Services, etc. b /970-80 /980-91 /970-80 1980-91 1970-80 /980-91 1970-80 1980-91 /970-80 /980-91 67 Azethaijan . . . . . . . . . . . 68 Turkmenistan . . . . . . . . . . . - . . . . 69 Turkey 5.9 5.0 3.4 3.0 6.6 6.0 6.1 7.2 6.5 5.0 70 PolandC 1.1 . . . . . . . 7! Bulgaria 1.9 . . -2.5 . . 3.2 . . . . . . 1.4 72 Costa Ricac 5.7 3.! 2.5 3.3 8.2 2.9 . . 3.1 5.9 3.2 73 Algeria 4.6 3.0 7.5 5.0 3.8 1.8 7.6 3.3 5.0 3.6 74 Panamac 4.4 0.5 2.0 2. I 3.9 -5.7 2.8 -0.4 5.0 1.6 75 Armenia Chilee .. . . . . . . . . . . . . . . . 76 1.4 3.6 3.1 4.! -0.! 3.6 -0.8 3.6 2.3 3.4 77 Iran, islamic Rep. 2.2 2.2 3.9 2.5 -4.8 3.9 6.4 -0.1 5.6 1.6 78 Moldova . . . . . 79 Ukraine . . . . . . . . . . . . . . . . . 80 Mauntius 6.8 6.7 -3.3 3.2 10.4 10.1 7.1 11.2 10.9 5.8 81 CzechoslovakiaC . 0.6 -0.4 . . 0.3 . . . . . . 1.2 82 Kazakhstan . . . . . . . . . . . . . . . . 83 Malaysia 7.9 5.7 . . 3.7 . . 7.7 . . 9.6 . . 4.7 Upper-middle-income 6.1 w 2.1 w 84 BotswanaC 14.5 9.8 8.3 3.0 17.6 10.7 22.9 7.5 14.5 10.3 85 South Africa 3.0 1.3 3.2 2.6 2.3 0.0 4.7 -0.1 3.8 2.5 86 Lithuania . . . . . . . . . . . . . . 87 HungaryC 5.2 0.6 2.8 0.9 6.3 -1.6 . . . . 5.2 2.4 88 VenezuelaC 3.5 1.5 3.4 2.8 0.5 2.1 5.7 1.3 6.3 0.9 89 Argentina 2.5 -0.4 2.5 1.5 2.1 -1.4 . . . . 2.9 0.1 90 UruguayC 3.0 0.6 0.8 0.2 4.1 0.0 . . 0.4 2.9 1.2 91 Brazil 8.1 2.5 4.2 2.6 9.4 1.7 9.0 1.7 7.8 3.2 92 MCXICOC 6.3 1.2 3.2 0.5 7.2 1.3 7.0 1.8 6.3 1.3 93 Belarus . . . . . . . . . . . . . 94 Russian Federation 95 Latvia 96 Trinidad and Tobago 5.9 -4.4 -1.4 -6.8 5.6 -6.6 1.7 -8.7 7.4 -2.3 97 Gabonv 9.0 0.2 0.9 . . 1.1 -1.5 -0.8 98 Estonia 99 Portugalv 4.3 2.9 .. . . . . . . . . . . . 100 Oman' 6.2 7.9 .. 7.1 . . 9.6 . . 18.3 . . 6.0 10! Puerto Rico 3.9 4.1 2.3 2.2 5.0 3.6 7.9 1.0 3.2 4.6 102 Korea, Rep.c 9.6 9.6 2.7 2.! 15.2 12.1 17.0 12.4 8.8 9.3 103 Greece 4.7 1.8 1.9 0.2 5.0 1.2 6.0 0.4 5.6 2.5 104 Saudi Arabiav 10.1 -0.2 5.3 14.0 10.2 -2.9 6.4 8.! 10.3 -0.2 105 Yugoslavia 6.0 0.8 3.! 0.6 8.0 0.8 .. . . 4.9 1.0 Low- and middle-income 5.3w 3.3w 2.7w 2.8w . . 3.3 w . . 6.2w 3.5w Sub-Saharan Africa 4.0 w 2.1 w 1.5 w 1.8 w 5.3 w 2.0 w 3.5 w . . 5.5 w 2.5 w East Asia & Pacific 6.6 w 7.7 w 3.1 w 4.3 w 9.4 w 9.4 w 10.5 w 10.6 w 7.1 w 8.6 w South Asia 3.5w 5.4w 1.8w 3.3w 4.6w 6.4w 4.6w 6.7w 4.7w 6.6w Europe and Central Asia . . 1.5 w . . . . . . . . . . . Middle East & N. Africa 5.2 w 2.1 w 4.2 w 3.6 w 3.2 w 0.9 w . . 3.5 w 7.2 w 2.1 w Latin America & Caribbean 5.5 w 1.7 w 3.5 w 1.9 w 6.1 w 1.4 w 8.0 w 1.3 w 6.1 w 2.0 w Severely indebted 6.1w 1.7w 3.8w 1.5w 6.7w 1.4w 8.2w 2.0w 6.4w 2.2w High-income economies 3.2 w 2.9 w 0.8 w 2.7 w 3.4 w 3.8 w OECD members 3.1 w 2.9 w 0.8 w 2.7 w 3.3 w 3.8 w 106 Ireland 4.9 3.5 . . . . . . . . . . . 107 tlsraelc 4.8 3.7 . . . . . . . . . 108 New Zealandc 1.9 1.5 . . 3.8 . . 1.3 . . 0.7 . . 1.6 109 Spain 3.5 3.2 . . . . . . . . . . . . . 110 tHong Kong 9.2 6.9 . . . . . . . . . . . . Ill tSingaporev 8.3 6.6 1.4 -6.6 8.6 5.8 9.7 7.0 8.3 7.3 112 United Kingdom 2.0 2.9 113 Australia 3.0 3.1 2.9 30 2.1 3.6 114 1taIy 3.8 2.4 0.9 0.5 3.6 2.1 2.9 4.0 2.7 115 Nethcrlandsc 2.9 2.! 3.9 3.7 2.3 3.8 1.8 116 Belgiumc 3.0 2.1 .. 1.5 . . 2.2 . . 3.0 . . 1.9 117 AustriaC 3.4 2.3 2.6 1.0 3.1 2.0 3.2 2.6 3.7 2.2 118 FtanceC 3.2 2.3 . . 1.9 . . 0.9 . . 0.6 . . 2.9 119 Canada 4.6 3.! 1.2 1.6 3.2 3.0 3.5 3.1 6.6 3.3 120 United States' 2.8 2.6 0.6 . . 2.1 3.0 . . 3.3 12! Germany 2.6 2.3 1.1 1.8 1.7 0.9 2.0 1.4 3.5 2.6 122 Denmark 2.2 2.3 2.3 3.2 1.1 2.9 2.6 1.4 2.7 2.1 123 Finland 3.1 3.0 0.2 -0.2 3.0 3.0 3.3 3.] 3.9 3.4 124 Norway 4.8 2.7 1.3 0.8 7.1 5.2 1.2 0.6 3.9 1.1 125 Sweden 1.9 2.0 . . 1.6 2.8 . . 2.5 . . 1.3 126 Japan 4.3 4.2 -0.2 1.2 4.0 4.9 4.7 5.6 4.9 3.7 127 Switzerlandc 0.5 2.2 World 3.5 w 3.9w 1.9 w 2.6 w 3.2 w . 3.9 w . . 4.1 w . Fuel exporters 5.0 w 1.3 w 3.2 w 3.6w 3.3 w -0.5 w 6.5w 2.3w 6.9w 1.5 w a. Because manufacturing is generally the most dynamic part of the industrial sector, its growth rate is shown separately. b. Services. etc. includes unallocated items. c. GDP and its components are at purchaser values. d. World Bank estimate. e. Data refer to the Federal Republic of Germany before unification. 241 Table 3. Structure of production GDP Distribution of gross domestic product (percent) (millions of dollars) Agriculture Industry Manufacturing a Services, etc. b 1970 1991 1970 1991 1970 1991 1970 1991 1970 1991 Lew-hicome ecoflomles 225,563 920,160 38w 29w 29w 34w 20w 26w 33w 38w China and India 146,193 591,577 38w 29w 32w 36w 24w 30w 29w 35w Other low-income 79,034 t 338,471 t 39 w 29 w 21 w 29 w 10 w 40 w 42 w 1 Mozambique . . 1,219 . . 64 . . 15 . . . . 21 2 Tanzania 1,174 2,223 41 61 17 5 10 4 42 34 3 Ethiopia 1,669 5,982 56 47 14 13 9 9 30 40 4 Uganda 1,286 2,527 51 . . 12 4 . . 37 5 Bhutan 240 43 . . 27 . . 10 29 6 Guinea-Bissau 79 211 47 46 2! 12 21 8 3! 42 7 Nepal 861 3,063 67 59 12 14 4 5 21 27 8 Burundi 225 1,035 71 55 10 16 7 12 19 29 9 Chadc 302 1,236 47 43 18 18 17 16 35 39 10 Madagascar 995 2,488 24 33 16 14 . . . 0 59 53 11 SierraLeone 383 743 28 43 30 14 6 3 42 43 12 Bangtadeshc 6,664 23,394 55 36 9 16 6 9 37 48 13 La0PDRC . 1,027 . . . . . . . . . . . . . 0 14 Malawi 271 1,986 44 35 17 20 . . 13 39 45 15 Rwanda' 220 1,579 62 38 9 22 4 20 30 40 16 MaliC 338 2,451 61 44 11 12 7 II 28 43 17 BurkinaFaso 335 2,629 42 44 21 20 14 12 37 37 18 Niger 647 2,284 65 38 7 19 5 8 28 42 19 India 52,949 221,925 45 31 22 27 15 18 33 4! 20 Kenya 1,453 7,125 33 27 20 22 12 12 47 5! 21 Nigeria 12,546 34,124 41 37 14 38 4 . 45 26 22 Chinac 93,244 369,651 34 27 38 42 30d 381 28 32 23 Haiti 394 2,641 . . . . . 24 Beninc 332 1,886 36 37 12 14 . 9 52 49 25 CentralAfricanRep. 169 1,202 35 41 26 16 7 . . 38 42 26 Ghana 2,214 6,413 47 53 18 17 II 10 35 29 27 Pakistan 9,102 40,244 37 26 22 26 16 17 41 49 28 Togo 253 1,633 34 33 2! 23 10 10 45 44 29 Guineac . 2,937 . 29 . . 35 . . 5 . 36 30 NicaraguaC 785 6,950 25 30 26 23 20 19 49 47 31 Sri Lanka 2,215 8,195 28 27 24 25 17 14 48 48 32 Mauritania 197 1,030 29 22 38 3! 5 . 32 47 33 Yemen, Rep.c . 7,524 . . 22 . . 26 . . 9 . . 52 34 Honduras 654 2,661 32 22 22 27 14 16 45 51 35 Lesotho 67 578 35 14 9 38 4 13 56 48 36 Indonesia' 9,657 116,476 45 19 19 41 tO 21 36 39 37 Egypt, Arab Rep. 6,598 30,265 29 18 28 30 . . . . 42 52 38 Zimbabwe 1,415 5,543 15 20 36 32 21 26 49 49 39 Sudan 1,901 . . 44 . . 14 . . 8 . . 42 40 Z.ambiac 1,789 3,831 II 16 55 47 10 36 35 37 Middle-income economies Lower-middle-income 1,167,639 41 BoliviaC 1,020 5,019 20 . . 32 . . 13 . 48 42 Côte d'Ivoire 1,147 7,283 40 38 23 22 13 21 36 40 43 SenegaN 865 5,774 24 20 20 19 16 13 56 62 44 Philippines 6,691 44,908 30 21 32 34 25 26 39 44 45 Papua New Guinea 646 3,734 37 26 22 35 5 10 41 38 46 Cameroonc 1,160 11,666 31 27 19 22 10 12 50 5! 47 Guatemalac 1,904 9,353 . . 26 . 20 . . . . . . 55 48 DominicanRep.0 1,485 7,172 23 18 26 25 19 13 51 57 49 Ecuadorc 1,673 11,595 24 15 25 35 18 2! 51 50 50 MoroccoC 3,956 27,652 20 19 27 31 16 18 53 50 51 Jordan 3,524 7 26 . 13 67 52 Tajikistan . 0 ' . . . . . . 53 PeraC 7,234 48,366 19 . . 32 . . 20 . . 50 54 EtSalvadorc 1,029 5,915 28 10 23 24 19 19 48 66 55 Congo' 274 2,909 18 12 24 37 . . 8 58 50 56 Syrian Arab Rep.0 2,140 17,236 20 30 25 23 . . . 0 55 47 57 Colombia 7,199 41,692 25 17 28 35 21 20 47 48 58 Pamguay' 595 6,254 32 22 21 24 17 18 47 54 59 Uzbekistan . . . . . . . . . . . . 0 60 Jamaicac 1,405 3,497 7 5 43 40 16 17 51 56 61 Romania 27,619 19 . . 49 . . . . 33 62 Namibia . 1,961 . . 10 . . 28 . 4 . . 62 63 Tunisia 1,244 11,594 20 18 24 32 10 17 56 50 64 Kyrgyzstanc 0 . . . . . 65 Thailandc 7,087 93,310 26 12 25 39 16 27 49 49 66 Georgia Note: For data comparability and coverage, see the technical notes. Figureu in italics are for years other than those specified. 242 Distribution of gross domestic product (percent) GDP (millions of dollars) Agriculture Industry Manufacturing a Sereices, etc. b 1970 1991 1970 1991 1970 1991 1970 1991 1970 1991 67 Azerbaijan . . . . . 68 Turkmenistan S S 69 Turkey 11,400 95,763 30 18 27 34 17 24 43 49 70 Polandc 78,031 7 50 43 7! Bulgaria . . 7,909 . . 13 . . 50 . . . . 37 72 Costa Ricac 985 5,560 23 18 24 25 . . 19 53 56 73 Algeria 4,541 32,678 II 14 41 50 15 10 48 36 74 Panamac 1,021 5,544 15 10 21 II 12 64 79 75 Armenia' . . 0 0 . 0 0 76 ChileC 8,186 31,311 7 41 . 26 . . 52 77 Iran, Islamic Rep. 10,914 96,989 19 21 43 21 14 9 38 58 78 Moldova . . . . 0 79 Ukraine . . . 0 ' 0 ' . . . . 80 Mauritius 184 2,253 16 II 22 33 14 23 62 56 81 Czechoslovakiac 33,172 . . 8 . . 56 . . 36 82 Kazakhstan .. . . . . .. .. . . 83 Malaysia' 4,200 46,980 29 . . 25 . . 12 46 Upper-middle-income 265,930 t 12 w 37 w . . . . 50 w 84 Botswana' 84 3,644 33 5 28 54 6 4 39 41 85 South Africa 16,293 91,167 8 5 40 44 24 25 52 51 86 Lithuania . . . . . . 20 . . 45 . . . 0 35 87 Hungalyc 5,543 30,795 18 10 45 34 . . 29 37 55 88 Venezuela 13,432 53,440 6 5 39 47 16 17 54 48 89 Argentina 20,526 114,344 13 15 38 40 27 . . 49 46 90 Uruguay 1,940 9,479 19 10 37 32 . . 25 44 58 91 Brazil 35,546 414,061 12 10 38 39 29 26 49 51 92 Mexicoc 38,318 282,526 12 9 29 30 22 22 59 61 93 Belarus . . . . . 0 94 Russian Federation . . . . . . 13 48 . . . . 39 95 Latvia . 0 . . 20 . . 48 . . 41 . 32 96 TrinidadandTobago 775 4,920 5 3 44 39 26 9 51 58 97 GabonC 322 4,863 19 9 48 45 7 6 34 46 98 Estonia . . . . . . . . . . . 99 Portugalc 6,184 65,103 . . . . . . . . . . . . . 100 Omanc 256 10,236 16 4 77 52 0 4 7 44 101 Puerto RiCOC 5,035 32.469 3 I 34 4! 24 39 62 57 102 Korea, Rep.0 8,887 282,970 26 8 29 45 21 28 45 47 103 Greece 8,600 57,900 IS 17 31 27 19 14 50 56 104 Saudi Arabia' 3,866 108,640 6 7 63 52 10 7 31 41 105 Yugoslavia 12,566 82,317 18 12 41 48 . . 41 40 Low- and midcfle-income . . . . 25 w . . 33 w . . . 42 w Sub-Saharan Africa 44,0731 164,339 1 35 w 31 w 23 w 29 w 8w . 41 w 40 w East Asia & Pacific 143,0541 961,7541 34 w 19 w 34 w 41 w 26 w 33 w 32 w 40 w South Asia 73,546 1 302,014 1 44 w 31 w 21 w 26 w 14 w 17 w 34 w 43 w Europe and Central Asia . . . . . . . . . Middle East & N. Africa 43,981) 1 413,241 1 16 w 14 w 42 w 11 w 42 w Latin America & Caribbean 154,857 1 1,203,873 1 13 w 34 w 24 w 53 w Severely indebted 147,806 t . . 14 w 35 w 24 w 52 w High-income economies 2,106,0851 17,053,7441 4w 39w 29 w 58w OECD members 2,078,008 1 16,626,259 1 4w 39 w 29 w 58 w 106 Ireland 3,323 39,028 17 11 37 9 24 3 46 80 107 lIsraeIc 5,603 62,687 . . . . . . . . . . 108 New Zealandc 6,415 42,86! 12 9 33 27 24 18 55 65 109 Spain' 37,909 527,131 . . . . . . . . . . . . 110 tHong Kong 3,463 67,555 2 0 36 25 29 17 62 75 III tSingaporec 1,896 39,984 2 0 30 38 20 29 68 62 112 United Kingdom 106,502 876,758 3 . . 44 . . 33 . . 53 113 Australia' 39,330 299,800 6 3 39 31 24 15 55 65 114 ItalyC 107,485 1,150,516 8 3 41 33 27 21 51 64 115 NetherlandsC 34,285 290,725 6 4 37 32 26 20 57 64 116 BelgiumC 25.242 196,873 . . 2 . . 30 . . 22 . . 68 117 Austrit 14,457 163,992 7 3 45 36 34 25 48 61 118 France' 142,869 1,199,286 . . 3 . . 29 . . 21 . . 68 119 Canada 73,847 510,835 4 36 23 59 120 United StatesC 1,011,563 5,610,800 3 35 . . 25 . 63 121 Germanyce 184,508 1,574,316 3 2 49 39 38 23 47 59 122 Denmark 13,511 112,084 7 5 35 28 22 19 59 67 123 Finland 9,762 110,033 12 6 40 34 27 24 48 60 124 Norway 11,183 105,929 6 3 32 36 22 14 62 62 125 Sweden 29,835 206,411 . . 3 S 34 . . 22 63 126 Japanc 203,736 3,362,282 6 3 47 42 36 25 47 56 127 Swit.zerlandc 20,733 232,000 . 0 0 World 2,792,7821 21,639,120 f 8w . . 38 w 27w 55w Fuel exporters 57,618: 458,283 1 17 w 12 w 39 w 11w 43w a. Because manufacturing is generally the most dynamic part of the industrial sector, its share of GDP is shown separately. b. Services, etc. includes unallocated items. c. GDP and its components are at purchaser values. d. World Bank estimate. e. Data refer to the Federal Republic of Germany before unification. 243 Table 4. Agriculture and food Food Fertilizer consumption production per Value added in agriculture Food aid in cereals (hundreds of grams capita (average Fish products (millions of current Cereal imports (thousands of ofplant nutrient per growth rate; (percentageof daily dollars) (thousands of metric tons) metric tons) hectare of arable land) 1979-81 100) protein supply) 1970 1991 1980 1991 1979/80 1990/91 1979/80 1990/91 /979-91 1970 1990 Low-income economies 85,549 t 276,360 1 35,359 1 36,510 1 6,913 1 7,373 1 474 w 993 w 5.8 w 6.3 w China and India 55,737 t 170,532 1 13,376 1 13,489 1 3551 351 I 669 w 1,478 w 2.3 w 2.9 w Other low-income 29,615 t 109,326 1 21,983 1 23,021 1 6,557 1 7,022 I 204 w 394 w 5.9 w 6.4 w I Mozambique . 854 368 479 151 454 78 8 -3.1 4.6 3.0 2 Tanzania 483 1,352 399 130 89 24 90 144 -1.4 8.1 7.7 3 Ethiopia 931 2,822 397 802 III 894 27 80 -1.4 0.2 0.0 4 Uganda 1,425 52 26 17 61 . . 0 -0.6 7.5 7.2 5 Bhutan 104 5 26 I 4 8 8 -0.6 6 Guinea-Bissau 37 96 21 64 18 7 5 17 0.3 3.1 2.1 7 Nepal 579 1,807 56 6 21 1 90 274 2.2 0 0.3 8 Bunindi 159 565 18 31 8 3 7 16 -0.6 1.6 1.3 9 Chad a 142 528 16 73 16 30 . . 18 -0.4 7 9.9 10 Madagascar 243 822 110 114 14 38 25 26 -1.4 3.6 4.4 11 SierraLeone 108 319 83 183 36 17 46 20 -1.1 10.9 10.8 12 Bangladesha 3,636 8,428 2,194 1,631 1,480 1,356 445 1,022 -0.6 7.4 4.8 13 Lao PDR . . . . 121 44 3 0 1 16 0.8 3.8 2.1 14 Malawi 119 701 36 120 5 181 110 198 -2.7 5.9 5.1 15 Rwandaa 135 812 16 19 14 9 3 26 -1.8 0.2 0.2 16 Malju 207 1,082 87 226 22 37 69 73 -0.7 4.6 3.5 17 Burkina Faso 139 1,074 77 177 37 56 26 39 2.4 0.7 0.9 18 Niger 420 877 90 143 9 79 5 3 -3.4 0.2 0.2 19 India 23,916 71,103 424 58 344 217 313 743 1.6 1.6 1.6 20 Kenya 484 1,895 387 330 86 63 169 477 0.5 1.5 2.9 21 Nigeria 4,787 12,271 1,828 763 . . . . 36 124 1.5 2.7 3.5 22 Chinau 31,821 99,429 12,952 13,431 12 134 1,273 2,777 3.0 3.1 3.9 23 Haiti . . . . 195 348 53 37 44 11 -1.2 0.7 2.8 24 Beninu 121 692 61 216 5 8 7 38 1.5 10 4.8 25 CentralAfrican Rep. 60 497 12 27 3 3 1 4 -1.0 4.2 3.0 26 Ghana 1,030 3,404 247 344 110 72 65 48 0.2 19.9 18.7 27 Pakistan 3,352 10,318 613 972 146 343 488 912 0.2 0.6 0.8 28 Togo 85 531 41 238 7 16 49 172 -1.0 5.7 8.4 29 Guineaa . 850 171 296 24 12 31 7 -0.5 1.5 4.5 30 Nicaraguaa 199 2,024 149 176 70 117 185 314 -5.1 1.1 0.4 31 SriLanka 627 2,203 884 918 170 200 776 901 -1.3 10 9.9 32 Mauritania 58 226 166 342 26 101 108 93 -1.5 9.2 3.3 33 Yemen,Rep.0 . 1,657 . . . . . . 149 98 . . . . . 34 Honduras 212 592 139 284 27 84 111 273 -1.6 0.9 1.8 35 Lesotho 23 82 107 100 29 31 144 144 -1.7 0 0.8 36 Indonesiau 4,340 22,465 3,534 2,795 831 45 440 1,141 2.2 7.9 8.7 37 Egypt, Arab Rep. 1,942 5,491 6,028 7,807 1,758 1,525 2,469 3,722 1.1 1.2 2.4 38 Zimbabwe 214 1,082 156 131 . . 8 443 606 -1.0 0.8 1.1 39 Sudan . . 2,625 236 1,188 212 453 27 63 -2.8 0.8 0.5 40 2a,nbiau 191 603 498 104 167 4 114 113 -0.7 8.3 4.3 Middle-income economies 73,2001 77,351 1 1,777 1 4,3941 715 w 697 w 6.5 w 6.8 w Lower-middle-income 40,1371 40,2791 1,292 1 4,1191 721 w 744 w 6.2 w 6.4 w 41 Boliviaa 202 . . 263 219 150 229 16 58 0.7 0.6 0.6 42 Côte d'Ivoire 462 2,754 469 644 2 59 165 97 -0.1 10.3 8.7 43 Senegalu 208 1,129 452 784 61 39 123 50 0.4 8 9.8 44 Philippinesa 1,975 9,489 1,053 1,848 95 81 444 738 -1.9 26.3 20.9 45 PapuaNewGuineaa 240 980 152 287 151 311 -0.1 II 11.8 46 Camemona 364 3,172 140 532 4 9 47 31 -1.8 5.5 6.7 47 Guatemalau . 2,410 204 410 10 170 582 700 -0.7 0.2 0.4 48 Dominican Rep.0 345 1,289 365 712 120 6 517 614 -2.6 5.7 2.8 49 Ecuador a 402 1,749 387 481 8 98 319 232 0.0 3.8 6.8 50 Mocoa 789 5,228 1,821 1,957 119 201 240 332 2.3 1.3 2.8 51 Jordan 263 505 1,539 72 481 433 475 -1.2 0.6 1.2 52 Tajikistan .. .. .. .. . . .. . . . 53 Pem 1,351 . . 1,309 1,432 109 371 338 336 -0.6 5.4 10.6 54 El Salvadoru 292 604 144 324 3 84 1,030 1,027 0.3 1.1 0.7 55 Congoa 49 356 88 96 4 15 6 119 -0.1 17.2 22.8 56 SyrianArabkep.a 435 4,091 726 - 1,741 74 30 224 539 -2.9 0.5 0.1 57 Colombia 1,806 7,258 1,068 780 3 1 603 1,112 0.6 1.7 1.4 58 Paraguaya 191 1,358 75 24 11 0 36 65 1.1 0.3 1.0 59 Uzbekistan .. .. .. .. .. .. .. .. 60 Jamaicaa 93 177 469 413 117 163 503 710 -0.5 12.2 8.9 61 Romania 5,121 2,369 1,834 . . 480 1,365 1,099 -2.3 1.8 3.3 62 Namjbia . . 194 . . 18 . . . . . . . . -2.9 2.8 3.5 63 Thnisia 245 2,084 817 920 165 348 122 181 0.5 1.9 3.7 64 Kyrgyzstan . . . . . . . . . . . . . . . . . . . . 65 Thailanda 1,837 11,063 213 521 3 104 160 471 0.5 14.4 12.0 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 244 Food Fertilizer consumption production per Value added in agriculture Food aid in cereals (hundreds of gram.s capita (average Fish products (millions of current Cereal imports (thousands of ofplant nutrient per growth rate; (percentage of daily dollars) (thousands of metric tons) metric tons) hectare of arable land) 1979-81 = 10)) protein supply) 1970 1991 1980 1991 1979/80 1990/91 1979/80 1990/9! 1979-91 1970 1990 67 Azerbaijan . . 68 Thrknsenistan . . . . ' . . . . . . . ' 69 Thrkey 3,383 17,090 6 638 16 4 451 676 -0.2 1.9 2.3 70 Polanda 5,342 7,811 166 742 2,425 1,046 1.1 4.2 4.8 71 Bulgaria 1,021 693 633 . . 100 1,928 1,728 -0.9 2.4 1.7 72 Cost.aRicaa 222 1,013 180 320 1 31 1,573 2,091 -0.7 2.9 2.2 73 Algeria 492 4,608 3,414 5,436 19 26 227 167 0.6 1.3 2.1 74 Panamaa 149 530 87 101 2 I 540 588 -2.0 5.7 7.4 75 Armenia S . . . . . . . 0 . . 76 Chilea 557 . . 1,264 588 22 11 333 653 1.5 4.2 7.8 77 Iran, Islamic Rep. 2,120 21,186 2.779 5,025 70 297 771 1.3 0.3 1.6 78 Moldova . 79 Ukraine . . . . . . . . . . . . . . . . . . . 80 Mauritius 30 248 181 183 22 7 2,564 2,616 -0.6 6.9 8.5 81 Czechoslovakia 2,102 136 . . . . 3,347 2,558 1.5 2.3 2.4 82 Kazakhstan .. .. .. .. .. .. .. 83 Malaysian 1,198 1,336 3,014 . . 4 912 1,950 4.1 13.7 13.8 Upper-middle-income . . . . 33,063 1 37,072 1 485 1 275 1 707 w 639 w . . 6.9 w 7.4 w 84 Botswanaa 28 190 68 99 20 0 8 7 -3.7 0.5 1.3 85 SouthAfrica 1,292 4,594 159 1,345 726 592 -1.1 4.6 3.8 86 Lithuania . . . . . . . . . . . . . . . 87 Hungaiy3 1,010 3,181 155 128 . . . . 2,805 1,269 1.2 1.1 1.3 88 Venezuelaa 835 2,662 2,484 1,468 . . . . 599 1,137 0.0 5.2 6.7 89 Argentinaa 2,693 16,588 8 31 . . . . 48 61 -0.6 1.1 1.7 90 Uniguay 378 926 45 83 7 20 633 551 0.8 0.8 1.1 91 Brazil 4,388 42,288 6,740 6,332 3 16 755 525 1.7 3.6 2.6 92 Mexicoa 4,462 25,221 7,226 5,433 . . 239 465 631 0.2 1.8 3.3 93 Belams . . . . . . . . . . . 94 Russian Federation . . . . . . . . . . . . . 95 Latvia . . . . . . . . . . . . . . . . . . . 96 Trinidadandlobago 40 144 252 201 . . . . 670 650 -1.7 6.2 3.6 97 Gabon 60 425 27 70 3 25 -l.4 15.6 12.9 98 Estonia . . . . . . . . 99 Portugal . . . . 3,372 1,369 267 877 877 1.5 15.7 15.0 100 Omana 40 374 120 345 306 1,554 . 101 PuertoRicoa 161 470 . . . . . . . . . . -1.4 . 102 Korea,Rep.a 2,311 22,793 5,143 10,411 184 3,857 4,601 -0.1 6.7 15.8 103 Greece 1,569 . . 1,199 753 . . 1,480 1,741 0.3 5 4.8 104 SaudiArabiaa 219 6,713 3,061 5,891 . . 115 2,068 9.4 2.5 2.3 105 Yugoslavia 2,212 9,641 1,420 217 1,102 991 -0.7 1.1 1.1 Low-and middle-income . . . . 108,5691 113,8751 8,6901 11,7671 576w 867 w 6.2 w 6.6w Sub-Saharan Africa 13,394 t 52,122 1 8,434 1 10,6261 1,602 1 3,221 1 59 w 90 w 6.1 w 6.2 w East Asia & Pacific 47,923 1 188,371 1 26,833 1 34,2641 1,525 1 486 I 953 w 1,902 w 10.2 w 11.4 w South Asia 32,7201 95,6451 4,2111 3,7871 2,3391 2,1611 328w 740w 11.7w 14.4w Europe and Central Asia . . . . 19,2691 6,265 1 284 1 1,326 I 1,446 w 1,094 w 3.6 w 4.0 w Middle East & N. Africa 7,243 1 57,1771 23,8811 34,9111 2,2201 2,8571 337 w 681 w . . 1.4 w 1.7 w Latin America & Caribbean 19,755 1 . . 25,782 1 22,6771 721 1 1,716 1 495 w 508 w 6.5 w 6.7 w Severely indebted 20,522 I . . 37,7231 30,8861 691 1 2,901 1 651 w 501 w 3.8 w 4.5 w High-income economies 85,4071 79,7981 75,0961 361 21 1,321 w 1,158w 8.2 w 8.6w OECDinembers 83,9851 70,7641 63,5361 . . 1,312 w 1,145w . . 7.5w 8.4w 106 Ireland 559 . . 553 341 . . 5,219 7,323 1.5 2.7 3.9 107 tlsrael3 295 . . l,601 1,635 31 2 1,885 2,343 -0.5 3.8 5.0 108 New Zealand3 914 . . 63 223 . . 12,060 8,7% 0.1 3.1 8.5 109 Spaina . 22,189 6,073 4,016 . . 821 979 1.3 9.2 9.8 110 tHongKong 62 185 812 785 . . . . . . 0.7 19 16.9 Ill tSingapore3 44 96 1,324 780 . . . . 5,375 56,000 -5.1 16.5 9.2 112 United Kingdom 2,975 . . 5,498 2,799 3,235 3,680 0.4 5.6 5.1 113 Australia3 2,277 9,718 5 36 . . 275 238 -0.4 3.3 4.1 114 Italy3 8,387 34,456 7,629 8,466 l,892 l,480 -0.5 3.9 5.6 115 Netherlands3 1,894 11,988 5,246 4,925 8,472 6,l60 0.8 4.6 2.9 116 Belgium . . 3,461 5,599b 6041b 5,282 4,902 1.4 5.1 5.0 117 Austria3 992 4,968 131 88 . . . . 2,484 1,997 0.7 2 2.7 118 France3 . . 40,012 1,570 l,206 3,120 2,953 0.3 5.1 5.8 119 Canada 3,238 1,383 448 . . 398 451 1.2 3.4 6.6 120 United States3 27,937 . . 199 2,834 . . . . 1,099 970 -0.6 3.3 4.3 121 Germany3 5,951c 23,867' 9,500 3,545 1.6 4,227 4.1 2,637 4.0 122 Denmark 882 5,082 355 207 . . . 2.6 . 2,627 9.1 2,463 10.5 123 Finland 1,205 7,485 367 58 1,892 1,819 0.1 . 7.2 8.7 124 Norway 624 3,084 725 196 . . . 3,220 2,355 0.5 14.5 15.2 125 Sweden . . 5,825 124 117 . 1,699 1,162 . -0.6 10.2 9.3 126 Japan3 12,467 73,671 24,473 27,474 . . 4,777 4,001 -0.1 . . 24.4 28.0 127 Switzerland3 1,247 493 . 4,654 4,075 -0.1 . . 2.7 3.7 World 238,777 1 . . 188,9711 188,9581 8,7261 11,769 1 813 w 957 w 6.7 w 7.1 is Fuel exporters 9,929 1 55,8281 18,772 1 24,0241 341 2101 167 w 404 w 6.4 w 6.1 is a. Value added in agriculture data are at purchaser values. b.Inc!udes Luxembourg. c. Data refer to the Federal Republic of Germany before unification. 245 Table 5. Commercial energy Energy consumption Energy imports Average annual growth raze (percent) per capita (kilograms as a percentage of Energy production Energy consumption of oil equivalent) merchandise exports 1970-80 1980-91 1970-80 1980-91 1970 1991 1970 1991 Low-income economies 7.3 w 4.8 w 7.0 w 5.6 w 166 w 376 w 6.0 w 10.0 w China and India 7.7 w 5.5 w 7.3 w 5.8 w 200 w 488 w 4.0 w 7.0 w Other low-income 6.7 w 3.2 w 6.0 w 4.7 w 94 w 173 w 7.0 w 13.0 w I Mozambique 32.3 -39.3 -1.6 1.0 113 59 2 Tanzania 7.3 3.2 -0.2 2.0 49 37 id.ó 65.0 3 Ethiopia 5.5 5.2 0.9 3.4 20 20 11.0 37.0 4 Uganda -3.2 2.8 -6.4 4.1 58 25 5 Bhutan Is 6 Guinea-Bissau . . . . 4.0 2.2 38 38 . 7 Nepal 12.7 10.9 6.0 8.0 10 22 19.0 38.0 8 Burundi . . 6.6 6.9 7.4 7 24 6.0 20.0 9 Chad S S 4.0 0.4 17 17 35.0 31.0 10 Madagascar 0.4 6.8 -2.5 1.8 60 39 9.0 36.0 II SierraLeone . . . . -1.7 0.1 158 75 8.0 32.0 12 Bangladesh . . 11.3 8.8 7.7 . . 57 26.0 13 La0PDR 41.3 -0.4 -4.0 2.3 73 42 . 14 Malawi 11.8 4.2 7.8 1.3 35 41 10.0 24.0 15 Rwanda 4.6 4.0 12.3 1.8 11 29 16 Mali 8.3 6.1 7.8 2.1 15 23 11.0 17 BurkinaFaso . . . . 12.0 1.1 8 17 21.0 35.0 18 Niger . . 13.5 11.8 2.3 16 41 7.0 22.0 19 India 6.8 6.6 6.4 7.2 113 337 8.0 26.0 20 Kenya 15.0 6.4 3.0 1.6 138 104 20.0 25.0 21 Nigeria 4.8 1.0 16.0 4.4 43 154 3.0 1.0 22 China 7.9 5.3 7.5 5.3 258 602 1.0 3.0 23 Haiti 14.0 5.7 9.8 1.7 27 49 7.0 43.0 24 Benin . . 7.4 2.8 3.7 41 46 7.0 29.0 25 CentralAfricankep. 4.3 2.6 -1.6 3.3 44 29 1.0 10.0 26 Ghana 5.8 -0.1 2.3 0.4 180 130 6.0 44.0 27 Pakistan 8.1 6.5 5.3 6.5 139 243 11.0 23.0 28 Too 9.2 . . 9.4 0.8 43 47 5.0 14.0 29 Guinea 15.8 4.1 2.3 1.4 68 68 . 30 Nicaragua 1.7 2.6 4.1 2.7 253 254 7.0 36.0 31 Sn Lanka 8.0 8.5 1.2 4.9 139 177 3.0 17.0 32 Mauritania 4.7 0.3 115 III 5.0 7.0 33 Yemen, Rep. 23.3 7.9 9 96 . 34 Honduras 13.9 4.1 6.1 2.0 191 181 9.0 20.0 35 Lesotho 36 Indonesia 7.9 1.6 9.9 4.8 99 279 2.0 8.0 37 Egypt, Arab Rep. 10.7 4.6 10.9 4.6 213 594 10.0 5.0 38 Zimbabwe -1.9 3.0 1.0 3.0 580 517 21.0 28.0 39 Sudan 21.0 2.2 -4.3 0.6 113 54 9.0 40 Zambia 16.0 1.8 7.7 1.3 299 369 5.0 21.0 Middle-income economies 3.1 w 2.3 w 5.4w 3.4w 918w 1,351 w 10.0w 11.0w Lower-middle-income 2.0 w 4.3 w 5.0w 2.9w 865w 1,102w 9.0w 11.0w 41 Bolivia 6.8 0.8 9.5 -0.1 183 251 1.0 1.0 42 Cbted'Ivoire 20.5 4.5 6.5 2.7 154 170 4.0 12.0 43 Senegal . . . . -2.9 -1.6 180 105 6.0 28.0 44 Philippines 13.6 6.3 4.0 1.9 235 218 14.0 20.0 45 PapuaNewGuinea 11.4 5.6 8.4 2.4 114 231 39.0 11.0 46 Cameroon 24.4 11.5 6.2 4.4 84 147 6.0 1.0 47 Guatemala 8.5 4.5 6.5 0.6 170 155 2.0 26.0 48 Dominican Rep. 17.0 4.2 6.7 0.9 216 341 14.0 49 Ecuador 36.3 2.9 13.5 3.7 216 598 9.0 1.0 50 Morecco 2.4 1.5 6.8 2.9 170 252 8.0 25.0 51 Jordan 13.1 5.3 322 856 40.0 41.0 52 Tajikistan .. .. .. .. .. 53 Pens 10.6 -1.6 4.4 1.4 468 451 1.0 9.0 54 El Salvador 14.5 3.7 7.7 2.4 157 230 2.0 31.0 55 Congo 41.0 6.9 6.2 3.3 138 214 4.0 1.0 56 Syrian Arab Rep. 9.3 8.0 14.4 3.9 313 955 17.0 10.0 57 Colombia 0.3 10.6 5.9 3.1 490 778 1.0 4.0 58 Paraguay 13.5 12.9 9.7 4.9 114 231 17.0 24.0 59 Uzbekistan 60 Jamaica -0.9 4.3 3.6 -1.4 968 858 25.0 35.0 61 Romania 3.3 -0.4 5.9 0.6 2,136 3,048 62 Namibiaa S S S S . . S S S 63 Tunisia 4.0 -0.2 8.7 4.5 230 556 8.0 13.0 64 Kyrgyzstan .. .5 .. .. .. 65 Thailand 5.5 24.1 7.9 7.4 150 438 17.0 12.0 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 246 Energy consumption Energy imports Average annual growth rate (percent) per capita (kilograms as a percentage of Energy production Energy consumption of oil equivalent) merchandise exports 1970-80 1980-91 1970-80 1980-91 1970 1991 1970 1991 67 Azerbaijan 68 Turkmenistan . . . . . . . . . . 69 Turkey 3.5 7.7 7.4 6.5 362 809 11.0 32.0 70 Poland 3.6 0.9 4.6 I.! 2,512 3,165 20.0 23.0 71 Bulgaria 2.0 2.3 5.1 0.8 2,657 3,540 . 72 CostaRica 6.9 6.3 6.6 3.7 378 570 5.0 21.0 73 Algeria 5.1 4.9 16.9 15.1 219 1,956 3.0 2.0 74 Panama 10.8 9.5 -6.6 0.2 2,524 1,661 75 Armenia . . . 0 . . . 76 Chile -0.1 3.1 0.1 3.1 867 892 5.0 13.0 77 Iran, Islamic Rep. -5.6 6.2 5.2 4.5 938 1,078 0.0 0.0 78 Moldova 79 Ukraine . . . . . . . . . . . . . 80 Mauritius 1.9 8.1 4.9 3.4 266 389 8.0 24.0 81 Czechoslovakia 0.8 0.0 2.8 0.6 3,893 4,681 10.0 5.0 82 Kazakhstan . . . . . . . . . . . . . 83 Malaysia 22.5 13.5 5.4 7.9 452 1,066 10.0 5.0 Upper-middle-income 3.9 w 0.9 w 5.8 w 3.8 w 989 w 1,701 w 10.0 w 11.0 w 84 Botswanaa 9.1 1.2 10.5 3.0 232 408 . 85 South Africaa 6.6 4.0 3.7 2.9 1,909 2,262 1.0 0.0 86 Lithuania . . . . . . . . . . . . . 87 Hungary 1.1 0.2 3.9 0.7 2,053 2,830 10.0 17.0 88 Venezuela -4.7 0.7 4.4 2.3 2,206 2,521 1.0 1.0 89 Argentina 3.3 3.2 3.3 3.6 1,208 1,764 5.0 6.0 90 Umguay 5.6 7.2 1.0 0.8 797 816 15.0 15.0 91 Brazil 8.7 7.3 8.8 4.7 410 908 13.0 19.0 92 Mexico 13.7 1.2 8.7 1.4 786 1,383 6.0 5.0 93 Belarus . 94 Russian Federation . . . 95 Latvia . . . . . . . . . 96 TrinidadandTobago 6.2 -2.5 4.5 1.7 4,795 4,907 . 97 Gabon 7.0 4.6 11.3 2.5 805 1,154 1.0 0.0 98 Estonia . . . . . . . . . . . 99 Portugal 4.5 3.2 5.1 2.9 747 1,584 15.0 15.0 100 Oman 0.3 8.6 5.9 10.1 660 2,859 101 PuertoRico -5.0 2.3 -1.5 0.5 3,497 2,015 . 102 Korea, Rep. 4.3 9.5 10.1 7.9 495 1,936 16.0 18.0 103 Greece 8.7 5.9 6.6 2.8 976 2,110 21.0 24.0 104 SaudiAmbia 8.7 -1.7 4.5 9.3 3,137 4,866 0.0 0.0 105 Yugoslavia 3.2 3.2 5.3 3.6 1,140 2,296 8.0 19.0 Low- and middle-income 4.2 w 3.1 w 6.0 w 4.3 w 362 w 631 w 9.0 w 11.0 H Sub-Saharan Africa 5.4 w 4.7 w 4.4 w 3.5 w 96 w 135 w 6.0 w 12.0 H East Asia & Pacific 7.8 w 5.0 w 7.1 w 5.3 w 255 w 571 w 9.0 w 10.0 H South Asia 6.7 w 6.5 w 6.4 w 7.2 w 102 w 289 w 8.0 w 25.0 H Europe and Central Asia 2.9 w 1.2 w 4.8 w 1.7 w 1,714 w 2,387 w 13.0 w 18.0 H Middle East & N. Africa 3.3 w 1.7w 7.5 w 7.3 w 545 w 1,185 w 2.0 w 4.0 H Latin America & Caribbean 2.1 w 2.5 w 5.6 w 2.7 w 722 w 1,051 w 11.0 w 13.0 s Severely indebted 6.9 w 3.4 w 6.3 w 3.1 w 818 w 1,252 w 10.0 w 11.0 ii High-income economies 1.4w 1.7w 1.8w 1.5w 4,463 w 5,106 w 11.0 w 11.0 w OECD members 1.5w 1.7w 1.7w 1.5w 4,572 w 5,122 w 11.0 w 11.0 w 106 Ireland 2.1 2.7 3.1 0.8 2,008 2,754 13.0 5.0 107 tlsrael -42.6 -7.4 3.0 2.3 1,876 1,931 9.0 11.0 108 New Zealand 4.5 6.1 2.6 5.0 2,834 4,893 7.0 7.0 109 Spain 3.7 2.6 5.1 1.7 1,276 2,229 26.0 17.0 110 tHongKong 5.4 3.1 973 1,438 4.0 7.0 Ill tSingapore . . . . 2.8 5.6 3,863 6,178 21.0 16.0 112 United Kingdom 7.2 0.4 -0.2 0.7 3,847 3,688 12.0 7.0 113 Australia 6.2 5.7 3.8 2.2 4,032 5,211 5.0 6.0 114 Italy 0.9 1.0 1.6 0.9 2,334 2,756 16.0 10.0 115 Netherlands 8.7 -2.4 2.6 1.4 4,531 5,147 12.0 9.0 116 Belgium . . . . . . . . . . 2,793 . 117 Austria 1.1 -0.2 2.0 1.4 2,773 3,500 10.0 7.0 118 France 0.8 6.3 2.3 1.2 3,182 3,854 13.0 10.0 119 Canada 3.2 3.3 3.5 2.0 7,467 9,390 5.0 5.0 120 United States 0.0 0.7 1.3 1.4 7,665 7,681 7.0 15.0 121 Germanyb _0.2b O.Ob 1.8b 0.4" 3,Q77b 3,463b 80b 122 Denmark 25.0 34.0 0.6 0.0 4,176 3,747 14.0 5.0 123 Finland 6.4 4.3 3.0 2.8 3,418 5,602 13.0 13.0 124 Norway 17.6 7.8 3.3 1.8 6,029 9,130 12.0 3.0 125 Sweden 6.7 4.2 1.2 1.3 5,398 5,901 11.0 8.0 126 Japan 1.6 3.9 2.2 2.2 2,654 3,552 20.0 17.0 127 Switzerland 4.5 1.0 1.2 1.4 3,186 3,943 7.0 5.0 World 2.7 w 2.4 w 2.9 w 2.5 w 1,343 w 1,195 w 10.0 w 11.0 w Fuel exporters 2.5w 1.6w 7.0w 6.8w 658w 1,261 w 3.0w 1.0w a. Figures for the South African Customs Union comprising South Africa, Namibia, Lesotho, Botswana, and Swaziland are included in South African data; trade among the component territories is excluded. b. Data refer to the Federal Republic of Germany before unification. 247 Table 6. Structure of manufacturing Distribution of manufacturing value added (percent; current prices) Value added in Fond, Machinery manufacturing (millions of beverages, Textiles and and transport current dollars) and tobacco clothing equipment Chemicals Othera 1970 1990 /970 1990 1970 /990 /970 /990 1970 /990 1970 /990 Low-income economies 44,177 t 240,456 China and India 35,483 t 181,072 1 Other low-income 8,256 I Mozambique . . . SI . 13 . 5 . 3 . . 28 2 Tanzania 118 86 36 30 28 18 5 7 4 /6 26 29 3 Ethiopia 149 614 46 48 31 19 0 2 2 4 21 28 4 Uganda 107 40 20 2 4 34 5 Bhutan . . 27 S . 6 Guinea-Bissau 17 18 , . . 7 Nepal 32 152 . 35 . 25 . . 2 . 8 . . 30 8 Bunrndi 16 99 53 25 0 6 . 16 9 Chad' 51 250 . . . . 10 Madagascar 36 39 28 36 6 3 7 7 23 14 II SierraLeone 22 52 . . . 12 Bangladeaht 387 1,959 30 24 47 35 3 5 II 17 10 18 13 La0PDR .. .. .. 14 Malawi . . 227 51 17 3 10 20 IS Rwandab 8 316 86 0 3 2 . . 8 16 Malib 25 286 36 . 40 . 4 . , 5 14 17 BurkillaFaso 47 325 69 9 2 . I 19 18 Nigert' 30 219 . . . . . 19 India 7,928 48,930 13 12 21 12 20 26 14 17 32 33 20 Kenya 174 862 33 38 9 10 16 10 9 9 33 33 21 Nigeria 426 36 . . 26 . . I . . 6 . . 31 22 China 27.555 132.142 15 15 24 . . 13 34 23 Haiti . 24 Benint 38 162 25 Central African Rep. 12 26 Ghanats 252 575 34 . 16 . 4 . 4 . . 41 27 Pakistan 1,462 6,184 24 29 38 19 6 7 9 15 23 30 28 Togo" 25 162 . -. 29 Guinea1' . . 123 . 30 Nicaragua" 159 260 53 14 2 . 8 . . 23 31 SriLanka 369 1,077 26 51 19 23 10 4 II 3 33 20 32 Mauritania 10 . . . 0 0 33 Yemen, Rep)' 0 549 20 . 50 . I . 28 34 Honduras 91 428 58 51 10 9 1 2 4 5 28 33 35 Lesotho 3 64 36 Indonesiat' 994 21,722 65 24 14 14 2 10 6 10 13 42 37 Egypt, Arab Rep. . . . . 17 31 35 16 9 9 12 8 27 36 38 Zimbabwe 293 1,508 24 30 16 17 9 8 II 10 40 34 39 Sudan . . 772 39 . . 34 . 0 3 . . 5 . . 19 40 Zambia 181 1,180 49 37 9 12 5 9 10 II 27 31 Middle-income economies Lower-middle-income 41 Boliviat 135 585 33 37 34 8 1 1 6 6 26 47 42 Côted'voire 149 . . 27 16 10 . . 5 42 43 Senegalt' 141 775 51 . . 19 . . 2 . . 6 . . 22 44 Philippines" 1,665 11,160 39 36 8 10 8 8 13 12 32 34 45 Papua New Guineat 35 320 23 1 35 . . 4 37 . 46 Cameroont' 119 1,363 50 61 15 -13 4 5 3 5 27 42 47 Guatemala' . . . . 42 43 14 9 4 3 12 16 27 28 48 Dominic9n Rep." 275 955 74 . . 5 . . I . . 6 . . 14 49 Ecuador 306 2,091 43 31 14 13 3 7 8 II 32 39 50 Monxcot' 641 4,886 31 25 6 16 . . 22 51 Jordan 441 21 26 14 7 7 4 6 15 52 49 52 Tajilcstan .. .. 53 Pens 1,430 . . 25 23 14 14 7 10 7 10 47 43 54 El Salvador1' 194 1,008 40 . . 30 . . 3 . . 8 . . 18 55 Congo" 220 65 58 4 4 1 3 8 10 22 24 56 Syrian Arab Rep." . . . . 37 24 40 31 3 6 2 5 20 34 57 Colombia 1,487 8,192 31 30 20 15 8 10 II 14 29 31 58 Paraguay° 99 994 56 . . 16 I 5 . . 21 59 Uzbekistan .. .. 60 Jamaicat' 221 793 46 7 II 5 . . 30 61 Romania . 13 . 19 25 5 38 62 Namibia . . 77 . 0 63 Tunisia 121 1,869 29 17 18 19 4 6 13 9 36 49 64 Kyrgyzstjs .. .. .. .. 65 Thailand 1,130 20,926 43 29 13 25 9 12 6 3 29 31 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 248 Distribution of manufacturing value added (percent; current prices) Value added in Food, Machinery manufacturing (millions of beverages, Textiles and and transport current dollars) and tobacco clothing equipment Chemicals Othera /970 /990 1970 /990 /970 1990 1970 1990 1970 1990 1970 1990 67 Azerbaijan . . . . 68 Thrkmenistan . . . . . . 69 Turkey 1.930 22,685 26 16 15 14 8 17 7 II 45 43 70 PoIand' 20 21 19 9 24 26 8 7 28 37 71 Bulgaria . . . . . . . . . 72 Costa Ricat 203 1,071 48 47 12 8 6 6 7 9 28 30 73 Algeria 682 4,816 32 20 . 9 . 4 . 35 74 Panamab 127 . 41 52 9 6 1 3 5 8 44 31 75 Armenia . . . . . . 76 Chile 2,088 . 17 24 12 7 II 5 5 9 55 56 77 Iran, Islamic Rep 1,501 8,819 30 . 20 . 18 6 . 26 78 Moldova . . . 79 Ukraine . . . . . . 80 Mauritius 26 496 75 23 6 50 5 3 3 5 12 19 81 Czechoslovakia1' . 9 10 12 II 34 35 6 7 39 37 82 Kazakhstn . . . . 83 Malaysia 500 . 26 13 3 6 8 31 9 II 54 39 Upper-middle-income 84 Botswanat' 5 128 . 52 . 8 . 9 . 31 85 SouthAfrica 3,892 23,197 IS 14 13 8 17 18 10 II 45 50 86 Lithuania . . . 87 Hungary" . . 8,831 12 10 13 9 28 27 8 13 39 41 88 Venezuela' 2,163 8,109 30 17 13 5 9 5 8 9 39 64 89 Argentina 5,523 . 18 20 17 10 17 13 8 12 40 44 90 Umguay 619 2,173 34 32 21 18 7 9 6 10 32 31 91 Brazil 10,421 108,789 16 13 13 12 22 23 10 12 39 40 92 Mexico' 8,449 55,621 28 22 15 9 13 15 II 14 34 40 93 Belarus . . . . . . 94 Russian Federation . . . . 95 Latvia . . 3,825 96 TrinidqdandTobago 198 435 18 3 . 7 2 70 97 Gabon" 22 264 37 . 7 . 6 . 6 . 44 98 Estonia . . . . . . . 99 Portugal1' . . . 18 18 19 20 13 14 10 10 39 39 100 Omarit 0 396 . . . . . . 101 PuertoRico 1,190 12,181 . 15 . . 5 . 18 . 45 . . 17 102 Korea,Rep.b 1,880 70,497 26 II 17 13 11 32 II 9 36 35 103 Greece 1,642 8,291 20 22 20 21 13 12 7 8 40 38 104 Saudi Arabiat' 372 7,962 . 7 . . 1 . 4 . 39 . . 50 105 Yugoslavia 10 16 15 19 23 24 7 8 45 31 Low- and middle-iacome Sub-Saharan Africa 3,046 East Asia & Pacific 36,524 287,606 South Asia 10,362 59,372 Europe and Central Asia Middle East & N. Africa 4,797 40,563 Latin America & Caribbean 36,590 I 254,873 Severely indebted 35,985 277,462 High-income economies 605,102 I OECD members 599,875 1 106 Ireland 786 1,535 31 26 19 4 13 32 7 15 30 23 107 tlsraelt' . . . . IS 13 14 9 23 32 7 10 41 37 108 New Zealandb 1,811 7,572 24 26 13 9 15 14 4 6 43 46 109 Spain" . . 124,454 13 18 15 8 16 25 11 10 45 38 110 tHongKong 1,013 11,403 4 7 41 38 16 20 2 2 36 33 Ill ISingaporeb 379 10,351 12 4 5 3 28 53 4 10 51 29 112 UnitedKingdom 35,415 . . 13 13 9 5 31 32 10 II 37 38 113 Aus1ra1iat 9,551 44,589 16 18 9 7 24 19 7 8 43 48 114 Italy" 29,093 242,899 10 8 13 13 24 33 13 10 40 36 115 Netherlandsb 8,861 58,147 17 16 8 3 27 25 13 18 36 39 116 Belgium" . . 43,260 17 17 13 8 25 23 9 II 37 41 117 Austria" 4,873 41,526 17 15 12 6 19 28 6 7 45 44 118 Franceb . 251,143 12 13 10 6 26 31 8 9 44 41 119 Canada 16,782 16 14 8 6 23 26 7 10 46 44 120 United Statest) 254,858 . . 12 12 8 5 31 31 10 12 39 40 121 Germany" 70,888 460,983 13 9 8 4 32 42 9 12 38 32 122 Denmark 2,929 21,376 20 21 8 4 24 23 8 II 40 40 123 Finland 2,588 26,170 13 13 10 4 20 25 6 8 51 51 124 Norway 2,416 14,472 15 21 7 2 23 24 7 9 49 44 125 Sweden . . 45,021 10 10 6 2 30 32 5 9 49 47 126 Japa&' 73,342 849,308 8 9 8 5 34 39 II 10 40 37 127 Swi1zer1andt . 10 . . 7 . . 31 9 . . 42 World 754,620 1 Fuel exporters 6,221 1 41,957 a. Includes unallocated data; see the technical notes. b. Value added in manufacturing data are at purchaser values. c. World Bank estimates. d. Data refer to the Federal Republic of Germany before unification. 249 Table 7. Manufacturing earnings and output Earnings per employee Total earnings as a Gross output per employee Growth rate Index (1980= IIX)) percentage of value added (1980=1(X)) 1970-80 1980-90 1988 1989 1990 1970 1988 1989 1990 1970 1988 1989 1990 Low-income economies China and India Other low-income 1 Mozambique 29 2 Tanzania -13.5 42 25 3 Ethiopia 0.5 ib 9 24 20 19 19 61 110 112 103 4 Uganda 5 Bhutan 6 Guinea-Bissau . 7 Nepal . 26 . 8 Bunindi -7.5 9 Chad .. 10 Madagascar -0.8 66 56 36 37 106 60 II SierraLeone . . . 12 Bangladesh- -3.0 0.6 99 98 95 26 34 34 33 206 110 III 113 13 La0PDR .. 14 Malawi -0.8 37 126 15 Rwanda 22 16 Mali . 46 139 17 BurkinaFaso 11.7 . - 18 Niger - . . 6 . . 19 India 0.4 3.4 127 134 - . 47 48 47 . - 83 175 179 . - 20 Kenya -3.4 -0.7 101 97 92 50 44 43 42 43 202 218 235 21 Nigeria -0.8 . . - 18 . . 182 . 22 China . - 3.5 220 244 251 23 Haiti -3.3 4.6 157 . 24 Benin 25 Central African Rep. . . . . . 124 160 26 Ghana -14.8 7.8 23 - . 193 . - 27 Pakistan 3.4 6.6 159 21 22 51 164 28 logo . . . .... - 29 Guinea . . . . 30 Nicaragua -2.0 - . 16 210 - 31 Sn Lanka 1.8 106 100 18 18 . . 70 137 134 32 Mauritania 33 Yemen, Rep. 34 Honduras 40 36 35 Lesotho 36 Indoneaia 5.2 5.1 149 155 186 26 23 20 21 42 180 204 211 37 Eyp1, Arab Rep. 4.1 -2.1 94 91 - 54 37 35 89 206 220 - 38 Zimbabwe 1.6 0.5 107 115 112 43 35 35 34 98 124 128 135 39 Sudan . . - . 31 40 Zambia -3.2 3.1 168 125 98 34 27 27 27 109 128 93 Middle-income economies Lower-middle-income 41 Bolivia 0.0 -6.4 64 55 49 43 27 27 27 65 41 42 Cbte d'Ivoire -0.9 . . - . 27 52 43 Senegal -4.9 0.5 44 Philippines -3.7 5.6 154 160 174 21 24 23 23 104 100 112 119 45 PapuaNewGuinea 2.9 -1.5 - - 40 - 46 Cameroon 3.2 - . 72 30 - 47 45 81 . . 153 184 47 Guatemala -3.2 -1.9 89 100 . - - 19 20 . . 48 Dominican Rep. -1.1 . . 35 63 49 Ecuador 3.3 -1.5 95 80 96 27 33 33 41 83 114 101 112 50 Morocco -3.5 - - . . 89 51 Jordan 8.6 -3.9 101 74 63 37 23 25 24 52 lajikistan . - . . 53 Peru . - -3.5 71 . 14 . - 80 54 ElSalvador 2.4 . . . . 28 71 55 Congo -2.6 34 43 . 56 Syrian Arab Rep. 2.6 -5.4 64 66 65 33 27 28 57 Colombia -0.2 1.6 114 117 116 25 15 15 86 148 158 164 58 Paraguay 59 Uzbekistan 60 Romania 39 61 62 63 Namibia Tunisia Jamaica 4.2 -0.2 -- -0.8 101 - . 44 43 95.::. 99 78 - 64 Kyrgyzstan 65 Thailand 0.3 5.9 160 160 158 25 28 28 28 77 109 112 113 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 250 Earnings per employee Total earnings as a Gross output per employee Growth rage Index (1980- 1(h)) percentage of value added (1980100) 1970-80 1980-90 1988 1989 1990 1970 1988 1989 1990 1970 1988 1989 1990 67 Azerbaijan 68 Turkmenistan . . . . 69 Turkey 6.1 -1.2 80 100 98 26 15 19 19 108 166 181 195 70 Poland 5.5 0.1 103 114 78 24 23 19 17 71 Bulgaria . . . . 72 Costa Rica . . -2.2 41 38 39 39 73 Algeria -1.0 . 45 120 74 Panama 0.2 2.2 123 126 134 32 37 37 37 67 73 74 81 75 Armenia . . . . . 76 Chile 8.1 -1.0 103 106 106 19 16 15 15 60 . 77 Iran, Islamic Rep. 7.9 -8.2 . . . . 25 84 78 Moldova . . . . . . . 79 Ukraine . . . . 80 Mauritius 1.8 -0.1 98 97 101 34 45 45 46 139 69 75 84 81 Czechoslovakia 2.3 0.4 106 107 100 49 39 42 43 . . 131 129 82 Kazakhstan . . . . 83 Malaysia 2.0 2.6 126 128 128 29 27 26 27 96 Upper-middle-income 84 Botswana 2.6 -5.5 66 . . 35 . . . . 67 . 85 SouthAfrica 2.7 0.2 104 106 106 46 47 47 . 64 88 88 86 Lithuania . . . . 87 Hungaly 3.6 2.3 125 127 120 28 39 36 40 41 105 103 100 88 Venezuela 4.9 -5.2 80 63 58 31 28 21 16 103 111 103 107 89 Argentina -2.1 -0.8 94 75 . . 28 20 16 75 74 73 90 Uruguay . . 0.8 116 107 . . 26 26 . . . Ill 112 91 Brazil 4.0 4.8 161 163 121 22 20 21 20 71 123 124 120 92 Mexico 1.2 -3.9 67 72 75 44 18 19 20 77 119 132 138 93 Belanis . . . . 94 Russian Federation 95 Latvia 96 Trinidad and Tobago -o. 97 Gabon 98 Estonia 99 Portugal 2.5 0.7 103 103 106 34 36 36 36 100 Oman .. 101 PuertoRico . . . . . 21 22 . . 102 Korea Rep. 10.0 7.4 161 191 189 25 28 31 31 40 177 193 204 103 Greece 4.9 0.6 104 112 110 32 39 40 40 56 109 115 104 Saudi Arabia . . . . 26 105 Yugoslavia 1.3 -0.7 88 102 . . 39 26 26 59 97 75 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland 4.1 1.9 Ill 112 114 49 29 27 26 107 tlsrael 8.8 -3.6 95 71 72 36 62 37 36 108 New Zealand 1.2 -0.7 94 90 95 62 55 53 55 141 140 109 Spain 4.4 0.9 106 109 109 52 38 39 39 110 tHong Kong 6.4 4.9 143 150 147 . . 55 55 55 Ill fSingapore 2.9 5.0 149 165 175 36 28 30 32 73 122 129 135 112 UnitedKingdom 1.7 2.6 123 124 125 52 40 41 40 113 Australia 2.9 -0.3 103 101 96 53 47 45 45 . . 128 139 136 114 Italy 4.1 1.0 109 110 . . 41 41 41 . . 50 136 141 115 Netherlands 2.5 1.1 107 108 109 52 49 48 48 116 Belgium 4.7 0.1 99 101 104 46 41 40 41 . . 141 146 142 117 Austria 3.4 1.8 114 116 121 47 54 53 54 65 119 127 134 118 France . . 2.2 117 121 123 63 63 63 . . 116 123 127 119 Canada 1.8 0.0 101 101 99 53 43 44 44 68 112 120 United States 0.1 0.7 107 106 104 47 36 35 36 64 . 121 Germanya 3.5 1.8 113 114 116 46 42 41 42 60 109 114 115 122 Denmark 2.5 0.6 105 104 . . 56 52 51 . . 64 103 107 123 Finland 2.6 2.8 122 126 130 47 44 43 47 73 132 140 148 124 Norway 2.6 1.6 110 110 112 50 56 54 57 74 118 127 135 125 Sweden 0.4 0.9 103 107 106 52 34 34 35 . . 126 131 132 126 Japan 3.1 2.0 117 120 121 32 34 33 33 48 120 131 137 127 Switzerland . . World Fuel exporters a. Data refer to the Federal Republic of Germany before unification. 251 Table 8. Growth of consumption and investment Average annual growth rate (percent) General government Private Gross domestic coeswnption consumption, etc. investment 1970-80 1980-91 1970-80 /980-9! 1970-80 1980-91 Low-income economies 5.4 w 6.5 w 4.1 w 4.8 w 7.0 w 7.0 w China and India 4.8 w 8.6 w 3.7 w 6.2 w 6.2 w 9.4 w Other low-income 6.2 w 3.5 w 4.9 w 2.7 w 8.5 w 1.7 w I Mozambique -2.1 1.4 3.0 2 Tanzania a 44 3.1 3 Ethiopia 2.4 2.1 8.3 5.2 -0.8 2.0 4 Uganda 0.9 3.3 11.9 5 Bhutan 6 Guinea-Bissau 1.3 3.1 -1.8 4.0 -1.7 4.3 7 Nepal 8 Burundi 35 16.3 2.7 9 Chad 10 Madagascar 1.5 -0.2 -0.8 0.4 2.6 11 Sierra Leone a -0.4 7.0 -1.3 -1.2 -5.3 12 Bangladesh a a 2.3 3.6 4.8 -1.0 13 Lao PDR 14 Malawi 5.8 3.5 2.7 4.2 15 Rwanda 7.5 6.8 4.3 -0.4 10.4 -1.0 16 Mali '.9 4.' 6.2 1.7 3.3 6.9 17 Burkina Faso 6.6 6.0 4.7 2.8 4.4 9.3 18 Niger 3.0 1.8 0.4 -0.9 7.6 -4.6 19 India 4.1 7.5 2.9 5.3 4.5 5.1 20 Kenya 9.2 3.4 6.4 5.l 2.4 0.6 21 Nigeria 11.4 -2.1 7.8 -1.9 11.4 -8.1 22 China 5.7 10.0 4.9 7.3 7.9 12.4 23 Haiti 1.6 -1.4 3.4 0.3 13.7 -3.4 24 Benin -1.9 0.3 3.2 0.9 11.4 -4.8 25 Central African Rep. -2.4 -0.9 5.2 1.9 -9.7 3.6 26 Ghana 5.1 0.5 1.7 2.6 -2.5 9.0 27 Pakistan 4.1 9.1 4.2 4.7 3.7 5.6 28 Togo 10.2 1.0 2.0 5.3 11.9 -0.7 29 Guinea 30 Nicaragua 10.7 -1.8 31 SriLanka 0.3 6.9 4.9 3.6 13.8 1.2 32 Mauritania 10.8 -4.7 0.2 3.7 19.8 -5.6 33 Yemen, Rep. 34 Honduras 2.0 2.8 91 35 Lesotho 17.8 2.9 10.6 -0.6 23.4 9.1 36 Indonesia 13.1 4.7 6.5 4.7 14.1 6.9 37 Egypt, Arab Rep. 3.3 7.4 3.3 18.7 -0.1 38 Zimbabwe 12.1 10.1 3.8 2.8 -4.2 0.3 39 Sudan -1.5 6.9 0.8 8.2 -1.2 40 Zambia 1.4 -3.2 0.2 3.7 -10.9 0.2 Middle-income economies Lower-middle-income iii w w - w 41 Bolivia 7.9 -1.2 4.5 2.2 2.3 -8.0 42 Côte d'Ivoire 12.1 -4.0 6.9 -1.7 12.6 -11.5 43 Senegal 5.9 2.5 3.0 2.6 0.3 3.7 44 Philippines 6.8 0.9 4.3 2.4 11.3 -1.8 45 Papua New Guinea -1.3 0.1 4.5 0.6 -5.4 -0.9 46 Cameroon 5.2 6.2 6.2 1.7 11.2 -5.4 47 Guatemala 6.5 2.2 5.3 1.2 7.9 -0.7 48 Dominican Rep. 2.7 0.2 5.8 1.7 9.4 4.5 49 Ecuador 14.5 -1.1 8.2 2.0 11.0 -2.4 50 Morocco 14.0 5.1 5.5 4.0 9.9 2.5 51 Jordan 1.7 1.9 -6.9 52 Tajikistan 53 Pent 4.0 2.2 54 El Salvador 6.8 2.8 4.2 0.6 7.3 2.3 55 Congo 4.1 3.3 1.5 3.1 1.5 -11.7 56 Syrian Arab Rep. -2.2 3.9 -6.8 57 Colombia 4.5 5.. 3.1 5o -0.2 58 Paraguay 4.8 1.6 8.6 1.7 18.6 0.3 59 Uzbekistan 60 Jamaica 1.4 1.8 61 Romania -2.1 62 Namibia 1.1 -5.0 63 Tunisia 3.7 8.6 3.3 6.8 64 Kyrgyzstan 65 Thailand 9.8 4.1 6.3 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 252 Average annual growth rate (percent) Genera! government Private Gross domestic consusnption consumption, etc. tnvestment 1970-80 /980-91 1970-80 1980-91 1970-80 1980-91 67 Azerbaijan 68 Turkmenistan . . . . . . . . . 69 Thrkey 6.3 3.5 4.8 6.1 6.9 2.8 70 Poland 0.2 1.1 -0.2 71 Bulgaria . . 0.8 . . 6.5 -2.1 72 Costa Rica 6.6 1.2 4.8 3.4 9.2 4.4 73 Algeria 10.8 4.7 4.1 2.3 13.6 -3.2 74 Panama 5.8 0.7 4.4 1.3 0.3 -7.5 75 Armenia .. .. .. 76 Chile 2.4 0.2 -0.5 2.2 1.0 5.1 77 Iran, Islamic Rep. 10.8 -3.8 2.3 5.5 10.4 -5.4 78 Moldova . 79 Ukraine .. .. .. 80 Mauritius 9.8 3.2 9.2 6.4 10.0 11.2 81 Czechoslovakia 2.6 1.7 0.2 82 Kazakhstan .. .. .. 83 Malaysia 9.3 3.2 7.5 5.1 10.8 4.4 Upper-middle-income 84 Botswana 15.4 12.5 10.7 5.6 6.9 0.4 85 SouthAfrica 5.5 3.4 2.3 1.8 2.5 -3.9 86 Lithuania . . . . . . . . . 87 Hungary 2.5 2.2 3.6 0.2 7.5 -1.6 88 Venezuela . . 2.5 . . 1.7 7.1 -3.9 89 Argentina 3.8 -9.3 2.4 0.0 3.1 -6.9 90 Uruguay 4.0 2.1 -1.9 1.0 . . -5.9 91 Brazil 6.0 8.7 8.0 1.6 8.9 -0.1 92 Mexico 8.3 1.8 5.9 1.8 8.3 -1.9 93 Belarus 94 Russian Federation 95 Latvia . . . . . . , . . 96 Trinidadandlobago 9.0 1.5 6.3 -7.4 14.2 -7.1 97 Gabon 10.2 -1.1 7.3 -2.6 13.6 -6.2 98 Estonia . . . . . . . . . 99 Portugal 8.6 2.5 4.5 5.0 3.1 -2.6 100 Oman . . . 101 PuertoRico . . 5.1 . , 2.7 . . 7.0 102 Korea, Rep. 7.4 6.6 7.4 8.3 14.2 13.0 103 Greece 6.9 2.3 4.0 2.5 2.1 1.6 104 Saudi Arabia . . . . . . . . . 105 Yugoslavia 4.5 0.3 5.6 -0.1 7.2 -3.3 Low- and middle-income 4.0 w 3.4w .. 1.4w Sub-Saharan Africa 0.9 w 5.1 w 0.7 w 7.3 w -3.3 w East Asia & Pacific 7.5 w 6.6 w 5.8 w 6.5 w 9.7 w 10.1 w South Asia 4.0 w 8.1 w 3.0 w 5.1 w 4.6 w 4.9 w Europe and Central Asia Middle East & N. Africa Latin America & Caribbean 6.0w 4.3w 5.5w 1.4w 7.0w -1.3 w Severely indebted 7.3w 3.9w 5.7 w 1.7 w 8.2 w -1.8 w High-income economies 2.6w 2.5w 3.5w 2.8w 2.0w 4.0w OECD members 2.6w 2.5w 3.5w 2.8w 1.9w 4.0w 106 Ireland 6.0 -0.1 4.3 2.2 5.2 0.0 107 tlsrael 3.9 0.6 5.8 5.2 0.6 3.6 108 New Zealand 3.6 1.1 1.7 1.9 -1.0 2.9 109 Spain 5.8 5.2 3.8 3.2 1.5 6.0 110 tHongKong 8.3 5.4 9.0 6.9 12.1 4.4 Ill iSingapore 6.2 6.5 5.8 6.0 7.8 4.3 112 UnitedKingdom 2.4 1.2 1.8 3.8 0.2 5.3 113 Australia 5.1 3.4 3.2 3.2 1.9 2.2 114 Italy 3.0 2.6 4.0 3.1 1.6 2.1 115 Netherlands 2.9 1.4 3.9 1.8 0.1 2.4 116 Belgium 4.1 0.5 3.8 1.9 2.1 3.8 117 Austria 3.8 1.3 3.8 2.6 2.7 3.0 118 France 3.4 2.2 3.3 2.4 1.4 2.9 119 Canada 3.8 2.4 5.3 3.4 5.6 4.5 120 UnitedStates 1.0 3.2 3.3 2.4 2.4 3.3 121 Germanyb 3.3 1.3 3.3 2.2 0.6 2.6 122 Denmark 4.1 0.9 2.0 1.7 -0.8 3.2 123 Finland 5.3 3.6 2.8 4.3 0.5 2.0 124 Norway 5.4 2.9 3.8 1.2 3.3 -0.2 125 Sweden 3.3 1.5 1.9 2.0 -0.7 3.5 126 Japan 4.9 2.4 4.7 3.7 2.5 6.0 127 Switzerland 1.8 3:0 1.1 1.7 -1.8 4.5 World 3.0w 2.7w 3.7w 3.0w 2.8w 3.4w Fuel exporters a. General government consumption figures are not available separately; they are included in private consumption, etc. b. Data refer to the Federal Republic of Germany before unification. 253 Table 9. Structure of demand Distribution of gross domestic product (percent) General Exports of goods government Private Gross domestic Gross domestic and nonfactor Resource consumption consumption, etc. investment savings services balance 1970 1991 1970 1991 1970 1991 1970 1991 1970 1991 1970 1991 Low-income economies 10w lOw 71w 64w 21w 27w 20w 27w 7w 19w -1w -1w China and India 8w lOw 68w 59w 24w 29w 24w 31w 3w 16w Ow 1w Other low-income 13w 12w 76w 71w 15w 22w 12w 17w 14w 26w -4w -6w 1 Mozambique 20 90 42 -to 23 -52 2 Tanzania Il 16 69 96 23 22 20 -II 26 20 -2 -33 3 Ethiopia 10 21 79 78 II 10 II 0 II 8 0 -10 4 Uganda a 8 84 93 13 12 164 -1 22 7 3 -13 5 Bhutan 0 S S S S S S 29 -12 6 Guinea-Bissau 20 17 77 85 30 30 3 -3 4 13 -26 -33 7 Nepal a 10 97 85 6 19 3 5 5 14 -3 -14 8 Bunsndi 10 16 87 85 5 17 4 -1 II 10 -1 -18 9 Chad 27 20 64 97 18 8 10 -17 23 19 -8 -25 10 Madagascar 13 9 79 92 10 8 7 -1 19 17 -2 -9 II Sierra leone 12 II 74 85 17 II IS 4 30 19 -2 -6 12 Bangladesh 13 II 79 86 II 10 7 3 8 9 -4 -7 13 LaoPDR II 15 -14 14 Malawi 16 14 73 77 26 20 II 9 24 24 -15 -II 15 Rwanda 9 20 88 78 7 13 3 I 12 12 -4 -II 16 Mali 10 12 80 82 16 23 10 6 13 17 -6 -16 17 BurkinaFaso 9 17 92 79 12 23 -1 4 7 12 -12 -19 18 Niger 9 8 89 86 tO 9 3 7 II 16 -7 -3 19 India 9 12 75 69 17 20 16 19 4 9 -1 -1 20 Kenya 16 17 60 63 24 21 24 19 30 27 -1 -1 21 Nigeria 8 13 80 65 IS 16 12 23 8 36 -3 6 22 China 8 9 64 52 28 36 29 39 3 20 0 3 23 Haiti 10 . . 83 . . II . . 7 . . 14 . . -4 24 Benin 10 12 85 85 12 12 5 3 22 24 -6 -9 25 CentralAfrican Rep. 21 15 75 86 19 II 4 -I 28 IS -15 -12 26 Ghana 13 9 74 83 14 16 13 8 21 17 -I -8 27 Pakistan 10 13 81 75 16 19 9 12 8 16 -7 -7 28 logo 16 15 58 74 IS 19 26 10 50 42 II -9 29 Guinea . . 10 . . 76 . . 18 . . 14 . . 26 . . -4 30 Nicaragua 9 21 75 89 18 21 16 -10 26 22 -2 -31 31 SriLanka 12 tO 72 77 19 23 16 13 25 28 -3 -10 32 Mauritania 14 9 56 81 22 16 30 tO 41 50 8 -6 33 Yemen,Rep. . . 28 . . 70 . . 13 . . 2 . . 29 . . -Il 34 Honduras 11 10 74 70 21 24 15 20 28 31 -6 -4 35 Lesotho 12 18 120 95 12 93 -32 -13 II 13 -44 -106 36 Indonesia 8 9 78 55 16 35 14 36 13 27 -2 37 Egypt, Arab Rep. 25 10 66 83 14 20 9 7 14 30 -5 -13 38 Zimbabwe 12 21 67 61 20 22 21 18 . . 33 . . -4 39 Spdan 21 . . 64 . . 14 . . IS . . 16 . . 2 40 Zambia 16 10 39 78 28 13 45 12 54 29 17 -1 Middle-income economies Lower-middle-income 41 Bolivia 10 15 66 77 24 14 24 9 25 18 0 -5 42 COted'Ivoire 14 18 57 67 22 10 29 15 36 37 7 5 43 Senegal 15 13 74 78 16 14 II 9 27 25 -5 -5 44 Philippines 9 9 69 72 21 20 22 19 22 30 I -1 45 PapuaNewGuinea 30 24 64 63 42 29 6 13 18 39 -35 -16 46 Camemon 12 14 70 71 16 IS 18 15 26 18 2 0 47 Guatemala 8 6 78 84 13 14 14 10 19 18 I -4 48 Dominican Rep. 12 9 77 77 19 17 12 14 17 27 -7 -3 49 Ecuador II 8 75 70 18 22 14 22 14 31 -5 0 50 Momcco 12 15 73 68 18 22 15 17 18 22 -4 -6 SI Jonlan 23 78 . . 21 . . -1 . . 57 . . -22 52 Tajikistan .. .. .. .. . . .. .. .. .. . 53 Peru 12 5 70 82 16 16 17 13 18 9 2 -3 54 ElSalvador II II 76 88 13 14 13 1 25 IS 0 -12 55 Congo 17 22 82 58 24 II 1 20 35 42 -23 9 56 SynanArabRep. 17 72 . . 14 . . 10 . . 18 . -4 57 Colombia 9 11 72 66 20 15 18 23 14 21 -2 8 58 Paraguay 9 8 77 75 15 25 14 17 15 26 -1 -8 59 Uzbekistan .. .. .. .. .. .. .. .. .. 60 Jamaica 12 12 61 68 32 20 27 20 33 64 -4 0 61 Romania 14 57 34 29 . . 17 . . -5 62 Namibia . . 27 . . 64 . . 14 . . 9 . . 58 . . -5 63 Tunisia 17 16 66 66 21 23 17 18 22 39 -4 -5 64 Kyrgyzstan . . 16 . . 50 . . 34 . . 34 . . . . . . -I 65 Thailand II 10 68 58 26 39 21 32 15 38 -4 -7 66 Georgia Note: For data comparability and coverage, see the technical notes. Figureu in italics ate for yearn other than those specified. 254 Distribution of gross domestic product (percent) General Exports of good.s government Private Gross domestic Gross domestic and nonfactor Resource consumption consumption, etc. investment savings services balance 1970 199! 1970 1991 1970 1991 1970 1991 1970 1991 1970 199! 67 Azerbaijan 68 Turkmenistan . . . . . . . . . . 69 Thrkey 13 Ii 70 66 20 20 17 17 6 20 2 -3 70 Poland 20 58 21 22 20 0 71 Bulgaria . . 13 . . 73 . . 13 . . 15 . . 63 . 2 72 Costakica 13 16 74 61 21 23 14 22 28 39 -7 -1 73 Algeria 15 16 56 48 36 30 29 36 22 31 -7 6 74 Panama 15 21 61 72 28 15 24 7 38 29 -3 -8 75 Armenia . . 0 0 . . . . . 0 76 Chile 13 10 70 66 16 19 17 24 15 36 1 5 77 Iran,lslamicRep. 16 13 59 77 19 20 25 10 24 20 6 -II 78 Moldova . . . . . . 79 Ukraine . . . . . . . . . . . . . . . . 80 Mauritius 14 12 75 65 10 28 11 23 43 64 1 -5 81 Czechoslovakia a . . 67 . . 31 . . . . . . 42 . . 2 82 Kazakhstan .. .. .. .. .. .. .. .. .. 83 Malaysia 16 14 58 56 22 36 27 30 42 81 4 -5 Upper-middle-income 11 w 66 w 24 w . 23 w 16 w -1 w 84 Botswana 20 . 78 42 2 23 -41 85 South Africa 12 21 63 58 28 16 24 21 22 25 -4 5 86 Lithuania . . 16 . 63 . . 21 . . 21 . . . . 87 Hungaly 10 13 58 67 34 19 31 19 30 34 -2 0 88 Venezuela II 9 52 67 33 19 37 23 21 31 4 5 89 Argentina 10 4 68 81 22 12 22 15 9 II a 2 90 Uruguay 19 13 83 70 a 13 -I 17 15 24 -1 4 91 Brazil Il 9 69 70 21 20 20 30 7 10 0 0 92 Mexico 7 8 75 72 21 23 19 20 6 16 -3 -3 93 Belanis . . a 71 30 . . 50 94 Russian Federation . . 20 . . 41 . . 39 . . 40 . . . . . . 0 95 Latvia . 10 . . 46 . . 34 . . 43 . . 35 . 10 96 TrinidadandTobago 13 15 60 59 26 18 27 26 43 41 I 8 97 Gabon 20 17 37 41 32 26 44 42 50 50 12 16 98 Estonia . 10 65 . . 29 . . 25 . . . . 0 -4 99 Portugal 14 . . 67 . 26 . . 20 . . 24 -7 100 Oman 13 35 19 38 14 17 68 26 74 . . 54 10 101 Puerto Rico 15 15 74 64 29 16 10 22 44 76 -18 6 102 Korea,Rep. 10 Il 75 53 25 39 15 36 14 29 -10 -3 103 Greece 13 20 68 72 28 17 20 8 10 23 -8 -9 104 Saudi Arabia 20 34 . 16 . 47 0 59 31 105 Yugoslavia 18 7 55 72 32 21 27 21 18 24 -5 -1 Low- and middle-income 11 w 68 w 23 w 24 w 21 w 13w -I w Sub-Saharan Africa 12w 15w 73 w 71 w 17w 16 w 16 w 14 w 21 w 28 w -1 w -3 w East Asia & Pacific 9w 10w 66 w 55 w 26 w 35 w 25 w 36 w 7w 30 w -I w -1 w South Asia 10 w 12w 76 w 72 w 16w 19 w 14w 17 w 5w lOw -2 w -3 w Europe and Central Asia Middle East & N. Africa w 57 w 19w 25 w 29 w 5w Latin America & Caribbean 10 w 13w 70 w 21 w 19 w 20 w 13w 18 w -2 w -1 w Severely indebted lOw .. 72w 22 w 20w 21w 16w High-income economies 16 w 17 w 60 w 61 w 23 w 21 w 24 w 22 w 14 w 20 w Iw Iw OECD members 16 w 17 w 60 w 61 w 23 w 21 w 24 w 22 w 13 w 19 w 1w Iw 106 Ireland 15 16 69 56 24 19 16 28 37 62 -8 9 107 flsrael 34 28 58 58 27 23 8 14 25 28 -20 -9 108 New Zealand 13 17 65 63 25 18 22 20 23 28 -3 2 109 Spain 10 16 64 62 27 25 26 22 13 17 -1 -3 110 tHong Kong 7 8 68 60 21 29 25 32 92 141 4 3 Ill tSingapore 12 II 70 43 39 37 18 47 102 185 -20 9 112 UnitedKingdorn 18 21 62 64 20 16 21 15 23 24 I -1 113 Australia 14 19 59 62 27 19 27 19 14 18 0 0 114 Italy 13 17 60 62 27 20 28 20 16 20 0 0 115 Netherlands 15 14 57 59 30 21 28 26 42 54 -2 5 116 Belgium 13 15 60 63 24 20 27 23 52 73 2 3 117 Austria 15 18 55 55 30 26 31 26 31 41 I 118 France 15 18 58 60 27 21 27 21 16 23 1 0 119 Canada 19 21 57 60 22 20 24 19 23 25 3 -I 120 UnitedStates 19 18 63 67 18 15 18 15 6 II 0 -1 121 Germanyt 16 18 55 54 28 21 30 28 21 34 2 6 122 Denmark 20 25 57 52 26 17 23 23 28 36 -3 6 123 Finland 14 24 57 56 30 21 29 20 26 22 -1 -1 124 Norway 17 21 54 51 30 19 29 28 42 45 -1 9 125 Sweden 22 27 54 54 25 17 24 19 24 28 -1 2 126 Japan 7 9 52 57 39 32 40 34 11 10 1 2 127 Switzerland 10 14 59 57 32 27 31 29 33 35 -2 World 15 w 16 w 61 w 62 w 23 w 22 w 24 w 23 w 14 w 21 w 0w Iw Fuel exporters 14 w . . 55 w 51 w 23 w . . 31 w . . 29 w 8w a. General government consumption figures are not available separately; they are included in private consumption. etc. b. Data refer to the Federal Republic of Germany before unification. 255 Table 10. Structure of consumption Percentage share of total household consu,npttonu Food Gross rents, Other consumption Ji4el and power Transport and Cereals Clothing Other co,nmuntcatton and and Fuel and Medical consusner Total tubers footwear Total power care Education Total Automobiles Total durables Low-income economies China and India Other low-income I Mozambique 2 Tanzania 64 32 10 8 l0 3 Ethiopia 49 24 6 14 17 4 Uganda 5 Bhutan 6 Guinea-Bissau S 7 Nepal 57 38 12 14 6 3 I I 0 3 8 Burundi S S S 9 Chad S S S 10 Madagascar 59 26 6 12 6 2 4 4 I 14 II Sierra Leone 56 22 4 15 6 2 3 12 8 12 Bangladesh 59 36 8 17 7 2 I 3 0 0 3 13 LaoPDR 14 Malawi 30 9 9 9 5 4 0 10 3 27 3 IS Rwanda 29 10 II 15 6 3 6 9 27 9 16 Mali 57 22 6 8 6 2 4 10 I 12 17 Burkina Faso S 0 - 18 Niger 19 India 52 18 II 10 3 3 4 7 0 13 3 20 Kenya 38 16 7 12 2 3 10 8 I 22 6 21 Nigeria 48 18 5 4 1 3 4 3 I 35 6 61b 22 China /3 8 3 I I / /5 23 Haiti 0 S S S S 24 Benin 37 12 14 12 2 5 4 14 2 IS 5 25 Central African Rep. S Sc 26 Ghana 50 13 11 3 3 /5 27 Pakistan 37 12 6 16 13 26 28 Togo 29 Guinea 30 Nicaragua 31 Sri Lanka 43 18 7 6 3 2 3 IS I 24 5 32 Mauritania S 33 Yemen, Rep. . . S . . S 5 34 Honduras 39 . . 9 2! 8 3 . IS 35 Lcsotho . S S ' . . . . . . . 36 Indonesia 48 21 7 13 7 2 4 4 0 22 5 37 Egypt,ArabRep. 49 10 II 9 3 3 6 4 1 18 3 38 Zimbabwe 40 9 II 12 5 4 7 6 1 20 3 39 Sudan 60 . . 5 /5 4 5 3 2 . /1 40 Zambia 36 8 10 Il 4 8 14 5 I 16 I Middle-income economies Lower-middle-income 4! Bolivia 33 . 9 /2 / 5 7 /2 22 42 Côte d'Ivoire 39 13 9 5 I 9 6 10 . . 22 3 43 Senegal 49 15 II II 4 2 6 5 0 14 2 44 Philippines 51 21 4 19 5 2 4 4 2 16 2 45 Papua New Guinea . . . . . S S S 46 Cameroon 24 7 7 16 3 12 9 12 I 20 3 47 Guatemala 36 lO lO 14 5 13 4 3 0 20 5 48 Dominican Rep. 46 13 3 IS 5 8 3 4 0 21 8 1d 6c 12c 49 Ecuador 30 . . 10 7' . 30 50 Morocco 38 12 II 9 2 5 8 8 I 21 5 SI Jordan 35 5 6 5 8 6 . . 35 52 Tajikistan S S S S S S S . S S S S S S S 53 Peru 35 8 7 IS 3 4 6 10 0 24 7 54 ElSalvador 33 12 9 7 2 8 5 10 I 28 7 55 Congo 37 16 6 9 3 6 8 15 1 19 4 56 Syrian Arab Rep. . . S S S S S S S S - S - - 57 Colombia 29 . . 6 12 2 7 6 13 - - 27 58 Paraguay 30 6 12 21 4 2 3 10 I 22 59 Uebekistan S S S S S S 5 0 5 5 5 5 5 5 5 . - 60 Jamaica 36 14 5 15 5 5 5 16 I 18 61 Romania 62 Namibia 63 Tunisia 37 7 10 13 4 7 10 7 I 18 64 Kyrgyzstan 65 Thailand 30 7 16 7 3 5 5 13 0 24 66 Georgia Note.- For data comparability and coverage. see the technical notes. Figures in italics are for years other than those specified. 256 Percentage share of total household consumption5 Food Gross rents, Other consumption fuel and power Transport and Cereals Clothing Other Comenuns Cation and and Fuel and Medical consumer Total tubers footwear Total power care Education Total Automobiles Total durables 67 Azerbaijan 68 Turkmenistan . . . . . . . S . 0 69 Turkey 40 9 15 13 7 4 1 5 0 23 70 Poland 29 4 9 6 2 6 7 8 2 35 9 71 Bulgaria . . S . . . . . 0 72 Costa Rica 33 8 8 9 I 7 8 8 0 28 9 73 Algeria . . . . . . . . . . . . . . . . 74 Panama 38 7 3 11 3 8 9 7 0 24 6 75 Armenia S S S . . . . . . . 76 Chile 29 7 8 13 2 5 6 11 0 29 5 77 Iran, Islamic Rep. 37 10 9 23 2 6 5 6 I 14 5 78 Moldova . - . . . . . 79 Ukraine . . . . . . . . . . . . . . . . . 80 Mauritius 24 7 5 19 3 5 8 II 1 28 4 81 Czechoslovakia . . . . . . . . . . . . . 82 Kazakhstan .. .. .. .. .. 83 Malaysia 23 4 9 5 7 19 33 Upper-middle-income 84 Botswana 25 12 8 8 2 8 18 8 2 26 7 7 51 85 South Africa 34 12 . 17 26 86 Lithuania . . . . . . . . . . . . . . . - . 87 Hungary 25 3 9 9 5 5 7 9 2 36 8 Sc 88 Venezuela 23 . . 7 10 . . 8 II . . 36 89 Argentina 35 4 6 9 2 4 6 13 0 26 6 90 Uruguay 31 7 7 12 2 6 4 13 0 27 5 91 Brazil 35 9 10 II 2 6 5 8 1 27 8 92 Mexico 35b 10 8 5 5 12 25 93 Belanis . . . 94 Russian Federation 95 Latvia S 96 TrinidadandTobago 19 3 14 18 8 i 97 Gabon 98 Estonia 0 99 Portugal 34 8 10 9 3 6 5 13 3 24 7 100 Oman 101 Puerto Rico . . . . . . . . . . . . . . . - 102 Korea,Rep. 35 14 6 Il 5 5 9 9 25 5 103 Greece 30 3 8 12 3 6 5 13 2 26 5 104 Saudi Arabia . . . . . . . . . . . . . . . . 105 }ltgoslavia 27 4 10 9 4 6 5 II 2 32 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECI) members 106 Ireland 22 4 5 II 5 10 7 II 3 34 5 107 tlsrael 21 . . 5 20 2 9 12 10 . . 23 108 New Zealand 12 2 6 14 2 9 6 19 6 35 9 109 Spain 24 3 7 16 3 7 5 14 3 27 6 110 tHongKong 12 I 9 15 2 6 5 9 1 44 15 III tSingapore 19 . . 8 II . . 7 12 13 . . 30 112 United Kingdom 12 2 6 17 4 8 6 14 4 36 7 113 Australia 13 2 5 21 2 10 7 13 4 31 7 114 Italy 19 2 8 14 4 10 7 II 3 30 7 115 Netherlands 13 2 6 18 6 II 8 10 3 34 8 116 Belgium 15 2 6 17 7 10 9 II 3 31 7 117 Austria 16 2 9 17 5 10 8 15 3 26 7 118 France 16 2 6 17 5 13 7 13 3 30 7 119 Canada Il 2 6 21 4 5 12 14 5 32 8 120 United States 10 2 6 18 4 14 8 14 5 30 7 121 Germany5 12 2 7 18 5 13 6 13 4 31 9 122 Denmark 13 2 5 19 5 9 9 13 5 33 7 123 Finland 16 3 4 IS 4 9 8 14 4 35 6 124 Norway 15 2 6 14 5 II 8 14 6 32 7 125 Sweden 13 2 5 19 4 Il 8 II 2 32 7 126 Japan 17 4 6 17 3 10 7 9 1 34 6 127 Switzerland 17 . . 4 17 6 15 . . 9 38 . World Fuel exporters a. Data refer to either 1980 or 1985. b. Includes beverages and tobacco. c. Refers to govemment expenditure. d. Excludes fuel. e. Includes fuel. f. Excludes $overnment expenditure. g. Data refer to the Federal Republic of Germany before unification. 257 Table 11. Central government expenditure Percentage of total expenditure Hoisting, Total Overall amenities; expenditure surplus/deficit social security Economic (percentage (percentage Deft nse Education Health and welfare services Othera of GNP) of GNP) 1980 1991 1980 199! 1980 199! 1980 199/ 1980 1991 1980 199) 1980 1991 1980 199! Low-income economies China and India I 2 Other low-income Mozambique Tanzania 9.2 13.3 . . .... 6.0 . . 2.5 . S . 42.9 . . 26.1 . 28.8 . . . . . -8.4 3 Ethiopia 10.1 . . 3.7 . . 5.4 . 23.8 . . 57.0 . . 23.4 . . -4.5 4 Uganda 25.2 . . 14.9 . . 5.1 . . 4.2 . . 11.1 . . 39.5 . . 6.1 . . -3.1 5 Bhutan 0.0 0.0 12.8 10.7 5.0 4.8 4.9 8.2 56.8 48.2 20.5 28.2 40.6 43.3 0.9 -2.6 6 Guinea-Bissau . 4.2 . 2.7 . 1.4 . . . 91.7 . . 63.0 . . -17.7 7 Nepal 6.7 5.9 9.9 10.9 3.9 4.7 1.7 6.8 58.8 43.0 19.1 28.8 14.2 18.4 -3.0 -6.2 8 Bunindi . 21.7 . . -3.9 9 Chad . . . 31.2 -7.3 10 Madagascar . 6.6 1.5 . 35.9 . . 16.1 . . -5.9 II SierraLeone' 4.1 9.9 14.9 13.3 9.1 9.6 3.6 3.1 . 29.0 68.3 35.2 29.8 9.8 -13.2 -2.9 12 Bangladeshb 9.4 10.1 11.5 11.2 6.4 4.8 5.3 8.0 46.9 34.4 20.4 31.5 10.0 15.0 2.5 -0.4 13 La0PDR .. .. 14 Malawi' 12.8 5.4 9.0 8.8 5.5 7.4 1.6 3.2 43.7 35.0 27.3 40.2 37.6 29.2 -17.3 -1.9 15 Rwanda 13.1 18.8 . 4.5 . 4.1 . 41.4 . 18.0 . 14.3 . . -1.7 16 Mali 11.0 . . 15.7 . . 3.1 . . 3.0 . . 11.2 . . 56.0 21.6 . . -4.7 17 BurkinaFaso 17.0 . . 15.5 . . 5.8 . . 7.6 . . 19.3 . . 34.8 . . 14.1 0.3 18 Niger 3.8 . . 18.0 . . 4.1 . . 3.8 . . 32.4 . . 38.0 . . 18.7 . . -4.8 19 India 19.8 17.0 1.9 2.5 1.6 1.6 4.3 6.9 24.2 20.8 48.3 51.2 13.2 17.5 -6.5 -7.0 20 Kenya' 16.4 10.0 19.6 19.9 7.8 5.4 5.1 3.9 22.7 20.7 28.2 40.1 26.1 28.3 -4.6 -5.8 21 Nigeria0 . . . 22 China . . . . . . 23 Haiti 9.6 . 6.6 4.5 . 5.0 . 28.0 . 46.3 17.5 . . -4.7 24 Benin . . . . . . 25 Central African Rep. 9.7 . 17.6 . 5.1 . 6.3 . 19.6 . 41.7 . 21.9 . . -3.5 26 Ghana' 3.7 . 22.0 . 7.0 . 6.8 . 20.7 . 39.8 . 10.9 . . -4.2 27 Pakistan 30.6 27.9 2.7 1.6 1.5 1.0 4.1 3.4 37.2 11.6 23.9 54.6 17.7 21.9 -5.8 -6.2 28 Togo 7.2 . 16.7 . 5.3 . 12.0 25.2 . 33.7 . 31.9 . . -2.0 29 Guinea . . . . . 24.9 . . -4.2 30 Nicaragua 11.0 11.6 . 14.6 . 7.4 . 20.6 . 34.9 32.6 33.8 -7.3 -15.2 31 SriLanka 1.7 9.4 6.7 8.3 4.9 4.8 12.7 18.4 l5.9 24.6 58.2 34.5 41.6 29.4 -18.4 -9.5 32 Mauritania 33 Yemen, Rep . 34 Honduras .. .. 35 Lesotho 0.0 6.5 15.3 21.9 6.2 11.5 1.3 5.5 35.9 3l.6 41.2 23.1 22.7 31.8 -3.7 -0.3 36 Indonesia 13.5 8.2 8.3 9.1 2.5 2.4 1.8 1.8 40.2 27.1 33.7 51.5 23.1 20.7 -2.3 0.4 37 Egypt, Arab Rep 11.4 12.7 8.1 13.4 2.4 2.8 l3.l 17.8 7.2 8.2 57.7 45.3 53.7 39.6 -12.5 -6.8 38 Zimbabwe 25.0 16.5 15.5 23.4 5.4 7.6 7.8 3.9 18.1 22.4 28.2 26.2 35.3 35.9 -11.1 -6.9 39 Sudant 13.2 . 9.8 . 1.4 . 0.9 19.8 . 54.9 . 19.8 . . -3.3 40 Zambiat' 0.0 . 11.4 . 6.1 . 3.4 32.6 . 4.6.6 . 40.0 21.9 -20.0 -5.0 Middle-income economies Lower-middle-income 41 Bolivia . . 13.1 . . 18.7 . . 3.3 . . 18.8 . . 16.9 . . 29.3 29.0 18.8 0 -0.1 42 COted'lvoire 3.9 . . 16.3 . . 3.9 4.3 13.4 . . 58.1 . . 32.4 30.1 -1 1.1 -3.6 43 Senegal 16.8 . . 23.0 . . 4.7 . . 9.5 . . 14.4 . . 31.6 . . 23.9 . . 0.9 44 Philippines' 15.7 10.9 13.0 16.1 4.5 4.2 6.6 3.7 56.9 24.7 3.4 40.3 13.4 19.1 -1.4 -2.1 45 PapuaNewGuinea' 4.4 . . 16.5 . . 8.6 . . 2.6 22.7 . . 45.1 . . 35.2 . . -2.0 46 Camemon 9.1 6.7 12.4 12.0 5.1 3.4 8.0 8.7 24.0 48.1 41.4 21.2 15.5 22.3 0.5 -3.5 47 Guatemala . . 13.3 . . 19.5 . . 9.9 . . 7.8 . . 21.7 . . 27.8 14.4 12.0 -3.9 -1.8 48 DominicanRep. 7.8 4.8 12.6 10.2 9.3 14.0 13.8 20.2 37.1 36.5 19.3 /4.2 17.5 12.3 -2.7 0.6 49 Ecuador1' 12.5 12.9 34.7 18.2 7.8 11.0 1.3 2.5 21.1 11.8 22.6 43.6 15.0 16.0 -1.5 2.1 50 Monicco 17.9 . . 17.3 . . 3.4 . . 6.5 . . 27.8 . . 27.1 . . 34.2 . . -10 SI Jotlan 25.3 21.3 7.6 14.8 3.7 5.0 14.5 17.7 28.3 10.4 20.6 30.9 . . 41.4 . . -4.0 52 Tajiki5stan . . . . . . . . 53 Pent 21.0 16.4 15.6 21.1 5.6 5.6 0.0 0.5 22.1 . 35.7 56.4 20.4 8.8 -2.5 -0.5 54 El Salvador b 8.8 20.6 19.8 14.4 9.0 7.7 2.1 1.4 21.0 19.1 39.3 36.7 17.6 10.4 -5.9 -2.1 55 Congo 9.7 . 11.0 . 5.1 . 7.0 . 34.2 . 33.0 . 54.6 . . -5.8 56 SyrianArabRep. 35.8 31.5 5.5 7.4 0.8 1.9 11.3 3.3 41.1 30.7 5.4 25.2 48.1 24.3 -9.7 0.4 57 Colombia 6.7 . 19.1 . 3.9 . 21.2 . . 27.1 . . 22.0 . . 13.5 15.1 -1.8 2.0 58 Paraguay 12.4 13.3 12.9 12.7 3.6 4.3 19.2 14.8 18.9 12.8 33.0 42.1 9.8 9.4 0.3 3.0 59 Uzbekistan 60 Jamaica . . . . 45.7 -17.1 61 Romanja . 10.3 . 10.0 . 9.2 . . 26.6 . . 33.0 . 10.9 . . 37.0 . . 2.0 62 Namjbja . 6.5 . . 22.2 . 9.7 . . 14.8 . 17.3 . . 29.5 . . 48.2 . . -7.6 63 Tunisia 12.2 5.6 17.0 17.5 7.2 6.3 13.4 18.3 27.8 24.4 22.4 27.9 32.5 34.6 -2.9 -4.3 64 Kyrgyzslan .. .. 65 Thailand 21.7 17.1 19.8 20.2 4.1 7.4 5.1 5.9 24.2 24.3 25.1 25.1 19.1 15.5 -4.9 5.0 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for yeara other than those specified. 258 Percentage oftotal expenditure Housing, Total Overall a,nenities; expenditure surplus/deficit social security Economic (percentage (percentage Deft nse Education Health and welfare services Other a of GNP) of GNP) 1980 1991 1980 1991 1980 1991 1980 1991 1980 1991 1980 199! 1980 199! 1980 1991 67 Azerbaijan 68 'l'urkmenistan . . . . . . . 69 Turkey 15.2 10.4 14.2 17.6 3.6 3.0 6.1 3.3 34.0 25.2 26.9 40.5 26.3 30.4 -3.8 -7.6 70 Poland . . . . . 71 Bulgaria . 5.6 . 6.2 . 4.8 . 23.9 . 46.6 . 12.8 . . 77.3 . . -9.9 72 CostaRica 2.6 24.6 19.1 28.7 32.0 9.5 13.3 18.2 8.6 16.4 27.0 26.3 25.9 -7.8 -1.4 73 Algeria . . . . . . . 74 Panama 0.0 5.3 13.4 /7/ 12.7 20.5 13.5 23.8 21.9 6.1 38.4 27.2 34.2 30.3 -5.8 3.5 75 Armenia . . . . 76 Chile 12.4 . 14.5 . 7.4 . 37.1 . 13.8 . 14.8 . 29.1 . . 5.6 77 Iran, Islamic Rep 15.9 9.6 21.3 20.9 6.4 7.9 8.7 15.5 24.0 16.1 23.7 29.9 35.7 22.8 -13.8 -2.8 78 Moldova . . . 79 Ukraine . . . . . . . 80 Mauritius 0.8 1.5 17.6 14.6 7.5 8.7 21.4 17.8 11.7 15.0 41.0 42.3 27.4 23.8 -10.4 0.0 81 Czechoslovakia . 7.1 . 1.9 . 0.4 . 27.0 . 40.2 . 23.4 . . 55.6 -6.9 82 Kazakhstan .. 83 Malaysia . . 29.6 30.6 -6.2 -2.3 Upper-middle-income 84 Botswanab 9.8 /3.3 22.2 20.5 5.4 5.! 7.9 16.2 26.9 16.8 27.9 28.2 36.5 41.9 -0.2 14.0 85 South Africa . 23.5 33.6 -2.5 -0.3 86 Lithuania . . . . . . . . 87 Hungary 4.4 3.6 1.8 3.3 2.7 7.9 22.3 35.3 44.0 22.0 24.7 27.9 58.3 54.7 -2.9 0.8 88 Venezuela 5.8 . 19.9 . 8.8 . 9.5 . 20.2 . 35.7 . 18.7 23.9 0 4.5 89 Argentina . 9.9 . 9.9 . 3.0 . . 39.4 . 16.0 . 2/.? 19.2 13.! -3.6 -0.5 90 Uruguay 13.4 9.2 8.8 7.4 4.9 4.5 48.5 50.3 11.4 8.7 13.0 20.0 22.7 27.2 0 0.4 91 Brazil 4.0 3.5 0.0 3.1 8.0 6.7 32.0 25.5 24.0 3.2 32.0 57.9 20.9 35.1 -2.5 -5.9 92 Mexico 2.3 2.4 18.0 13.9 2.4 1.9 18.5 13.0 31.2 13.4 27.6 55.5 17.4 /8.! -3.1 0.8 93 Belarus . . . 94 Russian Federation . . . . . . 95 Latvis . . . . . . . . . . 96 Trinidadandlobago 1.7 . . 11.6 . . 5.8 . . 15.9 43.5 21.5 32.5 . . 7.8 97 Gabon' . . . . . . 40.5 37.8 6.8 -2.0 98 Estonia . . . . . . . . . . . 99 Portugal 7.4 . 11.2 . 10.3 . 27.0 . 19.9 . 24.2 . 39.6 43.3 -10.1 -5 100 Oman 51.2 35.4 4.8 11.4 2.9 5.4 2.0 13.1 18.4 10.3 20.8 24.4 43.1 44.6 0.5 -8.1 101 Puerto Rico . . . . . . . . 102 Korea, Rep 34.3 22.2 17.1 15.8 1.2 2.0 7.5 11.3 15.6 19.2 24.3 29.5 17.9 17.3 -2.3 -1.7 103 Greece 12.6 6.7 10.0 6.0 10.3 8.7 31.3 0.8 16.6 8.8 19.2 68.9 34.4 600 -4.8 -26.2 104 Saudi Arabia . . . . . . . . lOS Yugoslavia 50.0 53.4 0.0 0.0 0.0 0.0 6.3 6.0 18.8 19.6 25.0 21.0 9.0 5.2 -1.1 0.3 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland 3.4 3.3 /1.4 12.2 13.7 13.0 27.7 29.1 18.4 12.8 25.4 29.4 48.9 47.5 -13.6 -2.4 107 tlsrael 39.8 22.4 9.9 10.4 3.6 3.7 14.4 30.5 13.4 10.1 19.0 22.9 72.4 36.2 -16.1 -5.7 108 NewZealandt 5.1 4.1 14.7 12.4 15.2 12.0 31.1 37.4 15.0 10.6 18.9 23.7 39.0 43.7 -6.8 1.5 109 Spain 4.3 5.4 8.0 5.6 0.7 13.7 60.3 37.7 11.9 11.0 14.8 26.6 26.6 34.0 -4.2 -2.3 110 tHong Kong . . . . . . . . Ill tSingapore 25.2 24.0 14.6 19.9 7.0 4.6 7.6 8.2 17.7 16.8 27.9 26.5 20.8 22.1 2.2 11.2 112 UnitedKingdom 13.8 11.1 2.4 3.2 13.5 13.3 30.0 31.8 7.5 8.5 32.9 32.0 38.2 38.2 -4.6 0.8 113 Australia 9.4 8.6 8.2 7.0 10.0 12.7 28.5 31.2 8.1 8.3 35.8 32.2 23.1 27.7 -1.5 0.6 114 Italy 3.4 . 8.4 . . 12.6 . . 29.6 . . 7.2 . 38.7 . 41.0 49.6 -10.7 -10.0 115 Netherlands 5.6 4.8 13.1 10.5 11.7 12.4 39.5 42.5 10.9 6.4 19.2 23.4 52.5 52.5 -4.5 -2.8 116 Belgium 5.7 . 15.0 . . 1.6 . . 44.7 . . 16.0 . 17.0 . 51.3 49.2 -8.2 -5.4 Ill Austria 3.0 2.4 9.7 9.4 13.3 12.9 48.7 47.9 11.7 9.1 13.5 18.3 37.7 39.8 -3.4 -.4.8 118 France 7.4 6.3 8.6 6.9 14.8 /5.3 46.8 46.4 6.8 5.1 15.6 20.0 39.3 43.7 -0.1 -1.4 119 Canada 7.7 7.4 3.8 2.9 6.7 5.2 35.4 36.4 19.4 /1.2 27.3 36.8 21.8 23.9 -3.6 -2.7 120 UnitedStates 21.2 21.6 2.6 1.7 10.4 13.8 37.8 28.7 9.7 10.1 18.2 24.1 21.7 25.3 -2.8 -.4.8 121 Germanyc 9.1 8.3 0.9 0.6 19.0 18.1 49.6 48.9 8.7 8.7 12.6 /5.4 30.3 32.5 -1.8 -2.5 122 Denmark 6.5 5.! 10.4 9.5 1.8 1.1 44.7 39.9 6.5 7.6 30.0 36.8 40.4 41.7 -2.7 -0.3 123 Finland 5.6 4.7 14.7 /4.9 10.5 11.2 28.2 36.7 27.0 18.8 14.0 /3.7 28.4 31.0 -2.2 0.1 124 Norway 7.7 8.0 8.7 9.4 10.6 10.3 34.7 39.3 22.7 17.5 15.6 /5.5 39.2 46.3 -2 0.7 125 Sweden 7.7 6.3 10.4 9.7 2.2 0.8 51.5 56.4 l0.9 8.0 17.3 18.8 39.8 44.2 -8.2 0.7 126 Japan" . . . . . 18.4 15.8 -7 -1.6 127 Switzerland 10.2 . 3.4 11.7 . . 49.3 14.2 11.2 . 19.5 . . -0.2 World Fuel exporters a. See the technical notes. b. Data are for budgetary accounts only. c. Data refer to the Federal Republic of Germany before unification. 259 Table 12. Central government current revenue Percentage of total cuj,'eng revenue Tax revenue Income, profit, Goods and International Total current and capital services trade and revenue gains Social security (domestic taxes) transactions Other a Nontax revenue (percentage of GNP) 1980 199! 19&) 1991 1980 1991 1980 1991 1980 1991 1980 199! 1980 199! Low-income economies China and India Other low-income I Mozambique 2 Tanzania 32.5 0.0 40.8 17.3 1.6 7.8 3 Ethiopia 20.9 0.0 24.3 35.7 . . 3.7 15.4 18.7 4 Uganda 11.5 . 0.0 . . 41.0 . . 44.3 . . 0.2 3.1 3.1 5 Bhutan 13.8 7.5 0.0 0.0 39.1 16.6 0.4 0.4 2.3 0.6 44.3 75.0 11.4 19.6 6 Guinea-Bissau . . . . . . . . . . . . . . . . . . . . . . . . . . 14.! 7 Nepal 5.5 9.9 0.0 0.0 36.8 36.7 33.2 30.8 8.2 5.5 16.2 17.1 7.8 9.5 8 Bwundi 19.3 . . 1.0 . . 25.3 . . 40.4 . . 8.4 . . 5.6 . . 14.0 9 Chad . . 22.6 . . 0.0 . . 33.7 . . 15.3 . . 6.6 . . 21.8 . . 8.9 10 Madagascar 16.6 15.3 11.3 0.0 39.3 19.5 27.6 44.5 2.7 1.1 2.4 19.5 13.4 9.1 11 SiereaLeo, 22.4 31.5 0.0 0.0 16.3 23.4 49.6 40.4 1.5 0.2 10.1 4.6 16.9 6.6 12 Bangladeshb 10.1 8.6 0.0 0.0 25.5 25.8 28.6 27.3 3.9 15.2 31.9 23.0 11.3 11.4 13 La0PDR .. .. .. .. .. .. .. .. .. .. .. 14 Malawi° 33.9 35.0 0.0 0.0 30.9 33.2 22.0 17.7 0.3 1.2 12.9 12.9 20.7 23.7 IS Rwanda 17.8 . . 4.1 19.3 . . 42.4 . . 2.4 14.0 . . 12.8 16 Mali 17.9 0.0 36.8 . . 17.9 19.5 8.0 11.0 17 Burkina Faso 17.8 . . 7.8 15.9 . . 43.7 . . 4.3 10.5 13.6 18 Niger 23.8 . . 4.0 . . 18.0 . . 36.4 . . 2.6 . . 15.3 . - 14.7 19 In&a 18.3 15.4 0.0 0.0 42.5 35.5 22.0 28.8 0.6 0.4 16.6 19.9 11.7 14.3 20 Kenya" 29.1 29.8 0.0 0.0 38.8 43.2 18.5 15.0 1.0 1.5 12.6 10.5 22.6 21.2 21 Nigeria" 22 China 23 Haiti 13.9 0.0 15.5 48.4 9.6 12.6 10.7 24 Benin 25 Central African Rep. 16.1 6.4 20.8 . . 39.8 . . 7.8 9.1 . . 16.4 26 Ghana' 20.5 . . 0.0 . . 28.2 . . 44.2 . . 0.2 . . 6.9 . . 6.9 27 Pakistan 13.8 10.0 0.0 0.0 33.6 32.2 3.4.4 30.2 0.2 0.3 17.9 27.2 16.4 16.9 28 Togo 34.4 . . 5.8 15.3 . . 32.0 . . -1.7 . . 14.2 . . 31.4 29 Guinea 28.1 . . 1.0 . . 6.4 17.1 279 74.4 0.7 2.4 35.8 6.1 . . 14.6 30 Nicaragua 7.8 16.9 8.9 11.8 37.3 37.5 25.2 17.6 10.7 10.5 10.1 5.8 24.9 16.8 31 SriLanka 15.5 12.8 0.0 0.0 26.8 46.1 50.5 25.9 1.9 4.7 5.3 10.5 20.3 20.4 32 Mauritania 33 Yemen. Rep. .. .. .. .. .. .. .. .. -. 34 Honduras 30.8 . . 0.0 . . 23.8 . . 37.2 . . 1.8 . . 6.5 . . 15.4 35 Lesotho 13.4 16.9 0.0 0.0 10.2 16.7 61.3 51.8 1.2 0.1 13.9 14.5 17.1 26.8 36 Indonesia 78.0 61.8 0.0 0.0 8.6 23.7 7.2 6.4 1.2 2.7 4.9 5.4 22.2 21.1 37 Egypt, Arab Rep. 16.2 15.9 9.1 14.2 15.1 11.9 17.3 14.0 7.7 8.2 34.6 35.8 47.1 35.4 38 Zimbabwe 46.2 44.4 0.0 0.0 27.9 26.3 4.4 19.0 1.2 1.0 20.2 9.3 24.4 31.5 39 Sudanb 14.4 . 0.0 26.0 . . 42.6 . . 0.7 . . 16.3 . . 14.0 40 Za.nbjab 38.1 0.0 43.1 . . 8.3 3.1 95.8 73 4.2 27.0 /1.9 Middle-income economies Lower-middle-income 41 Bolivia . . 5.1 . . 8.6 . . 34.0 . . 6.0 . . 8.4 . . 37.9 . . 16.6 42 C&ed'Ivoire 13.0 16.7 5.8 6.8 24.8 278 42.8 278 6.1 11.3 7.5 9.6 23.4 26.5 43 Senegal 18.4 . . 3.7 . . 26.0 . . 34.2 . . 11.4 . . 6.3 . . 24.9 44 Philippines 21.1 28.5 0.0 0.0 41.9 25.8 24.2 28.5 2.2 2.7 10.6 14.6 14.0 17.0 45 Papua New Guineati 60.5 . . 0.0 12.1 . . 16.4 . . 0.6 10.5 . . 23.5 46 Cameroon 21.7 45.2 8.0 6.4 18.0 20.2 38.4 14.0 5.9 9.! 7.9 5.1 16.2 19.0 47 Guatemala 11.2 18.1 11.2 0.0 26.4 23.2 30.2 33.8 11.1 72 9.9 177 11.3 9.7 48 DominicnRep. 19.3 21.4 3.9 4.5 21.6 22.5 31.2 40.3 1.7 1.3 22.4 10.0 14.7 12.7 49 Ecuador 44.6 56.9 0.0 0.0 17.4 21.5 30.8 14.3 3.0 5.5 4.3 1.7 13.5 18.1 50 Morocco 19.2 . . 5.4 . . 34.7 . . 20.8 . . 7.4 . . 12.5 . . 24.0 51 Jonlan 13.2 16.1 0.0 !.6 7.3 20.5 47.8 26.4 9.5 72 22.2 28.3 30.5 52 TajiIstan .. .. .. .. .. .. .. .. .. .. .. 53 Pen, 25.9 9.0 0.0 0.0 37.2 74.0 27.1 10.8 2.2 3.5 7.7 2.7 17.9 8.3 54 El Salvador b 23.2 23.1 0.0 0.0 29.8 45.5 37.0 20.6 5.6 5.2 4.5 5.5 11.7 9.1 55 Congo 48.8 . . 4.4 7.6 . . 13.0 . . 2.7 . . 23.5 39.1 56 SyrianArabRep. 9.7 30.7 0.0 0.0 5.3 3.2 14.3 7.4 10.1 35.1 60.7 23.5 26.8 24.4 57 Colombia 24.9 278 11.2 12.6 22.6 277 20.6 178 6.8 6.7 13.9 7.4 12.1 13.4 58 Paraguay 15.2 9.3 13.1 0.0 17.7 19.5 24.8 20.! 20.5 24.8 8.8 26.2 10.6 12.3 59 Uzbekistan .. .. .. .. .. .. .. 60 Jamaica 33.7 3.7 49.3 . . 3.1 6.3 4.0 . . 31.9 61 Romania . . 35.2 28.9 . . 23.2 . . 3.1 1.5 . . 8.1 . . 37.3 62 Natnibia . . 23.4 . . 0.0 . . 25.1 . . 37.5 . . 0.5 . . 13.5 . . 39.1 63 Thnisia 14.6 14.5 9.3 12.3 23.9 22.6 24.7 28.4 5.6 4.9 22.0 17.4 32.3 30.0 64 Kyrayzstan .. .. .. .. .. .. .. .. .. .. .. 65 l'ha,Jand 17.7 25.8 0.2 0.7 46.0 43.6 26.2 19.1 1.8 3.1 8.1 7.7 14.5 20.5 66 Georgia Note: For data compaibility and covelage, see the technical notes. Figures in italics are for yeats other than those specified. 260 Percentage of total current revenue Tax revenue Income, profit, Goods and International Total current and capital services trade and revenue gains Social security (domestic taxes) transactions Other a Nontax revenue (percentage of GNP) 1980 199! 1980 199! 1980 199! 1980 1991 1980 1991 1980 199! 1980 199! 67 Azerbaijan 68 Turkmenistan . . . 0 S ' . . . . . . . . S 69 Thrkey 49.1 44.9 0.0 0.0 19.7 34.6 6.0 5.1 4.6 2.9 20.7 12.6 22.3 20.7 70 Poland . . S S . . . . . . . . . . . . . 71 Bulgaria . . 31.7 . . 13.5 . . 12.7 1.2 0.4 . . 40.4 74.3 72 CostaRica 13.7 8.9 28.9 28.6 30.4 27.7 18.9 19.7 2.3 1.1 5.8 14.0 18.7 24.6 73 Algeria . . . . . . . . . . . . . . . . . . . . . . . . . 74 Panama 21.2 15.6 21.2 21.3 16.7 15.8 10.3 11.1 3.8 3.1 26.7 33.1 28.4 32.5 75 Armenia .. .. .. .. .. .. .. .. .. 76 Chile 17.6 . . 17.4 . . 35.8 . . 4.3 . . 4.9 . . 19.9 . . 33.2 77 Iran, Islamic Rep. 3.9 11.5 7.4 5.3 3.6 6.7 11.7 45.5 5.3 4.9 68.2 26.1 21.6 19.9 78 Moldova . . . 79 Ukraine . . . . . . . . . . . . . . . . . . . . . . . . . 80 Mauritius 15.3 13.7 0.0 4.2 17.2 22.6 51.6 46.7 4.3 6.8 11.6 5.9 21.0 24.2 81 Czechoslovakia . . 23.4 . . 0.0 . . 36.9 . . 6.4 . . 23.0 . . 10.2 49.2 82 Kazakhstan 83 Malaysia 375 331 04 00 168 209 330 180 18 24 105 256 273 281 Upper-middle-income 84 Botswanab 33.3 38.8 0.0 0.0 0.7 1.9 39.1 13.4 0.1 0.1 26.7 45.9 36.6 63.4 85 South Africa 55.8 46.8 1.1 1.9 23.8 32.7 3.3 8.8 3.2 2.7 12.7 7.1 25.0 32.3 86 Lithuania 87 Hungaly 185 179 153 292 383 313 69 58 48 02 161 155 555 556 88 Venezuela 67.4 61.4 4.6 5.3 4.2 3.4 6.8 8.2 1.8 1.0 15.2 20.8 22.2 24.3 89 Argentina 0.0 5.6 16.7 33.6 16.7 15.6 0.0 25.9 33.3 10.5 33.3 8.9 17.4 13.7 90 Umguay 10.9 6.7 23.4 27.0 43.3 35.9 14.2 9.8 2.7 15.5 5.5 5.1 23.1 27.8 91 Brazil 14.3 6.9 28.6 10.4 28.6 8.2 7.1 0.7 3.6 2.0 17.9 71.8 23.4 67.9 92 Mexico 36.7 36.5 14.1 13.6 28.9 56.0 27.6 4.6 -12.6 -18.3 5.3 7.7 15.6 14.7 93 Belanis 94 Russian Federation 95 Latvia 96 Trinidl and Tobago 69.8 10 40 70 07 175 428 97 Gabon 39.9 27.6 0.0 0.8 4.8 23.7 19.7 17.4 2.0 1.2 33.7 29.3 39.4 35.8 98 Estonia 99 Portugal 19.4 23.8 26.0 25.9 33.7 36.9 5.1 2.5 8.7 3.2 7.1 7.7 31.1 36.6 100 Oman 26.0 21.3 0.0 0.0 0.5 1.0 1.4 3.1 0.3 0.7 71.8 73.9 42.9 35.6 101 Puerto Rico . . . . . . . . . . . . . . . . . . . . . . . . . 102 Korea, Rep. 22.3 31.3 1.1 5.0 45.9 33.3 15.0 9.2 3.2 10.9 12.5 10.4 18.3 17.4 103 Greece 17.4 19.8 25.8 32.1 31.6 40.4 5.0 0.1 9.6 -2.3 10.6 9.9 29.7 32.2 104 Saudi Arabia . . . . . . . . . . . . . . . . . . . . . . . . . 105 Yugoslavia 0.0 0.0 0.0 0.0 64.3 66.4 35.7 31.3 0.0 0.0 0.0 2.3 7.9 5.5 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland 34.3 36.3 13.4 14.4 30.1 31.3 9.2 8.1 1.9 3.4 11.1 6.5 37.7 42.9 107 tlsrael 40.7 32.9 10.1 7.4 24.5 36.1 3.6 1.8 7.0 5.0 14.1 16.8 52.0 29.1 108 New Zealandb 67.3 57.3 0.0 0.0 18.0 26.1 3.2 1.9 1.3 3.0 10.3 11.8 34.9 39.1 109 Spain 23.2 32.1 48.0 36.9 12.6 22.6 3.8 2.1 4.4 0.8 8.0 5.6 24.0 32.0 110 tHong Kong . . . . . . . . . . . . . . . . . . . . . . . Ill tSingapore 32.5 25.6 0.0 0.0 15.8 16.0 6.9 2.0 13.9 13.8 30.9 42.6 26.3 27.7 112 UnitedKingdom 37.7 39.0 15.6 16.3 27.8 29.0 0.1 0.1 5.7 6.9 13.1 8.8 35.2 37.4 113 Australia 60.8 64.8 0.0 0.0 23.3 20.6 5.4 3.3 0.3 1.5 10.1 9.9 22.1 27.8 114 Italy 30.0 35.1 34.7 28.7 24.7 30.7 0.1 0.0 2.5 2.4 8.1 3.1 31.2 39.9 115 Netherlands 29.6 31.7 36.3 35.9 20.8 21.1 0.0 0.0 2.7 2.8 10.6 8.6 49.1 49.7 116 Belgium 38.5 34.9 30.6 35.3 24.2 23.4 0.0 0.0 2.5 3.2 4.3 3.2 44.0 44.3 117 Austria 21.1 19.6 35.0 36.4 25.6 24.7 1.6 1.5 9.1 8.7 7.7 8.9 34.9 35.6 118 France 17.7 18.2 41.2 43.8 30.9 27.2 0.1 0.0 2.7 3.8 7.4 7.0 39.4 40.9 119 Canada 52.6 52.4 10.4 13.5 16.6 18.8 7.0 3.5 -0.2 0.1 13.6 11.6 19.2 21.3 120 United States 56.6 50.7 28.2 35.1 4.4 3.7 1.4 1.5 1.2 1.0 8.2 8.0 19.9 19.8 121 Germanyc 18.7 16.0 54.2 51.0 23.1 27.5 0.0 0.0 0.1 -0.4 3.9 6.0 28.7 30.6 122 Denmark 35.9 37.5 2.3 3.8 46.9 40.8 0.1 0.1 3.3 3.3 11.6 14.6 36.4 40.7 123 Finland 26.7 28.4 11.5 11.6 49.1 45.5 2.0 0.9 3.0 3.4 7.7 10.1 27.5 31.3 124 Norway 27.4 16.6 22.3 24.2 39.6 34.4 0.6 0.5 1.1 1.3 8.9 23.0 42.4 47.4 125 Sweden 18.2 12.3 33.2 32.4 29.1 30.4 1.2 0.5 4.3 9.4 14.1 15.1 35.4 44.4 126 Japa&' 70.8 69.2 0.0 0.0 20.8 16.9 2.4 1.3 0.8 7.4 5.2 5.2 11.6 14.5 127 Switzerland 14.0 . . 48.0 19.3 . . 9.5 2.0 7.3 18.9 World Fuel exporters a. See the technical notes. b. Data are for budgetary accountu only. c. Data refer to the Federal Republic of Germany before unification. 261 Table 13. Money and interest rates Monetary holdings, broadly defined Average Nominal interest rates of banks Average annual o.nntwI (average annual percentage) nominal growth Average outstanding iijiatman rate (percent) as a percentage of GDP 1)epostt rate Lending rate (GOP def sor) /970-80 1980-91 /970 1980 1991 1980-91 /980 1991 1980 1991 Low-income economies China and India Other low-income I Mozambique 37.6 2 Tanzania 22.6 22.9 37.2 . . 25.7 4.0 17.0 11.5 31.0 Ethiopia 14.4 2.4 3 4 Uganda 28.1 12.5 14.0 I(, l7 25.3 56.9 6.8 1.0 31.2 . 10.8 . 6.0 34.4 5 Bhutan 32.3 22.0 8.3 6.5 15.0 6 Guinea-Bissau 56.3 7 Nepal '99 199 10.6 21.9 36.1 9.1 4.0 8.5 14.0 14.4 8 Burundi 20.1 9.9 9.1 13.5 4.3 2.5 . 12.0 9 Chad 15.2 9.0 9.4 20.0 19.4 1.1 5.5 7.5 11.0 10 Madagascar 13.8 16.0 17.3 22.3 19.0 16.8 5.6 9.5 II Sierra Leone 19.9 57.7 12.6 20.6 15.3 59.0 9.2 405 11.0 52.5 12 Bangladesh 21.0 16.7 30.0 9.3 8.3 12.1 11.3 15.9 13 Lao PDR 7.2 14.0 4.8 /5.0 14 Malawi 14.7 21.6 20.5 14.9 7.9 12.5 16.7 20.0 IS Rwanda 21.5 8.5 10.7 13.6 16.6 4.1 6.3 8.8 13.5 19.0 16 Mali 18.5 8.7 13.8 17.9 20.8 4.3 6.2 7.0 9.4 16.0 17 Burkina Faso 21.5 10.9 9.3 15.9 20.2 4.0 6.2 7.0 9.4 16.0 18 Niger 23.9 5.2 5.2 13.3 19.8 2.5 6.2 7.0 9.4 16.0 19 India 17.3 16.8 23.9 36.2 44.1 8.2 16.5 17.9 20 Kenya 19.8 15.1 31.2 36.8 40.7 9.3 5.8 1317 10.6 18.8 21 Nigeria 33.7 15.7 9.2 23.8 20.2 18.2 5.3 14.9 8.4 20.0 22 China 25.4 . . 33.6 84.6 5.8 5.4 5.0 11.2 23 Haiti 24.5 8.6 12.0 26.1 . . 7.1 10.0 . 24 Benin 19.0 5.4 10.1 17.1 26.5 1.6 6.2 7.0 . . 16.0 25 Central African Rep. 16.0 4.4 16.0 18.9 17.1 5.2 5.5 7.5 10.5 16.2 26 Ghana 36.4 42.9 18.0 16.2 12.5 40.2 11.5 21.3 19.0 27 Pakistan 17.1 13.3 41.2 38.7 35.9 7.0 . . . . . 28 logo 22.2 5.9 17.2 29.0 35.2 4.4 6.2 7.0 . . 16.0 29 Guinea 30 Nicaragua 1i2 0.0 00 583.4 7.5 . 31 SriLanka 23.1 15.2 22.0 35.3 33.1 11.2 14.5 18.5 19.0 13.8 32 Mauritania 21.5 11.6 9.5 21.3 26.5 8.6 5.5 5.0 12.0 10.0 33 Yemen, Rep. 18.7 . . . 9.3 34 Honduras 16.0 13.1 19.5 22.6 29.0 6.8 7.0 11.5 18.5 21.9 35 Lesotho 17.1 . . 34.6 13.8 . . 13.0 11.0 20.0 36 Indonesia 35.4 26.2 8.0 13.2 40.5 8.5 6.0 23.3 . . 20.6 37 Egypt, Arab Rep. 26.0 . 21.8 33.5 52.2 91.7 12.6 8.3 12.0 13.3 19.0 38 Zimbabwe . . . . . . . . 45.1 13.5 3.5 8.8 17.5 15.5 39 Sudan 28.3 28.0 17.5 32.5 . . 6.0 40 Zambia 10.7 29.9 32.6 . . 7.0 9.5 Middle-income economies Lower-middle-income 41 Bolivia 29.4 444.2 14.8 16.2 318.4 263.8 18.0 23.8 28.0 41.2 42 Côte d'Ivoire 22.6 3.7 24.7 25.8 31.5 3.9 6.2 7.0 9.4 16.0 43 Senegal 19.6 5.9 14.0 26.6 22.2 5.9 6.2 7.0 9.4 16.0 44 Philippines 19.2 16.8 29.9 26.4 32.2 14.6 12.3 18.8 14.0 23.1 45 Papua New Guinea 8.1 32.9 33.1 5.2 6.9 9.1 11.2 14.1 46 Cameroon 22.5 7.0 13.5 18.3 24.4 4.5 7.5 8.0 13.0 47 Guatemala 18.6 17.2 17.1 20.5 19.3 15.9 9.0 24.4 11.0 34' 48 Dominican Rep. 18.1 28.0 17.9 21.8 20.1 24.5 49 Ecuador 24.2 35.5 20.0 20.2 13.4 38.1 9.0 4.7 50 Morocco 18.7 14.5 31.1 42.4 58.3 7.1 4.9 8.5 7.0 9.0 51 Jonian 24.3 13.0 134.8 1.6 52 Tajikistan 53 Pent 33.6 224.0 17.8 16.4 . . 287.4 172.9 54 El Salvador 17.3 17.2 22.5 28.1 27.8 17.4 16.1 19.7 55 Congo 15.7 6.7 16.5 14.7 20.0 0.6 65 7.8 11.0 12.5 56 SyrianArabRep. 26.5 /9.2 34.8 40.9 14.4 5.0 57 Colombia 32.7 20.5 23.7 26.1 25.0 37.2 /9.0 47.1 58 59 60 Paraguay Uzbekistan Jamaica . 25.9 15.7 .. 24.4 16.9 31.4 19.8 35.4 40.3 25.1 19.6 10.3 27.4 13.b 35.6 61 Romania 10.3 33.4 36.6 6.2 62 Namibia . . . . . . . . 12.5 . . . . . 63 Tunisia 20.3 15.5 33.0 42.1 . . 7.3 2.5 7.4 7.3 9.9 64 Kyrgyzstan .. .. .. .. .. .. .. 65 Thailand 17.9 18.9 23.6 37.3 71.5 3.7 12.0 12.3 18.0 25.0 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 262 Monetary holdings, broadly defined Average Nominal interest rates ofbanks Average annual annual (average annual percentage) nominal growth Average outstanding inflation rate (percent) as a percentage of GDP Deposit rate Lending rate (GDP deflator) 1970-80 1980-9! 1970 1980 1991 1980-91 1980 199! /980 199! 67 Azerbaijan 68 Thrkmenistan 69 Turkey 32.9 5i.. 27.9 17.2 21.6 4 62.9 25.7 70 Poland 58.6 58.4 29.4 63.1 3.0 27.8 8.0 101.4 71 Bulgaria 7.9 1.6 5.' 72 Costa Rica 30.6 25.7 18.9 38.8 38.3 22.9 27.3 38.9 73 Algeria 24.1 14.3 53.6 58.5 10.2 74 Panama 36.9 2.4 75 Armenia 76 Chile 194.2 29.8 12.5 21.0 38.5 20.5 37.5 22.3 47.1 28.6 77 Iran, Islamic Rep. 33.2 16.7 26.1 54.5 14.1 78 Moldova 79 Ukraine 80 Mauritius 24.3 22.0 37.5 41.1 65.4 8.1 . . 12.3 17.8 81 Czechoslovakia 6.6 63.8 3.5 2.7 8.1 82 Kazakhstan 83 Malaysia 25.2 16 3 6 1.7 6.2 7.2 78 81 Upper-middle-income 84 Botswaña 25.8 28.2 27.0 13.3 5.0 11.4 8.5 11.8 85 South Africa 15.6 16.6 59.9 50.9 56.2 14.4 5.5 17.3 9.5 20.3 86 Lithuania 87 Hungary 10.3 3.0 23.0 28.0 88 Venezuela 26.4 20.2 24.1 43.0 21.2 31.1 29.8 89 Argentina 140.8 368.5 27.5 22.2 7.6 416.8 79.4 60.3 112.9 90 Uruguay 78.4 69.1 24.5 32.1 43.8 64.4 50.3 75.2 66.6 152.9 91 Brazil 9.7 18.4 327.7 115.0 913.2 92 Mexico -47.0 62.0 26.1 26.2 24.1 66.5 20.6 17.1 28.1 93 Belanis 94 Russian Federation 95 Latvia 96 Trinidad and Tobago 27.9 59 28.2 32.0 5.j 10.0 I 32 97 Gabon 31.3 5.2 14.5 15.2 21.7 1.3 7.5 8.8 12.5 16.0 98 Estonia 99 Portugal 20.2 18.6 87.6 80.8 74.7 17.4 19.0 14.6 18.8 22.9 100 Oman 29.4 11.0 13.8 30.6 -3.0 7.1 9.5 101 Puerto Rico 3.4 102 Korea, Rep. 30.4 21.3 32.1 31.7 52.3 5.7 19.5 10.0 18.0 10.0 103 Greece 23.9 22.3 42.9 61.6 79.3 17.7 14.5 20.7 21.3 29.5 104 Saudi Arabia 43.7 7.8 17.6 18.6 -3.1 105 Yugoslavia 28.4 119.0 54.8 59.1 122.9 5.9 11.5 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland 19.1 6.7 64.0 58.1 4.6.8 5.8 12.0 5.2 16.0 10.6 107 tlsrael 54.5 99.2 15.0 14.7 56.6 89.0 13.9 176.9 26.6 108 New Zealand 15.1 . . 51.4 50.9 . . 10.3 9.0 12.6 12.1 109 Spain 20.1 10.8 68.8 74.4 68.8 8.9 13.1 10.5 16.9 14.4 110 tHong Kong 69.5 . . 7.5 Ill tSingapore 17.1 13.5 66.2 74.4 126.1 1.9 9.4 4.6 11.7 7.6 112 United Kingdom 15.2 . . 49.2 460 . . 5.8 14.1 5.3 16.2 11.5 113 Australia 20.4 12.2 43.6 57.9 74.5 7.0 8.6 10.4 10.6 16.4 114 Italy 20.1 12.0 79.3 81.9 78.9 9.5 12.7 6.6 19.0 13.9 115 Netherlands 14.6 . . 53.9 77.2 1.8 6.0 3.2 13.5 12.4 116 Belgium 10.8 7.0 56.7 57.0 . . 4.2 7.7 6.3 12.9 Ill Austria 13.7 7.4 54.0 72.6 87.3 3.6 5.0 3.8 . 118 France 15.6 9.9 57.8 69.7 . . 5.7 6.3 6.7 18.7 16.0 119 Canada 17.5 8.4 48.4 65.0 75.7 4.3 12.9 8.6 14.3 9.9 120 United States 10.0 8.0 60.4 58.3 67.0 4.2 15.3 8.5 121 Germanya 9.4 6.4 52.8 60.7 68.2 2.8 8.0 7.6 12.0 12.5 122 Denmark 12.4 11.1 44.8 42.6 59.6 5.2 10.8 7.2 17.2 11.4 123 Finland 15.4 12.9 39.8 39.5 56.2 6.6 7.5 9.8 11.8 124 Norway 12.8 10.6 54.6 51.6 63.9 5.2 5.0 9.6 12.6 14.2 125 Sweden 11.4 9.2 48.1 46.5 45.5 7.4 11.3 8.0 15.1 16.1 126 Japan 16.0 8.9 94.7 134.0 183.1 1.5 5.5 3.3 8.4 7.5 127 Switzerland 5.4 6.8 109.8 107.4 112.0 3.8 7.6 7.8 World Fuel exporters a. Data refer to the Federal Republic of Germany before unification. 263 Table 14. Growth of merchandise trade Merchandise trade Average annual growth ratea (percent) (millions ofdollars) Terms of trade .spor1s Imports Exports Imports (1987 = 1991 1991 1970-80 1980-91 1970-80 1980-91 1985 1991 Low-income economies 161,496 1 167,270 1 3.5 w 6.6 w 6.0 w 2.7 w 106m 94m China and India 90,539 1 84,209 1 6.9 w 10.4 w 7.3 w 7.8 w 103m 106m Other low-income 70,957 1 83,062 I 2.0 w 3.3 w 5.4 w -1.3 w 106m 93m 1 Mozambique . . . . . . . . . 2 Tanzania 394 1,381 -7.5 -1.9 -0.6 2.8 101 84 3 Ethiopia 276 1,031 -2.3 1.9 -0.6 3.3 117 60 4 Uganda 200 550 -8.4 2.3 -1.5 3.6 143 48 5 Bhutan . . . 6 Guinea-Bissau 28 78 15.9 -2.5 -5.2 3.6 91 138 7 Nepal 238 740 10.9 8.1 8.8 4.9 98 85 8 Bunindi 91 254 0.2 8.6 5.0 -0.1 133 43 9 Chad 194 408 .. .. .. 10 Madagascar 344 523 -3.0 0.3 -0.8 0.5 98 85 II Sierra Leone 145 163 -5.7 -3.0 -2.0 -7.2 106 116 12 Bangladesh 1,718 3,470 3.8 7.2 -2.4 4.3 122 105 13 Lao PDR 97 228 . . . . . . . . . 14 Malawi 470 719 5.4 5.6 1.0 2.2 104 87 15 Rwand,a .. 16 Mali 354 638 8.3 6.7 5.2 3.5 95 99 17 Burkina Faso 116 602 7.3 6.5 6.4 3.3 108 89 18 Niger 385 431 21.0 1.8 10.9 -3.0 126 82 19 India 17,664 20,418 4.3 7.4 3.0 4.2 96 100 20 Kenya 1,203 2,034 2.9 2.9 1.9 1.0 114 87 21 Nigeria 12,071 6,525 0.4 1.2 19.4 -14.3 167 82 22 China* 72,875 63,791 8.7 11.5 11.3 9.5 109 III 23 Haiti 103 374 5.6 -4.5 6.3 -2.1 89 77 24 Benin 103 398 -11.6 11.3 4.0 -0.2 103 85 25 Central African Rep. 133 196 -0.6 1.3 -2.9 6.1 107 III 26 Ghana 992 1,418 -6.3 5.2 -2.2 1.8 106 62 27 Pakistan 6,528 8,439 0.7 9.9 4.2 2.6 90 80 28 Togo 292 548 4.9 6.5 11.2 2.5 118 80 29 Guinea . . . . . . . . . . . . . 30 Nicaragua 268 751 0.8 -4.2 0.1 -1.2 108 107 31 SriLanka 2,629 3,861 2.0 6.3 4.5 2.1 103 87 32 Mauritania 438 470 -2.0 5.6 1.4 3.1 113 109 33 Yemen, Rep. . . . . . . . . . . . . . 34 Honduras 679 880 3.8 -0.7 2.1 -1.2 III 113 35 Lesotho5 36 Indonesia 28,997 25,869 7.2 4.5 13.0 2.6 134 101 37 Egypt, Arab Rep. 3,887 7,862 -2.6 2.8 7.8 -2.3 131 93 38 Zimbabwe 1,779 2,110 2.3 0.4 -3.5 -1.0 100 101 39 Sudan 329 1,433 -3.5 -1.2 -0.6 -4.0 106 94 40 Zambia 1,082 1,255 -0.2 -3.2 -9.2 -1.8 90 116 Middle-income economies 524,9481 552,2571 4.1 w 3.4 w 6.1 w 1.1 w 109 m 103 m Lower-middle-income 214,9771 243,207 I 6.7 w 5.3 w 5.9 w 1.8 w 108 m 103 m 41 Bolivia 760 992 -0.8 4.5 7.3 0.2 167 73 42 Côted'Ivoire 3,011 1,671 4.8 4.5 9.1 -2.3 110 67 43 Senegal 977 1,407 1.8 5.6 3.7 3.4 106 93 44 Philippines 8,754 12,145 6.0 3.3 3.3 3.0 93 91 45 PapuaNewGuinea 1,361 1,614 10.6 6.8 1.8 1.6 Ill 80 46 Cameroon 2,022 1,448 4.2 11.5 5.4 -1.5 139 81 47 Guatemala 1,202 1,850 5.7 -0.7 5.8 -0.8 108 103 48 Dominican Rep. 658 1,729 -2.7 -1.5 2.0 1.4 109 112 49 Ecuador 2,957 2.328 12.5 4.8 6.7 -2.1 153 90 50 Morocco 4,278 6,872 3.9 5.9 6.6 3.8 88 98 SI Jordan 879 2,507 19.3 6.9 15.3 -0.8 95 116 52 Tajikistan .. .. .. .. .. 53 Pens 3,307 2,813 3.3 1.1 -1.7 -4.7 III 67 54 El Salvador 367 885 1.3 -2.7 4.6 -3.7 126 103 55 Congo 1,455 524 16.8 6.6 5.3 -1.9 145 84 56 SynanArabRep. 5,594 3,002 7.0 20.6 12.4 3.9 125 182 57 Colombia 7,269 4,967 1.9 12.0 6.0 -1.7 140 84 58 Paraguay 737 1.460 8.3 12.2 5.3 5.8 108 117 59 Uzbekistan .. .. .. .. .. 60 Jamaica 1,081 1,843 -1.7 0.8 -6.8 2.0 95 91 61 Romani% 62 Namibia . . . . . . . . . . . . . 63 Tunisia 3,709 5,180 7.5 5.6 12.5 1.5 105 95 64 Kyrgyzstan .. .. .. .. .. 65 Thailand 28,324 37,408 10.3 14.4 5.0 11.1 91 91 66 Georgia Data for Taiwan, China, are: 76,090 61,723 15.6 11.0 12.2 10.1 100 106 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.. 264 Merchandise trade Average annual growth ratea (percent) (millions of dollars) Termsof trade Esports Imports (1987= 100) Erporis Imports 1991 /991 1970-80 /980-9! 1970-80 1980-9! 1985 /991 67 Azerbaijan 68 Turkmenistan 69 Turkey 13,594 21,038 4.3 7.2 5.7 7.4 82 108 70 Poland 14,903 15,757 5.4 3.3 5.8 2.0 94 104 71 Bulgaria 72 Costa Rica 1,490 1,864 5.2 4.6 4.2 3.4 III 109 73 Algeria 11,790 7,683 -0.5 2.4 12.1 -5.6 174 95 74 Panama 333 1,681 -7.3 0.0 -5.1 -3.4 130 112 75 Armenia 76 Chile 8,552 7,453 10.4 5.2 2.2 1.9 102 122 77 Iran, Islamic Rep. 15,916 21,688 -6.8 14.7 10.3 7.9 160 88 78 Moldova 79 Ukraine 80 Mauntius 1,193 1,575 3.8 9.9 8.2 10.8 83 104 81 Czechoslovakia 16,317 7,947 6.4 0.1 5.7 -6.0 98 137 82 Kazakhstan 83 Malaysia 34,300 35,183 4.8 10.9 3.7 7.2 Ill 93 Upper-middle-income 309,972 1 309,0501 2.3 w 2.3 w 6.2w 0.6w 117m 105m 84 Botswanat' 85 South Africab 24,164 17,503 13.4 0.9 -2.0 -4.4 105 86 86 Lithuania 87 Hungary 10,180 11,370 3.8 2.2 2.0 1.0 104 102 88 Venezuela 15,127 10,181 -11.6 0.1 10.9 -6.6 174 101 89 Argentina 11,975 8,100 7.1 2.1 2.3 -5.5 110 113 90 Umguay 1,574 1,552 6.5 3.1 3.1 0.2 89 105 91 Brazil 31,610 22,959 8.5 4.3 4.0 0.8 92 119 92 Mexico 27,120 38,184 13.5 3.5 5.5 2.2 133 100 93 Belanis 94 Russian Federation 95 Latvia 96 Trinidad and Tobago 1,985 1667 -7.3 -2.6 -9.6 -10.8 156 97 97 Gabon 3,183 806 5.7 5.1 11.6 -3.0 140 79 98 Estonia 99 Portugal 16,326 26,329 1.2 II.! 1.0 10.0 85 112 100 Oman 101 Puerto Rico 102 Korea. Rep. 71,672 81,251 23.5 12.2 11.6 11.1 103 1(18 103 Greece 8,647 21,552 10.9 3.9 3.2 5.4 94 107 104 Saudi Arabia 54,736 25,540 5.7 -4.2 35.9 -9.0 176 79 105 Yugoslavia 13,953 14,737 5.3 -1.2 3.4 -1.2 95 107 Low- and middle-income 686,445 I 719,528 1 3.9 w 4.1 w 6.0 w 1.5w 107 m 100 m Sub-Saharan Africa 38,0851 35,207 1 0.2 w 2.7 w 5.2 w -4.0 w 107 m 87 m East Asia & Pacific 251,4481 265,796 1 9.5 w 10.2 w 7.6 w 8.4 w 96m 108 m South Asia 29,012 1 37,928 1 3.6 w 7.4 w 2.5 w 3.6 w 97 m 86m Europe and Central Asia 95,153 1 120,861 1 94 m 108 m Middle East & N. Africa 126,1361 119,0251 w - w V -3.2 w 129 m 95 m Latin America & Caribbean 122,446 1 123,207 1 -0.1 w 2.9 w 3.6 w -1.1 w Ill m 105 m Severely indebted 138,113 1 140,829 1 9.7 w 2.8 w 5.9 w -0.6 w hIm 98m High-income economies 2,650,106 I 2,788,686 1 5.3 w 4.1 w 2.3 w 5.1 w 97 m 101 m OECD members 2,441,157 1 2,520,853 1 5.6 w 4.1 w 2.0 w 5.1 w 94 m lOt m 106 Ireland 24,240 20,754 11.7 7.1 4.7 3.7 97 92 107 tlsrael 11,891 16,753 10.0 6.7 3.5 5.1 105 104 108 New Zealand 9,269 8,494 3.4 3.6 -0.3 3.2 88 94 109 Spain 60,134 93,062 9.1 7.5 1.9 9.4 91 108 110 tHong Kong 29,738 100,255 9.7 4.4 7.8 11.3 97 101 Ill tSingapore 58,871 65,982 4.2 8.9 5.0 7.2 99 101 112 United Kingdom 185,095 209,982 4.4 2.6 0.3 4.4 103 104 113 Australia 37,724 39,460 3.8 4.6 1.8 5.0 Ill 107 114 Italy 169,365 178,240 6.0 3.4 0.7 4.2 84 101 115 Netherlands 133,527 125,838 3.3 4.4 1.1 3.5 101 100 116 Belgiumc 118,222 121,038 5.6 4.6 2.9 3.3 94 95 117 Austria 41,082 50,697 6.2 6.2 4.0 5.3 87 89 118 France 212,868 230,257 6.6 3.5 2.4 3.3 96 102 119 Canada 124,797 117,633 2.0 5.7 0.4 7.8110 105 120 United States 397,705 506,242 6.5 4.0 4.3 7.0100 102 121 Germanyd 401,848 387,882 5.0 4.1 2.8 4.5 82 95 122 Denmark 35,687 32,158 4.3 5.0 -0.4 4.0 93 104 123 Finland 23,081 21,708 5.3 2.6 0.1 4.0 85 99 124 Norway 34,037 25,523 7.9 7.4 0.7 2.2 130 90 125 Sweden 55,042 49,760 2.5 3.8 -0.2 3.1 94 103 126 Japan 314,395 234,103 9.0 3.9 0.4 5.6 71 99 127 Switzerland 61,468 66,285 4.9 3.3 2.6 3.4 86 96 World 3,336,550 1 3,508,214 I 5.0 w 4.1 w 3.1w 4.3w 106m lOOm Fuel exporters 171,293 123,270 1 1.4 w -0.1 w 14.9 w -5.5 w 167m 85m a. See the technicai notes. b. Data are for the South African Customs Union comprising South Africa, Namibia, Lesotho, Botawana, and Swaziland; trade among the component territories is excluded. c. Includes Luxembourg. d. Data refer to the Federal Republic of Germany before unification. 265 Table 15. Structure of merchandise imports Percentage share of merchandise imports Other Machinery prtmar and transport Other Food Fuels commodities equipment manufactures 1970 1991 /970 199/ 1970 199/ 1970 /991 1970 /991 Low-income economies 15w lOw 6w 9w 8w 8w 31 w 34w 40w 38w China and India 14w 6w 4w 8w 14 w 10 w 32 w 35 w 36w 41w Other low-income 16w 14w 7w 11w Sw 7w 30w 32 w 42w 36w I Mozambique 2 Tanzania 7 II 9 19 2 4 40 33 42 33 3 Ethiopia 9 14 8 10 4 3 35 45 45 28 4 Uganda 6 8 2 30 3 2 34 27 55 34 5 Bhutan 6 Guinea-Bissau 28 32 7 7 4 3 16 IS 45 43 7 Nepal 5 9 It 12 0 14 25 24 60 41 8 Burundi 18 17 7 7 8 7 23 28 45 40 9 Chad 19 17 IS IS 4 3 23 27 38 38 10 Madagascar 12 13 7 24 3 3 30 32 48 28 II SierraLeone 23 24 7 28 4 3 22 19 44 26 12 Bangladesh 23 26 /3 13 /1 6 22 17 32 38 13 La0PDR -. .- - 0 - 14 Malawi 15 7 5 16 6 4 30 26 44 47 15 Rwanda - - - 16 Mali 29 18 9 28 6 2 21 25 36 28 17 BurkinaFaso 19 23 8 16 8 5 27 24 37 31 18 Niger 13 IS 4 20 6 6 26 28 51 31 19 India 21 5 8 23 19 12 23 18 29 42 20 Kenya 6 6 10 IS 4 4 34 38 4-6 37 21 Nigeria 8 18 3 I 3 5 37 36 48 41 22 China 7 6 I 3 10 9 39 41 43 41 23 Haiti 19 24 6 12 4 5 21 20 51 39 24 Benin 12 16 4 7 8 II 21 21 55 45 25 CentralAfricanRep. 17 17 I 7 2 5 36 33 44 38 26 Ghana 20 9 6 31 4 3 26 26 44 31 27 Pakistan 21 17 6 II 7 8 31 28 35 29 28 Togo 16 20 4 7 II 6 22 24 47 43 29 Guinea - - - - - - - - - - - - - - - - - - - 30 Nicaragua 10 16 6 13 3 2 28 33 54 36 31 SnLanka 47 17 3 II 4 4 18 19 29 49 32 Mauritania 23 23 8 7 2 I 38 40 29 29 33 Yemen, Rep. .. - - -. - - - - - - - - - - - - 34 Honduras II 13 7 16 2 3 29 25 SI 44 35 - - - - - - - - - - - - - - - - - 36 Indonesia II 5 2 9 4 9 35 45 47 32 37 Egyp,ArabRep. 21 29 9 3 14 10 27 24 29 34 38 Zimbabwe 5 5 24 24 6 6 29 29 37 37 39 Sudan 20 22 8 16 4 3 27 25 41 34 40 Zambia II 8 10 18 2 2 39 35 38 37 Middle-income economies 12 w lOw 9w 11 w 10 w 8w 35 w 36w 34w 36w Lower-middle-income 14 w 10 w 9w 10 w 9w 7w 33 w 38 w 34 w 35 w 41 Bolivia 20 14 1 I 3 3 37 43 40 39 42 Côted'Ivoire IS IS 5 21 3 3 33 23 44 34 43 Senegal 28 26 5 20 5 5 25 21 38 29 44 Philippines II 7 12 15 9 7 35 26 33 46 45 Papua New Guinea 23 17 10 9 2 2 30 38 36 34 46 Cameroon 12 14 5 I 2 3 32 35 49 47 47 Guatemala II 12 2 17 3 3 27 26 57 42 48 Dominican Rep. 17 17 14 26 5 4 27 21 38 32 49 Ecuador 7 8 6 I 3 5 35 41 49 46 50 Morocco 20 II 5 IS II 14 32 28 32 33 51 Jordan 30 26 6 14 5 4 17 18 42 37 52 Tajikistan - - - - - - - - - - - - - - - - - - 53 Peru 20 20 2 II 5 3 35 35 38 31 54 ElSalvador 13 16 2 13 5 6 23 24 56 41 55 Congo 19 IS 2 3 2 2 33 35 44 41 56 SyrianArabRep. IS 17 10 18 7 7 33 26 36 32 57 Colombia 7 7 I 6 9 8 46 33 37 47 58 Paraguay 13 8 IS 12 7 5 32 36 33 39 59 Uzbekistan - - - - - - - - - - - - - - - - - - - 60 Jamaica 22 20 IS 21 5 4 21 18 37 38 61 Romania - - - 62 Namibiaa 63 Tunisia 27 IS Ii 26 41 64 Kyrgyzstan - - - 65 Thailand 4 5 8 36 39 43 38 66 Georgia - - - - * Data for Taiwan. China, are: IS 6 5 10 18 12 35 36 28 36 Note: For data comparability and coverage, see the technical notes. Figures in italics ale for years other than those specified. 266 Percentage share of merchandise imports Other Machinery primary and transport Other Food Fuels commodities equipment manufactures 1970 1991 1970 199! 1970 1991 1970 1991 1970 199! 67 Azerbaijan 68 Turkmenistan 0 S 69 Turkey 8 7 8 21 8 12 41 29 36 31 70 Poland 14 7 18 22 II 8 27 34 24 28 71 Bulgaria . 72 Costa Rica II 9 4 17 3 4 29 20 53 50 73 Algeria 13 26 2 3 6 6 37 32 42 34 74 Panama 10 10 19 15 2 2 27 26 42 47 75 Armenia . 0 . . . . . 0 76 Chile 14 6 6 15 7 4 43 38 30 38 77 Iran, Islamic Rep. 7 13 0 0 8 5 41 44 45 38 78 Moldova . . 79 Ukraine . . . 0 S . . . . . . 80 Mauntius 36 27 7 18 3 5 13 12 41 37 81 Czechoslovakia 12 8 II 9 19 14 36 39 23 29 82 Kazakhstan . .. .. .. .. .. .. 83 Malaysia 20 6 12 4 9 5 28 55 31 30 Upper-middle-income II w 10w 9w 11 w 11 w 8w 37 w 34w 33w 36 w 84 Botswanaa .. .. .. .. .. 85 South Africaa 4 3 0 0 6 4 53 43 36 50 86 Lithuania . . . . 0 S . . . . . . . 87 Hungary 10 5 9 IS 19 8 31 30 31 42 88 Venezuela 10 12 I 2 5 9 45 44 38 33 89 Argentina 6 4 5 9 16 II 31 33 42 44 90 Uniguay II 7 15 16 14 6 31 30 29 41 91 Brazil II 10 12 26 8 8 35 27 34 29 92 Mexico 7 14 3 3 9 8 50 41 31 33 93 Belams . . 94 Russian Federation . . 95 Latvia . S . . . . . . . . 96 Trinidad andlobago II IS 53 15 2 6 13 26 22 38 97 Gabon 14 17 I I 1 2 39 40 44 39 98 Estonia . . S 0 . . 99 Portugal 13 13 9 9 14 5 30 36 34 37 100 Oman .. 101 Puerto Rico S . . . . . . . . . . . 102 Korea, Rep. 17 6 7 16 21 13 30 34 25 31 103 Greece II 13 7 10 10 6 48 33 25 38 104 Saudi Arabia 26 IS I 0 5 4 33 35 35 47 105 Yugoslavia 8 10 5 18 18 10 33 26 37 36 Low- and middle-income 13 w 10 w 8w lOw lOw 8w 34w 35w 35w 36w Sub-Saharan Africa 14 w 16 w 7w 12w 4w 4w 32w 31w 44w 37w East Asia & Pacific 13 w 6w 7w 9w lOw lOw 33w 39w 37w 36s South Asia 24 w 11 w 7w 19w 13w lOw 24w 20w 31w 40w Europe and Central Asia . Middle East & N. Africa 18 w 17 w 3w 4w 8w 6w 32w 35w 39w 38w Latin America & Caribbean 11 w 12 w 11w 13w 7w 7w 35w 34w 36w 35w Severely indebted 13 w 13 w 9w 11w 9w 7w 34w 35w 34w 34w High-income economies 15w 9w lOw lOw 16w 7w 25w 34w 33w 40w OECD members 15w 9w lOw lOw 17w 7w 25w 34w 33w 39w 106 Ireland 13 II 8 6 9 4 27 35 43 44 107 jIsrael 14 7 5 8 9 5 30 32 42 49 108 New Zealand 6 8 7 8 II 4 34 37 43 44 109 Spain 14 II 13 Il 18 7 26 37 28 34 110 tHongKong 19 6 3 2 10 5 16 27 52 60 Ill tSingapore IS 6 13 14 13 4 23 44 35 32 112 United Kingdom 23 10 10 6 21 6 17 36 29 41 113 Australia 5 5 5 6 7 3 41 42 42 45 114 Italy 19 13 14 10 21 10 20 32 26 36 115 Netherlands 14 13 II 9 II 5 25 31 39 42 116 Belgiumb 13 10 9 8 19 7 26 26 33 48 117 Austria 9 5 8 6 13 7 31 39 39 44 118 France 14 10 12 10 IS 6 25 34 33 40 119 Canada 9 6 6 5 6 4 49 51 31 34 120 United States 16 6 8 II 13 5 28 41 36 37 121 Germanyc 18 10 9 8 18 7 19 35 36 41 122 Denmark 10 12 10 5 9 5 28 31 42 46 123 Finland 9 6 II 13 9 7 33 34 37 39 124 Norway 8 6 8 4 13 7 35 38 36 44 125 Sweden 10 7 II 9 II 5 30 37 39 42 126 Japan 17 IS 21 23 37 15 II 16 14 30 127 Switzerland 12 6 5 5 10 5 27 32 46 52 World 15w 9w lOw lOw 15w 7w 27w 35w 34w 39w Fuel exporters 13 w IS w 5w 2w 5w 5w 36 w 40 w 41 w 39 w a. Data are for the South African Customs Union comprising South Africa, Namibia, Lesotho, Botswana, and Swaziland; trade among the component temtories is excluded. b. Includes Luxembourg. c. Data refer to the Federal Republic of Germnay before unification. 267 Table 16. Structure of merchandise exports Percentage share of merchandise exports Fuels, Other Machinery minerals, primary and transport Other Textiles and and metals co,nmodities equipment manufactures clothing a 1970 199! 1970 1991 1970 1991 1970 1991 1970 199/ Low- income economies 28w 23 w 44w 20w 3w lOw 24w 46w 13 w 23 w China and India 12 w 9w 27 w 16 w 10 w 17 w 51 w 58 w 27 w 28 w Other low-income 36 w 42 w 52 w 26 w 0w 2w 12 w 31 w 7w 17 w I Mozambique S 0 S 2 Tanzania 7 5 80 84 0 1 13 10 2 3 3 Ethiopia 2 3 97 94 0 0 2 3 0 1 4 Uganda 9 4 90 95 0 0 0 0 0 0 5 Bhutan S S S S S 6 Guinea-Bissau 0 0 98 97 1 . 1 0 0 7 Nepal 0 0 65 II 0 35 88 25 76 8 Bunindi 1 1 97 97 0 0 2 2 0 0 9 Chad 0 3 95 93 1 1 4 3 0 I 10 Madagascar 9 8 84 85 2 2 5 6 1 3 II Sierra Leone IS 34 22 33 . . . 63 32 0 12 Bangladesh 1 1 35 29 1 0 64 70 49 62 13 LaoPDR .. .. .. .. .. .. 14 Malawi 0 0 96 96 0 0 3 4 I 3 15 Rwanda . . . 16 Mali 1 0 89 93 0 0 10 7 8 6 17 BurkinaFaso 0 0 95 88 I 4 3 8 0 2 18 Niger 0 86 96 12 I 0 2 I 0 I 19 India 13 8 35 19 5 7 47 66 25 25 20 Kenya 12 16 75 64 0 5 12 15 I 2 21 Nigena 62 96 36 3 . 0 1 1 0 0 22 Chinat II 9 19 15 15 19 55 57 29 28 23 Haiti 17 12 57 46 . . 5 26 37 4 8 24 Benin 0 3 89 67 3 3 8 28 6 I 25 Central African Rep. 0 I 55 55 1 0 44 43 I 0 26 Ghana 13 15 86 84 0 0 I 1 0 0 27 Pakistan 2 I 41 26 0 0 57 72 47 60 28 logo 25 49 69 42 2 1 4 9 I 2 29 Guinea . . . . . . . . . . S S S S 30 Nicaragua 3 2 81 86 0 0 16 12 3 I 31 Sn Lanka I 1 98 34 0 2 1 62 0 43 32 Mauritania 88 86 II 9 0 4 0 I 0 0 33 Yemen, Rep. . S . S S 34 Hondura,s 9 5 82 89 0 0 8 6 2 I 35 Lesotho S . S 36 Indonesia 44 43 54 16 0 2 1 39 0 14 37 Egypt, Arab Rep. 5 40 68 20 I 1 26 40 19 27 38 Zimbabwe 18 17 47 51 2 4 33 28 4 6 39 Sudan 1 3 99 96 0 0 0 1 0 0 40 Zambia 99 98 I I 0 0 0 I 0 0 Middle-income economies 40 w 34 w Lower-middle-income 31 w 29 w 41w 27w 14w 17w 16w 29w 3w 9w 41 Bolivia 93 74 4 21 . . 0 3 4 0 I 42 Côte d'Ivoire 2 II 92 79 1 2 5 9 1 2 43 Senegal 12 22 69 56 4 3 15 20 6 2 44 Philippines 23 9 70 20 0 14 8 57 I 9 45 Papua New Guinea 42 62 55 35 0 2 3 I 0 0 46 Cameroon . . . . . . . . . . . . . . . . S 47 Guatemala 0 2 72 70 2 I 26 26 8 5 48 Dominican Rep. 4 I 77 79 0 3 20 17 0 0 49 Ecuador I 43 97 55 0 0 2 2 1 0 50 Morocco 33 20 57 29 0 3 9 48 4 20 51 Jordan 24 38 59 16 3 1 13 44 3 4 52 Tajikistan . . . . . . . . . . . . . . . . . 53 Pens 49 52 49 30 0 2 1 17 0 9 54 ElSalvador 2 3 70 56 3 3 26 37 II 15 55 Congo 1 92 70 5 I 0 28 3 0 0 56 SynanArabRep. 62 62 29 15 3 1 7 23 4 15 57 Colombia II 29 81. 38 I 3 7 31 2 II 58 Paraguay 0 0 91 89 . . 0 9 11 0 I 59 Uzbekistan .. .. .. .. .. .. .. 60 Jamaica 25 17 22 27 0 I 53 56 2 14 61 Romani% 62 Namibia 63 Tunisia 21 II 7 1 i 64 Kyrgyzstan 65 Thailand 15 77 32 0 22 1 66 Georgia * Data for Taiwan, China, are: 2 2 22 6 38 17 59 55 29 16 Note: Fordata comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 268 Percentage share of merchandise exports Fuels, Other Machinery minerals, primary and transport Other Textiles and and metals commodities equipment manufactures clothing a /970 1991 1970 199/ 1970 1991 1970 1991 /970 1991 67 Azerbaijan 68 Turkmenistan 0 S 0 . 69 Turkey 8 7 83 26 0 6 9 61 5 36 70 Poland 20 20 9 16 36 26 25 38 6 5 71 Bulgaria . . . 0 72 Costa Rica 0 2 80 72 3 3 17 23 4 5 73 Algeria 73 97 20 0 2 I 5 2 I 0 74 Panama 21 2 75 77 2 I 2 21 0 7 75 Armenia . . . . . . . S S S . . 76 Chile 88 50 7 35 I I 4 14 0 I 77 Iran, Islamic Rep. 90 90 6 7 0 0 4 4 3 3 78 Moldova . . 0 . . . 79 Ukraine . . . 0 . . . . . . . . 80 Mauritius 0 0 98 70 0 0 2 30 I 24 81 Czechoslovakia 7 4 6 6 50 54 37 36 7 6 82 Kazakhstan . . . . . . . . . . . . . . . 83 Malaysia 30 17 63 22 2 38 6 23 I 6 Upper-middle-income 84 Botswana' . . . . . . . . 85 South Africab . . . 86 Lithuania . . . . . . . . . 0 0 87 Hungary 7 8 26 28 32 22 35 42 8 9 88 Venezuela 97 86 2 2 0 I I II 0 I 89 Argentina 8 85 64 4 7 10 21 I 2 90 Uruguay I 79 59 I 2 20 38 14 16 91 Brazil II 16 75 28 4 18 II 38 I 4 92 Mexico 19 41 49 14 II 24 22 20 3 2 93 Belarus 94 Russian Federation . . . 0 0 . . . 95 Latvia . . . . . . . . . . . . . 96 Trinidad and Tobago 78 65 9 6 I I 12 28 97 Gabon 56 89 35 7 I 0 8 4 98 Estonia . . . . . . . 99 Portugal 5 5 31 13 8 19 56 63 25 30 100 Oman 101 Puerto Rico 102 Korea, Rep. . Si I 21 103 Greece 14 15 51 33 I 4 33 48 7 26 104 Saudi Arabia 100 99 0 0 0 I 0 0 0 0 105 Yugoslavia 15 9 26 12 23 29 37 50 10 7 Low- and middle-income 37 w 32 w 39w 20w lOw 16w 19w 35w 7w 13w Sub-Saharan Africa 41 w 58 w 51w 34w Ow 1w 7w 7w 1w 2w East Asia & Pacific 22 w 12w 46w 16w 6w 25w 27w 47w 13w 19K South Asia 9w 5w 44w 23w 3w 5w 45w 67w 28w 37w Europe and Central Asia Middle East & N. Africa w 81 w 20w 9w 4w 1w 7w lOw 3w 4w Latin America & Caribbean 40w 34w 51w 32w 2w 12w 9w 24w 1w 3w Severely indebted 21 w 34 w 56w 30w 11w 13w 15w 25w 3w 4w High-income economies 11w 8w 16w 11w 35w 42w 38w 39w 6w Sw OECD members 9w 7w 16w 11w 36w 43w 38w 39w 6w 4w 106 Ireland 8 2 52 24 7 30 34 44 10 4 107 tIsrael 4 2 26 10 5 24 66 64 12 7 108 New Zealand I 8 88 65 2 5 9 22 I 2 109 Spain 10 7 37 17 20 41 34 35 6 4 IlO tHong Kong I 2 3 3 12 24 84 72 44 40 Ill tSingapore 25 18 45 8 II 48 20 26 5 5 112 UnitedKingdom 8 10 9 8 41 41 42 41 6 4 113 Australia 28 35 53 28 6 7 13 29 1 I 114 Italy 7 3 10 8 37 38 46 52 13 12 115 Netherlands 14 12 29 25 20 22 37 41 8 4 116 Belgiumc 13 8 II 12 21 27 55 54 10 7 117 Austria 6 4 14 7 24 38 56 51 II 8 118 France 6 5 19 17 33 39 42 39 8 5 119 Canada 26 19 22 17 32 38 19 26 I I 120 United States 9 6 21 14 42 48 28 32 2 2 121 Germanyd 6 4 5 6 47 49 43 41 6 5 122 Denmark 4 5 42 31 27 25 27 39 6 5 123 Finland 4 7 29 II 16 28 50 55 6 2 124 Norway 25 58 20 9 23 15 32 18 2 I 125 Sweden 8 6 18 9 40 43 35 42 3 2 126 Japan 2 I 5 1 41 66 53 31 Il 2 127 Switzerland 3 3 8 4 32 32 58 62 8 5 World 16w 13w 21w 13w 30w 37w 34w 38w 6w 6w Fuelexporters 83w 89w 13w 6w 1w 1w 4w 4w 1w 1w a. See the technical notes. b. Data are for the South African Customs Union comprising South Afnca, Namibia, Lesotho. Botswana. and Swaziland; trade among the component territories is excluded. c. Includes Luxembourg. d. Data refer to the Federal Republic of Germany before unification. 269 Table 17. OECD imports of manufactured goods: origin and composition Value of imports of Composition of /991 imports of manufactures (percent) manufactures, by origin Elecincal (millions ofdollars) Textiles and machinery and Transport /970 /991 clothing Chemicals electronics equipment Others Low-income economies 1,266 t 73,602 t 40 w 4w 7w 2w 46 w China and India 777 1 55,576 t 37 w 5w 9w Iw 48 w Other low-income 4891 18,0261 51 w 3w 2w 4w 41 w I Mozansbique 7 10 60 0 0 0 40 2 Tanzania 9 48 56 4 0 2 38 3Ethiopia 4 59 12 9 2 5 73 4Uganda I 2 0 0 50 0 50 SBhutan 0 I 0 0 0 0 tOO 6 Guinea-Bissau 0 0 0 0 0 0 0 7 Nepal I 240 93 0 0 0 6 SBstnindi 0 2 50 0 0 0 50 9Chad 0 2 0 0 0 0 100 10 Madagascar 7 49 69 8 0 2 20 II Sierra Leone 2 135 I 0 1 0 99 12 Bangladesh 0 1.372 93 0 0 0 7 I3LaoPDR 0 26 92 0 0 0 8 14 Malawi I 18 67 0 0 0 33 l5Rwanda 0 2 0 0 0 50 50 16 Mali 2 51 0 0 20 0 80 17 BurkinaFaso 0 3 0 0 0 0 100 18 Niger 0 175 0 96 0 2 2 19 India 534 9,428 45 6 I I 47 20 Kenya 16 108 IS 5 3 6 72 21 Nigeria 13 238 6 16 2 2 74 22 China 243 46,148 35 5 II I 48 23 Haiti 17 301 60 I 10 0 29 24 Benin 0 4 50 0 0 25 25 25 Central African Rep. 12 80 0 0 0 0 100 26Ghana 8 98 0 1 I 0 98 27 Pakistan 207 3,234 85 0 0 0 IS 28Togo 0 9 II 0 0 0 89 29 Guinea 38 147 0 34 I 0 65 30 Nicaragua 6 7 14 14 14 0 57 31 Sn Lanka 9 1,346 73 I 2 0 24 32 Mauritania 0 4 25 0 25 0 50 33 Yemen, Rep. 0 0 0 0 0 0 0 34 Honduras 3 261 82 I 0 0 17 35 . . . . . . . . 36 Indonesia 15 7,302 36 2 3 I 58 37 Eypt,ArabRep. 33 793 61 5 1 15 18 38 Zimbabwe 0 261 22 0 2 1 74 39 Sudan I 7 0 0 0 14 86 40 Zambia 4 40 23 0 0 5 73 Middle-income economies 5,016 1 185,948 1 25w 7w 18w 7w 44w Lower-middle-income 1,778 I 66,559 1 33w 6w 18w 3w 41w 41 Bolivia I 54 20 6 0 0 74 42 Côted'Ivoire 7 239 21 2 I 2 75 43Senegal 4 34 3 12 15 3 68 44 Philippines 108 5,637 33 2 31 I 33 45 Papua New Guinea 4 35 6 0 0 3 91 46Caneroon 4 47 21 2 2 2 72 47 Guatemala 5 426 89 2 0 0 9 48 Dominican Rep. 10 1,807 55 I 7 0 38 49 Ecuador 3 83 17 4 5 6 69 50 Morocco 32 2,364 69 12 8 I 10 SI Jordan I 67 5 27 6 27 36 52 Tajikistan . . . . . . . . . 53 Peru 12 453 58 7 2 I 33 54 ElSalvador 2 184 65 I 22 0 13 55 Congo 4 205 0 0 0 0 100 56 Syrian Arab Rep. 2 53 68 0 2 8 23 57 Colombia 52 1,070 33 6 0 0 61 58 Paraguay 5 82 9 27 0 0 65 59 Uzbekistan .. .. .. .. 60 Jamaica 117 799 38 58 0 0 4 61 Rornania 188 1,648 32 6 4 3 56 62 Namibiaa . 63 Thnisia 19 2,135 70 7 9 2 12 64 Kyrgyzstan .. .. .. .. 65 Thailand 32 12.851 21 2 17 2 59 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 270 Value of imports of Composition sf1991 imports of manufactures (percent) manufactures by origin Electrical (millions of dollars) Textiles and machinery and Transport 1970 clothing Chemicals electronics equipment Others 67 Azerbaijan 68 Turkmenistan . . . . . 0 . . . 69 Thrkey 47 6,770 71 3 6 2 18 70 Poland 287 5,515 23 16 7 4 51 71 Bulgaria 68 553 29 18 6 1 47 72 Costa Rica 5 700 69 2 9 0 19 73 Algeria 39 1,686 0 4 0 0 96 74 Panama' 18 462 15 4 1 46 36 75 Armenia . . . . . . . . . . 76 Chile 15 697 9 26 0 2 63 77 Iran, Islamic Rep. 133 676 92 1 0 0 7 78 Moldova . 79 Ukraine . . . . . . . . . . . 80 Mauritius 1 750 86 1 0 0 13 81 Czechoslovakia 467 4,930 16 14 5 9 56 82 Kazakhstan .. .. .. .. 83 Malaysia 39 12,857 14 3 48 2 34 Upper-middle-income 3,238 I 119,3891 21 w 7w 17 w 10 w 45 w 84 Botswanaa . . 85 South Africaa 325 2,989 7 14 2 3 75 86 Lithuania . . . . . . . . . . . 87 Hungary 210 4,128 23 16 12 5 45 88 Venezuela 24 724 4 16 2 8 70 89 Argentina 104 1,375 8 23 2 4 64 90 Uruguay 23 300 51 3 0 1 45 91 Brazil 197 10,295 8 10 5 9 68 92 Mexico 508 26,519 5 4 32 21 38 93 Belarus 94 Russian Federation 95 Latvia 96 Trinidad and Tobago 0 31 97 Gabon 8 66 0 46 2 15 38 98 Estonia 99 Portugal 396 13,171 39 5 10 8 39 100 Oman 0 188 28 1 13 23 35 101 Puerto Rico . . . . . . . . . . . 102 Korea,Rep. 524 41,091 23 3 20 7 47 103 Greece 185 4,006 60 5 3 1 31 104 SaudiArabia 16 1,749 0 47 8 6 39 105 Yugoslavia 443 8,791 29 7 9 12 43 Low- and middle-income 6,282 259,562 29 w 6w 15w 6w 44w Sub-Saharan Africa 193 t 4,223: 21w 8w 1w 11w 58 w East Asia & Pacific 1,087 128,170 28 w 3w 19 w 3w 47 w South Asia 760 15,695 61 w 4w 1w 1w 34 w Europe and Central Asia 2,316 1 50,571 1 37 w 8w 9w 6w 40 w Middle East & N. Africa 306 1 10,385 1 42 w 17w 6w 4w 32 w Latin America & Caribbean 1,295 47,529 13 w 8w 20 w 14 w 45 w Severely indebted 1,420 I 50,899 12w 9w 19w 14w 47w High-income economies 120,492 1 1,578,136 t 6w 13 w 12w 19w 50 w OECD members 117,366: 1,472,714: 5w 13 w 11w 20 w 50 w 106 Ireland 439 15,906 7 29 11 2 51 107 tlsrael 308 7,878 10 14 10 4 62 108 New Zealand 121 1,967 9 23 7 4 57 109 Spain 773 33,133 4 9 8 36 43 110 tHong Kong 1,861 24,794 43 1 13 1 43 111 tSingapore 112 20,668 5 6 29 2 58 112 UnitedKingdom 10,457 108,160 5 17 10 14 53 113 Australia 471 6,676 4 32 5 11 49 114 Italy 7,726 113,636 16 8 8 11 57 115 Netherlands 5,678 74,299 7 27 9 10 47 116 Belgiumc 7,660 80,272 9 20 6 22 4.4 117 Austria 1,637 29,461 9 9 13 7 62 118 France 9,240 137,947 6 16 9 26 44 119 Canada 8,088 73,986 1 8 8 39 43 120 United States 21,215 213,854 2 12 13 22 51 121 Germany 23,641 266,516 5 15 11 20 50 122 Denmark 1,413 18,766 8 14 10 5 63 123 Finland 1,170 16,343 3 8 9 7 73 124 Norway 1,059 9,007 2 21 8 11 59 125 Sweden 4,143 39,320 2 11 9 19 60 126 Japan 8,851 184,917 1 4 20 31 45 127 Switzerland 3,568 48,409 5 22 10 3 60 World 126,774 1 1,837,698 1 9w 12w 12w 17w SOw Fuel exporters 290 I 7,401 15 w 23w 3w 3w 56w Note: Data cover high-income OECD countries only. a. Data are for the South African Customs Union comprising South Africa, Namibia, Lesotho, Botswana, and Swaziland; trade among the component territories is excluded. b. Excludes the Canal Zone. c. Includes Luxembourg. 271 Table 18. Balance of payments and reserves Current account balance Gross international reserves Net workers (millions of dollars) remittances Months of After official transfers Before official transfers (millions of dollars) Millions of dollars import coverage 1970 1991 1970 199/ /970 1991 /970 1991 1991 Low-income economies 3,907 86,647 1 5.1 w China and India 1,023 55,781 1 7.9 w Other low-income 2,884 I 30,8661 3.1 w I Mozambique . . -245 . . -783 -30° . . 240 2.3 2 Tanzania -36 _284a 37 _832a . 65 204 1.4 3 Ethiopia -32 _222a 43 _585a . 201 72 106 1.0 4 Uganda 20 _182a 19 _393a .5 . 57 59 1.0 . 5 Bhutan . l7a -36° . . 0 . . 99 9.3 6 Guinea-Bissau . . -19 . . -86 . . -2 . . . 7 Nepal _la -320° -25° -380° . . 0° 94 451 5.9 8 Burundi 2 3la _2l4a . 15 147 5.0 9 Chad 2 -80 -33 -347 -6 -39 2 124 2.6 10 Madagascar 10 -192 -42 -318 -26 I 37 89 1.2 11 Sierra Leone -16 -95 -20 -136 . . . . 39 10 0.2 -1 l4a _234a 932a 764a 12 Bangladesh -210° 0° 1,308 4.0 13 La0PDR . . -52 . . -121 . . . . 6 61 3.3 14 Malawi -35 I84a -46 -244 -4 0 29 158 2.5 IS Rwanda 7 -34 -12 -194 -4 -II 8 110 3.7 16 Mali -2 -37 -22 -344 -1 76 1 326 4.5 17 Burkina Faso 9 -90 -21 -426 16 79 36 350 4.8 18 Niger 0 -4 -32 -164 -3 -38 19 207 5.3 19 India _385a _3.026a _S9la .3,477a 80 2,540° 1.023 7,616 3.3 20 Kenya -49 -231 -86 -435 -2 220 145 0.6 21 Nigeria -368 1,203 -412 470 . . 12 223 4,678 4.4 22 Chinat _8la 13,272 _8la 12,885 0° 189 . . 48,165 10.1 23 Haiti II -II 4 -176 13 86 4 24 0.6 24 Benin -3 -89 -23 -174 0 70 16 196 3.6 25 Central African Rep. -12 -80° -24 _2l9a 4 36 I 107 4.4 26 Ghana -68 -220° -76 442a 9 0 43 644 4.4 27 Pakistan -667 -1,558 -705 -2,171 86 1,848 195 1,220 1.2 28 Togo 3 -83 -14 -170 -3 5 35 369 5.0 29 Guinea . . -236° . . -329° 0 . 30 Nicaragua -40 -5 -43 -849 . . 10 49 . 31 Sri Lanka -59 -268 -71 -472 3 442 43 724 2.5 32 Mauritania -5 -125 -13 -209 -6 0 3 72 1.2 33 Yemen, Rep. . 22a _06a 800° . . . 34 Honduras -64 -220 -68 -368 . . 0 20 112 1.0 35 Lesotho l8a 63 -1° -443 29° 0 . . 115 1.3 36 Indonesia -310 -4,080 -376 -4,212 . . 130 160 10,358 3.3 37 Egypt, Arab Rep. -148 2,404a -452 2,438a 29 3755a 165 6,185 4.4 38 Zimbabwe -l4 552a _26a _693a 0 59 295 1.4 39 Sudan -42 _l,652a ,g57a 62a 22 8 0.0 40 Za,nbia 108 Ia 107 _487a -48 0 515 186 1.4 Middle-income economies 23,267 1 241,422 1 3.6 w Lower-middle-income 13,049 1 103,643 1 3.6 w 41 Bolivia 4 -262 2 -422 . . -1 46 422 3.7 42 Côted'Ivoire -38 -1,451 -73 -1,614 -56 -491 119 29 0.1 43 Senegal -16 -133 -66 -503 -16 32 22 23 0.1 44 Philippines -48 -1,034 -138 -1,388 329 255 4,436 3.3 _838a 57a 45 Papua New Guinea -89° -239° - I ,053 345 1.6 _658a II 3 46 Cameroon -30 -658° -47 81 43 0.2 47 Guatemala -8 -184 -8 -186 . . 123 79 881 4.8 48 Dominican Rep. -102 -58 -103 -115 25 330 32 448 2.2 49 Ecuador -113 -467 -122 -577 . . 0 76 1,081 3.2 50 Morocco -124 -396 -161 -676 27 1,973 142 3,349 4.5 SI Jordan -20 _712a -130 -876° . . 450° 258 1,105 3.4 52 Tajikistan .. .. .. .. .. 53 Peru 202 -1,478 146 -1,794 . . . . 339 3,090 6.1 54 El Salvador 9 -168 7 -369 . . 468 64 453 3.1 55 Congo -45° -169 .53a -231 ...3 53a 9 9 0.1 56 Syrian Arab Rep. -69 1,827 -72 1,747 7 375 57 . 57 Colombia -293 2,349 -333. 2,363 6 866 207 6,335 8.6 -16 476a 476a 58 Paraguay -19 0 18 974 5.2 59 Uzbekistan .. .. .. .. .. .. 60 Jamaica -153 -198 -149 -303 29 144 139 106 0.5 61 Romania -23 -1,184 -23 -1,306 1,219 2.4 62 Namibia . 82a 257 . . . . . . . 63 TUnisia -53 -191 -88 -322 20 562 60 866 1.6 64 Kyrgyzstan .. .. .. .. .. 65 Thailand -250 -7,564 -296 -7,609 . . 0 911 18,393 4.8 66 Georgia .. * Data forlaiwan, China, are: I 12,015 2 12,036 . . 627 74,548 11.2 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 272 Current account balance Gross international reserves Net wor*ers (millions of dollars) remittances Months of After official transfers Before official transfers (millions of dollars) Millions of dollars import coverage 1970 1991 1970 1991 /970 1991 /970 /99/ 199/ 67 Azerbaijan S 68 Turkmenistan .. .. S 69 Turkey -44 272 -57 -1,973 273 2,819 6.616 2,9 70 Poland -1.282 -2,191 0 3,800 2.1 71 Bulgaria _718a 7lg 72 CostaRica -74 -82 -77 -165 . 0 16 931 4.5 73 Algeria -125 2.555a -163 2.555a 178 274 352 3.460 3.5 74 Panama -64 135 -79 -lOS . . 16 499 0.9 75 Armenia . . . . . . . . . 76 Chile -91 142 -95 -158 . 0 392 7,700 7.8 77 Iran, Islamic Rep. -507 _7,806a -511 _7,806a . 217 78 Moldova 79 Ukraine .. 80 Mauritius 8 -37 5 -39 46 915 5.5 81 Czechoslovakia 146 947a 156 96la 4,176 3.8 82 Kazakhstan .. 83 Malaysia 8 -4.530 2 -4,617 . 0 667 11.717 3.2 Upper-middle-income 10,219 r 137,779 1 3.6 w 84 Botswana -30 47 -35 -251 -9 0 . . 3,772 17.6 85 South Africa -1,215 2,664 -1.253 2,696 . . 1.057 3,187 1.5 86 Lithuania . . . . . . . . 87 Hungary 403 370k . . . . 4,028 37 88 Venezuela -104 1,663 -98 1,696 -87 -661 1,047 14,719 10.8 89 Argentina -163 -2,832 -160 -2,832 . . . . 642 8,073 5.5 90 Umguay -45 lOS -55 65 . . 186 1,146 5.9 91 Brazil -837 _3,071a -861 _3,Ø73a . 0 1,190 8,749 2.7 92 Mexico -1,068 -13.282 -1,098 -13,468 1,853 756 18,052 3.7 93 Belarus . . . . . . . . . . . . 94 Russian Federation . . . . . . . . . . . 95 Larvia . . . . . . . . . . . . . 96 Tnnidadandlobago -109 -17 -104 -20 3 5 43 358 1.9 97 Gabon -3 -160 -IS -185 -8 -125 15 332 3.6 98 Estonia . . . . . . . . . . . 99 Portugal _I58 -716 -l58 -2,098 504 4,517 1,565 26,239 10.5 100 Oman . . 1,095 . . /153 . . -845 13 1,765 5.5 101 Puerto Rico . . . . . . . . . . . . . . . 102 Korea, Rep. -623 -8.726 -706 -8,553 . . . . 610 13.815 3.8 103 Greece -422 -1,521 -424 -5,555 333 2,115 318 6,400 3.5 104 Saudi Arabia 71 -25,738 152 -19,250 -183 -13,746 670 13.298 2.4 105 Yugoslavia -372 -1,161 -378 -1,159 441 2,024 143 3,360 2.0 Low- and middle-income 27,175 1 328,069 1 4.0 w Sub-Saharan Africa 2,0281 14,735 1 3.0 w East Asia & Pacific 2,983 r 108,057 1 4.3 w South Asia 1,4041 12,018 1 2.8 w Europe and Central Asia 9,699 1 71,9871 4.3 w Middle East & N. Africa 4,477 t 43,101 I 3.0 w Latin America & Caribbean 5,527 1 74,986 1 4.6 w Severely indebted 11,7261 66,599 1 3.3 w High-income economies 71,7621 851,0611 2.7 w OECD members 69,820 1 797,852 1 2.6 w 306 Ireland -198 92/ -228 -1,761 698 5,867 2.! 107 tlsrael -562 -822 -766 -5,257 . . . . 452 6.428 3.0 108 New Zealand -232 -20 -222 30 16 245 258 2,950 2.6 109 Spain 79 -15.954 79 -19,810 469 1,603 1,851 71,345 7.1 110 tHong Kong 225 2,487 225 2,487 . . . . . . . III tSingapore -572 4.208 -585 4,350 . . . . 1,012 34,133 5.4 112 United Kingdom 1,970 -11.438 2,376 -9,575 . . 2,764 48,373 1.5 113 Australia -785 -9,853 -691 -9,655 . . . . 1,709 19,339 3.4 114 Italy 798 -21,454 1,094 -16,670 446 779 5,547 72,254 3.3 115 Netherlands -489 8.760 -513 11.950 -51 -315 3.362 33,335 2.4 116 Belgium' 717 4.731 904 6.201 38 -274 . . . 117 Austria -75 -252 -73 -144 -7 367 1,806 17,415 2.7 118 France -204 -6.148 IS -1,194 -641 -1,786 5,199 60,227 2.0 119 Canada 1,008 -25,529 960- -24,600 . . . . 4,733 20,836 1.4 120 United States 2,330 -3,690 4,680 -24.670 -650 -7,600 15.237 159.273 2.6 121 Germany' 837 -19.497 1,839 9.978 -1,366 -4,213 13,879 96,657 2.3 122 Denmark -544 2,167 -510 2.513 488 7,990 1.7 123 Finland -240 -6,695 -233 -5,998 . . . . 455 8,317 2.8 124 Norway -242 4,939 -200 6,125 . . -140 813 13.651 3.7 125 Sweden -265 -3.243 -160 -1.636 . . 20,477 20 775 3.0 126 Japan 1,990 72,905 2.170 84,740 . 4,876 . 80,626 . . 2.4 127 Switzerland 161 9,847 203 10,307 -313 -2,062 5,317 58,451 6.5 World 98,937 1 1,179,1301 3.0 w Fuel exporters 4,693 1 52,194 1 3.9 w a. World Bank estimate. b. Includes Luaembourg. c. Data prior to July 1990 refer to the Federal Republic of Germany before unification. 273 Table 19. Official development assistance from OECD and OPEC members OECD: Total net flows /965 /970 1975 1980 1985 /988 1989 1990 1991 Millions of US dollars 106 Ireland 0 0 8 30 39 57 49 57 72 108 New Zealand . 14 66 72 54 104 87 95 100 112 UnitedKingdom 472 500 904 1,854 1,530 2,645 2,587 2,638 3,348 113 Australia 119 212 552 667 749 1,101 1,020 955 1,050 114 Italy 60 147 182 683 1,098 3,193 3,613 3,395 3,352 115 Netherlands 70 196 608 1,630 1.136 2,331 2,094 2,592 2,517 116 Belgium 102 120 378 595 440 601 703 889 831 117 Austria 10 II 79 178 248 301 283 394 548 118 France 752 971 2,093 4,162 3,995 6,865 7,450 9,380 7,484 119 Canada 96 337 880 1,075 1,631 2,347 2,320 2,470 2,604 120 UnitedStates 4,023 3,153 4,161 7,138 9,403 10,141 7,676 11,394 11,362 121 Germanyb 456 599 1,689 3,567 2,942 4.731 4,949 6,320 6,890 122 Denmark 13 59 205 481 440 922 937 1,171 1,300 123 Finland 2 7 48 110 211 608 706 846 930 124 Norway II 37 184 486 574 985 917 1,305 1.178 125 Sweden 38 117 566 962 840 1,534 1,799 2,012 2,116 126 Japan 244 458 1,148 3,353 3,797 9.134 8,965 9,069 10,952 127 Switzerland 12 30 104 253 302 617 558 750 863 Total 6,480 6,968 13,855 27,396 29,429 48,114 46,713 55,632 55.519 As a percentage of donor GNP 106 Ireland 0.00 0.00 0.09 0.16 0.24 0.20 0.17 0.16 0.19 108 New Zealand . . 0.23 0.52 0.33 0.25 0.27 0.22 0.23 0.25 112 UnitedKingdom 0.47 0.41 0.39 0.35 0.33 0.32 0.31 0.27 0.32 113 Australia 0.53 0.59 0.65 0.48 0.48 0.46 0.38 0.34 0.38 114 Italy 0.10 0.16 0.11 0.15 0.26 0.39 0.42 0.32 0.30 115 Netherlands 0.36 0.61 0.75 0.97 0.91 0.98 0.94 0.94 0.88 116 Belgium 0.60 0.46 0.59 0.50 0.55 0.39 0.46 0.45 0.42 117 Austria 0.11 0.07 0.21 0.23 0.38 0.24 0.23 0.25 0.34 118 France 0.76 0.66 0.62 0.63 0.78 0.72 0.78 0.79 0.62 119 Canada 0.19 0.41 0.54 0.43 0.49 0.50 0.44 0.44 0.45 120 United States 0.58 0.32 0.27 0.27 0.24 0.21 0.15 0.21 0.20 121 Germanyt 0.40 0.32 0.40 0.44 0.47 0.39 0.41 0.42 0.41 122 Denmark 0.13 0.38 0.58 0.74 0.80 0.89 0.93 0.93 0.96 123 Finland 0.02 0.06 0.18 0.22 0.40 0.59 0.63 0.64 0.76 124 Norway 0.16 0.32 0.66 0.87 1.01 1.13 1.05 1.17 1.14 125 Sweden 0.19 0.38 0.82 0.78 0.86 0.86 0.96 0.90 0.92 126 Japan 0.27 0.23 0.23 0.32 0.29 0.32 0.31 0.31 0.32 127 Switzerland 0.09 0.15 0.19 0.24 0.31 0.32 0.30 0.31 0.36 National currencies 106 Ireland (millions of pounds) 0 0 4 15 37 37 34 35 41 108 New Zealand (millions of dollars) . . 13 55 74 109 158 146 160 185 112 UnitedKingdom(millionsofpounds) 169 208 409 798 1,180 1,485 1,577 1,478 1,736 113 Australia (millions of dollars) 106 189 402 591 966 1404 1,386 1.323 1,382 114 Italy (billions of lire) 38 92 119 585 2,097 4,156 4,958 4,068 3,859 115 Netherlands (millions ofguilders) 253 710 1,538 3,341 3,773 4,410 4,440 4,720 4,306 116 Belgium(millionsoffrancs) 5,100 6,000 13,902 17,399 26,145 22,088 27,714 29.720 26,050 117 Austria (millions ofschillings) 260 286 1,376 2,303 5,132 3,722 3,737 4.477 5,861 118 France(millionsoffrancs) 3,713 5,393 8,971 17,589 35,894 40,897 47,529 51,076 38,777 119 Canada(millionsofdollars) 104 353 895 1,357 2,327 2,888 2,747 2,882 3,009 120 UnitedStates(millionsofdollars) 4,023 3,153 4,161 7,138 9,403 10,141 7,676 11,394 11,362 121 Germany(millionsofdeutschemarks)b 1,824 2,192 4,155 6,484 8,661 8,319 9,302 10,311 10,446 122 Denmark(millionsofkroner) 90 443 1,178 2,711 4,657 6,304 6,850 7,347 7,096 123 Finland(millionsofmarkkaa) 6 29 177 414 1,308 2,542 3,031 3,336 3,845 124 Norway(millionsofkroner) 79 264 962 2,400 4,946 6,418 6,335 7,542 7,037 125 Sweden(millionsofkronor) 197 605 2,350 4,069 7,326 9,396 11,600 11,909 11,704 126 Japan(billionsofyen) 88 165 341 760 749 1,171 1,336 1,313 1,371 127 Switzerland(millionsoffrancs) 52 131 268 424 743 903 912 1,041 1,170 SUInmaI7 Billions of US dollars ODA (current prices) 6.5 7.0 13.9 27.3 29.4 48.1 46.7 55.6 55.5 ODA (1987 prices) 28.2 25.3 29.8 36.8 39.4 44.9 43.6 47.6 45.7 GNP (current prices) 1,374.0 2,079.0 4,001.0 7,488.0 8,550.0 13,547.0 13,968.0 15,498.0 16,818.6 Percent ODAasapercentageofGNP 0.47 0.34 0.35 0.35 0.34 0.34 0.32 0.33 0.33 Indec (1987 = 100) GDP deflator' 23.0 27.6 46.5 74.1 74.6 107.1 107.5 116.8 121.4 274 OECD: Total net bilateral flows to low-income economies' 1965 1970 1975 1980 1985 1986 /988 /989 /990 1991 As a percentage of donor GNP 106 Ireland . . 0.01 0.03 0.02 0.02 0.01 0.01 108 NewZealand . . . 0.14 0.01 0.00 0.01 0.01 0.01 0.00 112 United Kingdom 0.23 0.09 0.11 0.10 0.07 0.07 0.06 0.07 0.05 113 Australia 0.08 0.00 0.10 0.07 0.04 0.04 0.04 0.06 0.05 114 Italy 0.04 0.06 0.01 0.00 0.06 0.12 0.17 0.12 0.09 115 Netherlands 0.08 0.24 0.24 0.32 0.23 0.28 0.27 0.23 0.25 116 Belgium 0.56 0.30 0.31 0.13 0.13 0.12 0.09 0.05 0.09 117 Austna 0.06 0.05 0.02 0.11 0.05 0.03 0.03 0.07 0.10 118 France 0.12 0.09 0.10 0.06 0.11 0.10 0.12 0.14 0.13 119 Canada 0.10 0.22 0.24 0.13 0.14 0.13 0.13 0.09 0.10 120 United States 0.26 0.14 0.08 0.06 0.06 0.04 0.03 0.02 0.05 121 Germanyb 0.14 0.10 0.12 0.07 0.13 0.10 0.08 0.08 0.10 122 Denmarl 0.02 0.10 0.20 0.17 0.26 0.23 0.25 0.26 0.24 123 Finland . . . . 0.06 0.03 0.09 0.10 0.24 0.22 0.17 124 Norway 0.04 0.12 0.25 0.28 0.34 0.43 0.37 0.32 0.37 125 Sweden 0.07 0.12 0.41 0.26 0.24 0.30 0.21 0.23 0.25 126 Japan 0.13 0.11 0.08 0.12 0.10 0.10 0.13 0.13 0.10 127 Switzerland 0.02 0.05 0.10 0.07 0.11 0.10 0.10 0.12 0.11 Total 0.20 0.13 0.!! 0.08 0.08 0.08 0.09 0.08 0.09 OPEC: Total net flowsd 1976 1980 /984 /985 1986 /987 /988 1989 1990 /99/ Millions of US dollars 21 Nigeria 80 35 51 45 52 30 14 70 13 Qatar ISO 277 10 8 IS 0 4 -2 I 73 Algeria II 81 52 54 114 39 13 40 7 5 77 Iran, Islamic Rep. 751 -72 52 -72 69 -10 39 -94 2 88 Venezuela 109 135 90 32 85 24 55 52 15 Iraq 123 864 -22 -32 -21 -35 -22 21 55 0 Libya 98 376 24 57 68 66 129 86 4 25 104 SaudiArabia 2,791 5,682 3,194 2,630 3,517 2,888 2,048 1,171 3,692 1,704 United Arab Emirates 1,028 1,118 88 122 87 15 -17 2 888 558 Kuwait 706 1,140 1,020 771 715 316 108 169 1,666 387 Total OPEC 5,877 9,636 4,559 3,615 4,704 3,333 2,369 1,514 6,34! Total OAPECC 4,937 9,538 4,366 3,610 4,498 3,389 2,36! . As a percentage of donor GNP 21 Nigeria 0.19 0.04 0.06 0.06 0.13 0.12 0.05 0.28 0.06 Qatar 7.35 4.16 0.18 0.12 0.36 0.00 0.08 -0.04 0.02 0.01 73 Algeria 0.07 0.20 0.10 0.10 0.19 0,07 0.03 0.11 0.03 0.01 77 Iran, Islamic Rep. 1.16 -0.08 0.03 -0.04 0.03 0.00 0.01 -0.02 88 Venezuela 0.35 0.23 0.16 0.06 0.14 0.06 0.09 0.13 0.03 Iraq 0.76 2.36 -0.05 -0.06 -0.05 -0.08 -0.04 0.04 . . Libya 0.66 1.16 0.10 0.24 0.30 0.30 0.63 0.41 0.01 0.09 104 SaudiArabia 5.95 4.87 3.20 2.92 3.99 3.70 2.53 1.37 3.90 1.44 United Arab Emirates 8.95 4.06 0.3Z 0.45 0.41 0.07 -0.07 0.02 2.65 1.66 Kuwait 4.82 3.52 3.95 2.96 2.84 1.15 0.40 0.54 Tota!OPECd 2.32 1.85 0.76 0.60 0.78 0.52 0.34 0.21 Total OAPECC 4.23 3.22 1.60 1.39 1.80 1.10 0.86 a. Organization of Economic Cooperation and Development. b. Data refer to the Federal Republic of Germany before unification. c. See the technical notes. d. Organization of Petmleum Exporting Countries. e. Organization of Arab Peftoleum Exporting Countries. 275 Table 20. Official development assistance: receipts Net disbursement ofODA from all sources Per capita As percentage Millions of dollars (dollars) of GNP 1985 1986 1987 1988 1989 1990 199! 1991 /991 Low-income economies 17,432 t 19,484 t 21,412 t 24,513 1 24,763 t 30,653 1 31,921 t 10.2 w 3.0 w China and India 2,532 t 3,254 t 3,300 t 4,086 I 4,048 t 3,605 t 4,701 I 2.3 w 0.8 w Other low-income 14,900 t 16,230 t 18,112 t 20,427 t 20,715 I 27,047 1 27,220i 24.5 w 6.6 w I Mozambique 300 422 65l 893 772 935 920 57.0 69.2 2 Tanzania 484 681 882 982 920 1,141 1,076 42.7 39.2 3 Ethiopia 710 636 634 970 752 1,014 1,091 20.7 16.5 4 Uganda 180 198 280 363 443 551 525 31.1 19.5 5 Bhutan 24 40 42 42 42 48 64 43.8 25.9 6 Guinea-Bissau 58 71 III 99 lOt 117 101 101.3 48.2 7 Nepal 234 301 347 399 493 430 453 23.4 13.8 8 Burundi 139 187 202 189 l99 265 253 44.7 21.8 9 Chad 181 165 198 264 24! 303 262 44.9 20.2 10 Madagascar 185 316 321 304 321 386 437 36.3 16.4 II Siena Leone 65 87 68 102 l0O 65 105 24.7 12.9 12 Bangladesh 1,131 1,455 1,635 1,592 1,800 2,048 1,636 14.8 7.0 13 Lao PDR 37 48 58 77 140 152 131 30.8 12.8 14 Malawi 113 198 280 366 412 48! 495 56.2 22.6 IS Rwanda 180 211 245 252 232 293 351 49.2 22.2 16 Mali 376 372 366 427 454 467 455 52.2 l8.5 17 BurkinaFaso l95 284 281 298 272 336 409 44.1 l4.8 18 Niger 303 307 353 371 296 39! 376 47.5 l6.5 19 India 1,592 2,120 1,839 2,097 1,895 1,524 2,747 3.2 I.! 20 Kenya 430 455 572 808 967 l,053 873 34.9 10.6 21 Nigeria 1,032 59 69 l20 346 250 262 2.6 0.8 22 China 940 1,134 1,462 1,989 2,153 2,08! 1,954 1.7 0.5 23 Haiti 150 175 218 147 200 172 182 27.5 6.9 24 Benin 94 138 138 162 263 27! 256 52.4 13.6 25 CentralAfrican Rep. 104 139 176 196 192 244 174 56.5 13.8 26 Ghana 196 371 373 474 550 498 724 47.2 11.3 27 Pakistan 769 970 879 1,408 1,129 1,149 1,226 l0.6 2.7 28 Togo III 174 126 199 183 24! 204 54.0 12.5 29 Guinea 115 l75 213 262 346 296 371 63.1 12.6 30 Nicaragua 102 ISO 141 213 225 320 826 217.8 12.0 3l SriLanka 468 570 502 598 547 674 814 47.4 8.8 32 Mauritania 207 225 185 l84 242 202 208 l02.9 18.4 33 Yemen, Rep. 392 328 422 304 370 405 313 25.0 3.9 34 Honduras 270 283 258 32! 242 450 275 52.2 9.1 35 Lesotho 93 88 107 l08 127 139 123 67.9 18.9 36 Indonesia 603 7!! l,246 1,632 1,839 1,724 1,854 l0.2 1.6 37 Egypt, Arab Rep. 1,760 1,716 1,773 l,537 1,568 5,444 4,988 93.l l5.2 38 Zimbabwe 237 225 294 273 265 340 393 39.0 6.2 39 Sudan l,128 945 898 937 772 825 887 34.3 40 Zambia 322 464 430 478 392 486 884 106.3 Middle-income economies 9,037 1 9,439: 10,4301 9,621 1 10,013 t 15,412 1 15,500 1 16.2 w 0.7 w Lower-middle-income 7,049 t 8,087 1 9,027 t 8,257 1 8,533 1 13,629 t 13,6391 24.3 w 1.8w 41 Bolivia 197 322 318 394 440 506 473 64.4 9.4 42 Côted'!voire 117 186 254 439 403 693 633 51.2 6.6 43 Senegal 289 567 641 569 650 788 577 75.7 10.0 44 Philippines 460 956 770 854 844 1,279 1,051 l6.7 2.3 45 Papua New Guinea 257 263 322 380 339 416 397 100.l 10.6 46 Cameroon 153 224 213 284 458 431 501 42.2 4.3 47 Guatemala 83 135 241 235 261 203 197 20.8 2.1 48 DominicanRep. 207 93 I30 118 142 100 66 9.1 0.9 49 Ecuador 136 l47 203 137 160 155 220 20.4 1.9 50 Morocco 766 403 447 480 450 1,026 1,075 41.9 3.9 SI Jordan 538 564 577 417 273 884 905 247.1 22.0 52 Tajikistan .. .. .. .. . . .. .. 53 Peru 316 272 292 272 305 395 590 26.9 1.2 54 El Salvador 345 341 426 420 443 349 290 54.9 4.9 55 Congo 69 110 152 89 91 214 133 56.8 4.6 56 SyrianArabRep. 6l0 728 684 191 127 684 373 29.8 2.2 57 Colombia 62 63 78 61 67 88 123 3.8 0.3 58 Paraguay 50 66 8l 76 92 56 l44 32.6 2.3 59 Uzbekistan .. .. .. .. .. .. 60 Jamaica 169 178 168 193 262 273 l66 69.7 4.7 61 Romania . . . . . . . . . . . . . . . 62 Namibia 6 IS 17 22 59 123 184 124.0 8.1 63 Tunisia 163 222 274 316 283 393 322 39.l 2.5 64 Kyrgyzstan .. .. .. .. .. .. 65 Thailand 459 496 504 563 739 802 722 12.6 0.8 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 276 Net disbursement of ODA from all sources Per cc9sita As percentage Millions of dollars (dollars) of GNP 1985 1986 1987 1988 1989 1990 1991 1991 /991 67 Azerbaijan . . . . . . 68 Turkmenistan .. .. .. .. .. .. 69 Thrkey 179 339 376 267 140 1,219 1,675 29.2 1.6 70 Poland 71 Bulgaria 0 72 Costa Rica 280 196 228 187 226 227 173 56.4 3.1 73 Algeria 173 165 214 171 152 217 310 12.1 0.7 74 Panama 69 52 40 22 18 93 101 40.9 1.8 75 Armenia . . . . . . . . . . . S . . 76 Chile 40 -5 21 44 61 102 120 9.0 0.4 77 Iran, Islamic Rep. 16 27 71 82 96 105 194 3.4 0.2 78 Moldova . 79 Ukraine . . . . . . . . . . . . . . . 80 Mauritius 27 56 65 59 58 89 67 61.8 2.5 81 Czechoslovakia . 82 Kazakhstan .. .. .. .. .. .. 83 Malaysia 229 192 363 104 140 469 289 I5.9 0.6 Upper-middle-income 1,988 1 1,353 1 1,403 1 1,365 1 1,480 1 1,783 1 1,862 1 4.7 w 0.1 w 84 Botswana 96 102 156 151 160 149 135 102.5 3.7 85 South Africa . . . 86 Lithuania . 87 Hungary . . . . . . . . . . . . . . . 88 Venezuela 11 16 19 18 21 79 33 1.7 0.1 89 Argentina 39 88 99 152 211 171 253 7.7 0.2 90 Uruguay 5 27 18 41 38 47 51 16.3 0.5 91 Brazil 123 178 289 210 206 167 182 1.2 0.0 92 Mexico 144 252 155 173 86 141 185 2.2 0.1 93 Belanis 94 Russian Federation . . . . 95 Latvia . . . . . . . . . . . . . . . 96 TrinidadandTobago 7 19 34 9 6 18 -2 -1.3 0.0 97 Gabon 61 79 82 106 133 132 142 121.4 2.9 98 Estonia 99 Portugal . . . . . . . . . . . . . . . 100 Oman 78 84 16 I 18 66 14 8.8 0.1 101 PuertoRico . . . . . . 0 . . . . . 102 Korea, Rep. -9 -18 II 10 52 52 54 1.3 0.0 103 Greece Il 19 35 35 30 37 39 3.8 0.1 104 Saudi Arabia 29 31 22 19 36 44 45 2.9 0.0 105 Yugoslavia II 19 35 44 43 47 159 6.6 Low- and middle-income 26,469 1 29,155 32,027 I 34,286 1 34,934 1 46,127 I 47,453 I 11.6w 1.5 w Sub-Saharan Africa 9,522 t 10,587 1 11,9261 13,470 13,8481 16,538 I 16,158 I 33.1 w 10.0 w East Asia & Pacific 4,881 4,955 1 5,935 6,869 1 7,251 1 8,007 1 7,594 I 4.5 w 0.7 w South Asia 4,244 I 5,4741 5,307 1 6,236 6,101 6,030 I 7,488 1 6.5 w 2.1 w Europe and Central Asia 247 1 403 1 458 1 359 1 207 I 1,307 1,8961 20.6 w 1.0 w Middle East & N. Africa 4,710 1 4,474 t 4,700 1 3,670 I 3,517 I 9,747 1 9,300 1 38.1 w 2.3 w Latin America & Caribbean 3,024 1 3,262 1 3,701 I 3,682 I 4,010 1 4,498 5,017 1 11.4 w 0.5 w Severely indebted 3,633 1 3,851 1 4,166 I 3,5441 3,373 1 5,976 1 6,488 1 14.9w 0.6 w High-income economies 2,2321 2,3061 1,7461 1,6551 1,6671 1,8041 2,1501 55.3 w 0.4 w OECD members 106 Ireland .. .. .. .. .. .. 107 tlsrael 1,978 1,937 1,251 1,241 1,192 1,372 1,749 353.6 2.8 108 New Zealand 109 Spain . . . . . . . . . 0 . . 110 tHongKong 20 I8 19 22 40 38 36 6.3 0.0 Ill tSingapore 24 29 23 22 95 -3 8 2.8 0.0 112 United Kingdom 0 113 Australia . . . . 114 Italy . . . . 115 Netherlands . 116 Belgium 117 Austria 118 France 119 Canada 120 United States 121 Germany 122 Denmark 123 Finland 124 Norway 125 Sweden 126 Japan 127 Switzerland World 28,701 1 31,461 r 33,773 1 35,491 1 36,601 1 47,931 1 49,603 1 12.0 w 1.4 w Fuel exporters 1,606 1 805 1 1,033 I 781 1 1,076 1 1,468 1 1,948 1 7.5 w 0.3 w 277 Table 21. Total external debt Total arrears Use of IMF credit on LDOD Ratio of present Long-tenn debt (millions of Short-term debt Total external debt (millions of value to nominal (millions of dollars) dollars) (millions of dollars) (millions of dollars) dollars) value of debt 1980 199! 1980 1991 1980 1991 1980 199! 1980 1991 1991 low-income economies China and India Other low-income I Mozambique 0 4,055 0 118 0 527 0 4,700 0 1,442 82.7 2 Tanzania 1,999 5,798 171 143 306 519 2,476 6,460 3 1,257 74.7 3 Ethiopia 669 3,301 79 0 57 174 804 3,475 I 388 68.9 4 Uganda 542 2325 89 330 64 175 695 2.830 20 470 63.2 5 Bhuian 0 86 0 0 0 1 0 87 0 0 58.4 6 Guinea-Bissau 128 574 I 5 5 74 134 653 I 89 69.6 7 Nepal 156 1.705 42 39 7 26 205 1.769 0 12 46.8 8 Bunsndi 118 899 36 49 12 13 166 961 0 0 47.7 9 Chad 204 547 14 3! II 29 229 606 6 18 48.0 10 Madagascar 892 3,381 87 127 244 208 1,223 3,715 6 366 71.6 II Sierra Leone 323 642 59 101 53 547 435 1,291 7 396 84.8 12 Bangladesh 3,417 12,103 424 727 212 22! 4,053 13,051 0 18 49.1 13 La0PDR 277 1,096 16 21 I 4 295 1,121 I I 26.8 14 Malawi 625 1,530 80 115 116 31 821 1,676 3 7 49.8 IS Rwanda ISO 780 14 13 26 52 190 845 0 19 49.5 16 Mali 669 2,392 39 60 24 79 732 2,531 1 164 58.3 17 BurkinaFaso 28! 871 IS 9 35 76 330 956 0 37 59.9 18 Niger 687 1.503 16 73 159 77 863 1.653 0 77 65.5 19 India 18,709 64,315 977 3.451 926 3,791 20,611 71,557 0 0 78.9 20 Kenya 2,557 5,776 254 493 638 744 3,449 7,014 0 108 77.3 21 Nigeria 5.381 33,588 0 0 3,553 909 8,934 34,497 0 1.353 97.1 22 China 4,504 50,502 0 0 0 10,300 4,504 60,802 0 0 92.6 23 Haiti 242 610 46 33 14 105 303 747 0 57 56.5 24 Benin 334 1,221 16 22 73 57 424 1,300 5 38 53.6 25 CentralAfncanRep. 147 803 24 33 25 48 195 884 II 53 55.6 26 Ghana 1,171 2,992 lOS 834 131 384 1,407 4,209 5 77 63.9 27 Pakistan 8,525 17,745 674 1,068 737 4,157 9,936 22,969 0 0 74.8 28 Togo 899 1,143 33 79 113 134 1,045 1.356 8 25 66.8 29 Guinea 1,004 2,401 35 55 71 170 1,110 2,626 20 306 70.4 30 Nicaragu.a 1,661 8,703 49 24 466 1,718 2,176 10,446 6 3,743 89.3 3! Sri Lanka 1,231 5,758 391 401 220 394 1,841 6,553 0 0 60.8 32 Mauritania 718 1,912 62 57 65 330 844 2,299 10 335 77.1 33 Yemen, Rep. 1,453 5,207 48 0 183 1,264 1,684 6,471 I 1,597 83.5 34 Honduras 1.165 2,940 33 34 272 203 1,470 3,177 0 270 83.8 35 Lesotho 57 406 6 18 8 4 71 428 0 7 55.6 36 Indonesia 18,169 59,960 0 166 2,775 13,503 20,944 73,629 0 I 92.3 37 Egypt, Arab Rep. 16,477 36,978 411 127 4,027 3,466 20,915 40,571 383 1,739 52.8 38 Zimbabwe 696 2,868 0 0 90 561 786 3,429 0 0 88.1 39 Sudan 4,147 9,717 431 961 585 5,229 5,163 15,907 49 9,620 90.4 40 Zambia 2,227 4.958 447 918 586 1,403 3,261 7,279 6 1,268 86.0 Middle-income economies I .ower-middle-income 41 Bolivia 2,274 3,675 126 245 300 155 2,700 4,075 0 36 75.8 42 Côted'Ivoire 4,724 15,167 65 372 1,059 3,308 5,848 18,847 0 3,426 92.9 43 Senegal 1,114 2,890 140 327 219 305 1,473 3,522 0 29 69.2 44 Philippines 8.817 25,893 1,044 1,086 7,556 4,919 17,417 31,897 0 101 96.7 45 Papua New Guinea 624 2,566 31 61 64 128 719 2,755 0 2! 90.6 46 Cameroon 2,183 5,254 59 121 271 903 2,513 6,278 2 657 89.3 47 Guatemala 831 2,230 0 64 335 411 1,166 2,704 0 565 90.6 48 Dominican Rep. 1,473 3,554 49 89 480 849 2,002 4,492 7 1,314 92.1 49 Ecuador 4,422 10,094 0 182 1,575 2,192 5,997 12,469 0 3,654 97.6 50 Morocco 8.475 20,332 457 574 778 312 9,7)0 21,219 3 739 92.0 SI Jordan 1,486 7,570 0 95 486 977 1,972 8,641 6 1,134 91.5 52 Tajikistan .. .. .. .. . .. .. .. 53 Peru 6,828 15,298 474 706 2,084 4,705 9,386 20,709 0 7,852 92.6 54 ElSalvador 659 2.070 32 0 220 102 911 2.172 0 16 74.8 55 Congo 1,257 3,989 22 6 246 749 1,526 4,744 3 1,010 86.1 56 Syrian Arab Rep. 2,918 14,932 0 . 0 631 1,882 3,549 16,815 0 1,426 72.1 57 Colombia 4,604 15,617 0 0 2,337 1,752 6,941 17,369 0 167 100.4 58 Paraguay 780 1,799 0 0 174 377 954 2,177 0 532 90.8 59 Uzbekistan .. .. .. .. . . . . .. .. .. 60 Jamaica 1,496 3,779 309 391 98 286 1,904 4,456 0 353 91.2 61 Romania 7,131 334 328 809 2,303 770 9,762 1,913 0 0 99.7 62 Namibia . . . . . . . . . . . . . . . . . . . 63 Tunisia 3,390 7,369 0 258 136 670 3,526 8,296 0 78 90.) 64 Kyrgyzstan .. .. .. .. .. .. .. .. 65 Thailand 5,646 23,336 348 0 2,303 12,492 8,297 35,828 0 0 95.2 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 278 Tote.! arrears Use of 1MF credit on LDOD Ratio ofpresent Long-term debt (millions of Short-term debt Total external debt (millions of value to nominal (millions of dollars) dollars) (millions of dollars) (millions of dollars) dollars) value of debt 1980 1991 /980 /991 /980 199/ /980 /991 1980 /99/ /991 67 Azerbaijan . . S - S . - 68 Thrknsenistan . . . - . S - 0 . . 69 Thrkey 15,575 41.135 1,054 0 2,490 9,117 19,120 50,252 0 0 95.4 70 Poland 6,594 44,057 0 853 2,300 7,571 8,894 52,481 300 11,296 101.6 71 Bulgaria 272 11,023 0 414 0 487 272 11,923 0 2,673 97.1 72 CostaRica 2,112 3,620 57 83 575 340 2,744 4,043 0 219 91.7 73 Algeria 17,034 26,557 0 995 2,325 1,084 19,359 28,636 0 0 98.1 74 Panama 2,271 3,939 23 216 680 2,637 2,974 6,791 0 3.265 98.2 75 Armenia . . - . . . S S - . 76 Chile 9,399 14,744 123 958 2,560 2,200 12,081 17,902 0 1 99.9 77 Iran, Islamic Rep. 4,508 2,736 0 0 0 8,775 4,508 11,511 0 1,944 100.2 78 Moldova 79 Ukraine . . . . . . . . . . . . . . . . . . . 80 Mauritius 318 96! 102 0 47 31 467 991 0 17 87.2 81 Czechoslovakia 0 5,845 0 1,313 3,989 2,635 3,989 9,793 0 4 99.3 82 Kazakhstan .. .. .. .. . - .. .. . - 83 Malaysia 5,256 18,753 0 0 1,355 2,692 6,611 21,445 0 0 97.5 Upper-middle-income 84 Botswana 129 536 0 0 4 7 133 543 0 9 83.8 85 South Africa . . . 86 Lithuania - - . . . . S . . . . . . . . . 87 Hungary 6,416 19,221 0 1,259 3,347 2,177 9,764 22,658 0 0 100.3 88 Venezuela 13,795 28,839 0 3,249 15,550 2,284 29,345 34,372 IS 0 99.0 89 Argentina 16,774 47,188 0 2,483 10,383 14,036 27,157 63,707 0 13,818 103.9 90 Uruguay 1,338 3,128 0 58 322 1,003 1,660 4,189 0 0 99.4 91 Brazil 57,500 95,130 0 1,238 13,546 20,147 71,046 116,514 20 10,832 99.9 92 Mexico 41,215 83,891 0 6,766 16,163 11,080 57,378 101,737 0 0 96.9 93 Belams . . . . . . S 94 Russian Federation . . . . . . . . . 95 Latvia . - . . . - . S . . S 96 TrinidadandTobago 713 1,817 0 385 116 130 829 2,332 0 33 97.8 97 Gabon 1,271 2,935 15 121 228 787 1,513 3,842 0 466 95.7 98 Estonia S - . S S - . . . 99 Portugal 7,215 20,170 119 0 2,395 8,398 9,729 28,568 0 0 97.0 100 Oman 436 2,270 0 0 163 427 599 2,697 0 0 96.7 101 Puerto Rico . . . . . . . . . . . . . . . . . . . 102 Korea, Rep. 18,236 29,318 683 0 10,561 11,200 29,480 40,518 0 0 97.6 103 Greece . S . 104 Saudi Arabia . . . . . . . . . . . . . . . . . . . 105 Yugoslavia 15,586 15,872 760 307 2,140 293 18,486 16,471 0 636 102.5 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland 107 tlsrael 108 New Zealand 109 Spain 110 tHong Kong Ill tSingapore 112 United Kingdom 113 Australia 114 Italy 115 Netherlands 116 Belgium 117 Austria 118 France 119 Canada 120 United States 121 Germany 122 Denmark 123 Finland 124 Norway 125 Sweden 126 Japan 127 Switzerland World Fuel exporters 279 Table 22. flow of public and private external capital Disbursements Repayment ofprincipal Interest payments (millions of dollars) (millions of dollars) (millions of dollars) lirng-term public Long-term public Long-term public and publicly Private and publicly Private and publicly Private guaranteed non guaranteed guaranteed non guaranteed guaranteed nonguaranteed 1980 /991 1980 1991 1980 1991 1980 1991 1980 1991 1980 1991 Low-income economies China and India Other low-income I Mozambique o 141 0 0 0 23 0 3 0 10 0 0 2 Tanzania 373 257 31 0 26 64 16 0 38 33 7 0 3 Ethiopia 102 260 0 0 17 90 0 0 17 36 0 0 4 Uganda 92 179 0 0 32 68 0 0 4 24 0 0 5 Bhutan 0 10 0 0 0 4 0 0 0 2 0 0 6 Guinea-Bissau 69 29 0 0 3 2 0 0 I 2 0 0 7 Nepal 50 155 0 0 2 30 0 0 2 27 0 0 8 Burundi 39 85 0 0 4 26 0 0 2 12 0 0 9 Chad 6 97 0 0 3 4 0 0 0 5 0 0 10 Madagascar 350 198 0 0 30 48 0 0 26 67 0 0 II Sierra Leone 86 47 0 0 32 I 0 0 8 0 0 12 Bangladesh 657 873 0 0 63 259 0 0 47 153 0 0 13 Lao PDR 38 50 0 0 I 5 0 0 I 3 0 0 14 Malawi 153 170 0 0 33 61 0 0 35 41 0 0 15 Rwanda 27 100 0 0 3 13 0 0 2 7 0 0 16 Mali 95 95 0 0 5 0 0 3 6 0 0 17 BurkinaFaso 65 146 0 0 25 0 0 6 0 0 18 Niger 167 67 113 0 23 98 35 36 16 49 l0 19 India 1,895 7,147 285 317 664 3,403 91 278 502 2,585 30 116 20 Kenya 550 424 87 60 117 320 88 40 130 202 39 45 21 Nigeria 1,187 844 565 0 65 1,069 177 47 440 2.218 91 13 22 China 2,539 9,992 0 0 613 4,323 0 0 318 2,946 0 0 23 Haiti 47 34 0 0 15 6 0 0 5 6 0 0 24 Benin 62 99 0 0 6 16 0 0 3 12 0 0 25 Central African Rep. 25 114 0 0 I 4 0 0 0 5 0 0 26 Ghana 220 440 0 9 77 I0I 0 8 31 60 0 3 27 Pakistan 1,052 1,737 9 19 346 907 7 41 247 565 2 II 28 logo 100 65 0 0 19 19 0 0 19 18 0 0 29 Guinea 121 237 0 0 75 82 0 0 23 39 0 0 30 Nicaragua 266 134 0 0 45 168 0 0 42 218 0 0 31 SriLanka 269 755 2 0 51 162 0 3 33 127 0 2 32 Mauritania 130 57 0 0 17 40 0 0 13 20 0 0 33 Yemen, Rep. 566 163 0 0 25 83 0 0 10 28 0 0 34 Honduras 264 209 81 23 39 126 48 15 58 135 25 4 35 Lesotho 13 50 0 0 3 17 0 0 10 0 0 36 Indonesia 2,551 5,606 695 3,467 940 4,172 693 1,948 824 2,645 358 844 37 Egypt, Arab Rep. 2,803 1,799 126 120 368 1,247 46 170 378 580 23 72 38 Zimbabwe 132 379 0 112 40 227 0 33 I0 237 0 15 39 Sudan 711 130 0 0 53 13 0 0 49 10 0 0 40 Zambia 597 336 6 2 181 248 31 0 106 236 10 0 Middle-income economies Lower-middle-income 41 Bolivia 441 287 16 0 126 114 19 25 164 107 9 12 42 Côte d'Ivoire 1,413 401 262 900 517 226 38 529 353 273 31 228 43 Senegal 327 131 0 9 152 133 4 8 67 89 0 3 44 Philippines 1,382 1,682 472 261 221 1,392 320 112 375 1,395 204 66 45 Papua New Guinea 120 213 15 228 32 159 40 203 30 78 22 55 46 Cameroon 562 425 50 76 82 149 32 86 104 157 IS 18 47 Guatemala 138 80 32 3 15 157 62 3 30 97 30 10 48 Dominican Rep. 415 108 67 0 62 103 74 16 92 77 29 7 49 Ecuador 968 521 315 I 272 474 263 22 288 453 78 4 50 Morocco 1,703 1,276 75 8 565 940 25 8 607 1,083 II 5 SI Joidan 369 611 0 0 103 279 0 0 79 304 0 0 52 lajikistan 53 Peru 1,248 500 60 0 959 454 60 112 547 340 124 19 54 El Salvador 110 275 0 0 17 155 18 5 25 69 II 2 55 Congo 520 32 0 0 34 178 0 0 37 38 0 0 56 Syrian Arab Rep. 1,148 531 0 0 225 496 0 0 77 131 0 0 57 Colombia 1,016 1,643 55 298 250 1,906 13 307 279 1,183 31 99 58 Paraguay 158 130 48 8 44 109 36 7 35 76 9 0 59 60 Uzbekistan Jamaica 328 432 25 0 91 377 10 114 189 . i 61 Romania 2,797 304 0 0 824 23 0 0 332 6 0 0 62 Namibia 63 Tunisia 558 1,142 53 30 216 764 43 35 212 372 16 l 64 Kyrgyzstan 65 Thailand 1,315 1,453 1,288 3,846 172 1,147 610 1,140 269 709 204 1,069 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for yeam other than thoue specified. 280 Disbursements Repayment ofprincipa! Interest payments (millions of dollars) (millions of dollars) (millions of dollars) Long-term public Long-term public Long-term public and publicly Private and publicly Private and publicly Private guaranteed nonguaranteed guaranteed nonguaranteed guaranteed nonguaranteed 1980 /99! /980 1991 1980 1991 1980 1991 /980 1991 1980 199! 67 Azerbaijan 68 Turkmenistan , , . 69 Turkey 2,400 4,740 75 310 566 3,946 29 630 487 2,659 20 146 70 Poland 5,058 859 0 0 2,054 383 0 0 704 451 0 0 71 Bulgaria 222 674 0 0 5 802 0 0 23 274 0 0 72 Costa Rica 435 329 102 8 76 168 88 9 130 173 41 26 73 Algeria 3,398 6,391 0 0 2,529 7,712 0 0 1,439 1,952 0 0 74 Panama 404 I 0 0 215 73 0 0 252 108 0 0 75 Armenia . . S 0 . . 0 76 Chile 857 703 2,694 943 891 594 571 327 483 2,199 435 292 77 Iran, Islamic Rep. 264 1,086 0 0 531 174 0 0 432 25 0 0 78 Moldova . . . . . 79 Ukraine . . . . . . S . . . . . . 80 Mauritius 93 103 4 50 15 55 4 31 20 45 3 5 81 Czechoslovakia 0 1,358 0 0 0 984 0 0 0 329 0 0 82 Kazakhstan .. .. .. .. .. .. .. .. .. 83 Malaysia 1,015 1,646 441 747 127 1,606 218 296 250 1,119 88 112 Upper-middle-income 84 Botswana 27 68 0 0 6 47 0 0 7 32 0 0 85 SouthAfnca . . . . . . 86 Lithuania , . . . 0 . . . . . . . . . 87 Hungary 1,552 3,114 0 0 824 2,333 0 0 636 1,439 0 0 88 Venezuela 2,870 1,527 1,891 173 1,737 607 1,235 173 1,218 1,706 257 400 89 Argentina 2,839 1,641 1,869 0 1,146 2,432 707 0 841 3,017 496 133 90 Uruguay 293 507 63 299 93 468 37 124 105 170 17 16 91 Brazil 8,335 3,674 3,192 628 3,864 3,911 2,970 752 4,202 3,693 2,132 338 92 Mexico 9,131 5,819 2,450 1,771 4,010 3,602 750 1,491 3,880 5,776 700 595 93 Belanis . . . S . . . 94 Russian Federation . . . . S . . . . . 95 Laivia . . . . . . . . . S . . . . . . 96 TrinidadandTobago 363 93 0 0 176 173 0 0 50 143 0 0 97 Gabon 171 70 0 0 279 38 0 0 119 41 0 0 98 Estonia . . . S 99 Portugal 1,950 5,546 149 195 538 3,781 126 109 486 1,296 43 39 100 Oman 98 434 0 1) 179 385 0 0 44 156 0 0 101 Puerto Rico . . . . S . . . . . . . . . . . 102 Korea, Rep. 3,429 4,752 551 2,994 1,490 2,009 64 1,360 1,293 1,195 343 503 103 Greece . 104 Saudi Arabia . . . . . . . S . . . . . . 105 Yugoslavia 1.366 105 3,223 666 368 1,351 2,012 1,039 249 988 829 235 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland 107 flsrael 108 New Zealand 109 Spain 110 tHong Kong Ill tSingapore 112 United Kingdom 113 Australia 114 Italy 115 Netherlands 116 Belgium 117 Austria 118 France 119 Canada 120 United States 121 Germany 122 Denmark 123 Finland 124 Norway 125 Sweden 126 Japan 127 Switzerland World Fuel exporters 281 Table 23. Aggregate net resource flows and net transfers Net flows on long-term debt Foreign direct (millions of dollars) fficzal grants investment in the Aggregate net Aggregate net Public and pub- Private (millions of reporting economy resource flows transfers licly guaranteed nonguaranteed dollars) (millions of dollars) (millions of dollars) (millions of dollars) 1980 /991 /980 1991 1980 /991 /980 1991 1980 1991 /980 /991 Low-income economies China and India Other low-income 1 Mozambique 0 118 0 -3 76 752 0 23 76 889 76 879 2 Tanzania 348 193 15 0 485 688 0 0 848 880 804 847 3 Ethiopia 84 171 0 0 125 464) 0 0 209 631 192 595 4 Uganda 60 III 0 0 62 253 0 I 122 365 118 341 5 Rhutan 0 5 0 0 2 28 0 0 2 33 2 31 6 Guinea-Bissau 66 28 0 0 37 60 0 0 103 88 102 85 7 Nepal 48 125 0 0 79 160 0 0 127 285 125 259 8 Bumndi 35 59 0 0 39 126 0 I 74 186 72 171 9 Chad 3 93 0 0 22 106 0 0 25 199 25 194 10 Madagascar 319 150 0 0 30 352 0 14 349 516 322 447 11 SierraLeone 54 46 0 0 24 33 -19 0 59 79 46 78 12 Bangladesh 594 614 0 0 1,001 1,070 0 1 1,595 1,685 1,548 1,532 13 Lao PDR 37 45 0 0 16 54 0 0 53 99 52 96 14 Malawi 120 109 0 0 49 219 tO 0 178 328 135 287 IS Rwanda 25 87 0 0 68 181 16 5 109 273 98 261 16 Mali 89 89 0 0 104 209 2 4 195 302 192 278 17 BurkinaFaso 55 121 0 0 88 200 0 0 142 321 128 306 18 Niger 144 -32 79 -36 5! 249 49 0 324 181 248 160 19 India 1,231 3,744 194 39 649 562 0 0 2,073 4,345 1,541 1,643 20 Kenya 433 104 -1 20 121 836 79 43 632 1,003 312 696 21 Nigeria 1,122 -225 388 -47 3 141 -740 712 773 581 -1,357 -1,836 22 China 1,927 5,669 0 0 7 262 0 4,366 1,934 10,298 1,616 7,342 23 Haiti 32 28 0 0 30 142 13 14 75 183 59 169 24 Benin 56 83 0 0 41 147 4 0 101 231 96 218 25 CentralAfricanRep. 24 110 0 0 56 6! 5 -5 85 166 85 161 26 Ghana 143 340 0 1 23 581 16 0 181 922 135 859 27 Pakistan 706 830 2 -22 482 429 63 257 1,254 1,494 1,000 876 28 Togo 82 46 0 0 15 92 42 0 139 139 119 99 29 Guinea 47 155 0 0 25 138 0 0 72 293 49 254 30 Nicaragua 221 -33 0 0 48 730 0 0 269 696 207 478 31 SriLanka 219 593 2 -3 161 200 43 98 425 887 377 737 32 Mauritania 113 17 0 0 61 97 27 0 201 113 165 93 33 Yemen,Rep. 542 80 0 0 335 87 34 0 910 167 900 139 34 Honduras 225 83 33 9 20 475 6 45 283 611 123 40! 35 Lesotho 10 33 0 0 52 48 5 8 66 88 59 -72 36 indonesia 1,611 1,434 2 1,519 109 300 180 1,482 1902 4,735 -2,514 -1,056 37 Egypt, Arab Rep. 2,435 552 80 -50 165 3,355 548 253 3,229 4,110 2,813 3,451 38 Zimbabwe 93 152 0 80 127 231 2 0 22! 462 133 210 39 Sudan 658 117 0 0 388 416 0 0 1,046 533 997 523 40 Zambia 416 89 -25 2 71 697 62 0 524 788 324 552 Middle-income economies Lower-middle-income 41 Bolivia 315 173 -3 -25 48 599 47 52 407 800 214 663 42 Côted'Ivoire 896 175 224 371 27 264 95 46 1,241 856 670 309 43 Senegal 175 -2 -4 1 78 532 15 0 263 531 161 404 44 Philippines 1,161 290 152 149 59 400 -106 544 1,266 1,383 488 -381 45 PapuaNewGuinea 89 54 -25 25 279 280 76 0 418 359 163 227 46 Camemon 480 277 18 -10 29 269 130 0 656 536 422 361 47 Guatemala 123 -77 -30 0 14 51 III 91 217 66 114 -85 48 Dominican Rep. 353 5 -7 -16 14 40 93 145 454 174 267 89 49 Ecuador 696 46 52 -21 7 56 70 85 825 166 349 -419 50 Morocco 1,138 336 50 0 75 553 89 320 1,353 1,209 685 26 51 Jordan 266 332 0 0 1,127 407 34 -12 1,427 727 1,348 422 52 Tajikistan .. .. . - .. .. .. . . .. .. .. .. 53 Peru 289 46 0 -112 31 197 27 -7 347 124 -580 -269 54 ElSalvador 92 120 -18 -5 31 114 6 25 III 255 34 147 55 Congo 486 -147 0 0 20 38 40 0 546 -109 503 -147 56 SyrianArabRep. 924 35 0 0 1,651 109 0 0 2,574 144 2,497 13 57 Colombia 766 -263 42 -9 8 51 157 420 974 199 553 -2,030 58 Paraguay 114 21 13 I 10 22 32 80 168 124 70 47 59 Uzbekistan .. .. .. .. .. .. .. .. .. 60 Jamaica 236 55 15 -6 13 248 28 127 292 424 57 59 61 Romania 1,973 281 0 0 . . 0 40 1,973 321 1,641 315 62 Namibia . . . . . . . . . . . . . . . . . S - 63 Tunisia 342 378 10 -5 26 143 235 150 612 667 232 16 64 Kyrgyzstan .. .. .. .. .. .. .. .. .. 65 Thailand 1,143 306 678 2,706 75 220 190 2,014 2,087 5,245 1,576 3,412 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 282 Net flows on long-term debt Foreign direct (millions of dollars) Official grants investment in the Aggregate net Aggregate net Public wzd pub- Private (millions of reporting economy resource flows transfers licly guaranteed nonguaranteed dollars) (millions of dollars) (millions of dollars) (millions of dollars) 1980 199! 1980 1991 1980 1991 1980 1991 /980 1991 1980 1991 67 Azerbaijan 68 Turkmenistan 69 Thrkey 1,834 794 46 -319 185 1,147 18 810 2,083 2,432 1,545 -541 70 Poland 3,005 476 0 0 10 291 3,015 767 2,311 266 71 Bulgaria 217 -128 0 0 . . 0 0 217 -128 193 -402 72 Costa Rica 359 160 14 0 0 83 53 142 425 385 235 140 73 Algeria 869 -1,321 0 0 77 79 349 0 1,295 -1.242 -830 -3,194 74 Panama 189 -73 0 0 6 89 -47 -62 149 -45 -174 -193 75 Armenia 76 Chile -34 109 2,123 616 9 97 213 576 2,312 1,398 1,307 -1,738 77 Iran, Islamic Rep. -267 912 0 0 I 70 0 0 -265 982 -1,095 958 78 Moldova 79 Ukraine 80 Mauritius ' Ô 19 13 16 I 19 93 101 69 29 81 Czechoslovakia 0 374 0 0 0 600 0 974 0 3 82 Kazakhstan 83 Malaysia 889 41 223 451 6 57 934 3,455 2,052 4,003 524 837 Upper-middle-income 84 Botswana 21 21 0 0 51 62 112 0 184 83 69 51 85 South Africa 86 Lithuania 87 Hungary 728 781 0 0 . . : 0 1,462 728 2,243 92 760 88 Venezuela 1,133 920 656 0 0 5 55 1,914 1,844 2,839 47 505 89 Argentina 1,693 -791 1,162 0 2 40 678 2,439 3,535 1,688 1,593 -2,269 90 Uruguay 200 39 26 175 1 10 290 0 516 224 395 38 91 Brazil 4,472 -237 222 -124 14 46 1,911 1,600 6,618 1,286 -670 -4,545 92 Mexico 5,121 2,217 1,700 280 14 62 2,156 4,762 8,991 7,321 3,043 -556 93 Belarus 94 Russian Federation 95 Latvia 96 Trinidad and Tobago 187 -ó Ô 185 169 372 93 -I57 -286 97 Gabon -109 32 0 35 32 125 -73 192 -465 -32 98 Estonia 99 Portugal 1,411 1,765 23 86 28 15 157 2,021 1,620 3,887 1,074 2,488 100 Oman -81 49 0 0 157 3 98 0 174 51 -156 -105 101 Puerto Rico 102 Korea, Rep. 1,90 2,743 487 1,633 8 6 6 1,116 2,440 5,498 740 3,504 103 Greece 104 Saudi Arabia 105 Yugoslavia 998 -1,247 1,211 -373 0 0 2,208 -1,620 1,131 -2,843 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland 107 jIsrael 108 New Zealand 109 Spain 110 tHong Kong Ill tSingapore 112 United Kingdom 113 Australia 114 Italy 115 Netherlands 116 Belgium 117 Austria 118 France 119 Canada 120 United States 121 Germany 122 Denmark 123 Finland 124 Norway 125 Sweden 126 Japan 127 Switzerland World Fuel exporters 283 Table 24. Total external debt ratios Total debt service as Interest payments Concessional debt Multilateral debt Total external debt as a percentage of a percentage of as a percentage of as a percentage as a percentage Exports of goods exports of goods exports of goods of total of total and services GNP and services and services external debt external debt 1980 1991 1980 1991 1980 1991 1980 1991 1980 1991 1980 1991 Low-income economies 105.5 w 225.7 w 16.6 w 44.6 w 10.1 w 21.0 w 5.0 w 9.4 w 45.9 w 38.2 w 15.9 w 24.1 w China and India 69.1 w 140.7 w 5.3 w 21.5 w 6.4 w 16.9 w 2.6 w 7.4 w 61.7 w 30.6 w 24.2 w 23.8 w Other low-income 120.2 w 307.7 w 33.5 w 85.7 w 11.6 w 25.0 w 6.0 w 11.3 w 42.2 w 41.4 w 14.0 w 24.2 w I Mozambique 0.0 1,117.1 0.0 426.0 0.0 10.6 0.0 4.3 0.0 59.7 0.0 12.0 2 Tanzania 321.7 1,207.8 48.3 250.8 19.6 24.6 10.0 7.6 55.3 60.0 21.4 30.5 3 Ethiopia 131.4 464.7 l9.5 53.4 7.3 18.6 4.5 5.8 70.6 81.5 42.2 39.1 4 Uganda 210.3 1,429.4 55.1 109.2 17.4 70.0 3.7 18.8 26.9 58.7 12.4 51.2 5 Bhutan 95.4 0.0 38.8 7.2 2.4 0.0 80.1 0.0 57.3 6 Guinea-Bissau . . . 128.0 323.7 . . . 64.3 67.9 2l.3 42.9 7 Nepal 85.5 370.0 10.4 53.5 3.2 13.6 2.1 6.2 75.7 90.4 62.0 78.7 8 Burundi 180.1 758.8 18.2 83.8 9.5 31.5 4.8 10.9 62.6 88.2 35.7 74.9 9 Chad 320.2 251.1 31.6 47.0 8.3 4.5 0.7 2.7 50.9 80.1 32.6 63.9 10 Madagascar 235.7 744.6 30.6 148.3 17.1 32.0 10.9 15.6 39.3 57.6 14.9 37.4 II Sierra Leone 157.7 . . 40.7 167.5 23.2 . . 5.7 . . 32.8 31.0 14.2 14.4 12 Bangladesh 345.3 443.7 31.3 56.0 23.2 19.9 6.4 6.4 82.4 91.0 30.3 53.6 13 LeoPDR . . 996.2 . . 109.8 . . 7.6 . . 2.9 92.1 97.3 7.1 28.6 14 Malawi 260.8 318.8 72.1 78.5 27.7 25.0 16.7 9.4 33.8 76.8 26.7 71.5 15 Rwanda 103.4 591.8 16.3 53.7 4.2 17.6 2.8 8.6 74.4 91.9 47.8 73.9 16 Mali 227.3 442.7 45.4 104.8 5.1 4.6 2.3 2.1 84.5 91.9 23.7 38.0 17 Burkina Faso 88.0 188.8 22.3 34.9 5.9 9.1 3.1 4.2 66.9 77.5 42.9 67.0 18 Niger 132.8 466.8 34.5 72.9 21.7 50.4 12.9 9.3 18.0 49.9 16.5 43.7 19 India 136.2 295.3 11.9 29.3 9.3 30.7 4.2 13.6 75.1 41.6 29.5 33.5 20 Kenya 167.3 318.4 49.0 89.6 21.8 32.7 11.4 14.5 20.8 40.6 18.3 38.2 21 Nigeria 32.2 257.1 10.1 108.8 4.2 25.2 3.3 16.8 6.1 3.1 6.4 11.6 22 China 21.2 87.1 1.5 16.4 4.4 12.1 l.5 5.3 0.5 17.6 0.0 12.4 23 Haiti 72.9 186.5 20.9 28.8 6.2 6.6 1.8 4.0 70.7 75.1 43.8 67.5 24 Benin 133.1 262.2 30.2 70.1 6.3 6.2 4.5 3.0 39.2 78.2 24.5 45.7 25 Central African Rep. 94.7 671.9 24.3 71.5 4.9 11.4 1.6 5.9 30.1 78.1 27.4 57.8 26 Ghana 116.0 384.5 31.8 66.9 13.1 26.9 4.4 9.9 57.9 56.9 19.8 49.2 27 Pakistan 208.8 244.9 42.4 50.1 17.9 21.1 7.6 10.0 73.1 54.9 15.4 35.0 28 Togo 180.1 187.2 95.3 85.0 9.0 7.3 5.8 3.5 24.4 57.5 11.4 44.9 29 Guinea 200.5 351.0 94.8 19.8 17.9 6.0 5.8 59.7 70.3 11.7 31.5 30 Nicaragua 423.4 2,917.8 108.5 153.5 22.3 109.3 13.4 62.4 21.5 28.7 19.4 9.2 31 SriLanka 123.4 211.0 46.1 72.6 12.0 13.9 5.7 5.7 56.2 73.3 11.7 31.8 32 Mauritania 306.6 458.2 125.7 214.7 17.3 16.8 7.9 6.2 60.5 68.1 14.9 26.9 33 Yemen, Rep. 104.7 292.5 . . 88.1 4.5 7.3 2.3 3.5 83.9 73.8 14.9 16.4 34 Honduras 152.0 330.8 60.5 113.8 21.4 30.6 12.4 15.8 23.4 37.2 31.1 50.9 35 Lesotho 19.5 73.2 11.2 39.2 1.5 4.6 0.6 1.8 61.0 77.0 55.3 72.9 36 Indonesia 94.2 223.2 28.0 66.4 13.9 32.7 6.5 13.2 36.4 28.3 8.8 21.7 37 Egypt, Arab Rep. 227.0 . 280.0 97.5 133.1 14.7 16.7 9.1 6.3 46.1 37.6 12.6 8.2 38 Zimbabwe 45.4 164.9 14.9 57.0 3.8 27.2 1.5 14.4 2.3 29.1 0.4 21.2 39 Sudan 499.4 3,465.6 77.2 . . 25.5 . . 12.8 . . 34.4 29.0 12.3 11.7 40 Za,nbia 200.7 624.8 90.7 . . 25.3 50.3 8.7 26.1 25.4 35.9 12.2 20.9 Middle-income economies 132.5 w 159.8 w 31.9 w 41.2 w 23.9 w 20.3 w 12.2 w 9.3 w 8.1 w 9.9 w 6.4 w 13.0 w Lower-middle-income 100.2 w 157.6 w 28.7 w 53.2 w 16.7 w 19.5 w 8.0 w 8.3 w 14.8 w 16.5 w 8.6 w 14.9 w 41 Bolivia 258.2 432.4 93.3 85.3 35.0 34.0 21.1 14.7 24.7 42.1 16.6 42.2 42 Côte d'Ivoire 160.7 566.1 58.8 222.6 28.3 43.4 13.0 17.6 7.6 12.5 9.0 15.6 43 Senegal 162.7 224.6 50.5 63.1 28.7 19.9 10.5 7.9 27.9 55.0 17.8 40.1 44 Philippines 212.3 215.6 53.8 70.2 26.6 23.2 18.2 11.0 6.7 25.9 7.5 20.9 45 Papua New Guinea 66.0 160.7 28.9 84.6 13.8 29.6 6.6 8.5 12.2 25.5 21.2 31.6 46 Camemon 136.7 252.7 36.8 57.5 15.2 18.7 8.1 8.8 31.4 23.9 16.8 22.6 47 Guatemala 63.6 142.9 14.9 29.5 7.9 15.3 3.7 6.7 21.6 25.4 30.0 35.2 48 Dominican Rep. 133.8 193.1 31.2 65.7 25.3 11.6 12.0 4.6 20.5 27.5 10.2 19.6 49 Ecuador 201.6 362.7 53.8 114.5 33.9 32.2 15.9 14.7 5.0 8.1 5.4 17.9 50 Morocco 224.5 257.5 53.3 80.0 32.7 27.8 17.0 14.2 37.6 25.9 7.4 23.9 51 Jonlan 79.0 283.4 . . 226.9 8.4 20.9 4.3 11.8 41.5 34.1 8.0 10.8 52 Tajikistan .. .. .. . .. .. .. . . .. .. 53 Peru 194.2 483.6 47.6 44.3 44.5 27.7 19.9 13.4 15.1 11.5 5.5 9.2 54 El Salvador 71.1 155.4 26.2 37.4 7.5 17.2 4.7 5.8 25.9 59.3 28.3 38.7 55 Congo 148.1 386.3 99.0 181.7 10.6 21.3 6.6 6.4 26.4 36.4 7.7 12.1 56 SyrianArabRep. 106.2 . . 27.1 103.8 11.4 . . 4.7 . . 63.5 78.4 8.8 6.2 57 Colombia 117.1 167.7 20.9 43.5 16.0 35.2 11.6 13.8 16.3 5.6 19.5 35.5 58 Paraguay 121.9 125.7 20.7 35.0 18.6 11.9 8.5 5.3 31.9 30.7 20.2 33.2 59 Uzbekistan .. .. .. .. .. .. .. .. .. .. 60 Jamaica 129.3 186.3 78.3 134.9 19.0 29.4 10.8 9.7 20.9 27.4 15.0 26.8 61 Romania 80.3 39.3 . 6.9 12.6 2.0 4.9 1.5 0.0 l.0 8.3 1.4 62 Namibia . . . . . . . . . . . . . . . . . . . . 63 Tunisia 96.0 137.2 41.6 66.2 14.8 22.7 6.9 7.3 39.9 35.3 12.3 32.7 64 Kyrgyzstan . . . . . . . . . . . . . . . . . . . . . . . 65 Thailand 96.8 94.9 26.0 39.0 18.9 13.1 9.5 7.0 10.9 13.2 12.0 9.9 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 284 Total debt service as Interest payments Concessional debt Multilateral debt Total external debt as a percentage of a percentage of as a percentage of as a percentage as a percentage bports of goods exports of goods exports ofgoods of total of total and services GNP and services and services external debt external debt 1980 199) 1980 1991 1980 199! 1980 1991 1980 1991 1980 199! 67 Azerbaijan 68 Turkmenistan 69 Turkey 332.9 194.7 34.3 48.1 28.0 30.5 14.9 12.8 23.0 15.6 I 20.2 70 Poland 54.9 281.4 16.3 68.5 17.9 5.4 5.2 3.3 9.1 2.9 0.0 1.7 71 Bulgaria 2.9 237.9 1.4 151.7 0.3 22.1 0.2 6.1 0.0 0.0 0.0 4.6 72 Costa Rica 225.2 177.8 59.7 74.9 29.1 18.4 14.6 10.3 9.5 22.8 16.4 29.4 73 Algeria 129.9 214.8 47.0 70.4 27.4 73.7 10.4 15.8 6.5 3.4 1.5 9.5 74 Panama 38.4 106.4 87.5 130.1 6.3 3.9 3.3 1.9 9.0 6.7 11.0 14.1 75 Armenia 76 Chile 192.5 153.5 45.5 60.7 43.1 33.9 19.0 24.3 .1 2'i 24.1 77 Iran, Islamic Rep. 32.0 57.3 4.9 11.5 6.8 3.9 3.1 3.0 7.4 0.8 13.8 0.7 78 Moldova 79 Ukraine 80 Mauritius 80.7 53.2 41.6 37.0 9.1 8.8 5.9 3.0 15.6 39.3 16.6 29.7 81 Czechoslovakia 28.0 68.9 9.8 29.5 3.9 11.6 3.9 4.7 0.0 0.0 0.0 4.8 82 Kazakhstan 83 Malaysia 44.6 53.7 28.0 47.6 6.3 8.3 4.0 3.6 10.1 12.1 11.3 9.1 Upper-middle-income 173.1 w 162.2 w 34.4 w 33.2 w 33.0 w 21.1 w 17.5 w 10.4 w 3.5 w 2.9 w 4.8 w 11.1 w 84 Botswana 17.8 23.0 14.8 15.7 1.9 3.4 1.1 1.4 46.6 40.3 63.3 71.4 85 South Africa 86 Lithuania 87 Hungary 1808 448 770 325 132 56 04 00 148 88 Venezuela 132.0 187.0 42.1 65.3 27.2 18.7 13.8 13.9 0.4 0.2 0.7 6.4 89 Argentina 242.4 433.0 48.4 49.2 37.3 48.4 20.8 25.1 1.8 0.9 4.0 8.5 90 Uruguay 104.1 175.3 17.0 45.3 18.8 38.2 10.6 11.7 5.2 1.7 11.0 20.5 91 Brazil 305.2 324.9 31.3 28.8 63.1 30.0 33.8 15.4 2.5 2.5 4.4 9.5 92 Mexico 259.2 224.1 30.5 36.9 49.5 30.9 27.4 17.3 0.9 1.0 5.6 15.2 93 Belams 94 Russian Federation 95 Latvia 96 Trinidad and Tobago 24.6 105.4 14.0 48.2 6.8 16.2 1.6 8.4 4.7 2.4 8.6 6.7 97 Gabon 62.2 152.5 39.2 88.1 17.7 6.5 6.3 4.0 8.2 21.4 2.6 9.1 98 Estonia 99 Portugal 99.5 103.2 40.5 43.9 18.3 21.1 10.5 7.1 4.4 3.3 5.5 9.8 100 Oman 15.4 11.2 29.4 6.4 1.8 . . 43.6 10.3 5.8 4.9 101 Puerto Rico 102 Korea,Rep. 130.6 47.6 48.7 14.4 19.7 7.1 12.7 3.1 Il. .o 103 Greece 104 Saudi Arabia 105 Yugoslavia 103.1 87.8 25.6 20.8 20.4 7.2 6.9 7.9 4.3 7.6 16.9 Low- and middle-income 124.9 w 176.9 w 26.2 w 42.3 w 20.0 w 20.5 w 10.2 w 9.3 w 16.9 w 19.3 w 8.6 w 16.7 w Sub-Saharan Africa 96.6 w 329.4 w 28.6 w 107.9 w 10.9 w 20.8 w 5.7 w 10.5 w 26.9 w 33.8 w 13.4 w 22.7 w East Asia & Pacific 89.8 w 96.2 w 16.9 w 28.2 w 13.5 w 13.3 w 7.7 w 5.9 w 16.4 w 21.2 w 8.7 w 15.4 w South Asia 160.4w 287.1 w 17.0 w 35.6w 11.9w 26.0 w 5.1 w 11.5 w 74,4w 52.4w 25.0 w 36.7w Europe and Central Asia 81.2 w 152.2 w 24.3 w 51.4w 14.0w 20.0w 6.0 w 7.7 w 9.6w 5.8w 6.1 w 10.8 w Middle East & N. Africa 114.4w 185.8w 31.0w 58.8w 16.1 w 25.9 w 7.3 w 8.4w 31.8w 31.4w 8.3w 11.7w Latin America & Caribbean 195.5 w 256.0 w 35.1 w 41.3 w 37.1 w 29.2 w 19.6 w 15.8 w 4.5 w 5.4 w 5.8 w 14.2 w Severely indebted 176.6 w 285.9 w 34.0 w 46.4 w 34.0 w 30.8 w 17.1 w 14.1 w 6.7 w 8.3 w 5.0 w 10.9 w High-income economies OECD members 106 Ireland 107 flsrael 108 New Zealand 109 Spain 110 tHong Kong III tSingapore 112 United Kingdom 113 Australia 114 Italy 115 Netherlands 116 Belgium 117 Austria 118 France 119 Canada 120 United States 121 Germany 122 Denmark 123 Finland 124 Norway 125 Sweden 126 Japan 127 Switzerland World Fuel exporters 285 Table 25. Terms of external public borrowing Average Average Public loans with variable Commitments interest rate Average maturirv grace period interest rates, as a (millions of dollars) (percent) (rears) (years) percentage of public debt 1980 1991 1980 199) /980 1991 1980 /991 1980 1991 Low-income economies 30,186 1 37,643 1 6.3 w 5.3 w 23 w 22 w 6w 6w 16.6 w 22.0 w China and India 8,728 t 17,985t 7.6 w 6.1 w 24 w 18 w 6w 5w 14.8 w 26.3 w Other low-income 21,457 1 19,657 1 5.8 w 4.6 w 23 w 25 w 6w 7w 17.0 w 20.2 w I Mozambique 479 80 5.2 1.4 IS 34 4 9 0.0 3.7 2 Tanzania 710 246 4.1 1.9 23 42 8 12 4.4 5.5 3 Ethiopia 175 100 3.7 3.5 18 20 4 7 1.5 2.9 4 Uganda 209 437 4.6 2.2 25 37 6 9 1.3 1.3 5 Bhutan 7 7 1.0 1.0 50 39 10 10 0.0 0.0 6 Guinea-Bissau 38 16 2.4 0.7 18 39 4 10 1.6 0.2 7 Nepal 92 68 0.8 1.0 46 38 10 10 0.0 0.0 8 Burundi 102 102 1.3 1.2 42 40 9 10 0.0 0.0 9 Chad 0 73 0.0 0.7 0 50 0 10 0.2 0.0 10 Madagascar 445 53 5.6 0.7 II 40 5 10 8.3 6.1 II SierraLeone 70 46 5.2 0.7 26 48 7 10 0.0 1.1 12 Bangladesh 1,034 952 1.7 1.0 36 38 9 10 0.1 0.0 13 Lao PDR 94 147 0.2 0.8 40 40 31 10 0.0 0.0 14 Malawi 130 187 6.0 0.7 24 41 6 10 23.2 3.4 IS Rwanda 48 181 1.5 0.8 39 42 9 II 0.0 0.0 16 Mali 145 152 2.2 0.8 23 42 5 10 0.0 0.3 17 BurkinaFaso 115 167 4.3 1.0 21 37 6 10 4.3 0.0 18 Niger 341 48 7.4 2.7 18 30 5 10 56.4 16.5 19 India 4,902 8,538 5.4 5.8 34 20 7 6 4.2 21.0 20 Kenya 560 476 3.9 2.4 30 35 8 9 27.0 20.2 21 Nigeria 1,904 1,362 10.5 6.1 II 22 4 6 74.4 31.8 22 China 3,826 9,447 10.4 6.4 II 16 3 4 58.8 33.1 23 Haiti 51 52 5.5 1.3 20 39 6 10 3.1 0.6 24 Benin 448 145 8.3 1.1 12 36 4 9 0.4 2.1 25 Central African Rep. 38 118 0.6 2.9 13 32 4 8 1.9 0.1 26 Ghana 170 333 1.4 2.6 44 33 10 8 0.9 2.6 27 Pakistan 1,115 1,558 4.4 6.6 30 19 7 5 1.5 14.4 28 Togo 97 14 4.0 0.8 24 40 7 10 12.0 3.3 29 Guinea 269 100 4.6 2.6 19 24 6 5 0.3 7.2 30 Nicaragua 424 244 3.9 1.8 25 39 7 10 47.9 25.9 31 Sri Lanka 752 948 3.9 2.2 30 34 8 9 6.9 5.6 32 Mauritania 215 0 3.6 0.0 21 0 8 0 2.4 5.6 33 Yemen, Rep. 553 128 2.7 0.9 27 39 6 10 0.0 1.5 34 Honduras 495 254 6.8 3.1 24 27 7 8 34.3 21.7 35 Lesotho 59 176 5.9 5.6 24 25 6 7 3.5 0.0 36 Indonesia 4,277 7,840 8.1 6.1 19 20 6 6 30.7 43.1 37 Egypt, Arab Rep. 2,558 1,375 5.0 5.2 28 23 9 7 4.5 11.5 38 Zimbabwe 171 953 7.1 6.0 15 14 6 4 0.4 26.8 39 Sudan 905 16 6.1 0.7 18 40 5 10 10.6 19.6 40 Zambia 645 324 6.7 1.1 19 39 4 10 12.6 12.0 Middle-income economies 67,406 1 65,111 I 10.7 w 7.2 w 12 w 14 w 4w 4w 54.9 w 50.6 w Lower-middle-income 33,634, 34,445 I 9.5 w 6.8 w 14 w 15 w 5w 5w 38.8 w 45.3 w 41 Bolivia 370 408 8.4 5.1 15 25 5 6 31.6 24.2 42 Côted'Ivoire 1,685 362 11.4 6.0 10 23 4 6 42.4 65.7 43 Senegal 470 295 5.9 3.7 20 29 6 7 12.7 4.4 44 Philippines 2,143 2,751 9.9 4.9 17 24 5 7 49.9 41.7 45 Papua New Guinea 184 26! 11.2 4.5 18 24 5 7 43.5 52.7 46 Cameroon 164 353 6.9 6.2 24 18 6 5 22.9 18.6 47 Guatemala 247 35 7.9 2.3 15 32 4 3 35.6 16.8 48 Dominican Rep. 519 166 8.9 4.7 12 20 4 7 47.2 31.5 49 Ecuador 1,148 547 10.7 7.6 14 16 4 4 62.5 61.0 50 Morocco 1,686 1,834 8.0 6.5 15 19 5 6 31.0 52.5 51 Jordan 768 505 7.3 1.7 15 29 4 9 13.4 28.2 52 Tajikistan . . . . . . . . . . . . . . . . . . 53 Peru 1,614 328 9.4 3.1 12 30 3 10 31.2 27.8 54 El Salvador 225 379 4.2 7.4 28 17 8 5 27.4 14.1 55 Congo 966 0 7.7 0.0 II 0 3 0 6.6 27.3 56 SyrianArabRep. 1,168 401 1.3 6.0 24 27 5 5 0.0 0.0 57 Colombia 1,566 2,323 12.9 7.3 15 12 4 6 40.8 50.7 58 Paraguay 99 0 7.0 0.0 24 0 7 0 27.3 16.8 59 Uzbekistan .. .. .. .. .. .. .. 60 Jamaica 225 444 7.6 7.2 14 19 5 4 23.0 25.7 61 Romania 1,886 1,435 14.1 6.2 8 10 4 5 59.2 5.0 62 Namibia . . . . . . . . . . . . . . . . . . 63 Tunisia 777 1,366 6.7 6.7 18 16 5 5 20.0 23.4 64 Kyrgyzstan . . . . . . . . . . . . . . . . . . 65 Thailand 1,877 1,156 9.5 5.0 17 19 5 5 51.4 56.6 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 286 Average Average Public loans with variable Commitments interest rate Average ntoJuriiy grace period interest rates, as a (millions of dollars) (percent) (years) (years) percentage ofpublic debt /980 1991 1980 1991 1980 /991 1980 1991 1980 /991 67 Azetbaijan . . . 68 Thrkmenistan . . . . . . . . . . . . . . . . . . 69 Turkey 2.925 3.880 8.3 7.9 16 13 5 6 26.5 35.5 70 Poland 1,715 1,637 9.3 7.9 II 14 4 4 37.8 67.7 7! Bulgaria 1.578 269 13.9 7.7 9 17 6 5 76.7 73.0 72 CostaRica 62! 179 11.2 7.1 13 6 5 3 57.0 32.) 73 Algeria 3,538 8,429 8.1 7.4 12 8 4 2 25.0 41.6 74 Panama 534 0 11.3 0.0 II 0 5 0 52.7 61.3 75 Armenia . . . . . . . . . . . . . 0 76 Chile 835 1.223 13.9 7.0 8 IS 4 4 75.6 76.6 77 Iran, Islamic Rep. 0 759 0.0 6.1 0 21 0 2 37.8 84.7 78 Moldova . . . . . . . 79 Ukraine . . . . . . . . . . . . . . . . . 80 Mauritius 121 107 10.4 6.) 14 17 4 7 47.0 32.8 81 Czechoslovakia 8 1.423 8.2 8.3 12 8 4 4 0.0 33.3 82 Kazakhstan .. .. .. .. .. .. .. 83 Malaysia 1.423 868 11.2 7.4 14 13 5 7 50.7 52.2 Upper-middle-income 33,772/ 30,666/ 11.9 w 7.7 w II w 14 w 4w 4w 67.0 w 56.3 w 84 Botswana 69 28 6.0 0.5 18 48 4 10 0.0 14.5 85 South Afnca 86 Lithuania . . . . . . . . . . . . . . . . . 87 Hungarya 1,225 2,627 9.8 8.5 13 9 3 6 39.8 56.4 88 Venezuela 2,769 1,055 12.1 8.4 8 IS 3 5 81.4 62.7 89 Argentina 3,023 2.374 13.8 8.0 9 20 4 5 74.0 58.3 90 Uruguay 347 447 10.1 7.9 14 14 6 3 35.4 60.! 9! Brazil 9,638 3,975 12.5 7.6 10 II 4 4 72.2 71.8 92 Mexico 7,632 7,279 11.3 8.2 10 13 4 4 75.9 45.9 93 Belarus . 94 Russian Federation . . . . . . . . . . . . . 95 Latvia . . . . . . . . . . . . . . . . . 96 TrinidadandTobago 21) 283 10.4 7.7 9 20 4 5 31.9 51.7 97 Gabon 196 169 11.2 5.4 II 16 3 5 39.3 10.2 98 Estonia . . . . . . . . . . . 99 Portugal 2,015 6,531 10.9 7.0 10 19 3 2 30.6 27.0 100 Oman 454 362 7.9 5.8 9 7 3 5 0.0 59.7 101 Puerto Rico . . . . . . . . . . . . . . . . . 102 Korea, Rep. 4,928 4.910 11.3 7.7 15 II 4 4 36.4 41.8 103 Greece . . . . . . . 104 SaudiArabia 105 Yugoslavia 1,187 595 15.1 8.0 ' 14 ' 77.6 75.! Low- and middle-income 97,592 / 102,753 / 9.4 w 6.5 w 16 w 17w Sw 5w 45.0 w 40.7 w Sub-Saharan Africa 13,271 / 7,548 1 7.1 w 3.8 w 17w 28 w Sw 7w 23.7 w 21.2 w East Asia & Pacific 19,468 / 27,414 1 9.8 w 6.3 w 16w 17w 5w 5w 40.0 w 41.6 w South Asia 7,925 / 12,074 1 4.6 w 5.2 w 33 w 22 w 7w 7w 3.) w 16.1 w Europe and Central Asia 12,542 1 18,397 1 11.2 w 7.6 w 11 w 14 w 4w 4w 47,4 w 52.4 w Middle East & N. Africa 11,616/ 15,229 1 6.3 w 6.7 w 19w 14 w Sw 4w 18.2 w 27.0 w Latin America & Caribbean 32,770 1 22,092 1 11.6 w 7.5 w 11 w 15 w 4w 5w 68.0 w 55.1 w Severely indebted 37,501 1 28,948 / 10.7 w 7.4 w 11 w 13 w 4w 4w 59.4w 53.7w High-income economies OECD members 106 Ireland 107 (Israel 108 New Zealand 109 Spain 110 tHong Kong Ill tSingapore 112 United Kingdom 113 Australia 114 Italy 115 Netherlands 116 Belgium Ill Austria 1)8 France 119 Canada 120 United States 121 Germany 122 Denmark 123 Finland 124 Norway 125 Sweden 126 Japan 127 Switzerland World Fuel exporters a. Includes debt in convertible currencies only. 287 Table 26. Population growth and projections Hothetical size Age structure ofpopulation (percent) Average annual growth ofpopulation of stationary 1970-80 (percent) 1980-9! I99._2(x)oa ii Population (millions) 20O0 3,686! 2025a 5,1841 population (millions) 1991 35.4w 0-14 sears - 2025a 26.8w 15-64 sears 1991 2025a Low-income economies 2.2 w 2.0w 1.8 w 3,1271 60.6 w 65.6 w China and India 2.0 w 1.7 w 1.5 w 2,0161 2,307 t 2,9341 31.1 w 22.4 w 63.6 w 67.6 w Other low-income 2.6w 2.6w 2.4w 1,111 t 1,379 t 2,2501 43.1 w 32.4 w 55.2 w 63.0 w I Mozambique 2.6 2.6 2.9 16 21 43 113 44.5 42.2 52.1 55.0 2 Tanzania 3.0 3.0 3.0 25 33 59 116 46.7 36.4 50.5 60.9 3 Ethiopia 2.7 3.1 2.7 53 67 130 303 45.9 41.2 51.1 56.3 4 Uganda 2.7 2.5 3.3 17 23 48 137 48.7 45.1 48.6 53.4 5 Bhutan 1.8 2.1 2.4 I 2 3 6 40.6 34.3 55.8 61.7 6 Guinea-Bissau 4.7 1.9 2.0 I I 2 4 43.2 38.6 53.5 59.2 7 Nepal 2.6 2.6 2.5 19 24 38 65 43.4 30.8 53.7 64.7 8 Bunindi 1.6 2.9 2.9 6 7 14 33 45.6 40.8 50.7 56.6 9 Chad 2.1 2.4 2.6 6 7 14 29 42.2 39.0 53.4 57.4 10 Madagascar 2.6 3.0 2.8 12 15 26 49 45.1 33.9 52.5 62.5 II Sierra Leone 2.1 2.4 2.6 4 5 10 23 43.5 40.5 52.6 56.4 12 Bangladesh 2.7 2.2 1.9 III 131 180 268 42.3 26.2 56.8 69.0 13 Lao PDR 1.6 2.7 2.9 4 6 10 20 44.5 37.1 51.7 59.3 14 Malawi 3.1 3.3 3.1 9 12 24 73 46.9 45.0 49.7 52.8 IS Rwanda 3.4 3.0 2.3 7 9 17 37 48.9 40.8 50.6 57.2 16 Mali 2.1 2.6 3.1 9 II 24 59 46.8 41.0 49.8 56.5 17 BurkinaFaso 2.1 2.6 3.0 9 12 23 50 45.7 38.6 50.9 58.8 18 Niger 2.9 3.3 3.5 8 II 24 76 47.8 45.6 49.1 52.2 19 India 2.3 2.1 1.8 866 1,017 1,365 1,886 35.8 23.9 60.2 68.1 20 Kenya 3.8 3.8 3.5 25 34 73 192 48.8 41.0 49.4 56.3 21 Nigeria 2.9 3.0 2.8 99 128 217 382 46.5 31.9 52.3 63.9 22 China 1.8 1.5 1.3 1,150 1,290 1,569 1,890 27.0 21.2 66.4 67.1 23 Haiti 1.7 1.9 1.7 7 8 10 17 39.8 29.1 56.8 65.8 24 Benin 2.7 3.2 2.9 5 6 II 20 47.5 34.9 50.8 62.0 25 CentralAfricanRep. 2.2 2.7 2.5 3 4 7 18 42.3 41.4 53.5 56.3 26 Ghana 2.2 3.2 3.2 15 20 36 69 46.8 34.6 50.5 62.0 27 Pakistan 3.1 3.1 2.8 116 148 244 402 44.0 29.6 53.5 65.5 28 Togo 2.6 3.4 3.1 4 5 9 19 45.4 36.6 51.1 60.0 29 Guinea 1.4 2.6 2.9 6 8 14 33 46.7 40.6 50.7 56.8 30 Nicaragua 3.1 2.7 3.1 4 5 8 14 47.7 28.7 50.6 66.4 31 Sri Lanka 1.7 1.4 1.1 17 19 24 29 31.7 21.3 64.2 66.1 32 Mauritania 2.4 2.4 2.9 2 3 5 14 44.8 42.6 50.8 54.9 33 Yemen, Rep. 2.7 3.8 3.7 13 17 37 86 49.3 40.2 49.7 57.5 34 Honduras 3.4 3.3 2.9 5 7 II 18 39.8 28.1 56.8 66.9 35 Lesotho 2.3 2.8 2.4 2 2 3 6 41.9 29.0 55.0 65.7 36 Indonesia 2.4 1.8 1.4 181 206 265 354 35.8 23.6 60.2 68.4 37 Egypt, Arab Rep. 2.1 2.5 2.1 54 65 92 134 39.1 25.5 57.5 66.8 38 Zimbabwe 2.9 3.4 2.3 10 12 18 28 44.5 28.5 53.7 67.5 39 Sudan 3.0 2.7 3.0 26 34 60 117 45.2 35.5 52.4 60.9 40 Zambia 3.0 3.6 3.0 8 II 21 49 48.3 41.6 49.7 56.6 Middle-income economies 2.2 w 1.8w 1.5 w 1,401 t 1,561 1 2,1401 35.4 w 25.3 w 60.1 w 65.0 w Lower-middle-income 2.2 w 2.0 w 1.8 w 774t 894 1 1,302 1 36.8 w 27.0 w 58.7 w 64.8 w 41 Bolivia 2.6 2.5 2.4 7 9 14 22 41.2 28.1 55.9 66.3 42 COte d'Ivoire 4.1 3.8 3.3 12 17 32 67 48.2 38.2 49.7 59.0 43 Senegal 2.9 3.0 2.8 8 tO 18 38 45.5 38.1 51.6 59.3 44 Philippines 2.5 2.4 1.9 63 74 102 140 39.2 24.0 59.1 68.5 45 Papua New Guinea 2.4 2.3 2.3 4 5 7 12 40.3 29.0 57.2 66.5 46 Cameroon 3.0 2.8 3.1 12 16 29 56 44.7 35.7 51.6 61.0 47 Guatemala 2.8 2.9 2.9 9 12 21 36 45.2 30.3 52.3 65.0 48 Dominican Rep. 2.6 2.2 1.6 7 8 II 14 37.4 23.0 60.6 67.7 49 Ecuador 3.0 2.6 2.1 II 13 18 25 38.9 23.9 58.9 68.5 50 Morocco 2.4 2.6 2.2 26 31 45 65 40.7 24.6 57.5 68.7 51 Jordan 3.7 4.7 4.0 4 5 9 14 43.6 27.4 56.7 67.3 52 Tajikistan . . 3.0 3.1 5 7 13 21 44.9 30.3 50.2 64.2 53 Peru 2.8 2.2 1.9 22 26 36 48 37.1 23.4 60.8 68.4 54 ElSalvador 2.4 1.4 2.0 5 6 9 14 43.0 25.2 55.2 69.1 55 Congo 3.0 3.4 3.4 2 3 6 IS 45.5 40.8 49.5 56.9 56 Syrian Arab Rep. 3.3 3.3 3.4 13 17 34 69 48.1 35.5 49.3 61.0 57 Colombia 2.2 2.0 1.5 33 38 50 63 34.8 22.2 62.9 67.8 58 Paraguay 3.0 3.1 2.6 4 6 9 13 40.3 27.1 56.7 66.2 59 Uzbekistan . . 2.4 2.4 21 26 42 65 41.6 27.5 53.9 65.5 60 Jamaica 1.3 1.0 0.5 2 2 3 4 33.0 21.7 61.8 67.3 61 Romania 0.9 0.4 0.2 23 23 25 26 23.1 18.5 66.8 64.3 62 Namibia 2.8 3.1 3.2 I 2 3 7 44.8 34.2 51.5 61.9 63 Tunisia 2.2 2.4 1.9 8 10 13 18 37.0 23.3 60.2 68.5 64 Kyrgyzstan . . 1.9 1.4 4 5 7 II 38.2 25.8 55.6 66.3 65 Thailand 2.7 1.9 1.4 57 65 82 104 32.4 21.8 65.9 68.5 66 Georgia 0.7 0.4 5 6 6 7 23.9 19.6 64.5 62.1 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 288 Hypothetical size Age structure of population (percent) Average annual growth of population of stationary (percent) Population (millions) population 0-14 years 15-64 years 1970-80 1980-91 1991_2(xxJa 1991 2000a 2025a (millions) 1991 2025a 1991 2025a 67 Azethaijan 1.4 1.4 7 8 II 13 33.! 22.6 60.1 65.3 68 Turkmenistan . . 2.5 2.5 4 5 8 13 41.3 29.5 54.2 64.1 69 Turkey 2.3 2.3 1.9 57 68 91 121 35.2 23.! 61.8 68.0 70 Poland 0.9 0.7 0.3 38 39 43 49 24.8 19.9 65.7 62.9 71 Bulgaria 0.4 0.! -0.2 9 9 8 9 20.1 17.5 66.8 62.8 72 Costa Rica 2.8 2.7 2.0 3 4 5 6 48.2 22.5 49.7 66.2 73 Algeria 3.1 3.0 2.7 26 33 53 82 43.1 27.0 53.5 67.4 74 Panama 2.5 2.1 1.7 2 3 4 5 34.6 22.4 61.9 67.4 75 Armenia . . 0.9 1.5 3 4 5 6 30.! 21.8 62.3 64.1 76 Chile 1.6 1.7 1.3 13 15 19 23 30.6 21.8 63.3 65.9 77 Iran, Islamic Rep. 3.3 3.6 3.4 58 78 160 354 45.8 37.3 51.0 58.6 78 Moldova 0.9 0.! 4 4 5 7 31.6 23.0 60.0 63.4 79 Ukraine . . 0.4 0.0 52 52 52 55 21.2 18.4 65.6 62.2 80 Mauntius 1.5 1.0 1.! 1 1 1 2 29.0 20.4 67.5 66.7 81 Czechoslovakia 0.7 0.3 . . 16 . . . . . . 21.2 . . 67.5 82 Kazakhstan . . 1.2 0.7 17 18 22 28 31.6 22.3 62.4 64.6 83 Malaysia 2.4 2.6 2.2 18 22 3! 43 38.6 23.9 58.5 67.6 Upper-middle-income 2.2 w 1.5 w 1.1 w 627 1 667 t 839 1 33.5 w 22.6 w 62.1 w 65.5 w 84 Botswana 3.8 3.5 2.8 I 2 3 4 45.7 26.9 52.2 68.0 85 South Africa 2.8 2.5 2.2 39 47 69 103 38.6 25.6 58.3 67.4 86 Lithuania . . 0.8 0.2 4 4 4 5 22.5 19.6 65.6 62.5 87 Hungary 0.4 -0.2 -0.4 10 10 10 10 20.0 17.8 67.7 61.8 88 Venezuela 3.5 2.6 1.9 20 23 32 41 36.7 23.0 61.3 67.5 89 Argentina 1.7 1.3 1.0 33 36 43 53 29.4 21.7 62.2 65.2 90 Uruguay 0.4 0.6 0.6 3 3 4 4 25.4 20.2 63.3 63.8 9! Brazil 2.4 2.0 1.4 15! 172 224 285 34.2 22.6 62.8 67.5 92 Mexico 2.9 2.0 1.9 83 99 136 182 37.6 23.3 60.0 68.3 93 Belams . . 0.6 0.2 10 II II 12 22.9 19.0 64.8 62.1 94 Russian Federation 0.6 0.0 149 149 153 162 23.2 18.7 65.6 61.9 95 Latvia . . 0.3 -0.! 3 3 3 3 21.7 19.6 65.1 61.7 96 TrinidadandTobago 1.! 1.3 0.9 1 I 2 2 34.2 22.6 60.9 66.4 97 Gabon 4.7 3.5 2.9 I 2 3 7 39.7 39.8 53.4 56.2 98 Estonia 0.8 0.6 0.0 2 2 2 2 22.3 18.5 65.8 62.7 99 Portugal 1.3 0.! 0.0 10 10 10 9 20.2 16.0 67.9 65.0 100 Oman 4.2 4.3 3.9 2 2 5 10 46.6 37.1 51.1 58.3 101 Puerto Rico 1.7 0.9 0.9 4 4 5 5 27.5 19.7 64.6 64.0 102 Korea, Rep. 1.8 1.! 0.8 43 47 53 56 25.1 18.2 71.0 66.6 103 Greece 1.0 0.5 0.! 10 10 10 9 18.6 15.1 67.2 61.4 104 Saudi Arabia 5.0 4.6 3.5 15 2! 41 82 43.0 34.8 54.7 59.6 105 Yugoslavia 0.9 0.6 24 22.5 67.7 Low- and middle-income 2.2 w 2.0 w 1.7 w 4,528 t 5,247 1 7,325 1 35.4 w 26.3 w 60.5 w 65.4 w Sub-Saharan Africa 2,8 w 3.1 w 3.0 w 489 t 635 1 1,192 1 48.5 w 38.0 w 53.1 w 59.0 w East Asia & Pacific 1.9 w 1.6 w 1.4 w 1,667 1 1,891 1 2,3671 29.5 w 22.0 w 65.1 w 67.5 w South Asia 2.4 w 2.2 w 1.9 w 1,152 t 1,3681 1,908 1 38.0 w 25.4 w 57.7 w 67.4 w Europe and Central Asia 0.9 w 0.6 w 492 t 475 1 546 1 26.! w 20.9 w 64.8 w 63.9 w Middle East & N. Africa 2.9 w 3.2 w 2.9 w 244 t 315 1 552 i 42.9 w 32.1 w 53.4 w 62.7 w Latin America & Caribbean 2.4 w 2.0 w 1.6 w 445 I 515 1 691 1 36.0 w 23.3 w 60.8 w 67.3 w Severely indebted 2.3 w 2.1 w 1.8 w 486: 569 I 800 1 36.7 w 25.5 w 59.9 w 66.0 w High-income economies 0.8 w 0.6 w 0.5 w 822 1 864: 922 i 19.7 w 17.2 w 67.1 w 61.2 w OECI) niembers 0.7 w 0.6 w 0.5 w 783 1 820 1 869 1 19.3 w 17.2 w 67.0 w 61.2 w 106 Ireland 1.5 0.2 0.3 4 4 4 5 26.0 19.1 65.0 65.2 107 (Israel 2.7 2.2 2.8 5 6 8 tO 30.9 21.0 62.9 65.6 108 NewZealand 1.! 0.7 0.8 3 4 4 5 22.8 19.4 65.8 63.! 109 Spain 1.0 0.4 0.! 39 39 39 33 19.3 14.5 68.6 63.3 110 tHongKong 2.4 1.2 0.8 6 6 7 6 20.6 15.4 70.3 61.4 III (Singapore 2.0 1.7 1.5 3 3 4 4 22.9 18.3 70.7 63.5 112 United Kingdom 0. I 0.2 0.2 58 58 60 61 19.0 17.8 64.7 62.0 113 Australia 1.5 1.5 1.3 17 19 23 25 22.1 18.8 67.0 63.5 114 Italy 0.5 0.2 0.! 58 58 55 45 16.2 14.0 68.6 60.9 115 Netherlands 0.8 0.6 0.8 15 16 16 15 18.3 16.4 67.3 58.5 116 Belgium 0.2 0.1 0.2 10 10 10 10 17.7 16.7 66.0 60.6 117 Austria 0.! 0.2 0.4 8 8 8 7 17.3 15.7 66.8 61.5 118 France 0.6 0.5 0.4 57 59 63 63 19.9 17.3 65.6 60.9 119 Canada 1.2 1.2 0.8 27 29 34 35 20.9 18.0 67.1 61.2 120 UnitedStates 1.0 0.9 0.9 253 274 319 348 21.5 19.1 65.6 61.8 121 Germany 0.0 0.! 0.0 80 80 75 65 16.2 15.1 67.9 60.5 122 Denmark 0.4 0.! 0.2 5 5 5 5 17.0 16.2 67.7 60.8 123 Finland 0.4 0.4 0.3 5 5 5 5 19.2 17.2 66.7 59.2 124 Norway 0.5 0.4 0.4 4 4 5 5 18.9 17.7 64.5 61.5 125 Sweden 0.3 0.3 0.6 9 9 10 10 18.! 18.1 63.8 59.8 126 Japan 1.2 0.5 0.3 124 127 126 III 18.1 15.! 69.6 59.2 127 Switzerland 0.0 0.6 0.7 7 7 7 7 16.9 16.0 67.5 59.0 World 1.9 w 1.7 w 1.6 w 5,350 I 6,111 1 8,2471 32.8 w 25.3 w 61.5 w 65.0 w Fuel exporters 3.2 w 3.3 w 3.0 w 263 1 342 1 613 t 48.6 w 33.2 w 57.0 w 62.1 w a. For the assumptions used in the projections, see the technical notes. 289 Table 27. Demography and fertility - Women of Assumed year Married women of Crude birth rate Crude death rate childbearing age of reaching childbearing age (per /000 (per 1,0(k) as a percentage of net using contraception population) population) all women Totalfertilsy rate reproduction (percent)b 1970 1991 1970 199/ 1965 1991 1970 199/ 2tXjOa rate of I /989 Low-income economies 39 w 30 w 14 w lOw 46 w 51 w 6.0 w 3.8 w 3.2 w China and India 37 w 26 w 12 w 9w 46 w 53 w 5.8 w 3.1 w 2.5 w Other low-income 45 w 38 w 19 w 13w 45 w 47 w 6.3 w 5.2 w 4.5 w I Mozambique 48 45 24 19 47 45 6.7 6.5 6.7 2050 2 Tanzania 49 46 22 18 45 45 6.4 6.3 6.6 2035 3 Ethiopia 43 50 20 21 46 44 5.8 7.5 7.3 2045 4 Uganda 50 52 17 19 44 43 7.1 7.3 6.6 2050 II 5 Bhutan 41 39 22 17 48 47 5.9 5.9 5.4 2035 6 Guinea-Bissau 41 45 27 25 50 47 5.9 6.0 6.0 2040 7 Nepal 46 38 22 13 50 46 6.4 5.5 4.6 2030 8 Burundi 46 46 24 17 44 45 6.8 6.8 6.6 2045 9 9 Chad 45 44 26 18 47 46 6.0 5.9 6.1 2040 10 Madagascar 46 43 20 14 47 45 6.6 6.2 5.2 2035 11 Sierra Leone 49 48 30 22 47 45 6.5 6.5 6.5 2045 12 Bangladesh 48 34 21 13 44 48 7.0 4.4 3.3 2015 31 13 Lao PDR 44 44 23 16 47 45 6.1 6.7 6.0 2040 14 Malawi 56 53 24 21 46 45 7.8 7.6 7.4 2055 15 Rwanda 52 42 18 17 45 43 7.8 6.4 7.6 2040 16 Mali 51 50 26 19 46 45 6.5 7.0 7.0 2050 5 17 Burkina Faso 48 47 25 18 47 45 6.4 6.5 6.3 2045 18 Niger 50 52 28 19 45 44 7.2 7.4 7.3 2055 19 India 41 30 18 10 48 50 5.8 3.9 3.0 2015 45 20 Kenya 53 45 18 II 41 43 8.0 6.5 5.5 2050 27 21 Nigeria 51 44 21 14 45 45 6.9 5.9 5.0 2035 6 22 China 33 22 8 7 45 56 5.8 2.4 2.1 1995 72 23 Haiti 39 35 19 13 45 48 5.9 4.7 4.2 2025 II 24 Benin 50 45 22 15 44 44 6.9 6.3 5.2 2035 25 Central African Rep. 37 42 22 17 47 46 4.9 5.8 5.3 2045 26 Ghana 46 45 16 13 45 44 6.7 6.2 5.1 2035 13 27 Pakistan 48 41 19 II 43 46 7.0 5.7 4.6 2030 12 28 Togo 50 48 20 14 46 45 6.5 6.6 5.5 2040 33 29 Guinea 52 49 27 21 45 44 6.0 6.5 6.5 2045 30 Nicaragua 48 40 14 7 43 46 6.9 5.1 4.2 2025 31 Sri Lanka 29 21 8 6 47 54 4.3 2.5 2.1 2000 62 32 Mauritania 47 49 25 19 47 44 6.5 6.8 6.8 2050 33 Yemen, Rep. 53 52 23 14 47 44 7.8 7.5 7.5 2045 34 Honduras 49 38 15 7 44 46 7.2 5.0 4.1 2025 4' 35 Lesotho 43 35 20 II 47 46 5.7 5.1 4.5 2025 36 Indonesia 42 25 18 9 47 52 5.5 3.0 2.4 2005 50 37 Egypt, Arab Rep. 40 32 17 9 43 48 5.9 4.2 3.1 2020 38 38 Zimbabwe 53 36 16 8 42 47 7.7 4.7 3.4 2020 43 39 Sudan 47 44 22 15 46 45 6.7 6.3 5.4 2035 9 40 Zambia 49 47 19 15 46 45 6.7 6.5 6.1 2045 Middle-income economies 35w 25w 11w 8w 45w 49w 5.0w 3.2w 3.1w Lower-middle-income 36 w 28 w 12 w 8w 45 w 49 w 5.3 w 3.6 w 3.4 w 41 Bolivia 46 36 19 10 46 48 6.5 4.8 3.7 2025 30 42 Cóted'Ivoire 51 46 20 14 44 42 7.4 6.6 5.8 2040 43 Senegal 47 43 22 16 45 45 6.5 6.1 6.3 2040 44 Philippines 38 28 11 7 44 50 6.4 3.6 2.7 2010 45 Papua New Guinea 42 34 18 II 47 49 6.1 4.9 4.0 2025 46 Camemon 43 42 18 12 47 44 5.8 5.8 5.3 2035 /6 47 Guatemala 45 39 14 8 44 45 6.5 5.4 4.3 2030 23 48 Dominican Rep. 41 27 II 6 43 51 6.3 3.! 2.4 2005 56 49 Ecuador 43 30 12 7 43 50 6.3 3.7 2.8 2010 53 50 Morocco 47 32 16 8 45 49 7.0 4.3 3.4 2015 36 51 Jordan 37 5 45 46 . - 5.3 5.6 2025 35 52 Tajikistan . . 39 . . 6 . . 44 5.9 5.3 . . 2030 53 Peru 41 27 14 8 44 51 6.0 3.4 2.8 2010 54 ElSalvador 44 34 12 7 44 47 6.3 4.1 3.2 2015 47 55 Congo 43 49 16 16 45 43 5.9 6.6 6.3 2045 56 Syrian Arab Rep. 47 44 13 6 . . 43 7.7 6.3 5.4 2040 57 Colombia 36 24 9 6 43 54 5.3 2.7 2.2 2000 66 58 Paraguay 38 33 7 6 41 49 6.0 4.4 4.0 2025 48 59 Uzbekistan . . 35 . . 6 . . 47 5.7 4.3 . . 2025 60 Jamaica 34 24 8 7 42 52 5.3 2.7 2.1 2000 55 61 Romania 21 14 10 II 50 48 2.9 1.9 2.1 2030 62 Namibia 44 43 18 II 46 45 6.0 5.7 4.8 2035 63 Tunisia 39 27 14 6 43 50 6.4 3.5 2.7 2010 50 64 Kyrgyzstan . . 29 . . 8 . . 46 4.9 3.9 . . 2020 65 Thailand 39 21 9 6 44 55 5.5 2.3 2.1 1995 66 66 Georgia 15 8 48 2.6 2.1 . . 1995 Note: For data comparability, see the technical notes. Figures in italica are for yeara other than those specified. 290 Women of Assumed year Married women of Crude birth rate Crude death rate childbearing age of reaching childbearing age (per 1000 (per 1000 as a percentage of net using contraception population) population) all women Totalferttlzt' rate (percent)b reproduction 1970 1991 1970 199/ 1965 199/ 1970 /991 2000a rate oil J989 67 Azerbaijan 27 . 6 . . 50 4.7 2.8 . 2010 68 Turkmenistan . 34 . . 7 . . 47 6.0 4.5 . . 2030 69 Thrkey 36 28 12 7 45 50 4.9 3.4 2.7 2010 63 70 Poland 17 14 8 II 47 48 2.2 2.! 2.1 2030 71 Bulgaria 16 II 9 12 51 47 2.2 1.8 1.9 2030 72 Costa Rica 33 27 7 4 42 51 4.9 3.2 2.3 2005 73 Algeria 49 34 16 7 44 46 7.4 5.0 3.7 2025 36 74 Panama 37 25 8 5 44 52 5.2 2.9 2.2 2005 75 Armenia . . 23 . . 7 . . 50 3.2 2.7 . . 2005 76 Chile 29 23 10 6 45 53 4.0 2.7 2.1 2000 77 Iran, Islamic Rep. 45 44 16 9 42 44 6.7 6.2 5.6 2045 78 Moldova . . 17 . . II 43 2.6 2.5 2000 79 Ukraine . . 12 . . 13 . . 45 2.1 1.8 . . 2030 80 Mauritius 29 17 7 6 45 56 3.6 2.0 1.8 2030 81 Czechoslovakia 16 14 12 12 46 . . 2.1 1.9 2.0 82 Kazakhstun . . 21 . . 8 . . 49 3.4 2.8 . . 2000 83 Malaysia 36 29 10 5 44 50 5.5 3.7 3.0 2015 Upper-middle-income 32 w 21 w 10 w Sw 46 w 50 w 4.6 w 2.7 w 2.6 w 84 Botswana 53 36 17 6 45 46 6.9 4.8 3.1 2020 33 85 SouthAfrica 39 31 14 9 46 49 5.7 4.1 3.4 2020 86 Lithuania . . IS . . II . . 48 2.4 2.0 . . 2030 87 Hungaly 15 12 12 14 48 47 2.0 1.8 1.8 2030 88 Venezuela 38 29 7 5 44 52 5.3 3.7 2.7 2005 89 Argentina 23 21 9 9 50 47 3.1 2.8 2.3 2000 90 Uruguay 21 17 10 10 49 47 2.9 2.4 2.1 1995 91 Brazil 35 24 10 7 45 52 4.9 2.8 2.4 2000 6.5 92 Mexico 43 28 10 5 43 51 6.5 3.2 2.4 2010 53 93 Belarus . . 13 . . II . . 46 2.4 1.9 2030 94 Ruusian Federation . . 12 . . II . . 46 2.0 1.7 . . 2030 95 Latvia . . 14 . . 13 . . 45 1.9 2.0 . . 2030 96 TrinidadandTobago 28 24 8 6 46 51 3.6 2.8 2.3 2000 53 97 Gabon 31 42 21 15 48 46 4.2 5.8 6.1 2045 98 Eotonia 15 14 II 12 . . 46 2.1 2.1 . . 2030 99 Portugal 20 12 10 II 48 49 2.8 1.4 1.6 2030 100 Oman 50 41 21 5 47 42 7.2 6.8 5.9 2040 101 PuertoRico 25 18 7 7 48 52 3.2 2.3 2.1 1995 102 Korea Rep. 30 16 9 6 46 57 4.3 1.8 1.8 2030 77 103 Greece 17 10 8 9 51 47 2.3 1.4 1.6 2030 104 SaudiArabia 48 37 18 5 45 43 7.3 6.5 5.9 2040 105 Yugoslavia 18 14 9 9 50 2.3 2.0 2.0 Low- and middle-income 38 w 28 w 13w lOw 46 w 50 w 5.7 w 3.6 w 3.2 w Sub-Saharan Africa 48 w 46 w 21 w 16w 45 w 44w 6.6 w 6.4 w 5.9 w East Asia & Pacific 35 w 24 w 9w 7w 45 w 55 w 5.7 w 2.7 w 2.2 w South Asia 42 w 32 w 18 w 11 w 47 w 49 w 6.0 w 4.2 w 3.3 w Europe and Central Asia 22 w 17w lOw 10 w 48 w 47 w 2.9 w 2.3 w 2.3 w Middle East & N. Africa 45 w 38 w 16 w 8w 44w 46 w 6.8 w 5.3 w 4.6 w Latin America & Caribbean 36 w 26 w 10 w 7w 45 w 51 w 5.2 w 3.1 w 2.6 w Severely indebted 36 w 27 w 11 w 8w 46 w 50 w 5.2 w 3.5 w 3.0 w High-income economies 18 w 13 w 10 w 9w 47 w 50 w 2.4 w 1.8 w 1.7 w OECD members t7 w 13 w 10 w 9w 47 w 50 w 2.4 w 1.8 w 1.7 w 106 Ireland 22 15 II 9 42 49 3.9 2.1 2.1 2030 60 107 tlsrael 26 2! 7 6 46 50 3.8 2.8 2.3 2000 108 NewZealand 22 17 9 8 45 52 3.2 2.1 2.0 1995 09 Spain 20 10 8 9 49 50 2.8 1.3 1.5 2030 110 tHongKong 21 13 5 6 45 56 3.3 1.4 1.5 2030 81 III tSingapore 23 18 5 5 45 60 3.1 1.8 1.9 2030 112 United Kingdom 16 14 12 II 45 48 2.4 1.8 1.8 2030 113 Australia 21 15 9 7 47 53 2.9 1.9 1.9 2030 114 Italy 17 10 10 9 48 49 2.4 1.3 1.4 2030 115 Netherlands 18 13 8 9 47 SI 2.6 1.6 1.6 2030 116 Belgium 15 13 12 II 44 48 2.2 1.7 1.6 2030 117 Austria 15 12 13 II 43 49 2.3 1.6 1.6 2030 118 France 17 13 II 9 43 49 2.5 1.8 1.8 2030 119 Canada 17 15 7 7 47 53 2.3 1.9 1.7 2030 120 UnitedStates 18 16 tO 9 46 5! 2.5 2.1 1.9 1995 121 Germany 14 10 13 II 45 47 2.1 1.4 1.6 2030 122 Denmark 14 13 10 12 47 50 1.9 1.7 1.6 2030 123 Finland 14 13 10 10 48 49 1.8 1.9 1.8 2030 124 Norway 17 14 10 II 45 49 2.5 1.9 1.8 2030 84 125 Sweden 14 14 10 II 47 47 1.9 2.1 1.9 1995 126 Japan 19 10 7 7 56 50 2.! 1.5 1.6 2030 56 127 Switzerland 16 13 9 9 48 50 2.1 1.6 1.7 2030 World 34 w 26 w 13 w 9w 46 w 50 w 5.0 w 3.3 w 3.0 w Fuel exporters 47w 41 w 18w II w 44w 45w 6.8w 5.7w 5.0w a. For assumptions used in the projections, see the technical notes to Table 26. b. Data include women whose husbands practice contraception; see the technical notes. 291 Table 28. Health and nutrition Births Years of Prevalence Babies with low Infant mortaltt:i rate Population per attended by le lost of (per /,IXXI live health staff birth weight per i , malnutrition Physician Nursing person births) (percent) (percent) population (under 51 /970 1990 1970 /990 1985 1985 /970 /991 /990 /990 Low-income economies 14,080 w 6,760 w 5,580 w 109 w 71 w China and India 4,890 w 2,460 w 2,994) H' 96w 60w Other low-income 22,380 w 11,730w 11,580w 136 w 91 w I Mozambique 18,860 4,280 28 15 171 149 141 2 Tanzania 22,600 24,880 3,310 5,470 74 14 132 115 112 20 3 Ethiopia 86,120 32,650 58 158 130 107 4 Uganda 9,210 . 0 10 109 118 107 45 5 Bhutan 13,110 3 182 132 6 Guinea-Bissau 17,500 2,820 16 20 185 148 7 Nepal 51,360 16,830 17,700 2,760 10 157 101 67 . 8 Burundi 58,570 . 6,870 . . 12 14 138 107 81 38 9 Chad 61,900 30,030 8,010 . . II 171 124 106 35 10 Madagascar 10,120 8,130 240 . 62 10 181 114 63 53 II SierraLeone 17,830 2,700 . 25 14 197 145 188 12 Bangladesh 8,450 . . 65,780 . . . . 31 140 103 69 60 13 Lao PDR 15,160 4,380 1,390 490 . 39 146 100 93 14 Malawi 76,580 45.740 5,330 1,800 59 10 193 143 110 60 15 Rwanda 59,600 72,990 5,610 4,190 17 142 III 124 33 16 Mali 44,090 19,450 2,590 1,890 27 17 204 161 108 31 17 BurkinaFaso 97,120 57,320 . . 1,680 . 18 178 133 114 46 18 Niger 60,090 34,850 5,6/0 650 47 20 170 126 121 49 19 India 4,890 2,460 3,710 33 30 137 90 20 Kenya 8,000 10,130 2,520 . . 13 102 67 45 21 Nigeria 19,830 4,240 . . 25 139 85 98 22 China . S . 2,500 . . . 6 69 38 . 23 Haiti 12,520 7,410 . 20 17 141 94 69 24 Benin 28,570 . . 2,6(X) . . 34 10 155 III 89 35 25 Central African Rep. 44,740 25,930 2,460 . . . . 15 139 lOb 74 . 26 Ghana 12,910 22,970 690 1,670 73 17 III 83 55 36 27 Pakistan 4,310 2,940 6,600 5,040 24 25 142 97 61 57 28 logo 28,860 . . 1,590 . . . . 20 134 87 79 14 29 Guinea 50,010 . . 3,720 . . 18 181 136 125 30 Nicaragua 2,150 1,450 . . S 15 106 56 45 . 31 SriLanka 5,900 . . 1,280 . 87 28 53 18 14 45 32 Mauritania 17,960 . . 3,740 . . 23 10 165 119 . . 30 33 Yemen, Rep. 34,790 . . . . . . . . . . 175 109 104 34 Honduras 3,770 3,090 1,470 . 50 20 110 49 27 21 35 Lesotho 30,400 . . 3,860 . . 28 10 134 81 . . 27 36 Indonesia 26,820 7,030 4,810 . . 43 14 118 74 36 14 37 Egypt, Arab Rep. 1,900 1,320 2,320 490 24 7 158 59 33 13 38 Zimbabwe 6,300 7,180 640 1,000 69 15 96 48 37 12 39 Sudan 14,520 . . 990 . . 20 IS 149 101 84 55 40 Zambia 13,640 11,290 1,730 600 14 106 106 86 Middle-income economies 3,640 w 2,060w 1,640w SOw 38w Lower-middle-income 5,000 w 2,850 w 1,300 w 87w 42w 41 Bolivia 2,020 3,070 36 15 153 83 59 18 42 Cbted'Ivoire 15,520 . . 1,930 . . 20 14 135 95 50 12 43 Senegal 15,810 17,650 1,670 10 135 81 99 22 44 Philippines 9,270 8.120 2,690 . . . . 18 66 41 27 19 45 PapuaNewGuinea 11,640 12,870 1,710 1,180 34 25 112 55 79 46 Cameroon 28,920 12,190 2,560 1,690 . . 13 126 64 67 47 Guatemala 3,660 . . . . . . 19 10 100 60 41 34 48 Dominican Rep. . . . . 1,400 . . 57 16 90 54 24 /3 49 Ecuador 2,910 980 2,680 620 27 10 100 47 21 38 50 Morocco 13,090 4,840 1,050 . . 9 128 57 43 12 51 Jordan 2,480 770 870 500 75 7 29 18 52 Tajikistan . . 350 . . S S . 50 24 53 Peru 1,920 . . . . 55 9 108 53 32 13 54 ElSalvador 4,100 890 . . 35 15 103 42 28 55 Congo 9,510 780 12 126 115 24 56 SyrianArabRep. 3,860 1,160 1,790 870 37 9 96 37 25 57 Colombia 2,260 . . . . 51 15 77 23 II 12 58 Paraguay 2,300 . . 2,2/0 . . 22 6 57 35 22 4 59 Uzbekistan . . 280 . . S S S S . 44 20 60 Jamaica 2,630 530 . . 89 8 43 15 8 61 Rornania 840 560 430 99 6 49 27 19 . 62 Namibia . . 4,620 . . . . . . . . 118 72 . . . 63 l'unisia 5,930 1,870 940 3(X) 60 7 127 38 21 JO 64 Kyrgyzstan . . 280 . . . S S . . 40 20 65 Thailand 8,290 5,000 1,170 550 33 12 73 27 22 26 66 Georgia 170 . . . 5 16 15 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 292 Births Years of Prevalence attended by Babies with low Infant mortaltr rare life lost of Population per (Per I . live health staff birth weight per I,(XX) malnutrition Phisician Nursing person births) @ercent) (percent) population (under 5) /970 I /970 / 990 /985 1985 1970 I 1990 1990 67 Azerbaijan 250 33 16 68 Turkmenistan 290 ' 56 29 69 Turkey 2,230 1,260 1,010 78 7 147 58 31 70 Poland 700 490 250 8 33 15 16 71 Bulgaria 540 320 240 100 27 17 15 72 Costa Rica 1,620 /030 460 . . 93 9 62 14 73 Algeria 8.100 2,330 . . 330 . . 9 139 64 27 74 Panama 1,660 840 1,560 . . 83 8 47 21 . . 25 75 Armenia . . 250 . . . . . . . . . 22 14 76 Chile 2,160 2,150 460 340 97 7 78 17 13 2 77 Iran, Islamic Rep. 3,270 3/40 1,780 1,150 9 131 68 32 78 Moldova . . 250 . . . . . . . . . . 23 19 79 Ukraine . . 230 . . . . . . . . . . 18 16 80 Mauritius 4,190 1.180 610 . . 90 9 60 19 . . 24 81 Czechoslovakia 470 310 170 . . 100 6 22 II 16 82 Kazakhstan . . 250 . . . . . . . . . . 32 19 83 Malaysia 4,310 2,700 /270 380 82 9 45 15 15 24 Upper-middle-income 1,740 w 640 w 2,010 w 72 w 34 w 84 Botswana 15.220 5,150 1,900 . . 52 8 101 36 . . 15 85 South Africa 1,750 300 12 79 54 44) 86 Lithuania . . 220 . . . . . . . . 14 19 87 Hungary 510 340 210 . . 99 10 36 16 IS 88 Venezuela 1,120 630 440 330 82 9 53 34 13 5 89 Argentina 530 960 6 52 25 12 90 Umguay 910 . . . . . . . . 8 46 21 IS 9 91 Brazil 2,030 . . 4,140 . . 73 8 95 58 26 13 92 Mexico 1,480 . . 1,610 . . IS 72 36 17 14 93 Belarus . . 250 . . . IS 14 94 Russian Federation 210 20 17 95 Latvia 200 . 23 16 96 Trinidad and Tobago 2,250 190 90 44 19 9 97 Gabon 5,250 . . 570 . . 92 16 138 95 25 98 Estonia 210 . 20 14 99 Portugal 1,110 490 820 8 56 II 12 100 Oman 8,380 1,060 3,420 60 14 159 31 101 Puerto Rico 29 14 10 102 Korea, Rep. 2,220 1,370 1,190 65 9 SI 16 10 103 Greece 620 580 990 6 30 10 10 104 Saudi Arabia 7,460 660 2,070 420 78 6 119 32 37 105 Yugoslavia 1,000 530 420 I/O . . 7 56 21 16 Low- and middle-income 10,260 w 4,970 w 4,640 w 102 w 61 w Sub-Saharan Africa 31,730 w 23,540 w 3,460 w 144 w 104 w East Asia & Pacific 15,760 w 6,170 w 2,720 w 76 w 42 w South Asia 6,120 w 2,930 w 10,150 w 138 w 92 w Europe and Central Asia 1,070 w 420 w 520 w 63 w 26 w Middle East & N. Africa 6,410 w 2,240 w 1,940w 135 w 60w Latin America & Caribbean 2,020 w 1,180 w 2,640 w 82 w 44w Severely indebted 2,910 w 1,680 w 2,330 w 85 w 48w High-income economies 710 w 420 w 220 w 20 w 8w OECD members 700 w 420 w 220 w 20 w 8w 106 Ireland 980 630 160 . . 4 20 8 II 107 tlsrael 410 . . 99 7 25 9 9 108 New Zealand 870 . . /50 99 5 17 9 Il 109 Spain 750 280 . . . . 96 . . 28 8 10 110 tHongKong 1,510 . . 560 . . . . 4 19 7 7 Ill tSingapore 1,370 820 250 . . 100 7 20 6 9 112 UnitedKingdom 810 240 . . 98 7 19 7 12 113 Australia 830 . . 99 6 18 8 9 114 Italy 550 210 . . 7 30 8 10 115 Netherlands 800 410 300 . . 4 13 7 10 116 Belgium 650 310 . . . . 100 5 21 8 II 117 Austria 540 230 300 70 . . 6 26 8 II 118 France 750 350 270 . . 5 18 7 10 119 Canada 680 450 140 99 6 19 7 9 120 United States 630 420 160 . . 100 7 20 9 II 121 Germany 580° 370° . . . . 5° 23 7 12 122 Denmark 690 390 . . . . . . 6 14 8 12 123 Finland 960 410 130 . . . . 4 13 6 II 124 Norway 720 . . 160 . . 100 4 13 8 10 125 Sweden 730 370 140 . . 100 4 Il 6 II 126 Japan 890 610 3/0 . 100 5 13 5 8 127 Switzerland 700 (,30 . . 5 15 7 10 World 7,640 w 3,980 w 3,940 w 85 w 53 w Fuel exporters 10,730 w 2,030 w 2,770 w 128 w 70 w a. Data refer to the Federal Republic of Germany before unification. 293 Table 29. Education Percentage ofage group enrolled in education Primary net PrtmarY Secondary enrollment Primary pupil! Total Female Total Female Tertiary (total) (percent) teacher ratio 1970 1990 /970 /990 /970 1990 1970 1990 /970 /990 /975 /990 /970 /990 Low-income economies 74 w 105 w 98 w 21 w 41 w 34 w 2w 36 w 38 w China and India 83w 119w 109w 25w 46w 38w 1w .. 34w 38w Other low-income SSw 79w 44w 73w 13w 28w 8w 24w 3w 4w 72w 39w 39w I Mozambique 47 58 48 5 7 5 0 41 69 58 2 Tanzania 34 63 27 63 3 4 2 4 0 47 47 35 3 Ethiopia 16 38 10 30 4 15 2 12 0 1 28 48 36 4 Uganda 38 76 30 4 13 2 I 1 34 35 5 Bhuian 6 26 1 20 1 5 0 2 0 21 37 6 Guinea-Bissau 39 59 23 42 8 7 6 4 0 0 59 45 7 Nepal 26 86 8 57 10 30 3 17 3 6 64 22 37 8 Bumndi 30 72 20 64 2 5 I 4 1 I 37 67 9 Chad 35 57 17 35 2 7 0 3 1 65 67 lO Madagascar 90 92 82 90 12 19 9 18 3 64 65 40 II Sierral.eone 34 48 27 39 8 16 5 l2 I I 32 34 12 Bangladesh 54 73 35 68 . . 17 . . II 3 3 . 65 46 63 13 Lao PDR 53 104 40 91 3 26 2 21 I / 69 36 28 14 Malawi . . 71 . 64 . . 4 . 3 I / 54 43 64 IS Rwanda 68 69 60 68 2 7 I 6 0 1 65 60 57 16 Mali 22 24 15 17 5 6 2 4 . . . l9 40 42 17 Burkina Faso 13 36 10 28 I 7 I 5 0 1 . . 29 44 57 18 Niger 14 29 10 21 I 7 I 4 0 I . 25 39 42 19 India 73 97 56 83 26 44 IS 33 6 . . . 41 61 20 Kenya 58 94 48 92 9 23 5 19 1 2 88 34 3l 21 Nigeria 37 72 27 63 4 20 3 17 2 3 . . 34 41 22 China 89 135 . 129 24 48 . . 41 I 2 100 29 22 23 Haiti 53 0 . . 6 . . 4 . . . . . . . . 47 2! 24 Benin 36 61 22 44 5 II 3 6 2 3 . 52 4l 35 25 CentralAfiicanRep. 64 67 41 51 4 11 2 6 I 2 55 64 90 26 Ghana 64 75 54 67 14 39 8 31 2 2 30 29 27 Pakistan 40 37 22 26 13 22 5 13 . . 3 . . 41 41 28 Togo 71 103 44 80 7 22 3 10 2 3 72 58 59 29 Guinea 33 37 21 24 13 10 5 5 5 I . . 26 44 40 30 Nicaragua 80 98 81 101 18 38 l7 44 14 . 65 75 37 33 31 SriLanka 99 107 94 105 47 74 48 77 3 4 . . . 0 32 Mauritania 14 51 8 42 2 16 0 10 4 . . . . 24 49 33 Yemen, Rep. 22 . . 7 . . 3 . . S . . 34 Honduras 87 108 87 109 14 . . 13 . . 8 9 . . 35 . 35 Lesotho 87 107 101 115 7 26 7 31 2 5 70 46 55 36 Indonesia 80 117 73 114 16 45 11 41 . 72 98 29 23 37 Egypt, Arab Rep. 72 98 57 90 35 82 23 71 18 19 38 25 38 Zimbabwe 74 117 66 116 7 50 6 46 I 5 36 39 Sudan 38 49 29 . . 7 20 4 . . 2 3 . . 47 34 40 Zambia 90 93 80 91 13 20 8 14 2 2 . 80 47 44 Middle-income economies 94 w 103 w 88 w 99 w 33 w 126 w 28 w 59 w 14 w 16 w 89w 33w 25w Lower-middle-income 93 w 100 w 83 w 97 w 31 w 172 w 24 w 57 w 12 w 16 w 87 w 33 w 25 w 41 Bolivia 76 82 62 78 24 34 20 31 17 23 73 82 27 25 42 Côte d'Ivoire 58 . . 45 . . 9 . . 4 . . 3 . . . . . . 45 36 43 Senegal 41 58 32 49 10 16 6 II 3 3 . . 48 45 58 44 Philippines 108 111 . . 110 46 73 . . 75 3 27 95 99 29 33 45 PapuaNewGuinea 52 71 39 65 8 12 4 10 2 . . 73 30 32 46 Cameroon 89 101 75 93 7 26 4 21 2 4 69 75 48 51 47 Guatemala 57 79 51 . . 8 8 8 53 36 48 Dominican Rep. 100 95 100 96 21 . . . . . . . . 55 47 49 Ecuador 97 . . 95 . . 22 . . 23 . . 37 20 78 . . 38 50 Morocco 52 68 36 55 13 36 7 30 6 10 47 55 34 27 51 Jordan . . . . . . . . 39 17 52 Tajikistan .. .. .. .. .. .. .. .. .. .. 53 Peru 107 126 99 . . 31 70 27 . . 19 36 95 35 28 54 El Salvador 85 78 83 78 22 26 21 26 4 17 . . 70 36 40 55 Congo . . 6 . . 62 66 56 SyrianArabRep. 78 109 59 102 38 52 21 43 18 20 87 98 37 25 57 Colombia 108 110 110 III 25 52 24 57 10 14 . . 73 38 30 58 Paraguay 109 107 103 106 17 30 17 30 9 8 83 95 32 25 59 Uzbekistan .. .. .. .. .. .. .. .. .. .. .. 60 Jamaica 119 105 119 105 46 60 45 63 7 5 90 99 47 37 61 Romania 112 91 113 96 44 92 38 90 II 9 . . . 62 Namibia . . 94 . . 99 . . 34 . . 38 . . . . . 64 47 28 63 Tunisia 100 116 79 109 23 45 13 40 5 9 95 64 Kyrgyzstan .. .. .. .. .. .. .. .. .. 65 Thailand 83 85 79 85 17 32 15 32 13 16 . . . . 35 18 66 Georgia Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 294 Percentage of age group enrolled in education PnmI) net Primary Secondary enrollment Primary pupill Total Female Total Female Tertiary (total) (percent) teacher ratio /970 1990 /970 1990 /970 1990 1970 /990 1970 1990 1975 1990 1970 1990 67 Azerbaijan . - 68 Turkmenistan . - 69 Turkey 110 110 94 105 27 54 15 42 6 14 99 38 30 70 Poland 101 98 99 98 62 82 65 84 18 22 96 97 23 16 7! Bulgaria 101 96 00 95 79 74 . . 75 16 31 96 85 22 15 72 CostaRica 110 102 09 101 28 42 29 43 23 26 92 87 30 32 73 Algena 76 95 58 88 II 60 6 53 6 12 77 88 40 28 74 Panama 99 /07 97 105 38 59 40 62 22 21 87 92 27 20 75 Armenia . . . . . . . . . . . . . . . . . . . . 76 Chile 107 98 107 97 39 74 42 77 13 19 94 86 50 29 77 Iran, Islamic Rep. 72 112 52 106 27 56 18 47 6 . . 94 32 28 78 Moldova . . . . . 79 Ukraine . . . . . . . . . . . . . . . . . . 15 8 80 Mauritius 94 106 93 /04 30 52 25 53 I 2 82 92 32 21 Czechoslovakia 98 93 98 93 84 39 87 18 20 . 81 31 17 . . . . 19 82 Kazakhstan .. .. .. .. .. .. .. .. .. .. .. 83 Malaysia 87 93 84 93 34 56 28 58 4 7 . . . . 3! 20 Upper-middle-income 95 w 106 w 93 w 105 w 35 w 54 w 31 w . 15 w 17 w 80 w 91 w 34 w 25 w 84 Botswana 65 110 67 112 7 46 6 47 I 3 58 91 36 32 85 South Africa 99 99 18 17 34 86 Lithuania . . . . . . . . . . . . . 87 Hungary 97 94 97 94 63 79 55 79 13 IS . . 90 18 12 88 Venezuela 94 92 94 94 33 35 34 41 21 29 81 61 35 23 89 Argentina 105 Ill 106 114 44 . . 47 22 . . 96 19 19 90 Uruguay 112 106 109 106 59 77 64 18 50 . . . . 29 23 91 Brazil 82 108 82 . . 26 39 26 . . 12 12 71 88 28 23 92 Mexico 104 112 101 110 22 53 17 53 14 14 98 46 31 93 Belanis 94 Russian Federation . . . 95 1..atvia . . . . 96 Trinidad and Tobago 106 95 107 96 42 80 82 87 90 34 26 97 Gabon 85 . 81 . . 8 5 . . 4 . . . 46 98 Estonia 99 Portugal 98 119 96 117 57 59 51 59 II 18 91 99 34 14 100 Oman 3 103 I 99 . . 54 48 . - 5 32 84 18 28 101 PuertoRico 117 . . . . . 71 . . . . . . . . . . . . . 30 102 Korea,Rep. 103 108 103 110 42 87 32 85 16 39 99 100 57 36 103 Greece 107 1(X) 106 101 63 99 55 97 17 29 97 96 31 2) 104 SaudiArabia 45 78 29 72 12 48 5 41 7 14 42 62 24 16 105 Yugoslavia 106 95 103 95 63 79 58 79 22 18 27 23 Low- and middle-income 79 w 104 w 64 w 98 w 24 w 61 w 18 w 39 w 6w 7w 91 W 35 w 35 is Sub-Saharan Africa 46 w 68 w 36 w 61 w 6w 17 w 4w 16 w Iw 2w . - 46 w 43 w 41 is East Asia & Pacific 88 w 127 w 77 w 123 w 24 w 49 w 16 w 44 w 4w 5w 100 w 30 w 23 is South Asia 67 w 88 w 50 w 75 w 25 w 39 w 14 w 29 w 3w . . , 67 w 42 w 58 is Europe and Central Asia 105 w 101 w 100 w 100 w 50 w 71 w 44 w 70 w 14 w 16 w 89 w 27 w 19 is Middle East & N. Africa 68 w 97 w 50 w 90 w 24 w 365 w IS w SO w 10 w 12 w 85 w 35 w 26 is Latin America & Caribbean 95 w 107 w 94 w 106 w 28 w 49 w 26 w 57 w 15 w 16 w . . 88 w 34 w 26 is Severely indebted 90 w 104 w 85 w 98 w 31 w 209 w 27 w 56 w 14 w 15 w 78 w 88 w 32 w 25 is High-income economies 106 w 104 w 106w 104w 73 w 92w 71 w 96w 36 w w 88w 97 w 26 w 17 w OECD members 106 w 104 w 106 w 104 w 74 w 93 w 73 w 96 w 36 w 33 w 88 w 97 w 26 w 17 w 106 Ireland 106 100 106 /01 74 98 77 102 20 26 91 88 24 27 107 tlsrael 96 93 95 95 57 83 60 86 29 33 . . . . 17 18 108 New Zealand 110 106 109 lOS 77 89 76 91 29 41 100 100 21 19 109 Spain 123 109 125 108 56 107 48 112 24 34 100 1(X) 34 21 110 tHong Kong 117 106 115 . 36 . . 31 . II . . 92 . . 33 Ill tSingapore lOS 110 101 109 46 69 45 71 . . 8 100 1(X) 30 26 112 United Kingdom 104 107 104 107 73 84 73 85 20 25 97 1(X) 23 20 113 Australia 115 lOS 115 lOS 82 83 80 85 25 35 98 97 28 17 114 Italy 110 97 109 96 61 79 55 78 28 31 97 . . 22 12 115 Netherlands 102 117 102 118 75 103 69 101 30 34 92 100 30 13 116 Belgium 103 102 104 103 81 104 80 104 26 37 . . 99 20 10 117 Austria 104 103 103 102 72 83 73 85 23 33 89 93 21 II 118 France Ill III 117 110 74 99 77 100 26 40 98 100 26 12 119 Canada 101 lOS 100 104 65 106 65 107 42 70 . . 96 23 IS 120 United States . . 105 104 . 92 91 56 75 72 99 27 121 Germany . . 105 . 105 . 97 . . 103 . . 32 87 . . 18 122 Denmark 96 98 97 98 78 109 75 110 29 32 . . 9 II 123 Finland 82 99 79 99 102 114 106 124 32 47 . - 10() 22 18 124 Norway 89 99 94 99 83 100 83 102 26 43 100 98 20 6 125 Sweden 94 107 95 107 86 91 85 93 31 33 100 100 20 6 126 Japan 99 101 99 101 86 96 86 97 31 31 99 100 26 21 127 Switzerland 18 26 World 83w 104w 71w 99w 31w 65w 28w 46w 13w 11w 92w 33w 33w Fuel exporters 59w 89w 44w 83w 15w 39w 11 w 35w Sw 11 w 90w 34w 32 is' 295 Table 30. Income distribution and PPC estimates of GDP PPC estimates of GDP per capita a Percentage share ofincome or conswnption United Slates = 100 Current international dollars Lowest Second Third Fourth Highest Highest Year 20 percent quintile quintile quintile 20 percent 10 percent /987 /99/ Observeda Regression Low-income economies China and India Other low-income I Mozambique . . . . . . . . . . . 2.7' 2.7' . . 600 2 Tanzania 1991 d.c 2.4 5.7 10.4 18.7 62.7 46.5 2.5 2.6 570 640 3 Ethiopia I98l_82'g 8.6 12.7 16.4 21.1 41.3 27.5 1.9 1.7 37O 620 4 Uganda l989_90d.e 8.5 12.1 16.0 21.5 41.9 27.2 47c SIC 1,120 5 Bhutan . 2.8' 2.8C . 620 6 Guinea-Bissau . . . . . . . . . . . . 3.1 C 3.1 C 690 7 Nepal 1984-85"' 9.1 12.9 16.7 21.8 39.5 25.0 4.8c 5.1 C 1,130 8 Burundi 3.2c 33C . 720 9 Chad 2.9c 33C . 730 10 Madagascar . . . . . . 3.6 3.2 7lO 1,120 II Sierra Leone . . . . . . . . . . . . 3.5 3.6 800 1,190 12 Bangladesh l98889 9.5 13.4 17.0 21.6 38.6 24.6 5.0 5.2 I,l60 990 13 LaoPDR . 8.3C 8.7c . 1,930 14 Malawi . . . . . . . . . . . . 3.5 3.6 8O0 570 IS Rwanda l98385 9.7 13.1 16.7 21.6 38.9 24.6 3.9 3.1 680 750 16 Mali . . . . . . . . 2.3 2.2 480 730 17 BurkinaFaso . . . . . . 34C 34C 750 18 Niger . . . . . . . . . . . 3.9 3.6C , 790 19 India l989_90d.c 8.8 12.5 16.2 21.3 41.3 27.1 4.6 5.2 i,I50 1,900 20 Kenya l981_83i.k 2.7 6.4 11.1 18.9 60.9 45.4 6.1 6.1 l,3SO 1,490 21 Nigeria . . . . . . . . . . . . 5.5 6.1 l,360 1,900 22 China I990" 6.4 11.0 16.4 24.4 41.8 24.6 6.5 7.6 1,680' 2,040 23 Haiti . . . 6.7C . . 1,220 24 Benin . . . . . . . . 7.3 6.8 I,500 1,180 25 Central African Rep. . . . . . . . 55C 49C . 1,090 26 Ghana I988_89d.c 7.0 11.3 15.8 21.8 44.1 29.0 8.9C 9.00 . 2,000 27 Pakistan 199ld.e 8.4 12.9 16.9 22.2 39.7 25.2 8.3 8.9 l,970 1,570 28 Togo 6.4k' 59C 1,310 29 Guinea . . . . , . . . . . . . . . . 30 Nicaragua . . . l5.5 ll.5' 2,550 31 Sn 1985_86h.i 4.9 8.4 12.4 18.2 56.2 43.0 11.0 12.0 2,65O 2,580 32 Mauritania . . . 6.9' 6.3' . . 1,390 33 Yemen, Rep. . . . . . . . . . . . . . . . . . 34 Honduras 1989h. 2.7 6.0 10.2 17.6 63.5 47.9 8.5 8.2 I,820m 2,670 35 L.esotho 1986_87h. 4.5 6.5 10.0 17.6 61.3 45.0 7.2C 8.5C . 1,890 36 Indonesia I990d.c 8.7 12.1 15.9 21.1 42.3 27.9 10.5 12.3 2,730m 2,720 37 Egypt, Arab Rep. . . 16.3 16.3 3,600 3,140 38 Zimbabwe . . . . . . . 9.2 9.8 2,1601 2,580 39 Sudan . . . . . . . . . . . . . . . 40 Za,nbia . . . . . . . . . . . 5.3 4.6 I,OIO 1,010 Middle-income economic Lower-middle-income 41 Bolivia . . . . . . . . . . . . 9.6 9.8 2,170m 2,260 42 Côted'Ivoire l988d.e 7.3 11.9 16.3 22.3 42.2 26.9 9.1 6.8 i,5i0 1,680 43 Senegal . . . . . . . . . . . . 7.9 7.6 I,68O 1,600 44 Philippines l988 6.5 10.1 14.4 21.2 47.8 32.1 10.8 11.0 2,440 2,900 45 Papua New Guinea . . . 8.8 8.3' . . 1,830 46 Cameroon 15.1 10.8 2,400 1,970 47 Guatemala I989" 2.1 5.8 10.5 18.6 63.0 46.6 14.4 14.4 3,180m 2,600 48 Dominican Rep. l989 4.2 7.9 12.5 19.7 55.6 39.6 15.5 13.9 3,080m 3040 49 Ecuador 17.8 18.7 4,140m 3,950 50 Morocco l990_9ld.e 6.6 10.5 15.0 21.7 46.3 30.5 13.8 15.1 3,340 2,800 51 Jordan 29.3 C 22.00 4,870 52 Tajikistan 11.9 9.9 2.180' 53 Peru 1985_86cj.e 49 9.2 13.7 21.0 51.4 35.4 19.7 14.1 3,IIOm 2,930 54 ElSalvador 9.5 9.5 2,IIOm 2,750 55 Congo 13.1 12.7 2,800f 56 SynanArabRep. 20.9 23.6 5,220 4,630 57 Colombia 1988 h.i 4.0 8.7 ' 13.5 20.8 53.0 37.1 23.7 24.7 5.460m 4,080 58 Paraguay 14.9 15.5 3,420m 2,900 59 Uzbekistan 12.1 12.6 2.790' 60 Jamaica l990 6.0 9.9 14.5 21.3 48.4 32.6 15.1 16.6 3,670 4,050 61 Romania 42.3 31.2 6,900 62 Namibia 63 Tunisia 1990 d.e 5.9 10.4 15.3 22.1 46.3 30.7 20.2 21.2 4.6901 3,780 64 Kyrgyzst.an 14.2 14.8 3,2801 65 Thailand l988d.1 6.1 9.4 13.5 20.3 50.7 35.3 17.1 23.8 3,740 66 Georgia .. .. .. .. 24.7 16.6 3,670 Note: For data comparability and coverage, see the technical notes. Figums in italics am for years other than specified. 296 PPC estimates of GDPper capita a Percentage share of income or consumption United States = 100 Current international dollars Lowest Second Third Fourth Highest Highest Year 20 percent quintile qaintile qaintile 20 percent JO percent /987 /991 Observed a Regression b 67 Azerbaijan . . . . 20.2 16.6 3,670 68 Thrkmenistan . . . 17.3 16.0 3,540' 69 Thrkey . . . . . . . . . . 21.0 21.9 4,840 3,950 70 Poland I989i 9.2 13.8 17.9 23.0 36.1 21.6 24.8 20.3 4,SOO 4,720 71 Bulgaria . . S . . 31.lc 22.5c . 4,980 72 Costa Rica 1989h,i 4.0 9.1 14.3 21.9 50.8 34.1 22.5 23.0 5,100w 4,320 73 Algeria . . . 0 S S 28.6' 25.5' . 5,640 74 Panama I989 2.0 6.3 11.6 20.3 59.8 42.1 25.6 22.2 4,910w 5,030 75 Armenia S . . . . . . 24.3 20.8 4,610' 76 Chile 1989h.j 3.7 6.8 10.3 16.2 62.9 48.9 27.3 31.9 7060w 5,380 77 Iran, Islamic Rep. . . . . . . 22.1 21.l 4,67o 6,760 78 Moldova . . . . . . . . 23.1 21.0 4,640 79 Ukraine . . . . . . . . 25.7 23.4 5,180' 80 Mauritius . . . . . . . . 40.8 50.5 I l,I8O 5,480 81 Czechoslovakia . . . . . . . . . 35.0' 28.4' . . 6,280 82 Kazakhstan . . . . . . . . . . . . 23.0 20.3 4,490 83 Malaysia l9891 4.6 8.3 13.0 20.4 53.7 37.9 26.5 33.4 7,400k 6,530 Upper-middle-income 84 Botswana 95_it.g 1.4 4.6 9.4 18.2 66.4 49.6 16.7 21.2 4,690 4,080 85 South Africa . . . . . . . . . . . . . . . 86 Lithuania . . . . . . . . . . . . 29.4 24.4 5,4l0 87 Hungary 1989h 10.9 14.8 18.0 22.0 34.4 20.8 31.9 27.5 6,O80 5,260 88 Venezuela 1989h.i 4.8 9.5 14.4 21.9 49.5 33.2 36.3 36.7 8.120w 6.990 89 Argentina . . . . . . . . . . 25.6 23.1 5,120m 5,840 90 Umguay . . . . . . . . . . . . 30.4 30.1 6,670w 5,690 91 Brazil 1989h,i 2.1 4.9 8.9 16.8 67.5 51.3 26.0 23.7 5,240w 4,180 92 Mexico l984 4.1 7.8 12.3 19.9 55.9 39.5 31.4 32.4 7,l70 5,190 93 Belan.is . . . . 29.7 31.0 6,850 94 Russian Federation . . . . . . . . 35.2 31.3 6,930' 95 Latvia . . . . . . . . 37.2 34.1 7,540 96 TrinidadandTobago . . . . . . . 42.5c 379C 8,380 97 Gabon . . . . . . . . . . . . . 98 Estonia . . . . . . . . . . . . 45.8 36.6 8,090' 99 Portugal . . . . . . . . . . . . 35.9 42.7 9,450 7,730 100 Oman . . . . . . . . 38.5 40.6C . 8,990 101 PuertoRico . . . . . . . . . . . 447C 493C . 10,920 102 Korea, Rep. . . . . . . . . . . . . 28.6 37.6 8,320 10,070 103 Greece . . . . . . . . . . . . 33.8 34.7 7,680 9,700 104 Saudi Arabia . . . . . . . . . . . . 45.1 49.0c . 10,850 105 Yugoslavia 198951 5.3 10.7 16.2 23.7 44.2 27.4 28.4c Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe and Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECD members 106 Ireland . . . . . . . . . . . . 42.3 51.6 11,430 15,060 107 tlsrael 1979J.k 6.0 12.1 17.8 24.5 39.6 23.5 60.2 60.8 13,460w 12,980 108 New Zealand l98I_82j.' 5.t 10.8 16.2 23.2 44.7 28.7 68.0 63.1 13,970 12,660 109 Spain 1980_81j.k 6.9 12.5 17.3 23.2 40.0 24.5 50.5 57.3 12,670 13,760 110 tHongKong l980i 5.4 10.8 15.2 21.6 47.0 31.3 74.0 83.7 l8,520 14,470 Ill tSingapore l982_83j.k 5.1 9.9 14.6 21.4 48.9 33.5 57.1 C 71 .2c . 15,760 112 United Kingdom 1979J.k 5.8 11.5 18.2 25.0 39.5 23.3 73.0 73.8 16,340 15,470 113 Australia l985J' 4.4 11.1 17.5 24.8 42.2 25.8 76.7 75.4 16,680 15,820 114 Italy 1986J5 6.8 12.0 16.7 23.5 41.0 25.3 71.4 77.0 17,040 15,960 115 Netherlands l983J.k 6.9 13.2 17.9 23.7 38.3 23.0 70.0 76.0 16,820 19,110 116 Belgium 1978_79J,k 7.9 13.7 18.6 23.8 36.0 21.5 71.5 79.1 17,510 18,470 117 Austria . . . . . . . . . . . . 72.6 79.9 17,690 17,850 118 France I979J5 6.3 12.1 17.2 23.5 40.8 25.5 78.1 83.3 18,430 18,990 119 Canada 1987J.k 5.7 11.8 17.7 24.6 40.2 24.1 90.7 87.3 19,320 19,370 120 United States 1985jk 4.7 11.0 17,4 25.0 41.9 25.0 100.0 100.0 22,130 22,130 121 Germany 1984J.k 6.8 12.7 17.8 24.1 38.7 23.4 80.5 89.3 19,770 21,130 122 Denmark 1981J.k 5.4 12.0 18.4 25.6 38.6 22.3 79.1 80.8 17,880 20,780 123 Finland l98lJ,k 6.3 12.1 184 25.5 37.6 21.7 73.0 72.9 16,130 19,850 124 Norway 1979i.k 6.2 12.8 18.9 25.3 36.7 21.2 79.8 77.6 17,170 20,290 125 Sweden l98lik 8.0 13.2 17.4 24.5 36.9 20.8 80.3 79.0 17,490 19,510 126 Japan 1979i5 8.7 13.2 17.5 23.1 37.5 22.4 74.4 87.6 19,390 23,830 127 Switzerland 1982J.k 5.2 11.7 16.4 22.1 44.6 29.8 95.6 98.4 21,780 World Fuel exporters a. Extrapolated from t990 ICP estimates unless noted otherwise; b. See technical notes; c. Obtained from the regression estimates; d. Data refer to expenditure shares by fractiles of persons; e Data ranked by per capita expenditure; f. Extrapolated from 1985 ICP estimates; g. Data ranked by household expendi- ture. h. Data refer to income shares by fractiles of persons; i. Data ranked by per capita income; j. Data refer to income shares by fractiles of households; k. Data ranked by household income; I. These values are subject to more than the usual margin of error (see technical notes); m and n are extrapolated, respectively, from 1980 and 1975 ICP estimates and scaled up by the corresponding US deflator. 297 Table 31. Urbanization Population in Urban population Population in ctttes of! millton or more tn capital dcc as a /990, as a percentage of As a percentage of Average annual growth percentage of total population rate (percent) Urban Total Urban Total 1970 1991 1970-80 1980-9! /990 1990 1965 /990 /965 1990 Low-income economies 18w 39 w 3.7w 11w 3w 41w 31w 7w 9w China and India 18 w 46w 3.2w 3w 1w 42w 29w 8w 9w Other low-income 18 w 28 w 4.7w 5.0w 27w 7w 38w 35w 6w lOw I Mozambique 6 28 11.5 10.1 38 10 68 38 3 10 2 Tanzania 7 34 12.7 10.1 21 7 38 18 2 6 3 Ethiopia 9 13 4.8 5.3 29 4 27 30 2 4 4 Uganda 8 II 3.6 4.5 41 4 5 Bhutan 3 6 4.1 5.7 22 I . . 6 Guinea-Bissau 15 20 5.8 3.7 36 7 7 Nepal 4 10 7.3 7.3 20 2 8 Bumndi 2 6 7.7 5.7 81 4 9 Chad II 30 8.1 6.3 43 13 10 Madagascar 14 25 5.3 6.2 23 6 II SienaLeone 18 33 5.2 5.3 52 17 . . . . 12 Bangladesh 8 17 6.8 6.1 37 6 50 47 3 8 13 La0PDR 10 19 5.1 6.0 53 10 . 14 Malawi 6 12 7.5 6.0 31 4 . . 15 Rwanda 3 8 8.1 7.6 56 4 . . . . 16 Mali 14 20 4.1 3.8 41 8 . . . . . 17 BurkinaFaso 6 9 4.3 5.2 51 5 . . . . . l8 Niger 9 20 7.5 7.4 39 8 . . . . . 19 India 20 27 3.9 3.7 4 1 32 32 6 9 20 Kenya 10 24 8.5 7.8 26 6 41 27 4 6 21 Nigeria 20 36 6.1 5.8 23 8 23 24 4 8 22 China 18 60 2.6 . . 2 I 49 27 9 9 23 Haiti 20 29 3.5 3.8 56 16 47 56 8 16 24 Benin 18 38 8.4 SI 12 4 . 25 Central African Rep. 30 48 4.5 4.8 52 24 26 Ghana 29 33 2.7 4.1 22 7 27 22 7 7 27 Pakistan 25 33 4.4 4.6 I 0 44 42 tO 13 28 Togo 13 26 6.4 6.6 55 l4 . . . . . 29 Guinea l4 26 4.8 5.6 89 23 47 88 5 23 30 Nicaragua 47 60 4.4 3.9 46 28 36 44 15 26 31 SriLanka 22 22 1.5 1.5 17 4 32 Mauritania 14 48 lO.4 7.3 83 39 33 Yemen,Rep. 13 30 7.0 7.3 Il 3 34 Honduras 29 45 5.7 5.4 35 15 35 Lesotho 9 21 7.1 7.0 17 4 36 Indonesia 17 31 5.1 5.0 17 5 42 33 7 lO 37 Egypt, Arab Rep. 42 47 2.5 3.2 37 17 53 52 22 24 38 Zimbabwe Il 28 5.6 5.8 31 9 . . . . . 39 Sudan 16 22 4.9 3.9 35 8 30 35 4 8 40 Zambia 30 51 5.9 6.0 25 12 Middle-income economies 46w 62w 3.7w 3.2w 25w 14w 40w 40w 17w 24w Lower-middle-income 41 w 54 w 3.6 w 3.3 w 28 w 13 w 36 w 36 w 13 w 19 w 41 Bolivia 41 52 3.4 4.0 34 17 28 33 II 17 42 Côted'lvoire 27 41 7.5 4.7 45 18 30 45 7 18 43 Senegal 33 39 3.4 4.0 52 20 40 53 13 20 44 Philippines 33 43 3.8 3.7 32 14 28 32 9 14 45 Papua New Guinea 10 16 5.3 4.3 33 5 46 Cameroon 20 42 7.6 5.6 16 7 . . . 47 Guatemala 36 40 3.3 3.5 23 9 . . . . . 48 DominicanRep. 40 61 4.9 3.9 52 31 46 SI 16 31 49 Ecuador 40 57 4.8 4.4 21 12 50 49 19 28 50 Morocco 35 49 4.1 4.3 9 4 39 36 12 17 5 j.1b 50 69 5.7 5.7 46 31 33 38 IS 26 52 Tajikistan . . 32 . . . . . . . . . . . . . 53 Peni 57 71 4.0 3.1 41 29 37 41 19 29 54 ElSalvador 39 45 2.9 2.1 26 II 55 Congo 33 41 4.0 4.7 68 28 56 SyrianArabkep. 44 51 4.1 4.1 34 17 58 60 23 30 57 Colombia 57 71 3.3 2.9 21 IS 38 39 20 27 58 Paraguay 37 48 4.2 4.4 48 23 59 Uzbekistan . . 41 . . . . . . . 60 Jamaica 42 53 2.6 2.1 52 27 . 61 Romania 42 53 2.5 1.2 18 9 21 18 8 9 62 Namibia 19 28 4.9 5.1 36 10 . . . . . 63 Tunisia 44 55 4.1 2.8 37 20 35 37 14 20 64 Kyrgyzstan . . 38 . . . . . . . . . . . . . 65 Thailand 13 23 5.4 4.6 56 13 66 57 8 13 66 Georgia 56 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 298 Population in Urban population Population in cities of! million or more in capital citi as a 1990. as a percentage of As a percentage of Average annual growth percentage of total population rate (percent) Total Urban Total Urban /970 199! 1970-80 1980-9! 1990 /990 1965 1990 1965 1990 67 Azerbaijan . 54 68 Turkmenistan . . . . . . . . . . . . . 69 Turkey 38 63 3.7 5.8 8 5 41 35 14 22 70 Poland 52 62 2.0 1.3 9 6 32 28 16 18 71 Bulgaria 52 68 2.1 1.1 20 13 21 19 10 13 72 Costa Rica 40 48 3.6 3.7 72 34 62 72 24 34 73 Algeria 40 53 4.1 4.8 23 12 24 23 9 12 74 Panama 48 54 2.9 2.9 37 20 75 Armenia . 68 . . . . . . . 0 0 76 Chile 75 86 2.4 2.2 42 36 39 42 28 36 77 Iran, Islamic Rep. 41 57 5.2 5.0 21 12 43 41 16 23 78 Moldova . . 47 . 79 Ukraine . . 67 . . . . . . . . . 80 Mauritius 42 41 1.7 0.5 36 IS . . . . . 81 Czechoslovakia 55 . . 2.7 1.7 II 8 IS II 8 8 82 Kazakhstan . . 57 . . . . . . . . . . . . . 83 Malaysia 27 44 5.0 4.8 22 10 16 22 4 10 Upper-middle-income 53 w 73 w 3.9 w 3.0 w 21 w 15 w 46 w 44 w 23 w 32 w 84 Botswana 8 29 10.0 10.0 37 10 . . . . . 85 South Africa 48 60 3.8 3.8 10 6 40 30 19 18 86 Lithuania . . 68 . . . . . . . . . . . . . 87 Hungary 46 62 2.0 1.1 33 20 43 33 19 20 88 Venezuela 72 85 5.0 2.7 25 21 34 29 24 27 89 Argentina 78 87 2.2 1.8 41 36 53 49 40 42 90 Umguay 82 86 0.6 0.8 45 39 53 45 43 39 91 Brazil 56 76 4.1 3.3 2 2 48 47 24 35 92 Mexico 59 73 4.1 2.9 34 25 41 45 22 32 93 Belarus 66 . . 2.3 94 Russian Federation . . 74 . 95 Latvia . . 71 . . . . . 96 Trinidad and Tobago 39 70 5.0 3.3 12 8 97 Gabon 26 47 8.3 6.0 57 26 98 Estonia . . 72 . . . . . 99 Portugal 26 34 2.6 1.4 48 16 44 46 II 16 100 Oman 5 II 8.0 8.3 41 4 . . . . . lOt PuertoRico 58 75 3.1 1.9 53 39 46 54 24 40 102 Korea,Rep. 41 73 5.3 3.5 36 26 74 69 24 50 l03 Greece 53 63 1.9 1.3 54 34 59 55 28 34 104 SaudiArabia 49 78 8.4 6.1 17 13 23 29 9 23 105 Yugoslavia 35 . . 3.6 2.8 12 7 II 12 3 7 Low- and middle-income 25 w 46w 3.7 w 6.3 w 15w 6w 41 w 33 w 10 w 13 w Sub-Saharan Africa 16 w 29 w 5.8 w 5.8 w 33 9w 30 w 30 w 4w 9w East Asia & Pacific 19w 52 w 3.2 w 11.1 w lOw 4w 47 w 30 w 9w 10 H' South Asia 19 w 26 w 4.1 w 3.9 w 8w 2w 35 w 35 w 6w 9w Europe and Central Asia 44w 64w 16 w 9w 31 w 28 w 12w 16 H' Middle East & N. Africa 41 w 55 w 4.5 w 4.5 w 26 w 14 w 43 w 41 w 17 w 22 H' Latin America & Caribbean 57 w 72 w 3.7 2.9 w 24 w 16w 44w 44w 24w 33 w Severely indebted 54w 68w 3.7w 3.0w 21w 14w 41 w 42w 21 w 29w High-income economies 74w 77w 1.1w 0.8w 12w 9w 38w 37w 27w 29w OECI) members 74w 77w 1.0w 0.8w 11w 7w 37w 36w 27w 28w 106 Ireland 52 57 2.2 0.5 46 26 . . . . . 107 tlsrael 84 92 3.2 2.3 12 II 43 45 34 41 108 NewZealand 81 84 1.4 0.8 12 10 . . . . . 109 Spain 66 79 2.0 1.1 17 13 26 28 16 22 ItO tHongKong 90 94 2.6 1.5 101 95 90 99 81 93 Ill tSingapore tOO l00 2.0 1.7 101 101 73 100 73 100 112 United Kingdom 89 89 0.1 0.2 14 13 33 26 28 23 113 Australia 85 86 1.6 1.5 2 1 60 59 50 51 114 Italy 64 69 0.9 0.6 8 5 42 37 26 25 115 Netherlands 86 89 1.1 0.6 8 7 18 16 16 14 116 Belgium 94 97 0.3 0.3 10 10 . . . . . 0 117 Austria 52 59 0.7 0.9 47 27 51 47 26 28 118 France 71 74 0.9 0.6 20 15 30 26 20 19 119 Canada 76 77 1.2 1.2 4 3 37 39 27 30 120 United States 74 75 1.0 1.1 2 1 49 48 35 36 121 Germany 80 . . 0.3 . . . . I 19 15 IS 13 122 Denmark 80 87 0.9 0.4 31 27 38 31 29 27 123 Finland 50 60 2.1 0.4 34 20 27 34 12 20 124 Norway 65 75 1.3 1.0 21 16 . . . 0 0 125 Sweden 81 84 0.6 0.4 23 19 17 23 13 20 126 Japan 7l 77 1.8 0.6 19 15 37 36 25 27 127 Switzerland 55 60 0.4 II 7 4 World 35 w 51 w 2.7 w 4.7 w 14 w 6w 40 w 34 w 14 w 16 w Fuel exporters 35 w 52 w 5.6 w 4.9 w 25 w 13 w 30 w 31 w 10 w 16 w 299 Table 32. Women in development Health and welfare Education Percentage of cohort Under-5 mona/itt rate Life expectancy at birth (cears) Maternal persisting to grade 4 Females per 1(X) males (per / (XX) (tie births) morialir (per Secondarra Female Male Female Male Primary Female Male /C0.(kX) l:i'e 1991 /991 1970 /991 /970 /99/ births) /988 1970 /986 1970 /986 /970 /990 1970 /990 Low-income economies 96w 104w 54w 58w 53w 61w 308w 78 w 65 w China and India 75w 80w 57w 60w 57w 64w 115w .. 79 w 65 w Other low-income 135w 148w 47w 57w 46w 54w 587 w 65 w 66 w 74 w 70 w 61 w 76 w 44w 66w I Mozambique 265 294 42 48 39 45 76 . . 61 2 Tanzania 153 171 47 49 44 46 342 82 90 88 89 65 98 38 74 3 Ethiopia 185 204 44 50 43 47 57 56 56 56 46 64 32 67 4 Uganda 175 95 SI 47 49 46 550 65 . . 31 5 Bhutan 200 88 41 49 39 47 1,305 5 59 3 4! 6 Guinea-Bissau 236 262 36 39 35 38 43 56 62 53 7 Nepal 139 125 42 53 43 54 833 . . . . . 18 47 16 8 Burundi 169 189 45 50 42 46 47 84 45 84 49 84 17 57 9 Chad 197 219 40 49 37 46 77 . . 81 34 44 9 22 10 Madagascar 156 174 47 52 44 50 333 65 . . 63 . . 86 97 70 99 II Sierra Leone 34I 377 36 45 33 40 67 70 40 56 12 Bangladesh 136 130 44 52 46 53 60 . 43 . 43 47 81 . . 49 13 Lao PDR I53 I72 42 52 39 49 56! 59 77 36 66 14 Malawi 185 205 41 45 40 44 350 55 67 60 72 59 8/ 36 54 IS Rwanda 209 234 46 48 43 45 300 63 76 65 75 79 99 44 56 16 Mali ISO 205 41 50 40 47 2.325 52 68 89 75 55 58 29 48 17 BurkinaFaso 189 209 42 50 39 46 8/0 71 86 68 84 57 62 33 50 18 Niger 303 337 40 48 37 44 75 93 74 78 53 57 35 42 19 India 125 123 49 60 50 60 45 60 71 39 55 20 Kenya 97 113 52 61 48 57 84 78 84 76 71 95 42 78 21 Nigeria 77 95 43 53 40 50 800 64 66 59 76 49 74 22 China 37 48 63 71 61 67 /15 76 . . 81 86 . . 72 23 Haiti 45 164 49 56 46 53 600 0) -- .. 96 24 Benin 57 175 45 52 43 49 /61 71 75 . . 45 . . 44 37 25 Central African Rep. 122 136 45 50 40 45 67 81 67 85 49 63 20 38 26 Ghana 122 140 SI 57 48 53 /000 77 82 75 82 35 63 27 Pakistan 139 137 47 59 49 59 270 56 is') 36 52 25 41 28 Togo I3I 149 46 56 43 52 85 78 86 45 65 26 34 29 Guinea 2I5 239 37 44 35 44 1,247 . . 77 87 46 46 26 31 30 Nicaragua 59 72 55 68 52 64 300 48 62 45 59 101 104 89 138 31 Sn Lanka 19 25 66 74 64 69 80 94 97 73 99 89 93 101 105 32 Mauritania 188 209 41 49 38 45 8(X) . . 83 . . 83 39 69 13 45 33 Yemen, Rep. 148 I66 42 52 41 52 330 . . . . . . . . . . . . . 34 Honduras 54 66 55 68 51 63 22/ . . . . . . . . 99 98 79 35 Lesotho 146 167 50 58 48 55 220 87 87 70 76 ISO 121 Ill 149 36 Indonesia 102 120 49 61 46 58 450 67 82 89 99 84 93 59 82 37 Egypt. Arab Rep. 82 96 52 62 50 60 85 . 93 61 80 48 76 38 Zimbabwe 50 63 52 62 49 59 77 74 81 80 81 79 99 63 88 39 Sudan 156 76 43 53 4l 50 61 75 40 80 40 Zambia I66 86 48 50 45 47 93 99 80 9/ 49 59 Middle-income economies 44w 54w 62w 71w 58w 65w 107w 78w 87w 76w 90w 86w 91w 94w 104w Lower-middle-income 50 w 60w 61w 69w 57w 64w 111w 79w 87w 80w 88w 80w 90w 89w 104w 41 Bolivia 117 127 48 61 44 57 37! . . . . . . . . 69 90 64 42 Cbte divoire 144 163 46 53 43 50 . . 77 83 83 88 57 71 27 45 43 Senegal 140 160 44 49 42 46 .. .. 90 .. 94 63 72 39 51 44 Philippines 53 68 59 67 56 63 74 . . 85 . . 84 . 94 . 45 PapuaNewGuinea 67 81 47 56 47 55 7(X) 76 . . 84 . . 57 80 37 62 46 Cameroon 112 130 46 57 43 54 59 85 58 86 74 85 36 68 47 Guatemala 76 84 54 67 51 62 33 73 79 . . 65 48 Dominican Rep. 66 72 61 69 57 65 300 55 13 00 OR 49 Ecuador 56 62 60 69 57 64 156 69 70 76 50 Morocco 66 79 53 65 50 61 78 80 83 81 51 66 40 69 SI Jordan 30 33 . . 70 66 90 97 92 99 78 94 53 96 52 Tajikistan 60 66 .. 72 67 39 53 Peru 62 76 56 66 52 62 165 85 . . 74 54 El Salvador 46 50 60 68 56 63 /48 61 . 62 . . 92 98 77 95 55 Congo 59 177 49 54 43 49 86 90 89 98 78 87 43 72 56 Syrian Arab Rep. 37 47 57 69 54 65 143 92 93 95 95 57 87 36 71 57 Colombia 23 29 63 72 59 66 200 57 74 SI 72 101 98 73 100 58 Paraguay 38 46 67 69 63 65 300 70 77 7/ 77 89 93 91 102 59 Uzbekistan 47 59 73 66 43 60 Jamaica 16 20 70 76 66 71 115 100 . 98 100 99 103 61 Romania 28 38 71 73 67 67 90 . 89 97 106 151 174 62 Namibia 85 97 49 60 47 56 08 . . 127 63 Tunisia 40 51 55 68 54 67 /27 . . 91 64 85 38 77 64 Kyrgyzstan 45 58 . . 70 . . 62 43 . . 65 Thailand 30 40 61 72 56 66 37 71 . . 69 88 95 69 97 66 Georgia 18 23 . . 77 . . 69 55 Noic: For data comparability and coverage. sec the technical notes. Figures in italics are for years other than those specified. 300 Health and welfare Education Percentage of co/tort Under-S mona/it raze Life expectancY at birth (years) Maternal persisting to grade 4 Females per 1(X) males (per I,CWX) live births) mortality (per Secondarru Female Male Female Male Primary Female Male 100.(XX) live 1991 /991 1970 /99! 1970 1991 births) /988 1970 1986 1970 /986 /970 /990 /970 1990 67 Azerbaijan 34 45 75 - 67 29 . . . . . . - 68 Turkmenistan 68 83 . 70 . 62 55 - . . . S S - - S - S - - 69 Turkey 70 77 59 70 55 64 146 76 98 81 98 73 89 37 63 70 Poland IS 21 74 75 67 67 . - 99 . - 97 . . 93 95 251 266 71 Bulgaria 18 23 74 75 69 68 9l 9l 100 93 94 93 . - 198 72 Costa Rica 13 16 69 78 65 74 18 93 91 91 90 96 94 III 103 73 Algeria 77 85 54 67 52 65 . . 90 95 95 97 60 SI 40 79 74 Panama 24 28 67 75 64 71 60 97 88 97 85 92 93 99 103 75 Armenia 24 30 . 75 . . 68 35 . . . . - - - . - S - 76 Chile 18 22 66 76 59 68 40 86 . . 83 - - 98 95 130 115 77 Iran. Islamic Rep. 83 91 54 65 55 65 120 75 92 74 93 55 86 49 74 78 Moldova 24 32 - . 72 - . 65 34 . . . - . . - . - - . . - - 79 Ukraine 18 26 74 75 67 66 33 - . . . . . . - 96 . . 127 - 80 Mauritius 22 28 65 73 60 67 99 97 99 97 99 94 98 66 100 81 Czechoslovakia 12 17 73 76 67 68 . . 96 97 98 97 96 97 183 132 82 Kazakhstan 33 44 . 73 . . 64 53 . . . . . . - - - - - - 83 Malaysia IS 21 63 73 60 68 26 . . - 88 95 69 104 Upper-middle-income 36 w 46 w 64 w 72 w 59 w 65 w 104 w 75 w 70 w 94 w 94 w 95 w 101 w 102 w 84 Botswana 36 44 51 70 48 66 . . 97 96 90 97 113 107 88 114 85 South Africa 65 79 56 66 50 59 . . . . . . - - 98 - - 95 86 Lithuania IS 21 75 76 67 65 29 . . . . . - - . . . - . . 87 Hungary 17 23 73 74 67 66 . - 90 97 99 97 93 95 202 198 88 Venezuela 35 44 68 73 63 67 55 84 91 61 81 99 99 102 137 89 Argentina 28 32 70 75 64 68 140 92 . - 69 . - 98 103 156 90 Uruguay 21 25 72 77 66 70 36 - . 98 - - 96 91 95 129 - - 91 Brazil 64) 73 61 69 57 63 140 56 . - 54 - . 99 . . 99 - - 92 Mexico 38 50 64 73 60 67 200 - - 73 - - 94 92 94 - - 92 93 Belarus IS 21 76 76 68 66 25 . - - - - . . - . - . - 94 Russian Federation 2l 29 - . 74 - - 64 49 . - - . . - 95 Latvia 17 23 - - 75 - 64 57 - - - - - . - - - - - - - 96 TrinidadandTobago 21 25 68 74 63 69 89 78 - 74 - - 97 97 113 102 97 Gabon 144 163 46 55 43 52 . - 73 80 78 78 91 43 98 Estonia 13 19 74 75 66 65 41 . . . . . . . - - - - 99 Portugal II IS 71 77 64 70 . 92 . . 92 - - 95 9! 98 116 100 Oman 33 43 49 71 46 67 . . . . 97 1(X) l6 89 82 101 Puerto Rico 16 20 75 80 69 72 21 . . . . . . . - . . - . . - 102 Korea, Rep. 16 22 62 73 58 67 26 96 100 96 100 92 94 65 87 103 Greece 12 14 74 80 70 75 . . 97 99 96 99 92 94 98 103 104 SaudiArabia 33 44 54 71 51 68 . 93 91 - 46 84 16 79 105 Yugoslavia 19 25 70 76 65 70 . - 91 - 99 - 91 94 86 98 Low- and middle-income 80 w 89 w 56 w 63 w 54 w 62 w 238 w 61 w 76 w 64 w 80 w 70 w 81 w 60 w 73 w Sub-Saharan Africa 167 w 186 w 45 w 52 w 42 w 49 w 686 w 66 w 71 w 69 w 72 w 60 w 76 w 40 w 67 iv East Asia & Pacific 46 w 58 w 60 w 66 w 58 w 66 w 195 w 78 w -. 82 w 88 w 75 H South Asia 129 w 127 w 48 w 59 w 50 w 59 w 444 w 45 w 48 w 55 w 69 w 38 w 54 H Europe and Central Asia 28 w 35 w 69 w 74 w 64 w 66 w 60 w 90 w 97 w 92 w 98 w 89 w 94 w 137 w 143 iv Middle East & N, Africa 73 w 84 w 54 w 65 w 52 w 63 w 151 w 83 w 90 w 87 w 92 w 54 w 79 w 41 w 72 iv Latin America & Caribbean 48 w 58 w 63 w 7l w 58 w 65 w 162 iv' 66 w 76 w 60 w 85 w 96 w 97 w 101 w 103 H Severely indebted 55w 66w 62 w 69w 58w 64w 171 w 75w 80w 73w 89w 87w 88w 109w 115 iv High-income economies 8w 11w 75w 80w 68w 73w 95w 98w 93w 97w 96w 95w 95w 100w OECI) members 8w 11w 75w 80w 68w 73w 95w 98w 93w 97w 96w 95w 95w 100w 106 Ireland 9 II 73 78 69 72 . . . - 98 - - 97 96 96 124 /01 107 tlsrael 10 14 73 78 70 74 . 96 97 96 97 92 98 131 1l6 108 New Zealand 9 13 75 79 69 73 . . - 98 - - 98 94 94 94 98 109 Spain 9 II 75 80 70 74 . . 76 98 76 97 99 93 84 102 110 tHong Kong 5 7 73 80 67 75 4 94 . . 92 . . 90 - - 74 Ill tSingapore 7 9 70 77 65 72 10 99 1(X) 99 100 88 90 103 1(X) 112 United Kingdom 8 10 75 79 69 72 . . . . . . . . . . 95 96 94 96 113 Australia 8 10 75 80 68 73 . . 76 97 74 94 94 95 91 99 114 Italy 10 12 75 81 69 74 . . . . . . . . . . 94 95 86 97 115 Netherlands 8 10 77 80 71 74 . . 99 . - 96 - . 96 99 91 109 116 Belgium 10 12 75 80 68 73 . . . - 87 - - 85 94 97 87 - - 117 Austria 9 II 74 80 67 73 . . 95 99 92 98 95 95 95 94 118 France 8 10 76 81 68 73 . . 97 - . 90 - - 95 94 107 106 119 Canada 8 10 76 81 69 74 . 95 97 92 93 95 93 95 96 120 United States 9 l3 75 79 67 72 - . - . . . . . 95 95 . . - 121 Germany' 8 10 74 79 67 73 . . 97 99 96 97 96 96 93 98 122 Denmark 9 II 76 78 71 72 . . 98 100 96 l00 97 96 102 106 123 Finland 7 9 74 79 66 73 . . . 98 . . 99 90 95 112 III 124 Norway 9 II 77 80 71 74 . . 99 . 98 lOS 95 97 lOS 125 Sweden 7 9 77 8! 72 75 . . 98 . . 96 96 95 92 109 126 Japan 5 7 75 82 69 76 . 100 100 100 100 96 95 101 99 127 Switzerland 8 10 76 81 70 74 . . 94 - - 93 98 96 93 99 World 69w 77w 60w 65w 57w 64w 237w 67w 78w 70w 82w 77w 84w 68w 76w Fuelexporters 114 iv' 127w SOw 61w 48w 58w 492w 75w 88w 75w 88w 60w 81w 51w 80w a. See the technical notes. b. Data refer to the Federal Republic of Germany before unification. 301 Table 33. Forests, protected areas, and water resources Forest area (thousands of sq. IS, Freshwater resources: annual withdrawal (/970-89) Nationally protected areas (1992) Annual deforestation Per capita (cubic meters) As a As a Total area /981-85 Area percentage Total percentage of Industrial (thousands of total (cubic total water and Total Closed Total Closed ofsq. km) Number area kilometers) resources Total Domestic agricultural I,ow-incotue economies China and India Other low-income I Mozambique 154 9 1.2 0.1 0.0 0.0 0.8 I 53 13 40 2 Tanzania 420 14 3.0 . . 130.0 28 13.8 0.5 I 36 8 28 3 Ethiopia 272 44 0.9 0.! 25.3 II 2. I 2.2 2 49 5 44 4 Uganda 60 8 0.5 0.1 18.7 32 7.9 0.2 0 20 6 14 5 Bhutan 21 2! 0.0 0.0 9.1 5 19.3 0.0 0 14 5 9 6 Guinea-Bissau 21 7 0.6 0.2 0.0 0 0.0 0.0 0 II 3 8 7 Nepal 21 19 0.8 0.8 11.1 12 7.9 2.7 2 151 6 145 8 Bumndi 0 0 0.0 0.0 0.9 3 3.2 0.1 3 20 7 13 9 Chad 135 5 0.8 . . 29.8 7 2.3 0.2 0 34 5 29 10 Madagascar 132 103 1.6 1.5 11.1 36 1.9 16.3 41 1.642 16 1.626 II Sierra Leone 21 7 0.1 0.1 0.8 2 1.1 0.4 0 96 7 89 12 Bangladesh 9 9 0.! 0.1 1.0 8 0.7 22.5 l' 211 6 205 13 La0PDR 136 84 1.3 1.0 0.0 0 0.0 1.0 0 260 21 239 14 Malawi 43 2 1.5 . . 10.6 9 8.9 0.2 2 20 7 13 IS Rwanda 2 I 0.! 0.0 3.3 2 12.4 0.2 2 23 6 17 16 Mali 73 5 0.4 . . 40.! II 3.2 1.4 2 162 3 159 17 Burkina Faso 47 3 0.8 0.0 26.6 12 9.7 0.2 I 18 5 3 IS Niger 26 I 0.7 0.0 97.0 6 7.7 0.3 I' 41 9 32 19 India 640 378 050 . 138.4 332 4.2 380.0 18d 612 18 594 20 Kenya 24 II 0.4 0.2 34.7 36 6.0 1.1 7 50 14 37 2! Nigeria 148 60 4.0 3.0 28.7 21 3.1 3.6 I' 37 II 26 22 China 1,150 978 0.0 . . 286.5 402 3.0 460.0 16 462 28 434 23 Haiti 0 0 0.0 0.0 0.1 3 0.3 0.0 0 7 2 5 24 Benin 39 0 0.7 0.0 8.4 2 7.5 0.1 0 26 7 19 25 Central African Rep. 359 36 0.6 0.1 61.1 13 9.8 0.1 0 25 5 20 26 Ghana 87 17 0.7 0.2 10.7 8 4.5 0.3 35 12 23 27 Pakistan 25 22 0.1 0.1 36.6 53 4.6 53.4 33d 2.053 21 2.032 28 Togo 17 3 0.1 0.0 6.5 II 11.4 0.1 28 17 II 29 Guinea 107 21 0.9 0.4 1.6 3 0.7 0.7 0 40 14 126 30 Nicaragua 45 45 1.2 1.2 9.5 21 7.3 0.9 370 93 278 31 Sri Lanka 17 17 0.6 0.6 7.8 43 11.9 6.3 IS 503 10 493 32 Mauritania 6 0 0.0 17.5 4 0.7 lO 494 0.1 1.7 59 435 33 Yemen, Rep. 0 0 0.0 . . 0.0 0 0.0 1.5 147 179 7 172 34 Honduras 40 38 0.9 0.9 5.0 35 4.4 1.3 I 510 20 490 35 Lesotho 0 0 . . . . 0. I I 0.2 0. I I 31 7 24 36 Indonesia 1,169 1,139 10.00 . 93.4 186 10.2 16.6 95 12 83 37 Egypt, Arab Rep. 0 0 . . . . 8.0 13 0.8 56.4 97d 1,213 85 1.128 38 Zimbabwe 198 2 0.8 0.0 30.7 25 7.9 1.2 5 138 19 119 39 Sudan 477 7 5.0 0.0 93.6 14 3.7 18.6 .092 II 1.081 40 Zambia 295 30 0.7 0.4 63.6 20 8.5 0.4 0 86 54 32 Middle-income economies Lower-middle-income 41 Bolivia 668 440 1.2 0.9 98.6 27 9.0 1.2 0 184 IS 166 42 Côted'lvoire 98 45 2.6a 19.9 12 6.2 0.7 I 66 5 51 43 Senegal 110 2 0.5 . . 21.8 9 11.1 1.4 4d 202 10 192 44 Philippines 95hi 95 1.40 14a 5.7 27 1.9 29.5 9 693 125 568 45 Papua New Guinea 382 342 0.2 0.2 0.3 6 0.1 0.1 0 28 8 20 46 Cameroon 233 165 l.9a 1.00 20.5 14 4.3 0.4 0 37 17 20 47 Guatemala 45 44 0.9 0.9 8.3 17 7.6 0.7 I 139 13 126 48 Dominican Rep. 6 6 0.0 0.0 9.7 18 19.9 3.0 IS 443 22 421 49 Ecuador 147 143 3.4 3.4 107.5 18 37.9 5.6 2 567 40 527 50 Morocco 32 15 0.1 . . 3.6 10 0.8 11.0 37 499 30 469 SI Jordan I 0 . . . . 1.0 8 1.1 0.5 41d 173 50 123 52 Tajikistan . . . . . . . . 0.9 3 0.6 . . . . 53 Peru 706 697 2.7 2.7 27.0 22 2.1 6.1 15 301 57 244 54 El Salvador I I 0.1 0.1 0.2 5 0.9 1.0 5 245 17 228 55 Congo 213 213 0.2 0.2 11.8 10 3.4 0.0 0 19 12 7 56 Syrian Arab Rep. 2 I 0.0 . . 0.0 0 0.0 3.3 9d 434 30 404 57 Colombia 517 464 8.9 8.2 90.5 41 7.9 0d 172 71 101 58 Paraguay 197 41 4.5 . 12.0 14 3.0 0.4 0" 110 17 94 59 Uzbekistan . . . . . . . . 2.4 10 0.5 . . . . 60 Jamaica I I 0.0 0.0 0.0 I 0.1 0.3 4 159 II 148 61 Romania 67 63 . . . . 10.9 40 4.6 25.4 I2d 1,144 92 1,052 62 Namibia 184 . . 0.3 . . 103.7 II 12.6 0.1 2 84 5 79 63 Tunisia 3 2 0.! . . 0.4 6 0.3 2.3 52d 317 41 276 64 Kyrgyzstan . . . . . . . . 2.0 5 1.0 . . . . 65 Thailand 157 92 2.4a I .6 56.6 92 11.0 31.9 l8 600 24 576 66 Georgia 1.9 15 2.7 Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified. 302 Forest area (thousands ofsq. km) Freshwater resources: annual withdrawal (1970-89) Nationally protected areas (1992) Annual deforestation Per capita (cubtc meters) Area As a As a Total area /981-85 percentage percentage of Industrial (thousands Total ofsquare of total (cubic iota! water and Total Closed Total Closed kilometers) Number area kilometers) resources Total Domestic agricultural 67 Azerbaijan . . . . I .8 Il 2.0 68 Turkmenistan . . . . . . . . 11.1 8 2.5 . . 0 69 Turkey 202 89 . . . . 2.7 18 0.3 23.8 l8' 434 104 330 70 Poland 87 86 . . . . 22.4 80 7.2 16.8 30 472 76 396 7(1 71 Bulgaria 37 33 . . . . 2.6 50 2.4 14.2 1,600 112 1,488 72 Costa Rica 18 16 0.4 0.4 6.3 27 12.4 1.4 I 780 31 749 16d 73 Algeria 18 IS 0.4 . . 127.0 18 5.3 3.0 160 35 125 74 Panama 42 42 0.4 0.4 13.2 14 17.2 1.3 1 744 89 655 75 Armenia . . . . . . . . 2.2 4 7.4 . . . . 76 Chile 76 76 0.5 . . 137.2 65 18.1 16.8 4 1,623 97 1,526 77 Iran, Islamic Rep. 38 28 0.2 . . 79.8 62 4.8 45.4 39 1,362 54 1,308 78 Moldova . . . . . . . . 0.0 0 0.0 . . . . . 79 Ukraine . . . . . . . . 4.6 17 0.8 . . . . 80 Mauritius 0 0 0.0 0.0 0.0 0 0.0 0.4 16 410 66 344 81 Czechoslovakia 46 44 . . . 20.6 65 16.1 5.8 6' 379 87 292 82 Kazakhstan . . . . . . . . 8.4 8 0.3 . . . . Malaysia 210b 210 2.7a 14.9 48 4.5 9.4 768 177 591 83 . 2 Upper-middle-income 84 Botswana 326 0 0.2 . . 102.3 9 17.6 0.1 l 100 5 95 85 SouthAfrica 3 3 . . . . 73.9 229 6.1 9.2 18 410 66 344 86 Lithuania . . . . . . . 0.0 0 0.0 . . . . 87 Hungary 16 16 . . . . 5.8 54 6.2 5.4 5" 502 45 457 88 Venezuela 339 319 2.5 1.3 275.3 104 30.2 4.1 0 387 166 221 89 Argentina 445 445 I .8 . 93.9 112 3.4 27.6 3d 1,042 94 948 90 Uruguay 5 5 . 138a.b . . . 0.3 8 0.2 0.7 I' 241 14 227 91 Brazil 5,145 3575 . 257.6 186 3.0 35.0 l' 248 107 141 92 Mexico 484 463 l0.Oa . 98.1 60 5.0 54.2 15 875 53 823 93 Belarus . . . . . . . . 2.4 4 1.1 . . . . . 94 Russian Federation . . . . . . . . 200.3 75 1.2 . . . 95 Latvia . . . . . . . . 1.7 21 2.6 . . . . 96 Trinidad and Tobago 2 2 0.0 0.0 0.2 13 3.9 0.2 3 148 40 108 97 Gabon 206 205 0.2 0.2 10.5 6 3.9 0.1 0 57 41 16 98 Estonia . . . . . . 3.2 36 7.1 . . . . 99 Portugal 30 26 . . . . 5.5 22 6.0 10.5 16 1,075 161 914 00 Oman 0 0 . . . 0.5 2 0.3 0.4 22 561 17 544 101 Puerto Rico . . . . . . . . 0.4 29 4.0 . . . . 102 Korea, Rep. 49 49 . . . . 7.6 26 7.6 10.7 17 299 33 266 103 Greece 58 25 . . . . 1.0 18 0.8 7.0 l2' 720 58 662 104 Saudi Arabia 2 0 . . . . 212.0 9 9.9 2.3 106 321 144 177 3d 105 Yugoslavia 105 91 . . . . 7.9 61 3.1 8.8 393 63 330 Low- and middle-income Sub-Saharan Africa East Asia & Pacific South Asia Europe aird Central Asia Middle East & N. Africa Latin America & Caribbean Severely indebted High-income economies OECI) members 106 Ireland 4 3 . . . . 0.4 6 0.6 0.8 2 235 38 l97 107 jIsrael I I 2.1 21 10.0 1.9 86d 441 71 370 . . . . 108 New Zealand 95 72 . . . . 29.0 124 10.7 1.9 0 585 269 316 109 Spain 108 69 . . . . 35.0 161 6.9 45.9 4I' 1,184 142 1,042 110 tHong Kong . . . . . . 0.4 12 36.3 . . . . Ill tSinapore 0 0 . . . . 0.0 1 4.4 0.2 32 84 38 46 112 UnttedKingdom 22 20 . . . . 46.3 131 18.9 14.5 12 253 51 202 113 Australia 1,067 417 . . 812.4 733 10.6 17.5 5 1,280 832 448 114 Italy 81 64 . . . 20.1 143 6.7 56.2 3051 984 138 846 115 Netherlands 4 3 . . . 3.5 67 9.4 14.5 l6' 993 50 943 116 Belgium 8 7 . . . 0.8 3 2.5 9.0 72d 917 101 816 2d 117 Austria 38 38 . . . 21.2 187 25.3 2.1 279 53 226 118 France 151 139 . . . . 53.6 79 9.7 43.7 24' 783 125 658 119 Canada 4,364 2,641 . . . . 494.5 411 5.0 43.9 2 1,684 303 1,381 120 UnitedStates 2,960 2,096 I.6a . 984.6 937 10.5 467.0 19 1,952 234 1,718 121 Germany 72C 70' . . . . 87.8 472 24.6 44.6' 28c.d 729 73C 656' 122 Denmark 4.1 9.5 9d 228 160 5 5 . . . . 65 1.2 68 311 123 Finland 232 199 . . . . 8.1 34 2.4 3.0 605 73 532 124 Norway 87 76 . . . 14.9 80 4.6 2.0 0 490 98 392 125 Sweden 278 244 . . . . 29.2 189 6.5 3.0 2" 356 128 228 126 Japan 253 239 . . . . 46.7 685 12.3 89.3 16 733 125 608 127 Switzerland II 9 . . . . 7.5 112 18.2 1.1 2" 170 39 131 World Fuel exporters a. Data are for the periods as follows: Tanzania 1989, India 1983-87, Indonesia 1982-90, Cole d'lvoire 1981-86. Philippines 1981-88. Cameroon 1976-86. Paraguay 1989-90, Thailand 1985-88. Costa Rica 1973-89, Malaysia 1979-89. Argentina 1980-89, Brazil (Legal Amazon only) 1989-90, Mexico 1981-83. United States 1977- 87. b. See the technical notes for alternative estimates. c. Data refer to the Federal Republic of Germany before unification. d. Total water resources include nver flows from other countries in addition to internal renewable resources. 303 Table la. Basic indicators for other economies GNPper capitaa Average Life Area annual Average annual Adult illiteracy expectancy Population (thousands growth rate rate of inflat,ona (percent) at birth (thousands) of square Dollars (percent) (p'rt'hlt) years) Female Total mid-1991 kilometers) 1991 1980-91 /970-80 1980-9/ 199/ /990 1990 I Cambodia 8,790 181 200 . 50 78 . . 65 2 Equatorial Guinea 427 28 330 2.8 -0.9 47 . . 63 50 3 Gambia,The 902 II 360 -0.1 10.6 18.2 44 84 73 4 SàoloméandPnncipe 118 I 400 -3.3 4.0 21.5 67 5 Guyana 802 215 430 -4.5 9.8 35.0 65 5 4 6 Maldives 221 b 460 6.7 62 7 Comonss 492 2 500 -1.0 . . 56 8 Solomon Islands 325 29 690 3.5 8.4 12.4 65 9 Kiribati 73 I 720 . . 10.6 5.4 56 10 Cape Verde 380 4 750 2.3 9.4 9.4 67 . II Western Samoa 161 3 960 . . . . 11.6 66 12 Swaziland 828 17 1,050 3.1 12.3 10.3 57 13 Vanuatu 151 12 1,150 5.0 65 14 Tonga 100 I 1,280 . . . . . . 67 IS St.Vincent 108 b 1,730 5.2 13.8 4.4 71 16 Fiji 741 18 1,930 -0.2 12.8 6.1 71 17 Belize 194 23 2.010 2.5 8.7 2.9 68 18 Grenada 91 b 2,180 . . . . . 70 19 Dominica 72 I 2,440 4.7 16.8 6.0 72 20 St. Lucia 153 I 2,490 . . . 72 21 Suriname 457 163 3,630 -4.5 11.8 9.0 68 5 5 22 St. Kjus and Nevis 39 b 3,960 5.8 7.2 70 23 Antigua and Barbuda 80 b 4,430 3.8 . . 6.9 74 24 Seychelles 69 b 5,110 3.2 16.9 3.5 71 25 Barbados 258 b 6,630 1.3 13.5 5.2 75 26 Bahrain 516 7,130 -3.8 -0.3 1 . . 69 31 23 27 Malta 357 b 7,280 3.8 4.2 2.1 76 28 Cyprus 710 9 8,640 4.9 . . 5.5 77 29 Bahamas, The 259 14 11,750 1.3 6.4 5.9 69 30 Qatar 506 II 14,770 -12.2 . . 70 31 UnitedArabEmirates 1,629 84 20,140 -6.3 . . 1.1 72 32 Iceland 258 103 23,170 1.8 35.1 30.0 78 33 Luxembourg 385 3 31,780 3.5 6.9 4.2 75 . 34 Afghanistan 20,979 652 f . . 43 86 71 35 Albania 3,301 29 e -0.4 73 36 American Samoa 40 b c 37 Andorra 58 b c . . 38 Angola 9,461 1,247 e . . 46 72 58 39 Aruba 61 b d 40 Bermuda 61 b c . . 8.4 41 Bninei 265 6 c 74 42 Channel Islands 145 b c 77 43 Cuba 10,736 III e . . 76 44 Djibouti 452 23 e . 49 45 Faeme Islands 47 1 c 46 Fed. Sts. of Micmnesia 105 b e 47 French Guiana 123 90 d 48 French Polynesia 202 4 c 68 49 Gibraltar 30 b d 50 Greenland 56 342 c SI Guadeloupe 395 2 c . 74 52 Guam 145 I c . . . 72 . 53 Iraq 18,578 438 d 17.9 10.3 65 51 40 54 Isle of Man 70 1 c . 55 Korea,Dem.Rep. 22,191 121 e . . 71 56 Kuwait 1,460 18 c 21.9 -2.7 75 33 27 57 Lebanon 3,708 10 e . 66 58 Liberia . 27 20 2,639 III f . . 9.2 . . 55 71 61 59 Libya 4,706 1,760 d 18.2 0.2 63 50 36 60 Macao 476 b e . . 73 61 Marshall Islands 48 b e 62 Martinique 363 I d 76 63 Mayotte 76 b' d 64 Mongolia 2,250 1,567 e 1.0 63 . . 65 Myanmar 42,758 677 f 59 28 19 66 Netherlands Antilles 192 c I . 77 67 New Caledonia 171 19 d . . 70 68 Reunion 602 3 d 72 69 San Marino 23 b c . . 70 Somalia 8,051 638 f 15.2 49.7 48 86 76 71 VietNam 67,679 332 f . . . 67 16 12 72 Virgin Islands (U.S.) 99 b c 12.5 3.9 . . . 74 73 Zaire 38.631 2,345 f 31.4 60.9 52 39 28 a. See the technical note for Table 1. b. Less than 500 square kilometers. c. GNP per capita estimated to be in the high-income range. d. GNP per capita estimated to be in the upper-middle-income range. e. GNP per capita estimated to be in the lower-middle income range. f. GNP per capita estimated to be in the low-income range. 304 Technical notes The main criterion for country classification is gross dated, some figuresespecially those relating to national product (GNP) per capita. With the addition current periodsmay be extrapolated. Several esti- of the recently independent republics of the former mates (for example, life expectancy) are derived from Soviet Union, the main tables now include country models based on assumptions about recent trends data for 127 economies, listed in ascending GNP per and prevailing conditions. Issues related to the re- capita order. A separate table, showing basic indica- liability of demographic indicators are reviewed in tors for economies with sparse data or with popula- the U.N. 's World Population Trends and Policies. tions of less than 1 million, Table la covers a further Readers are urged to take these limitations into ac- seventy-three economies. Other changes are outlined count in interpreting the indicators, particularly in the Introduction. when making comparisons across economies. Data reliability Base years Considerable effort has been made to standardize To provide long-term trend analysis, facilitate inter- the data, but full comparability cannot be ensured national comparisons and include the effects of and care must be taken in interpreting the indicators. changes in intersectoral relative prices, constant price Many factors affect availability and reliability; the sta- data for most economies are partially rebased to three tistical systems in many developing economies are base years and linked together. The year 1970 is the still weak, statistical methods, coverage, practices, base year for data from 1960 to 1975, 1980 for 1976 to and definitions differ widely among countries, and 1982, and 1987 for 1983 and beyond. These three pe- cross-country and cross-time comparisons involve riods are "chain-linked," to obtain 1987 prices complex technical problems that cannot be unequivo- throughout all three periods. cally resolved. For these reasons, while the data are Chain-linking is accomplished for each of the three drawn from the sources thought to be most authorita- subperiods by rescaling; this moves the year in which tive they should be construed only as indicating current and constant price versions of the same time trends and characterizing major differences among series have the same value, without altering the trend economies rather than offering precise quantitative of either. Components of gross domestic product measures of those differences. In particular, data is- (GDP) are individually rescaled and summed to pro- sues have yet to be resolved for the fifteen economies vide GDP and its subaggregates. In this process a of the former Soviet Union. Coverage is sparse, and rescaling deviation may occur between the constant the data are subject to more than the normal range of price GDP by industrial origin and the constant price uncertainty. GDP by expenditure. Such rescaling deviations are Most social and demographic data from national absorbed under the heading private consumption, etc. sources are drawn from regular administrative files, on the assumption that GDP by industrial origin is a although some come from special surveys or periodic more reliable estimate than GDP by expenditure. census inquiries. In the case of survey and census Because private consumption is calculated as a re- data, figures for intermediate years have to be inter- sidual, the national accounting identities are main- polated or otherwise estimated from the base refer- tained. Rebasing does involve incorporating in pri- ence statistics. Similarly, because not all data are up- vate consumption whatever statistical discrepancies 305 arise for expenditure. The value added in the services growth rates are calculated for two periods, 1970-80 sector also includes a statistical discrepancy, as re- and 1980-91, and are computed, unless otherwise ported by the original source. noted, by using the least-squares regression method. Because this method takes into account all observa- Summary measures tions in a period, the resulting growth rates reflect The summary measures are calculated by simple general trends that are not unduly influenced by ex- addition when a variable is expressed in reasonably ceptional values, particularly at the end points. To comparable units of account. Economic indicators exclude the effects of inflation, constant price eco- that do not seem naturally additive are usually com- nomic indicators are used in calculating growth rates. bined by a price-weighting scheme. The summary Details of this methodology are given at the begin- measures for social indicators are weighted by ning of the technical notes. Data in italics indicate population. that they are for years or periods other than those The World Development Indicators, unlike the specifiedup to two years earlier for economic indi- World Tables, provide data for, usually, two reference cators and up to three years on either side for social points rather than annual time series. For summary indicators, since the latter tend to be collected less measures that cover many years, the calculation is regularly and change less dramatically over short pe- based on the same country composition over time riods of time. and across topics. The World Development Indicators All growth rates shown are calculated from con- permit group measures to be compiled only if the stant price series and, unless otherwise noted, have country data available for a given year account for at been computed using the least-squares method. The least two-thirds of the full group, as defined by the least-squares growth rate, r, is estimated by fitting a 1987 benchmarks. As long as that criterion is met, least-squares linear regression trend line to the log- noncurrent reporters (and those not providing ample arithmic annual values of the variable in the relevant history) are, for years with missing data, assumed to period. More specifically, the regression equation behave like the sample of the group that does provide takes the form log X = a + bt + e where this is estimates. Readers should keep in mind that the pur- equivalent to the logarithmic transformation of the pose is to maintain an appropriate relationship across compound growth rate equation, X, = X,, (1 + r)t. In topics, despite myriad problems with country data, these equations, X is the variable, t is time, and a = and that nothing meaningful can be deduced about log Xo and b = log (1 + r) are the parameters to be behavior at the country level by working back from estimated; e is the error term. If b* is the least-squares group indicators. In addition, the weighting process estimate of b, then the average annual percentage may result in discrepancies between summed sub- growth rate, r, is obtained as [antilog (b*)] and group figures and overall totals. This is explained multiplied by 100 to express it as a percentage. more fully in the introduction to the World Tables. Table 1. Basic indicators Sources and methods Data on external debt are compiled directly by the For basic indicators for economies with sparse data or World Bank on the basis of reports from its develop- with populations of less than 1 million, see Table A.1. ing member countries through the Debtor Reporting Population numbers for mid-1991 are World Bank System. Other data are drawn mainly from the estimates. These are usually projections from the United Nations and its specialized agencies, the In- most recent population censuses or surveys; most are ternational Monetary Fund, and country reports to from 1980-91, and, for a few countries, from the the World Bank. Bank staff estimates are also used to 1960s or 1970s. Note that refugees not permanently improve currentness or consistency. For most coun- settled in the country of asylum are generally consid- tries, national accounts estimates are obtained from ered to be part of the population of their country of member governments through World Bank economic origin. missions. In some instances these are adjusted by The data on area are from the Food and Agriculture Bank staff to provide conformity with international Organization (FAQ). Area is the total surface area, definitions and concepts, consistency, and measured in square kilometers, comprising land area currentness. and inland waters. GNP per capita figures in U.S. dollars are calculated Growth rates according to the World Bank Atlas method, which is For ease of reference, only ratios and rates of described below. growth are usually shown; absolute values are gener- GNP per capita does not, by itself, constitute or ally available from other World Bank publications, measure welfare or success in development. It does notably the 1993 edition of the World Tables. Most not distinguish between the aims and ultimate uses 306 of a given product, nor does it say whether it merely tionally large margin from the rate effectively applied offsets some natural or other obstacle, or harms or to foreign transactions. This applies to only a small contributes to welfare. For example, GNP is higher in number of countries. For all other countries the Bank colder countries, where people spend money on calculates GNP per capita using the World Bank Atlas heating and warm clothes, than in balmy climates, method. where people are comfortable wearing light clothes in The Atlas conversion factor for any year is the aver- the open air. age of a country's exchange rate for that year and its More generally, GNP does not deal adequately exchange rates for the two preceding years, after ad- with environmental issues, particularly natural re- justing them for differences in relative inflation be- source use. The World Bank has joined with others to tween the country and the United States. This three- see how national accounts might provide insights year average smooths fluctuations in prices and ex- into these issues. The possibility of developing "sat- change rates for each country. The resulting GNP in ellite" accounts is being considered; such accounts U.S. dollars is divided by the midyear population for could delve into practical and conceptual difficulties, the latest of the three years to derive GNP per capita. such as assigning a meaningful economic value to Some sixty low- and middle-income economies suf- resources that markets do not yet perceive as fered declining real GNP per capita in constant prices "scarce" and allocating costs that are essentially during the 1980s. In addition, significant currency global within a framework that is inherently national. and terms of trade fluctuations have affected relative GNP measures the total domestic and foreign value income levels. For this reason the levels and ranking added claimed by residents. It comprises GDP (de- of GNP per capita estimates, calculated by the Atlas fined in the note for Table 2) plus net factor income method, have sometimes changed in ways not neces- from abroad, which is the income residents receive sarily related to the relative domestic growth perfor- from abroad for factor services (labor and capital) less mance of the economies. similar payments made to nonresidents who contrib- The following formulas describe the procedures for uted to the domestic economy. computing the conversion factor for year t: In estimating GNP per capita, the Bank recognizes that perfect cross-country comparability of GNP per i ps\ capita estimates cannot be achieved. Beyond the clas- (e2,) = - Ee_2 I - ) + e1 t-2 sic, strictly intractable index number problem, two obstacles stand in the way of adequate comparability. and for calculating per capita GNP in U.s. dollars for One concerns the GNP and population estimates year t: themselves. There are differences in national ac- counting and demographic reporting systems and in (Y) (Y / N ± e7_2,,) the coverage and reliability of underlying statistical where information among various countries. The other ob- stacle relates to the use of official exchange rates for = current GNP (local currency) for year converting GNP data, expressed in different national = GNP deflator for year currencies, to a common denominationconven- e = average annual exchange rate (local currency to the U.S. dollar) for year tionally the U.S. dollarto compare them across = midyear population for year countries. P = U.S. GNP deflator for year Recognizing that these shortcomings affect the comparability of the GNP per capita estimates, the Because of problems associated with the availabil- World Bank has introduced several improvements in ity of comparable data and the determination of con- the estimation procedures. Through its regular re- version factors, information on GNP per capita is not view of member countries' national accounts, the shown for some economies. Bank systematically evaluates the GNP estimates, fo- The use of official exchange rates to convert na- cusing on the coverage and concepts employed and, tional currency figures to U.S. dollars does not reflect where appropriate, making adjustments to improve the relative domestic purchasing powers of curren- comparability. As part of the review, Bank staff esti- cies. The U. N. International Comparison Programme mates of GNP (and sometimes of population) may be (ICP) has developed measures of real GDP on an in- developed for the most recent period. ternationally comparable scale, using purchasing The World Bank also systematically assesses the power of currencies (PPCs) instead of exchange rates appropriateness of official exchange rates as conver- as conversion factors. Table 30 shows the most recent sion factors. An alternative conversion factor is used PPC estimates. Information on the ICP has been pub- (and reported in the World Tables) when the official lished in four studies and in a number of other re- exchange rate is judged to diverge by an excep- ports. The most recent study is Phase VI, for 1990, a 307 part of which has already been published by the Or- other international agencies but more often collected ganization for Economic Cooperation and Develop- during World Bank staff missions. ment (OECD). World Bank staff review the quality of national ac- The ICP figures reported in Table 30 are prelimi- counts data and in some instances, through mission nary and may be revised. The United Nations and its work or technical assistance, help adjust national se- regional economic commissions, as well as other in- ries. Because of the sometimes limited capabilities of ternational agencies, such as the EC, the OECD, and statistical offices and basic data problems, strict inter- the World Bank, are working to improve the meth- national comparability cannot be achieved, especially odology and to extend annual purchasing power in economic activities that are difficult to measure, comparisons to all countries. However, exchange such as parallel market transactions, the informal sec- rates remain the only generally available means of tor, or subsistence agriculture. converting GNP from national currencies to U.S. GDP measures the total output of goods and ser- dollars. vices for final use produced by residents and nonresi- Average annual rate of inflation is measured by the dents, regardless of the allocation to domestic and growth rate of the GDP implicit deflator for each of foreign claims. It is calculated without making deduc- the periods shown. The GDP deflator is first calcu- tions for depreciation of "manmade" assets or deple- lated by dividing, for each year of the period, the tion and degradation of natural resources. Although value of GDP at current values by the value of GDP at SNA envisages estimates of GDP by industrial origin constant values, both in national currency. The least- to be at producer prices, many countries still report squares method is then used to calculate the growth such details at factor cost. International comparability rate of the GDP deflator for the period. This measure of the estimates is affected by differing country prac- of inflation, like any other, has limitations. For some tices in valuation systems for reporting value added purposes, however, it is used as an indicator of infla- by production sectors. As a partial solution, GDP es- tion because it is the most broadly based measure, timates are shown at purchaser values if the compo- showing annual price movements for all goods and nents are on this basis, and such instances are foot- services produced in an economy. noted. However, for a few countries in Tables 2 and 3, Life expectancy at birth indicates the number of years GDP at purchaser values has been replaced by GDP a newborn infant would live if prevailing patterns of at factor cost. mortality at the time of its birth were to stay the same The figures for GDP are U.S. dollar values con- throughout its life. Data are World Bank estimates verted from domestic currencies using single-year of- based on data from the U.N. Population Division, the ficial exchange rates. For a few countries where the U.N. Statistical Office, and national statistical offices. official exchange rate does not reflect the rate effec- Adult illiteracy is defined here as the proportion of tively applied to actual foreign exchange transactions, the population over the age of fifteen who cannot, an alternative conversion factor is used (and reported with understanding, read and write a short, simple in the World Tables). Note that this table does not use statement on their everyday life. This is only one of the three-year averaging technique applied to GNP three widely accepted definitions, and its application per capita in Table 1. is subject to qualifiers in a number of countries. The Agriculture covers forestry, hunting, and fishing as data are from the illiteracy estimates and projections well as agriculture. In developing countries with high prepared in 1989 by UNESCO. More recent informa- levels of subsistence farming, much agricultural pro- tion and a modified model have been used; therefore, duction is either not exchanged or not exchanged for the data for 1990 are not strictly consistent with those money. This increases the difficulty of measuring the published in last year's World Development contribution of agriculture to GDP and reduces the Indicators. reliability and comparability of such numbers. The summary measures for GNP per capita, life Industry comprises value added in mining; manu- expectancy, and adult illiteracy in this table are facturing (also reported as a separate subgroup); con- weighted by population. Those for average annual struction; and electricity, water, and gas. Value added rates of inflation are weighted by the 1987 share of in all other branches of economic activity, including country GDP valued in current U.S. dollars. imputed bank service charges, import duties, and any statistical discrepancies noted by national com- Tables 2 and 3. Growth and structure of production pilers, are categorized as services, etc. Partially rebased, chain-linked 1987 series in do- Most of the definitions used are those of the U.N. mestic currencies, as explained at the beginning of System of National Accounts (SNA), Series F, No. 2, the technical notes, are used to compute the growth Revision 3. Estimates are obtained from national rates in Table 2. The sectoral shares of GDP in Table 3 sources, sometimes reaching the World Bank through are based on current price series. 308 In calculating the summary measures for each indi- relation to the average produced annually in 1979-81 cator in Table 2, partially rebased constant 1987 U.s. (1979-81 = 100). The estimates are derived by divid- dollar values for each economy are calculated for each ing the quantity of food production by the total popu- year of the periods covered; the values are aggre- lation. For the index, food is defined as comprising gated across countries for each year; and the least- nuts, pulses, fruits, cereals, vegetables, sugar cane, squares procedure is used to compute the growth sugar beet, starchy roots, edible oils, livestock, and rates. The average sectoral percentage shares in Table livestock products. Quantities of food production are 3 are computed from group aggregates of sectoral measured net of annual feed, seeds for use in agricul- GDP in current U.S. dollars. ture, and food lost in processing and distribution. Fish products are measured by the level of daily pro- Table 4. Agriculture and food tein supply derived from the consumption of fish in relation to total daily protein supply from all food. The basic data for value added in agriculture are from This estimate indirectly highlights the relative impor- the World Bank's national accounts series at current tance or weight of fish in total agriculture, especially prices in national currencies. Value added in current since fish is not included in the index of food prices in national currencies is converted to U.S. dol- production. lars by applying the single-year conversion pro- The summary measures for fertilizer consumption cedure, as described in the technical note for Tables 2 are weighted by total arable land area; the summary and 3. measures for food production are weighted by The figures for the remainder of this table are from population. the Food and Agriculture Organization (FAO). Cereal imports are measured in grain equivalents and de- Table 5. Commercial energy fined as comprising all cereals in the Standard Interna- tional Trade Classification (SITC), Revision 2, Groups The data on energy are primarily from U.N. sources. 041-046. Food aid in cereals covers wheat and flour, They refer to commercial forms of primary energy bulgur, rice, coarse grains, and the cereal component petroleum and natural gas liquids, natural gas, solid of blended foods. The figures are not directly compa- fuels (coal, lignite, and so on), and primary electricity rable because of reporting and timing differences. Ce- (nuclear, geothermal, and hydroelectric power)all real imports are based on calendar-year data reported converted into oil equivalents. Figures on liquid fuel by recipient countries, and food aid in cereals is based consumption include petroleum derivatives that have on data for crop years reported by donors and inter- been consumed in nonenergy uses. For converting national organizations, including the International primary electricity into oil equivalents, a notional Wheat Council and the World Food Programme. Fur- thermal efficiency of 34 percent has been assumed. thermore, food aid information from donors may not The use of firewood, dried animal excrement, and correspond to actual receipts by beneficiaries during a other traditional fuels, although substantial in some given period because of delays in transportation and developing countries, is not taken into account be- recording or because aid is sometimes not reported to cause reliable and comprehensive data are not the FAO or other relevant international organiza- available. tions. Food aid imports may also not show up in Energy imports refers to the dollar value of energy customs records. The time reference for food aid is importsSection 3 in the Standard International Trade the crop year, July to June. Classification, Revision 1and are expressed as a per- Fertilizer consumption measures the plant nutrients centage of earnings from merchandise exports. Be- used in relation to arable land. Fertilizer products cause data on energy imports do not permit a distinc- cover nitrogenous, potash, and phosphate fertilizers tion between petroleum imports for fuel and those (which include ground rock phosphate). Arable land for use in the petrochemicals industry, these percent- is defined as land under temporary crops (double- ages may overestimate the dependence on imported cropped areas are counted once), temporary energy. meadows for mowing or for pasture, land under mar- The summary measures of energy production and ket or kitchen gardens, and land temporarily fallow consumption are computed by aggregating the re- or lying idle, as well as land under permanent crops. spective volumes for each of the years covered by the The time reference for fertilizer consumption is the periods and applying the least-squares growth rate crop year, July to June. procedure. For energy consumption per capita, Average growth rate of food production per capita has population weights are used to compute summary been computed from the index of food production measures for the specified years. per capita. The index relates to the average annual The summary measures of energy imports as a per- growth rate of food produced per capita in 1979-91 in centage of merchandise exports are computed from 309 group aggregates for energy imports and merchan- manufacturing, with 1980 as the base year. To derive dise exports in current dollars. this indicator, UNIDO data on gross output per em- ployee in current prices are adjusted using the im- Table 6. Structure of manufacturing plicit deflators for value added in manufacturing or in The basic data for value added in manufacturing are industry, taken from the World Bank's national ac- from the World Bank's national accounts series at cur- counts data files. rent prices in national currencies. Value added in cur- To improve cross-country comparability, UNIDO rent prices in national currencies is converted to U.S. has, where possible, standardized the coverage of es- dollars by applying the single-year conversion pro- tablishments to those with five or more employees. cedure, as described in the technical note for Tables 2 The concepts and definitions are in accordance and 3. with the International Recommendations for Industrial The data for distribution of manufacturing value added Statistics, published by the United Nations. Earnings among industries are provided by the United Nations (wages and salaries) cover all remuneration to em- Industrial Development Organization (UNIDO), and ployees paid by the employer during the year. The distribution calculations are from national currencies payments include (a) all regular and overtime cash in current prices. payments and bonuses and cost of living allowances; The classification of manufacturing industries is in (b) wages and salaries paid during vacation and sick accordance with the U.N. International Standard Indus- leave; (c) taxes and social insurance contributions and trial Classification of All Economic Activities (ISIC), Revi- the like, payable by the employees and deducted by sion 2. Food, beverages, and tobacco comprise ISIC Divi- the employer; and (d) payments in kind. sion 31; textiles and clothing, Division 32; machinery and The term "employees" in this table combines two transport equipment, Major Groups 382-84; and chemi- categories defined by the U.N., regular employees cals, Major Groups 351 and 352. Other comprises and persons engaged. Together these groups com- wood and related products (Division 33), paper and prise regular employees, working proprietors, active related products (Division 34), petroleum and related business partners, and unpaid family workers; they products (Major Groups 353-56), basic metals and exclude homeworkers. The data refer to the average mineral products (Divisions 36 and 37), fabricated number of employees working during the year. metal products and professional goods (Major "Value added" is defined as the current value of Groups 381 and 385), and other industries (Major gross output less the current cost of (a) materials, Group 390). When data for textiles, machinery, or fuels, and other supplies consumed; (b) contract and chemicals are shown as not available, they are also commission work done by others; (c) repair and included in other. maintenance work done by others; and (d) goods Summary measures given for value added in man- shipped in the same condition as received. ufacturing are totals calculated by the aggregation The value of gross output is estimated on the basis method noted at the beginning of the technical notes. of either production or shipments. On the production basis it consists of (a) the value of all products of the Table 7. Manufacturing earnings and output establishment; (b) the value of industrial services ren- dered to others; (c) the value of goods shipped in the Four indicators are showntwo relate to real earn- same condition as received; (d) the value of electricity ings per employee, one to labor's share in total value sold; and (e) the net change in the value of work-in- added generated, and one to labor productivity in the progress between the beginning and the end of the manufacturing sector. The indicators are based on reference period. In the case of estimates compiled on data from the United Nations Industrial Develop- a shipment basis, the net change between the begin- ment Organization (UNIDO), although the deflators ning and the end of the reference period in the value are from other sources, as explained below. of stocks of finished goods is also included. Earnings per employee are in constant prices and are derived by deflating nominal earnings per employee Tables 8 and 9. Growth of consumption and by the country's consumer price index (CPI). The CPI investment; structure of demand is from the International Monetary Fund's Interna- tional Financial Statistics. GDP is defined in the note for Tables 2 and 3, but Total earnings as a percentage of value added are de- here it is in purchaser values. rived by dividing total earnings of employees by General government consumption includes all current value added in current prices to show labor's share in expenditure for purchases of goods and services by income generated in the manufacturing sector. Gross all levels of government. Capital expenditure on na- output per employee is in constant prices and is pre- tional defense and security is regarded as consump- sented as an index of overall labor productivity in tion expenditure. 310 Private consumption, etc. is the market value of all than who pays for consumption goods, and it im- goods and services, including durable products (such proves international comparability because it is less as cars, washing machines, and home computers) sensitive to differing national practices regarding the purchased or received as income in kind by house- financing of health and education services. holds and nonprofit institutions. It excludes pur- Cereals and tubers, a major subitem of food, comprise chases of dwellings but includes imputed rent for the main staple products: rice, flour, bread, all other owner-occupied dwellings (see the technical note for cereals and cereal preparations, potatoes, yams, and Table 10 for details). In practice, it includes any statis- other tubers. For high-income OECD members, how- tical discrepancy in the use of resources. At constant ever, this subitem does not include tubers. Gross prices, it also includes the rescaling deviation from rents, fuel and power consist of actual and imputed partial rebasing, which is explained at the beginning rents and repair and maintenance charges, as well as of the technical notes. the subitem fuel and power (for heating, lighting, cook- Gross domestic investment consists of outlays on ad- ing, air conditioning, and so forth). Note that this ditions to the fixed assets of the economy plus net item excludes energy used for transport (rarely re- changes in the level of inventories. ported to be more than 1 percent of total consurnp- Gross domestic savings are calculated by deducting tion in low- and middle-income economies). As men- total consumption from GDP. tioned, medical care and education include government Exports of goods and nonfactor services represent the as well as private consumption expenditure. Transport value of all goods and nonfactor services provided to and communication also include the purchase of auto- the rest of the world; they include merchandise, mobiles, which are reported as a subitem. Other con- freight, insurance, travel, and other nonfactor ser- sumption, the residual group, includes beverages and vices. The value of factor services, such as investment tobacco, nondurable household goods and house- income, interest, and labor income, is excluded. Cur- hold services, recreational services, and services (in- rent transfers are also excluded. cluding meals) supplied by hotels and restaurants; The resource balance is the difference between ex- carry-out food is recorded here. It also includes the ports of goods and nonfactor services and imports of separately reported subitem other consumer durables, goods and nonfactor services. comprising household appliances, furniture, floor Partially rebased 1987 series in constant domestic coverings, recreational equipment, and watches and currency units are used to compute the indicators in jewelry. Table 8. Distribution of GDP in Table 9 is calculated Estimating the structure of consumption is one of from national accounts series in current domestic cur- the weakest aspects of national accounting in low- rency units. and middle-income economies. The structure is esti- The summary measures are calculated by the mated through household expenditure surveys and method explained in the note for Tables 2 and 3. similar survey techniques. It therefore shares any bias inherent in the sample frame. Since, conceptually, Table 10. Structure of consumption expenditure is not identical to consumption, other apparent discrepancies occur, and data for some Percentage shares of selected items in total household countries should be treated with caution. For exam- consumption expenditure are computed from details ple, some countries limit surveys to urban areas or, of GDP (expenditure at national market prices) de- even more narrowly, to capital cities. This tends to fined in the U.N. System of National Accounts (SNA), produce lower than average shares for food and high mostly as collected from the International Compari- shares for transport and communication, gross rents, son Program (ICP) Phases IV (1980) and V (1985). For fuel and power, and other consumption. Controlled countries not covered by the ICP, less detailed na- food prices and incomplete national accounting for tional accounts estimates are included, where avail- subsistence activities also contribute to low food able, to present a general idea of the broad structure shares. of consumption. The data cover eighty-four countries (including Bank staff estimates for China) and refer to the most recent estimates, generally for 1980 and Table 11. Central government expenditure 1985. Where they refer to other years, the figures are shown in italics. Consumption here refers to private The data on central government finance in Tables (nongovernment) consumption as defined in the 11 and 12 are from the IMF's Government Finance Sta- SNA and in the notes for Tables 2 and 3, 4, and 9, tistics Yearbook (1992) and IMF data files. The accounts except that education and medical care comprise gov- of each country are reported using the system of com- ernment as well as private outlays. This ICP concept mon definitions and classifications found in the IMF's of "enhanced consumption" reflects who uses rather Manual on Government Finance Statistics (1986). 311 For complete and authoritative explanations of con- colleges; and of vocational, technical, and other train- cepts, definitions, and data sources, see these IMF ing institutions. Also included is expenditure on the sources. The commentary that follows is intended general administration and regulation of the educa- mainly to place these data in the context of the broad tion system; on research into its objectives, organiza- range of indicators reported in this edition. tion, administration, and methods; and on such sub- The shares of total expenditure and current revenue by sidiary services as transport, school meals, and category are calculated from series in national curren- school medical and dental services. Note that Table 10 cies. Because of differences in coverage of available provides an alternative measure of expenditure on data, the individual components of central govern- education, private as well as public, relative to house- ment expenditure and current revenue shown in hold consumption. these tables may not be strictly comparable across all Health covers public expenditure on hospitals, ma- economies. ternity and dental centers, and clinics with a major Moreover, inadequate statistical coverage of state, medical component; on national health and medical provincial, and local governments dictates the use of insurance schemes; and on family planning and pre- central government data; this may seriously under- ventive care. Note that Table 10 also provides a mea- state or distort the statistical portrayal of the alloca- sure of expenditure on medical care, private as well tion of resources for various purposes, especially in as public, relative to household consumption. countries where lower levels of government have Housing, amenities; social security and welfare cover considerable autonomy and are responsible for many expenditure on housing (excluding interest subsidies, economic and social services. In addition, "central which are usually classified with other) such as in- government" can mean either of two accounting con- come-related schemes; on provision and support of cepts: consolidated or budgetary. For most countries, housing and slum-clearance activities; on community central government finance data have been consoli- development; and on sanitation services. These cate- dated into one overall account, but for others only the gories also cover compensation for loss of income to budgetary central government accounts are available. the sick and temporarily disabled; payments to the Since all central government units are not always in- elderly, the permanently disabled, and the unem- cluded in the budgetary accounts, the overall picture ployed; family, maternity, and child allowances; and of central government activities is usually incom- the cost of welfare services, such as care of the aged, plete. Countries reporting budgetary data are the disabled, and children. Many expenditures rele- footnoted. vant to environmental defense, such as pollution Consequently, the data presented, especially those abatement, water supply, sanitary affairs, and refuse for education and health, are not comparable across collection, are included indistinguishably in this countries. In many economies, private health and ed- category. ucation services are substantial; in others, public ser- Economic services comprise expenditure associated vices represent the major component of total expen- with the regulation, support, and more efficient oper- diture but may be financed by lower levels of ation of business; economic development; redress of government. Caution should therefore be exercised regional imbalances; and creation of employment op- in using the data for cross-country comparisons. Cen- portunities. Research, trade promotion, geological tral government expenditure comprises the expendi- surveys, and inspection and regulation of particular ture by all government offices, departments, estab- industry groups are among the activities included. lishments, and other bodies that are agencies or Other covers interest payments and items not in- instruments of the central authority of a country. It cluded elsewhere; for a few economies it also in- includes both current and capital (development) cludes amounts that could not be allocated to other expenditure. components (or adjustments from accrual to cash Defense comprises all expenditure, whether by de- accounts). fense or other departments, on the maintenance of Total expenditure is more narrowly defined than the military forces, including the purchase of military measure of general government consumption given supplies and equipment, construction, recruiting, in Table 9 because it excludes consumption expendi- and training. Also in this category are closely related ture by state and local governments. At the same items such as military aid programs. Defense does time, central government expenditure is more not include expenditure on public order and safety, broadly defined because it includes government's which are classified separately. gross domestic investment and transfer payments. Education comprises expenditure on the provision, Overall surplus/deficit is defined as current and capi- management, inspection, and support of preprimary, tal revenue and official grants received, less total ex- primary, and secondary schools; of universities and penditure and lending minus repayments. 312 Table 12. Central government current revenue and similar bank accounts that the issuer will readily exchange for money. Where nonmonetary financial Information on data sources and comparability is institutions are important issuers of quasi-monetary given in the note for Table 11. Current revenue by liabilities, these are also included in the measure of source is expressed as a percentage of total current monetary holdings. revenue, which is the sum of tax revenue and nontax The growth rates for monetary holdings are calcu- revenue and is calculated from national currencies. lated from year-end figures, while the average of the Tax revenue comprises compulsory, unrequited, year-end figures for the specified year and the pre- nonrepayable receipts for public purposes. It includes vious year is used for the ratio of monetary holdings interest collected on tax arrears and penalties col- to GDP. lected on nonpayment or late payment of taxes and is Nominal interest rates of banks, also from IFS, repre- shown net of refunds and other corrective transac- sent the rates paid by commercial or similar banks to tions. Taxes on income, profit, and capital gains are taxes holders of their quasi-monetary liabilities (deposit levied on the actual or presumptive net income of rate) and charged by the banks on loans to prime individuals, on the profits of enterprises, and on capi- customers (lending rate). The data are, however, of tal gains, whether realized on land sales, securities, limited international comparability, partly because or other assets. Intragovernmental payments are coverage and definitions vary and partly because eliminated in consolidation. Social security contribu- countries differ in the scope available to banks for tions include employers' and employees' social secu- adjusting interest rates to reflect market conditions. rity contributions as well as those of self-employed Because interest rates (and growth rates for mone- and unemployed persons. Domestic taxes on goods and tary holdings) are expressed in nominal terms, much services include general sales and turnover or value of the variation among countries stems from differ- added taxes, selective excises on goods, selective ences in inflation. For easy reference, the Table I indi- taxes on services, taxes on the use of goods or prop- cator of recent inflation is repeated in this table. erty, and profits of fiscal monopolies. Taxes on interna- tional trade and transactions include import duties, ex- Table 14. Growth of merchandise trade port duties, profits of export or import monopolies, The main data source for current trade values is the exchange profits, and exchange taxes. Other taxes in- U.N. Commodity Trade (COMTRADE) data file sup- clude employers' payroll or labor taxes, taxes on plemented by World Bank estimates. The statistics on property, and taxes not allocable to other categories. merchandise trade are based on countries' customs They may include negative values that are adjust- returns. ments, for instance, for taxes collected on behalf of Merchandise exports and imports, with some excep- state and local governments and not allocable to indi- tions, cover international movements of goods across vidual tax categories. customs borders; trade in services is not included. Non tax revenue comprises receipts that are not a Exports are valued f.o.b. (free on board) and imports compulsory nonrepayable payment for public pur- c.i.f. (cost, insurance, and freight) unless otherwise poses, such as fines, administrative fees, or entrepre- specified in the foregoing sources. These values are in neurial income from government ownership of prop- current U. S. dollars. erty. Proceeds of grants and borrowing, funds arising The growth rates of merchandise exports and im- from the repayment of previous lending by govern- ports are based on constant price data, which are ments, incurrence of liabilities, and proceeds from obtained from export or import value data as deflated the sale of capital assets are not included. by the corresponding price index. The World Bank uses its own price indexes, which are based on inter- Table 13. Money and interest rates national prices for primary commodities, and unit value indexes for manufactures. These price indexes The data on monetary holdings are based on the are country-specific and disaggregated by broad com- IMF's International Financial Statistics (IFS). Monetary modity groups. This ensures consistency between holdings, broadly defined, comprise the monetary and data for a group of countries and those for individual quasi-monetary liabilities of a country's financial in- countries. Such consistency will increase as the stitutions to residents but not to the central govern- World Bank continues to improve its trade price in- ment. For most countries, monetary holdings are the dexes for an increasing number of countries. These sum of money (IFS line 34) and quasi money (IFS line growth rates can differ from those derived from na- 35). Money comprises the economy's means of pay- tional practices because national price indexes may ment: currency outside banks and demand deposits. use different base years and weighting procedures Quasi money comprises time and savings deposits from those used by the World Bank. 313 The terms of trade, or the net barter terms of trade, 27 and 28. Machinery and transport equipment are the measure the relative movement of export prices commodities in SITC Section 7. Other manufactures against that of import prices. Calculated as the ratio represent SITC Sections 5 through 9, less Section 7 of a country's index of average export prices to,its and Division 68. Textiles and clothing, representing average import price index, this indicator shows SITC Divisions 65 and 84 (textiles, yarns, fabrics, changes over a base year in the level of export prices made-up articles, and related products and clothing), as a percentage of import prices. The terms of trade are a subgroup of other manufactures. index numbers are shown for 1985 and 1991, where The summary measures in Table 15 are weighted 1987 = 100. The price indexes are from the source by total merchandise imports of individual countries cited above for the growth rates of exports and in current U.S. dollars and those in Table 16 by total imports. merchandise exports of individual countries in cur- The summary measures for the growth rates are rent U.S. dollars. (See the technical note for Table 14.) calculated by aggregating the 1987 constant U.S. dol- lar price series for each year and then applying the Table 17. OECD imports of manufactured goods: least-squares growth rate procedure for the periods shown. origin and composition Tables 15 and 16. Structure of merchandise imports The data are from the United Nations, reported by and exports high-income OECD economies, which are the OECD members excluding Greece, Portugal, and Turkey. The shares in these tables are derived from trade The table reports the value of imports of man ufac- values in current dollars reported in the U.N. trade tures of high-income OECD countries by the economy data system and the U.N. 's Yearbook of International of origin, and the composition of such imports by Trade Statistics, supplemented by World Bank major manufactured product groups. These data are estimates. based on the U.N. COMTRADE databaseRevision Merchandise exports and imports are defined in the 1, SITC for 1970, and Revision 2 SITC for 1991. technical note for Table 14. The table replaces one in past editions on the origin The categorization of exports and imports follows and destination of manufactured exports, which was the Standard International Trade Classification (SITC), based on exports reported by individual economies. Series M, No. 34, Revision 1. For some countries, Since there was a lag of several years in reporting by data for certain commodity categories are unavailable many developing economies, estimates based on var- and the full breakdown cannot be shown. ious sources were used to fill the gaps. Until these In Table 15, food commodities are those in SITC estimates can be improved, the current table, based Sections 0, 1, and 4 and Division 22 (food and live on up-to-date and consistent but less comprehensive animals, beverages and tobacco, animal and vegeta- data, is included instead. Manufactured imports of ble oils and fats, oilseeds, oil nuts and oil kernels). the predominant markets from individual economies Fuels are the commodities in SITC Section 3 (mineral are the best available proxy of the magnitude and fuels, and lubricants and related materials). Other pri- composition of the manufactured exports of these mary commodities comprise SITC Section 2 (inedible economies to all destinations taken together. crude materials, except fuels), less Division 22 (oil- Manufactured goods are the commodities in the Stan- seeds, oilnuts, and oil kernels) and Division 68 (non- dard International Trade Classification (SITC), Revision ferrous metals). Machinery and transport equipment are 1, Sections 5 through 9 (chemical and related prod- the commodities in SITC Section 7. Other manufac- ucts, basic manufactures, manufactured articles, ma- tures, calculated residually from the total value of chinery and transport equipment, and other manu- manufactured imports, represent SITC Sections 5 factured articles and goods not elsewhere classified), through 9, less Section 7 and Division 68. excluding Division 68 (nonferrous metals). This defi- In Table 16, fuels, minerals, and metals are the com- nition is somewhat broader than the one used to de- modities in SITC Section 3 (mineral fuels, and lubri- fine exporters of manufactures. cants and related materials), Divisions 27 and 28 The major manufactured product groups reported (crude fertilizers and crude minerals, excluding coal; are defined as follows: textiles and clothing (SITC Sec- petroleum and precious stones, and metalliferous tions 65 and 84), chemicals (SITC Section 5), electrical ores and metal scrap), and Division 68 (nonferrous machinery and electronics (SITC Section 72), transport metals). Other primary commodities comprise SITC Sec- equipment (SITC Section 73), and others, defined as the tions 0, 1, 2, and 4 (food and live animals, beverages residual. SITC Revision 1 data are used for the year and tobacco, inedible crude materials, except fuels, 1970, whereas the equivalent data in Revision 2 are and animal and vegetable oils and fats), less Divisions used for the year 1991. 314 Table 18. Balance of payments and reserves The summary measures are computed from group aggregates for gross international reserves and total The statistics for this table are mostly as reported by imports of goods and services in current dollars. the IMF but do include recent estimates by World Bank staff and, in rare instances, the Bank's own cov- Table 19. Official development assistance from erage or classification adjustments to enhance inter- OECD and OPEC members national comparability. Values in this table are in U.S. dollars converted at current exchange rates. Official development assistance (ODA) consists of net The current account balance after official transfers is the disbursements of loans and grants made on conces- difference between (a) exports of goods and services sional financial terms by official agencies of the mem- (factor and nonfactor) as well as inflows of unre- bers of the Development Assistance Committee quited transfers (private and official) and (b) imports (DAC) of the Organization for Economic Cooperation of goods and services as well as all unrequited trans- and Development (OECD) and members of the Or- fers to the rest of the world. ganization of Petroleum Exporting Countries (OPEC) The current account balance before official transfers is to promote economic development and welfare. Al- the current account balance that treats net official un- though this definition is meant to exclude purely mili- requited transfers as akin to official capital move- tary assistance, the borderline is sometimes blurred; ments. The difference between the two balance of the definition used by the country of origin usually payments measures is essentially foreign aid in the prevails. ODA also includes the value of technical form of grants, technical assistance, and food aid, cooperation and assistance. All data shown are sup- which, for most developing countries, tends to make plied by the OECD, and all U.S. dollar values are current account deficits smaller than the financing converted at official exchange rates. requirement. Total net flows are net disbursements to developing Net workers' remittances cover payments and re- countries and multilateral institutions. The disburse- ceipts of income by migrants who are employed or ments to multilateral institutions are now reported expect to be employed for more than a year in their for all DAC members on the basis of the date of issue new economy, where they are considered residents. of notes; some DAC members previously reported on These remittances are classified as private unrequited the basis of the date of encashment. transfers and are included in the balance of payments The nominal values shown in the summary for current account balance, whereas those derived from ODA from high-income OECD countries were con- shorter-term stays are included in services as labor verted at 1987 prices using the dollar GDP deflator. income. The distinction accords with internationally This deflator is based on price increases in OECD agreed guidelines, but many developing countries countries (excluding Greece, Portugal, and Turkey) classify workers' remittances as a factor income re- measured in dollars. It takes into account the parity ceipt (hence, a component of GNP). The World Bank changes between the dollar and national currencies. adheres to international guidelines in defining GNP For example, when the dollar depreciates, price and, therefore, may differ from national practices. changes measured in national currencies have to be Gross international reserves comprise holdings of adjusted upward by the amount of the depreciation monetary gold, special drawing rights (SDRs), the to obtain price changes in dollars. reserve position of members in the IMF, and holdings The table, in addition to showing totals for OPEC, of foreign exchange under the control of monetary shows totals for the Organization of Arab Petroleum authorities. The data on holdings of international re- Exporting Countries (OAPEC). The donor members serves are from IMF data files. The gold component of, OAPEC are Algeria, Iraq, Kuwait, Libya, Qatar, of these reserves is valued throughout at year-end Saudi Arabia, and United Arab Emirates. ODA data (December 31) London prices: that is, $37.37 an for OPEC and OAPEC are also obtained from the ounce in 1970 and $353.60 an ounce in 1991. The re- OECD. serve levels for 1970 and 1991 refer to the end of the year indicated and are in current U.S. dollars at pre- Table 20. Official development assistance: receipts vailing exchange rates. Because of differences in the definition of international reserves, in the valuation Net disbursements of ODA from all sources consist of of gold, and in reserve management practices, the loans and grants made on concessional financial levels of reserve holdings published in national terms by all bilateral official agencies and multilateral sources do not have strictly comparable significance. sources to promote economic development and wel- Reserve holdings at the end of 1991 are also ex- fare. They include the value of technical cooperation pressed in terms of the number of months of imports and assistance. The disbursements shown in this ta- of goods and services they could pay for. ble are not strictly comparable with those shown in 315 Table 19 since the receipts are from all sources; dis- long-term debt, use of IMF credit, and short-term bursements in Table 19 refer only to those made by debt. high-income members of the OECD and members of Total arrears on LDOD denotes principal and interest OPEC. Net disbursements equal gross disbursements due but not paid. less payments to the originators of aid for amortiza- Present value is the discounted value of the future tion of past aid receipts. Net disbursements of ODA debt service payments. are shown per capita and as a percentage of GNP. The summary measures of per capita ODA are Table 22. Flow of public and private external capital computed from group aggregates for population and for ODA. Summary measures for ODA as a percent- Data on disbursements, repayment of principal (am- age of GNP are computed from group totals for ODA ortization), and payment of interest are for public, and for GNP in current U.S. dollars. publicly guaranteed, and private nonguaranteed long-term loans. Table 21. Total external debt Disbursements are drawings on long-term loan com- The data on debt in this and successive tables are mitments during the year specified. from the World Bank Debtor Reporting System, sup- Repayments of principal are actual amount of princi- plemented by World Bank estimates. That system is pal (amortization) paid in foreign currency, goods, or concerned solely with developing economies and services in the year specified. does not collect data on external debt for other groups Interest payments are actual amounts of interest paid of borrowers or from economies that are not members in foreign currency, goods, or services in the year of the World Bank. The dollar figures on debt shown specified. in Tables 21 through 25 are in U.S. dollars converted Table 23. Aggregate net resource flows and net at official exchange rates. transfers The data on debt include private nonguaranteed debt reported by twenty-seven developing countries Net flows on long-term debt are disbursements less the and complete or partial estimates for an additional repayment of principal on public, publicly guaran- twenty others that do not report but for which this teed, and private nonguaranteed long-term debt. of- type of debt is known to be significant. ficial grants are transfers made by an official agency in Public loans are external obligations of public cash or in kind in respect of which no legal debt is debtors, including the national government, its agen- incurred by the recipient. Data on official grants ex- cies, and autonomous public bodies. Publicly guaran- clude grants for technical assistance. teed loans are external obligations of private debtors Net foreign direct investment is defined as investment that are guaranteed for repayment by a public entity. that is made to acquire a lasting interest (usually 10 These two categories are aggregated in the tables. percent of the voting stock) in an enterprise operating Private nonguaranteed loans are external obligations of in a country other than that of the investor (defined private debtors that are not guaranteed for repay- according to residency), the investor's purpose being ment by a public entity. an effective voice in the management of the enter- Use of IMF credit denotes repurchase obligations to prise. Aggregate net resource flows are the sum of net the IMF for all uses of IMF resources, excluding those flows on long-term debt (excluding use of IMF resulting from drawings in the reserve tranche. It is credit), plus official grants (excluding technical assis- shown for the end of the year specified. It comprises tance) and net foreign direct investment. Aggregate purchases outstanding under the credit tranches, in- net transfers are equal to aggregate net resource flows cluding enlarged access resources, and all special fa- minus interest payments on long-term loans and re- cilities (the buffer stock, compensatory financing, ex- mittance of all profits. tended fund, and oil facilities), Trust Fund loans, and operations under the enhanced structural adjustment Table 24. Total external debt ratios facilities. Use of IMF credit outstanding at year-end (a stock) is converted to U.S. dollars at the dollar-SDR Total external debt as a percentage of exports of goods and exchange rate in effect at year-end. services represents public, publicly guaranteed, pri- Short-term debt is debt with an original maturity of vate nonguaranteed long-term debt, use of IMF one year or less. Available data permit no distinctions credit, and short-term debt drawn at year-end, net of between public and private nonguaranteed short- repayments of principal and write-off s. Throughout term debt. this table, goods and services include workers' remit- Total external debt is defined here as the sum of pub- tances. For estimating total external debt as a percentage lic, publicly guaranteed, and private nonguaranteed of GNP, the debt figures are converted into U.S. do!- 316 lars from currencies of repayment at end-of-year offi- Table 26. Population growth and projections cial exchange rates. GNP is converted from national currencies to U.S. dollars by applying the conversion Population growth rates are period averages calcu- procedure described in the technical note for Tables 2 lated from midyear populations. and 3. Population estimates for mid-1991 and estimates of Total debt service as a percentage of goods and services is fertility and mortality are made by the World Bank the sum of principal repayments and interest pay- from data provided by the U.N. Population Division, ments on total external debt (as defined in the note the U.N. Statistical Office, and country statistical of- for Table 21). It is one of several conventional mea- fices. Estimates take into account the results of the sures used to assess a country's ability to service latest population censuses, which in some cases are debt. neither recent nor accurate. Note that refugees not Interest payments as a percentage of exports of goods and permanently settled in the country of asylum are gen- services are actual payments made on total external erally considered to be part of the population of their debt. country of origin. The summary measures are weighted by exports of The projections of population for 2000, 2025, and goods and services in current dollars and by GNP in the year in which the population will eventually be- current dollars, respectively. come stationary (see definition below) are made for Concessional debt as a percentage of total external debt each economy separately. Information on total popu- conveys information about the borrower's receipt of lation by age and sex, fertility, mortality, and interna- aid from official lenders at concessional terms as de- tional migration is projected on the basis of gener- fined by the DAC, that is, loans with an original grant alized assumptions until the population becomes element of 25 percent or more. stationary. Multilateral debt as a percentage of total external debt A stationary population is one in which age- and conveys information about the borrower's receipt of sex-specific mortality rates have not changed over a aid from the World Bank, regional development long period, and during which fertility rates have re- banks, and other multilateral and intergovernmental mained at replacement level; that is, when the net agencies. Excluded are loans from funds adminis- reproduction rate (defined in the note for Table 27) tered by an international organization on behalf of a equals 1. In such a population, the birth rate is con- single donor government. stant and equal to the death rate, the age structure is constant, and the growth rate is zero. Table 25. Terms of external public borrowing Population projections are made age cohort by age cohort. Mortality, fertility, and migration are pro- Commitments refer to the public and publicly guaran- jected separately, and the results are applied iter- teed loans for which contracts were signed in the year atively to the 1990 base-year age structure. For the specified. They are reported in currencies of repay- projection period 1990 to 2005, the changes in mortal- ment and converted into U.S. dollars at average an- ity are country specific: increments in life expectancy nual official exchange rates. and decrements in infant mortality are based on pre- Figures for interest rates, maturities, and grace periods vious trends for each country. When female second- are averages weighted by the amounts of the loans. ary school enrollment is high, mortality is assumed to Interest is the major charge levied on a loan and is decline more quickly. Infant mortality is projected usually computed on the amount of principal drawn separately from adult mortality. Note that the data and outstanding. The maturity of a loan is the inter- reflect the potentially significant impact of the human val between the agreement date, when a loan agree- immunodeficiency virus (HIV) epidemic. ment is signed or bonds are issued, and the date of Projected fertility rates are also based on previous final repayment of principal. The grace period is the trends. For countries in which fertility has started to interval between the agreement date and the date of decline (termed "fertility transition"), this trend is the first repayment of principal. assumed to continue. It has been observed that no Public loans with variable interest rates, as a percentage country where the population has a life expectancy of of public debt refer to interest rates that float with less than 50 years has experienced a fertility decline; movements in a key market rate; for example, the for these countries, fertility transition is delayed, and London interbank offered rate (LIBOR) or the U.S. the average decline of the group of countries in fertil- prime rate. This column shows the borrower's expo- ity transition is applied. Countries with below- sure to changes in international interest rates. replacement fertility are assumed to have constant The summary measures in this table are weighted total fertility rates until 1995-2000 and to regain re- by the amounts of the loans. placement level by 2030. 317 International migration rates are based on past and derstate contraceptive prevalence because they do present trends in migration flows and migration p01- not measure use of methods such as rhythm, with- icy. Among the sources consulted are estimates and drawal, or abstinence, nor use of contraceptives not projections made by national statistical offices, inter- obtained through the official family planning pro- national agencies, and research institutions. Because gram. The data refer to rates prevailing in a variety of of the uncertainty of future migration trends, it is years, generally not more than two years before the assumed in the projections that net migration rates year specified in the table. will reach zero by 2025. All summary measures are country data weighted The estimates of the size of the stationary popula- by each country's share in the aggregate population. tion are very long-term projections. They are in- cluded only to show the implications of recent fertil- Table 28. Health and nutrition ity and mortality trends on the basis of generalized assumptions. A fuller description of the methods and The estimates of population per physician and per nurs- assumptions used to calculate the estimates is con- ing person are derived from World Health Organiza- tained in World Population Projections, 1992-93 Edition. tion (WHO) data and are supplemented by data ob- tained directly by the World Bank from national Table 27. Demography and fertility sources. The data refer to a variety of years, generally no more than two years before the year specified. The crude birth rate and crude death rate indicate re- Nursing persons include auxiliary nurses, as well as spectively the number of live births and deaths occur- paraprofessional personnel such as traditional birth ring per thousand population in a year. They come attendants. The inclusion of auxiliary and paraprofes- from the sources mentioned in the note to Table 26. sional personnel provides more realistic estimates of Women of childbearing age are those in the 15-49 age- available nursing care. Because definitions of doctors group. and nursing personnel varyand because the data The total fertility rate represents the number of chil- shown are for a variety of yearsthe data for these dren that would be born to a woman if she were to two indicators are not strictly comparable across live to the end of her childbearing years and bear countries. children at each age in accordance with prevailing Data on births attended by health staff show the per- age-specific fertility rates. The rates given are from centage of births recorded where a recognized health the sources mentioned in the note for Table 26. service worker was in attendance. The data are from The net reproduction rate (NRR), which measures the WHO, supplemented by UNICEF data. They are number of daughters a newborn girl will bear during based on national sources, derived mostly from off i- her lifetime, assuming fixed age-specific fertility and cial community reports and hospital records; some mortality rates, reflects the extent to which a cohort reflect only births in hospitals and other medical in- of newborn girls will reproduce themselves. An NRR stitutions. Sometimes smaller private and rural hos- of 1 indicates that fertility is at replacement level: at pitals are excluded, and sometimes even relatively this rate women will bear, on average, only enough primitive local facilities are included. The coverage is daughters to replace themselves in the population. therefore not always comprehensive, and the figures As with the size of the stationary population, the should be treated with extreme caution. assumed year of reaching replacement-level fertility Babies with low birth weight are children born weigh- is speculative and should not be regarded as a ing less than 2,500 grams. Low birth weight is fre- prediction. quently associated with maternal malnutrition. It Married women of childbearing age using contraception tends to raise the risk of infant mortality and lead to are women who are practicing, or whose husbands poor growth in infancy and childhood, thus increas- are practicing, any form of contraception. Contracep- ing the incidence of other forms of retarded develop- tive usage is generally measured for women age 15 to ment. The figures are derived from both WHO and 49. A few countries use measures relating to other UNICEF sources and are based on national data. The age groups, especially 15 to 44. data are not strictly comparable across countries since Data are mainly derived from demographic and they are compiled from a combination of surveys and health surveys, contraceptive prevalence surveys, administrative records that may not have representa- World Bank country data, and Mauldin and Segal's tive national coverage. article "Prevalence of Contraceptive Use: Trends and The infant mortality rate is the number of infants Issues" in volume 19 of Studies in Family Planning who die before reaching one year of age, per thou- (1988). For a few countries for which no survey data sand live births in a given year. The data are from the are available, and for several African countries, pro- U.N. publication Mortality of Children under Age 5: gram statistics are used. Program statistics may un- Projections, 1950-2025 as well as from the World Bank. 318 The years of life lost (per 1,000 population) conveys the varies among countries. The youth populationthat burden of mortality in absolute terms. It is composed is, 20 to 24 yearshas been adopted by UNESCO as of the sum of the years lost to premature death per the denominator since it represents an average ter- 1,000 population. Years of life lost at age x are mea- tiary level cohort even though people above and be- sured by subtracting the remaining expected years of low this age group may be registered in tertiary life, given a life expectancy at birth fixed at 80 years institutions. for men and 82.5 for women. This indicator depends Primary net enrollment is the percentage of school- on the effect of three variables: the age structure of age children who are enrolled in school. Unlike gross the population, the overall rate of mortality, and the enrollment, the net ratios correspond to the country's age structure of mortality. primary-school age group. This indicator gives a Child malnutrition measures the percentage of chil- much clearer idea of how many children in the age dren under five with a deficiency or an excess of nu- group are actually enrolled in school without the trients that interfere with their health and genetic po- numbers being inflated by over- or under-age tential for growth. Methods of assessment vary, but children. the most commonly used are the following: less than The primary pupil-teacher ratio is the number of pu- 80 percent of the standard weight for age; less than pils enrolled in school in a country, divided by the minus two standard deviation from the 50th percen- number of teachers in the education system. tile of the weight for age reference population; and The summary measures in this table are country the Gomez scale of malnutrition. Note that for a few enrollment rates weighted by each country's share in countries the figures are for children of three or four the aggregate population. years of age and younger. The summary measures in this table are country data weighted by each coun- Table 30. Income distribution and PPC estimates of try's share in the aggregate population. GDP Table 29. Education The first six columns of the table report distribution of income or expenditure accruing to percentile The data in this table refer to a variety of years, gener- groups of households ranked by total household in- ally not more than two years distant from those spe- come, per capita income, or expenditure. The last cified; however, figures for females sometimes refer four columns contain estimates of per capita GDP to a year earlier than that for overall totals. The data based on purchasing power of currencies (PPCs) are mostly from UNESCO. rather than exchange rates (see below for the defini- Primary school enrollment data are estimates of tion of PPC). children of all ages enrolled in primary school. Fig- The first six columns of the table give the shares of ures are expressed as the ratio of pupils to the popu- population quintiles and the top decile in total in- lation of school-age children. Although many coun- come or consumption expenditure for 36 low- and tries consider primary school age to be 6 to 11 years, middle-income countries, and 20 high-income coun- others do not. For some countries with universal pri- tries. The rest of this note refers to the former set of mary education, the gross enrollment ratios may ex- countries. The data sets for these countries refer to ceed 100 percent because some pupils are younger or different years between 1981 and 1991, and are drawn older than the country's standard primary school from nationally representative household surveys. age. The data sets have been compiled from two main The data on secondary school enrollment are calcu- sources: government statistical agencies (often using lated in the same manner, but again the definition of published reports), and the World Bank (mostly data secondary school age differs among countries. It is originating from the Living Standards Measurement most commonly considered to be 12 to 17 years. Late Study). In cases where the original unit record data entry of more mature students as well as repetition from the household survey were available, these and the phenomenon of "bunching" in final grades have been used to calculate directly the income (or can influence these ratios. expenditure) shares of different quantiles; otherwise, The tertiary enrollment ratio is calculated by divid- the latter have been estimated from the best available ing the number of pupils enrolled in all post-second- grouped data. For further details on both the data ary schools and universities by the population in the and the estimation methodology, see Chen, Datt, and 20-24 age group. Pupils attending vocational schools, Ravallion, 1993. adult education programs, two-year community col- There are several comparability problems across leges, and distance education centers (primarily cor- countries in the underlying household surveys, respondence courses) are included. The distribution though these problems are diminishing over time as of pupils across these different types of institutions survey methodologies are both improving and be- 319 coming more standardized, particularly under the and expressed in 1991 international dollars, are pre- initiatives of the United Nations (under the House- sented in the tenth column. The adjustments do not hold Survey Capability Program) and the World Bank take account of changes in the terms of trade. The (under the Living Standard Measurement Study and observed figures should be used wherever available. the Social Dimensions of Adjustment Project for Sub- Where both observed and regression numbers are Saharan Africa). The data presented here should available a comparison between the two indicates the nevertheless be interpreted with caution. In particu- range of errors associated with the regression esti- lar, the following three sources of noncomparability mates. For countries that do not have PPC-based ob- ought to be noted. First, the surveys differ in using served data, there is no alternative to the use of re- income or consumption expenditure as the living gression estimates, but the extent and direction of standard indicator. For 17 of the 36 low- and middle- errors cannot be inferred in these cases. income countries, the data refer to consumption ex- ICP recasts traditional national accounts through penditure. Typically, income is more unequally dis- special price collections and disaggregation of GDP tributed than consumption. Second, the surveys dif- by expenditure components. ICP details are prepared fer in using the household or the individual as their by national statistical offices, and the results are coor- unit of observation; in the former case, the quantiles dinated by the U.N. Statistical Division (UNSTAT) refer to percentage of households, rather than per- with support from other international agencies, par- centage of persons. Third, the surveys also differ ticularly the Statistical Office of the European Com- according to whether the units of observation are munities (Eurostat) and the Organization for Eco- ranked by household income (or consumption) or by nomic Cooperation and Development (OECD). The per capita income (or consumption). The footnotes to World Bank, the Economic Commission for Europe, the table identify these differences for each country. and the Economic and Social Commission for Asia The 1987 indexed figures on PPC-based GDP per and the Pacific (ESCAP) also contribute to this exer- capita (US=100) are presented in the seventh col- cise. A total of sixty-four countries participated in ICP umn. They include: (i) results of the International Phase V. For one country (Nepal), total GDP data Comparison Programme (ICP) Phase VI for 1990 for were not available, and comparisons were made for OECD countries extrapolated backward to 1987; (ii) consumption only. Luxembourg and Swaziland are results of ICP Phase V for 1985 for non-OECD coun- the only two economies with populations under I tries extrapolated to 1987; (iii) the latest available re- million that have participated in ICP; their 1987 re- sults from either Phase IV for 1980 or Phase III for sults, as a percentage of the U.S. results, are 83.1 and 1975 extrapolated to 1987 for countries that partici- 15.0, respectively. The figures given here are subject pated in the earlier phases only; (iv) World Bank esti- to change and should be regarded as indicative only. mates for the economies of the Former Soviet Union The next round of ICP surveys for 1993 is expected (FSU) based on partial and preliminary ICP data for to cover more than eighty countries, including China the former U.S.S.R. for 1990 extrapolated to 1987; (v) and several FSU economies. a World Bank estimate for China; and (vi) ICP esti- The "international dollar" (1$) has the same pur- mates obtained by regression for the remaining coun- chasing power over total GDP as the U.S. dollar in a tries that did not participate in any of the phases. given year, but purchasing power over subaggregates Economies whose 1987 figures are extrapolated from is determined by average international prices at that another year or imputed by regression are footnoted level rather than by U.S. relative prices. These dollar accordingly. values, which are different from the dollar values of The blend of extrapolated and regression-based GNP or GDP shown in Tables 1 and 3 (see the techni- 1987 figures underlying the seventh column is extrap- cal notes for these tables), are obtained by special olated to 1991 using Bank estimates of real per capita conversion factors designed to equalize the purchas- GDP growth rates and expressed as an index ing powers of currencies in the respective countries. (US = 100) in the eighth column. For countries that This conversion factor, the Purchasing Power of Cur- have ever participated in ICP, as well as for China and rencies (PPC), is defined as the number of units of a the economies of the FSU, the latest available PPC- country's currency required to buy the same amounts based values are extrapolated to 1991 by Bank esti- of goods and services in the domestic market as one mates of growth rates and converted to current "in- dollar would buy in the United States. The computa- ternational dollars" by scaling all results up by the tion involves deriving implicit quantities from na- U.S. inflation rates; these are presented in the ninth tional accounts expenditure data and specially col- column. Footnotes indicate which year PPC-based lected price data and then revaluing the implicit data were extrapolated. Regression estimates of all quantities in each country at a single set of average countries except FSU economies, whether or not they prices. The average price index thus equalizes dollar participated in ICP, extrapolated from 1987 to 1991 prices in every country so that cross-country compar- 320 isons of GDP based on them reflect differences in equation is first obtained by fitting the following re- quantities of goods and services free of price-level gression to 1987 data: differentials. This procedure is designed to bring ln (r) = .5603 ln (ATLAS) + .3136 In (ENROL) + .5706; cross-country comparisons in line with cross-time (.0304) (.0574) (.1734) real value comparisons that are based on constant RMSE = .2324; Adj.R-Sq = .95; N = 78. price series. The ICP figures presented here are the results of a where all variables and estimated values are ex- pressed as US = 100; two-step exercise. Countries within a region or group such as the OECD are first compared using their own r = ICP estimates of per capita GDP converted to group average prices. Next, since group average U.S. dollars by PPC, the array of r consisting of ex- prices may differ from each other, making the coun- trapolations of the most recent actual ICP values tries in different groups not comparable, the group available for countries that ever participated in ICP; prices are adjusted to make them comparable at the ATLAS = per capita GNP estimated by the Atlas world level. The adjustments, done by UNSTAT and method; Eurostat, are based on price differentials observed in ENROL = secondary school enrollment ratio; and a network of "link" countries representing each RMSE = root mean squared error. group. However, the linking is done in a manner that retains in the world comparison the relative levels of ATLAS and ENROL are used as rough proxies of GDP observed in the group comparisons, called intercountry wage differentials for unskilled and "fixity." skilled human capital, respectively. Following Isen- The two-step process was adopted because the rel- man (see Paul Isenman, "Inter-Country Compari- ative GDP levels and rankings of two countries may sons of 'Real' (PPP) Incomes: Revised Estimates and change when more countries are brought into the Unresolved Questions," in World Development, 1980, comparison. It was felt that this should not be al- vol. 8, pp.61-72), the rationale adopted here is that lowed to happen within geographic regions; that is, ICP and conventional estimates of GDP differ mainly that the relationship of, say, Ghana and Senegal because wage differences persist among nations due should not be affected by the prices prevailing in the to constraints on the international mobility of labor. A United States. Thus overall GDP per capita levels are technical paper providing fuller explanation is avail- calculated at "regional" prices and then linked to- able on request (Sultan Ahmad, "Regression Esti- gether. The linking is done by revaluing GDPs of all mates of Per Capita GDP Based on Purchasing Power the countries at average "world" prices and reallocat- Parities," Working Paper Series 956, International ing the new regional totals on the basis of each coun- Economics Department, World Bank, 1992. For fur- try's share in the original comparison. ther details on ICP procedures, readers may consult Such a method does not permit the comparison of the ICP Phase IV report, World Comparisons of Purchas- more detailed quantities (such as food consumption). ing Power and Real Product for 1980 (New York: United Hence these subaggregates and more detailed expen- Nations, 1986). diture categories are calculated using world prices. These quantities are indeed comparable interna- Table 31. Urbanization tionally, but they do not add up to the indicated GDPs because they are calculated at a different set of Data on urban population and agglomeration in large prices. cities are from the U.N. 's World Urbanization Prospects, Some countries belong to several regional groups. supplemented by data from the World Bank. The A few of the groups have priority; others are equal. growth rates of urban population are calculated from Thus fixity is always maintained between members of the World Bank's population estimates; the estimates the European Communities, even within the OECD of urban population shares are calculated from both and world comparison. For Austria and Finland, sources just cited. however, the bilateral relationship that prevails Because the estimates in this table are based on within the OECD comparison is also the one used different national definitions of what is urban, cross- within the global comparison. But a significantly dif- country comparisons should be made with caution. ferent relationship (based on Central European The summary measures for urban population as a prices) prevails in the comparison within that group, percentage of total population are calculated from and this is the relationship presented in the separate country percentages weighted by each country's publication of the European comparison. share in the aggregate population; the other sum- To derive ICP-based 1987 figures for countries that mary measures in this table are weighted in the same are yet to participate in any ICP survey, an estimating fashion, using urban population. 321 Table 32. Women in development Maternal mortality refers to the number of female deaths that occur during childbirth per 100,000 live This table provides some basic indicators disaggre- births. Because deaths during childbirth are defined gated to show differences between the sexes that il- more widely in some countries to include complica- lustrate the condition of women in society. The mea- tions of pregnancy or the period after childbirth, or of sures reflect the demographic status of women and abortion, and because many pregnant women die their access to health and education services. Statisti- from lack of suitable health care, maternal mortality is cal anomalies become even more apparent when so- difficult to measure consistently and reliably across cial indicators are analyzed by gender, because re- countries. The data in these two series are drawn porting systems are often weak in areas related from diverse national sources and collected by the specifically to women. Indicators drawn from cen- World Health Organization (WHO), although many suses and surveys, such as those on population, tend national administrative systems are weak and do not to be about as reliable for women as for men; but record vital events in a systematic way. The data are indicators based largely on administrative records, derived mostly from official community reports and such as maternal and infant mortality, are less reli- hospital records, and some reflect only deaths in hos- able. More resources are now being devoted to de- pitals and other medical institutions. Sometimes velop better information on these topics, but the re- smaller private and rural hospitals are excluded, and liability of data, even in the series shown, still varies sometimes even relatively primitive local facilities are significantly. included. The coverage is therefore not always com- The under-5 mortality rate shows the probability of a prehensive, and the figures should be treated with newborn baby dying before reaching age 5. The rates extreme caution. are derived from life tables based on estimated cur- Clearly, many maternal deaths go unrecorded, par- rent life expectancy at birth and on infant mortality ticularly in countries with remote rural populations; rates. In general, throughout the world more males this accounts for some of the very low numbers are born than females. Under good nutritional and shown in the table, especially for several African health conditions and in times of peace, male chil- countries. Moreover, it is not clear whether an in- dren under 5 have a higher death rate than females. crease in the number of mothers in hospital reflects These columns show that female-male differences in more extensive medical care for women or more com- the risk of dying by age 5 vary substantially. In indus- plications in pregnancy and childbirth because of trial market economies, female babies have a 23 per- poor nutrition, for instance. (Table 28 shows data on cent lower risk of dying by age 5 than male babies; low birth weight.) the risk of dying by age 5 is actually higher for fe- These time series attempt to bring together readily males than for males in some lower-income econ- available information not always presented in inter- omies. This suggests differential treatment of males national publications. WHO warns that there are in- and females with respect to food and medical care. evitably gaps in the series, and it has invited coun- Such discrimination particularly affects very young tries to provide more comprehensive figures. They girls, who may get a smaller share of scarce food or are reproduced here, from the 1991 WHO publication receive less prompt costly medical attention. This pat- Maternal Mortality: A Global Factbook. The data refer to tern of discrimination is not uniformly associated any year from 1983 to 1991. with development. There are low- and middle-in- The education indicators, based on UNESCO come countries (and regions within countries) where sources, show the extent to which females have equal the risk of dying by age 5 for females relative to males access to schooling. approximates the pattern found in industrial coun- Percentage of cohort persisting to grade 4 is the per- tries. In many other countries, however, the numbers centage of children starting primary school in 1970 starkly demonstrate the need to associate women and 1986, respectively, who continued to the fourth more closely with development. The health and wel- grade by 1973 and 1989. Figures in italics represent fare indicators in both Table 28 and in this table's earlier or later cohorts. The data are based on enroll- maternal mortality column draw attention, in particu- ment records. The slightly higher persistence ratios lar, to the conditions associated with childbearing. for females in some African countries may indicate This activity still carries the highest risk of death for male participation in activities such as animal women of reproductive age in developing countries. herding. The indicators reflect, but do not measure, both the All things being equal, and opportunities being the availability of health services for women and the gen- same, the ratios for females per 100 males should be eral welfare and nutritional status of mothers. close to 100. However, inequalities may cause the ra- Life expectancy at birth is defined in the note to tios to move in different directions. For example, the Table 1. number of females per 100 males will rise at second- 322 ary school level if male attendance declines more rap- timated to be between 68,000 and 71,000 square kilo- idly in the final grades because of males' greater job meters in 1987. The most recent estimate for Malaysia opportunities, conscription into the army, or migra- is 185,000 square kilometers. tion in search of work. In addition, since the numbers Total annual deforestation refers to both closed and in these columns refer mainly to general secondary open forest. Open forest is defined as at least a 10 education, they do not capture those (mostly males) percent tree cover with a continuous ground cover. In enrolled in technical and vocational schools or in full- the ECE countries, open forest has 5-20 percent time apprenticeships, as in Eastern Europe. crown cover or a mixture of bush and stunted trees. All summary measures are country data weighted Deforestation is defined as the permanent conversion by each country's share in the aggregate population. of forest land to other uses, including pasture, shift- ing cultivation, mechanized agriculture, or infrastruc- Table 33. Forests, protected areas, and water ture development. Deforested areas do not include resources areas logged but intended for regeneration, nor areas degraded by fuelwood gathering, acid precipitation, This table on natural resources represents a step to- or forest fires. In temperate industrialized countries ward including environmental data in the assessment the permanent conversion of remaining forest to of development and the planning of economic strate- other uses is relatively rare. Assessments of annual gies. It provides a partial picture of the status of for- deforestation, both in open and closed forest, are dif- ests, the extent of areas protected for conservation or ficult to make and are usually undertaken as special other environmentally related purposes, and the studies. The estimates shown here for 1981-85 were availability and use of fresh water. The data reported calculated in 1980, projecting the rate of deforestation here are drawn from the most authoritative sources during the first five years of the decade. Figures from available. Perhaps even more than other data in this other periods are based on more recent or better as- Report, however, these data should be used with cau- sessments than those used in the 1980 projections. tion. Although they accurately characterize major dif- Special note should be taken of Brazilthe country ferences in resources and uses among countries, true with the world's largest tropical closed forestwhich comparability is limited because of variation in data now undertakes annual deforestation estimates. The collection, statistical methods, definitions, and gov- estimate of deforestation is the most recent. Brazil is ernment resources. unique in having several assessments of forest extent No conceptual framework has yet been agreed and deforestation that use common methodology upon that integrates natural resource and traditional based on images from Landsat satellites. Closed for- economic data. Nor are the measures shown in this est deforestation in the Legal Amazon of Brazil dur- table intended to be final indicators of natural re- ing 1990 is estimated at 13,800 square kilometers, source wealth, environmental health, or resource de- down from the 17,900 square kilometers estimated in pletion. They have been chosen because they are 1989. Between 1978 and 1988, deforestation in this available for most countries, are testable, and reflect region averaged about 21,000 square kilometers per some general conditions of the environment. year, having peaked in 1987 and declined greatly The total area of forest refers to the total natural thereafter. By 1990, cumulative deforestation (both stands of woody vegetation in which trees predomi- recent and historical) within the Legal Amazon to- nate. These estimates are derived from country statis- taled 415,000 square kilometers. Deforestation out- tics assembled by the Food and Agriculture Organiza- side the Legal Amazon also occurs, but there is much tion (FAO) in 1980. Some of them are based on more less information on its extent. A 1980 estimate, that recent inventories or satellite-based assessments per- open forest deforestation in Brazil totaled about formed during the 1980s. In 1993 the FAO will com- 10,500 square kilometers, is the most recent available. plete and publish an assessment of world forest ex- Nationally protected areas are areas of at least 1,000 tent and health that should modify some of these hectares that fall into one of five management catego- estimates substantially. The total area of closed forest ries: scientific reserves and strict nature reserves; na- refers to those forest areas where trees cover a high tional parks of national or international significance proportion of the ground and there is no continuous (not materially affected by human activity); natural ground cover. Closed forest, for members of the Eco- monuments and natural landscapes with some nomic Commission for Europe (ECE), however, is de- unique aspects; managed nature reserves and wild- fined as those forest areas where tree crowns cover life sanctuaries; and protected landscapes and sea- more than 20 percent of the area. These natural scapes (which may include cultural landscapes). This stands do not include tree plantations. More recent table does not include sites protected under local or estimates of total forest cover are available for some provincial law or areas where consumptive uses of countries. Total forest area in the Philippines was es- wildlife are allowed. These data are subject to varia- 323 tions in definition and in reporting to the organiza- tween 1970 and 1989. Data for small countries and tions, such as the World Conservation Monitoring countries in arid and semiarid zones are less reliable Centre, that compile and disseminate these data. To- than those for larger countries and those with higher tal surface area is used to calculate the percentage of rainfall. total area protected. Total water resources include both internal renewable Freshwater withdrawal data are subject to variation resources and, where noted, river flows from other in collection and estimation methods but accurately countries. Estimates are from 1992. Annual internal show the magnitude of water use in both total and renewable water resources refer to the average an- per capita terms. These data, however, also hide nual flow of rivers and of aquifers generated from what can be significant variation in total renewable rainfall within the country. The total withdrawn and water resources from one year to another. They also the percentage withdrawn of the total renewable re- fail to distinguish the variation in water availability source are both reported in this table. Withdrawals within a country both seasonally and geographically. include those from nonrenewable aquifers and de- Because freshwater resources are based on long-term salting plants but do not include evaporative losses. averages, their estimation explicitly excludes decade- Withdrawals can exceed 100 percent of renewable long cycles of wet and dry. The Département Hydro- supplies when extractions from nonrenewable aqui- géologie in Orleans, France, compiles water resource fers or desalting plants are considerable or if there is and withdrawal data from published documents, in- significant water reuse. Total per capita water with- cluding national, United Nations, and professional drawal is calculated by dividing a country's total literature. The Institute of Geography at the National withdrawal by its population in the year that with- Academy of Sciences in Moscow also compiles global drawal estimates are available. Domestic use includes water data on the basis of published work and, where drinking water, municipal use or supply, and uses for necessary, estimates water resources and consump- public services, commercial establishments, and tion from models that use other data, such as area homes. Direct withdrawals for industrial use, includ- under irrigation, livestock populations, and precipita- ing withdrawals for cooling thermoelectric plants, are tion. These and other sources have been combined by combined in the final column of this table with with- the World Resources Institute to generate (un- drawals for agriculture (irrigation and livestock pro- published) data for this table. Withdrawal data are for duction). Numbers may not sum to the total per cap- single years and vary from country to country be- ita figure because of rounding. 324 Data sources Production U.N. Department of International Economic and Social Affairs. Various years. Statistical Yearbook. New and domestic York. absorbtion Various years. Energy Statistics Yearbook. Statistical Papers, series J. New York. U.N. International Comparison Program Phases IV (1980), V (1985), and Phase VI (1990) reports, and data from ECE, ESCAP, Eurostat, OECD, and U.N. FAO, IMF, UNIDO, and World Bank data; national sources. Fiscal and International Monetary Fund. Government Finance Statistics Yearbook. Vol. 11. Washington, D.C. monetary Various years. International Financial Statistics. Washington, D.C. accounts U.N. Department of International Economic and Social Affairs. Various years. World Energy Supplies. Statistical Papers, series J. New York. IMF data. Core International Monetary Fund. Various years. International Financial Statistics. Washington, D.C. international U.N. Conference on Trade and Development. Various years. Handbook of International Trade and transactions Development Statistics. Geneva. U.N. Department of International Economic and Social Affairs. Various years. Monthly Bulletin of Statistics. New York. Various years. Yearbook of International Trade Statistics. New York. FAO, IMF, U.N., and World Bank data. External Organization for Economic Cooperation and Development. Various years. Development Co-operation. finance Paris. 1988. Geographical Distribution of Financial Flows to Developing Countries. Paris. Human IMF, OECD, and World Bank data; World Bank Debtor Reporting System. and natural Bos, Eduard, Patience W. Stephens, and My T. Vu. World Population Projections, 1992-93 Edition resources (forthcoming). Baltimore, Md.: Johns Hopkins University Press. Chen, S., G. Datt, and M. Ravallion. 1993. "Is Poverty Increasing in the Developing World?" Working Paper Series 1146. World Bank, Policy Research Department, Washington D.C. Institute for Resource Development/Westinghouse. 1987. Child Survival: Risks and the Road to Health. Columbia, Md. Mauldin, W. Parker, and Holden J. Segal. 1988. "Prevalence of Contraceptive Use: Trends and Issues." Studies in Family Planning 19, 6: 335-53 Sivard, Ruth. 1985. WomenA World Survey. Washington, D.C.: World Priorities. U.N. Department of International Economic and Social Affairs. Various years. Demographic Yearbook. New York. Various years. Population and Vital Statistics Report. New York. Various years. Statistical Yearbook. New York. 1989. Levels and Trends of Contraceptive Use as Assessed in 1988. New York. 1988. Mortality of Children under Age 5: Projections 1950-2025. New York. 1991. World Urbanization Prospects 1991. New York. 1991. World Poptlation Prospects: 1990. New York. 1992. World Population Prospects: 1992 Revision. New York. U.N. Educational Scientific and Cultural Organization. Various years. Statistical Yearbook. Paris. 1990. Compendium of Statistics on Illiteracy. Paris. UNICEF. 1989. The State of the World's Children 1989. Oxford: Oxford University Press. World Health Organization. Various years. World Health Statistics Annual. Geneva. 1986. Maternal Mortality Rates: A Thbulation of Available Information, 2nd edition. Geneva. 1991. Maternal Mortality: A Global Factbook. Geneva Various years. World Health Statistics Report. Geneva. World Resources Institute data (unpublished). FAO and World Bank data World Conservation Monitoring Center data (unpublished). 325 Part 1 Classification of economies by income and region Sub-Saha ran African Asia Europe and Central Asia Middle East and North Africa East & Eastern Europe Income Southern East Asia and and Central Rest of Middle North group Subgroup Africa West Africa Pacific South Asia Asia Europe East Africa Americas Burundi Benin Cambodia Afghanistan Comoros Yemen, Rep. Egypt, Arab Guyana Burkina Faso China Bangladesh Ethiopia Central Rep. Haiti Indonesia Bhutan Kenya African Lao PDR Honduras India Nicaragua Lesotho Rep. Myanmar Maldives Madagascar Chad Solomon Nepal Malawi Equatorial Islands Pakistan Mozambique Guinea Viet Nam Sri Lanka Rwanda Gambia, The Somalia Ghana Low- Sudan Guinea income Tanzania Guinea-Bissau Uganda Liberia Zaire Mali Zambia Mauritania Zimbabwe Niger Nigeria São Tome and Principe Sierra Leone Togo Angola Cameroon Fiji Albania Djibouti Turkey Iran, Islamic Algeria Belize Cape Verde Kiribati Armenia Rep. Mauritius Morocco Bolivia Congo Korea, Dem. Azerbaijan Iraq Namibia Côte d'Ivoire Tunisia Chile Rep. Bulgaria Jordan Swaziland Senegal Malaysia Colombia Czecho- Lebanon Costa Rica Marshall slovakia" Syrian Arab Islands Cuba Georgia Rep. Dominica Micronesia, Kazakhstan Fed. Sts. Dominican Kyrgyzstan Rep. Lower Mongolia Moldova Ecuador Papua New Poland El Salvador Guinea Romania Philippines Grenada Tajikistan Guatemala Thailand Turknienistan Tonga Jamaica Ukraine Panama Vanuatu Uzbekistan Western Paraguay Samoa Peru St. Lucia Middle- St. Vincent . income Botswana Gabon Amencan Belarus . Gibraltar Mayotte Bahram Libya Antigua and Samoa Estonia Greece Oman Reunion Guam Barbuda Hungary Isle of Man Saudi Arabia Argentina Seychelles Korea, Rep. Latvia Malta South African Macao Aruba Lithuania Portugal Barbados New Russian Brazil Caledonia Federation Yugoslaviac French Guiana Guadeloupe Upper Martinique Mexico Netherlands Antilles Puerto Rico St. Kitts and Nevis Suriname Trinidad and Tobago Uruguay Venezuela No.oflow-&rniddle- income economies: 162 26 23 25 8 22 6 9 5 38 326 Part 1 (continued) Sub-Saha ran African Asia Europe and Central Asia Middle East and North Africa East& Eastern Europe Income Southern East Asia and and Central Rest of Middle North group. Subgroup Africa West Africa Pacific South Asia Asia Europe East Africa Americas Australia Andorra Canada Japan Austria United New Belgium States Zealand Denmark Finland France Germany Iceland OECD Ireland countries Italy Luxembourg High- Netherlands income Norway San Marino Spain Sweden Switzerland United Kingdom Brunei Channel Israel Bahamas French Islands Kuwait Bermuda Non-OECD Polynesia Cyprus Qatar Virgin countries Hong Kong Faeroe United Arab Islands (US) Singapore Islands Emirates OAE Greenland Total no. of economies: 201 26 23 33 8 22 28 13 5 43 For some analysis, South Africa is not included in Sub-Saharan Africa. Refers to the former Czechoslovakia; disaggregated data are not yet available. Refers to the former Socialist Federal Republic of Yugoslavia; disaggregated data are not yet available. Other Asian economiesTaiwan, China. - Definitions of groups These tables classify all World Bank member economies, plus all upper-middle-income, $2,556$7,910; and high-income, $7,911 other economies with populations of more than 30,000. or more. Income group: Economies are divided according to 1991 GNP per The estimates for the republics of the former Soviet Union should capita, calculated using the World Bank Atlas method. The groups be regarded as very preliminary; their classification will be kept are: low-income, $635 or less; lower-middle-income, $636-2,555; under review. 327 Part 2 Classification of economies by major export category and indebtedness Low- and middle-income Low-income Middle-income High-income Severely Moderately Less Severely Moderately Less Not classified Group indebted indebted indebted indebted indebted indebted by indebtedness OECD non-OECD China Bulgaria Hungary Czecho- Armenia Belgium Hong Kong Poland slovakia' Belarus Canada Israel Korea, Dem. Estonia Finland Sing,pore Rep. Georgia Germany OAE Korea, Rep. Kyrgyzstan Ireland Exporters of Lebanon Latvia Italy manufactures Macao Lithuania Japan Romania Moldova Luxembourg Russian Sweden Federation Switzerland Ukraine Uzbekistan Afghanistan Guinea Chad Albania Chile Botswana American Iceland Faeroe Burundi Malawi Solomon Argentina Costa Rica French Samoa New Islands Equatorial Rwanda Islands Bolivia Guatemala Guiana Zealand Greenland Guinea Togo Zimbabwe Côte d'Ivoire Papua New Guadeloupe Ethiopia Cuba Guinea Namibia Ghana Mongolia Paraguay Guinea-Bissau Peru Reunion Guyana St. Vincent Honduras Suriname Liberia Swaziland Exporters of nonfnel Madagascar Mauritania primary Myanmar products Nicaragua Niger São Tome and Principe Somalia Tanzania Uganda Viet Nam Zaire Zambia Nigeria Algeria Gabon Iran, Islamic Turkmenistan Brunei Angola Venezuela Rep. Qatar Exporters Congo Libya United Arab of fuels Iraq Oman Emirates (mainly oil) Saudi Arabia Trinidad and Tobago Cambodia Benin Bhutan Jamaica Dominican Antigua and Aruba United Bahamas Egypt, Arab Gambia, The Burkina Faso Jordan Rep. Barbuda Kingdom Bermuda Rep. Haiti Lesotho Panama Greece Barbados Cyprus Sudan Maldives Cape Verde French Nepal Djibouti Polynesia Yemen, Rep. El Salvador Fiji Grenada Kiribati Malta Exporters Martinique oj services Netherlands Antilles Seychelles St. Kitts and Nevis St. Lucia Tongs Vanuatu Western Samoa Kenya Bangladesh Brazil Cameroon Bahrain Azerbaijan Australia Kuwait Lao PDR Central Ecuador Colombia Belize Kazakhstan Austria Mali African Rep. Mexico Philippines Dominica Tajikistan Denmark Mozambique Comoros Morocco Senegal Malaysia France Diversied Syrian Arab Tunisia Mauritius Netherlands Sierra Leone India exporters Turkey Portugal Norway Indonesia Rep. Pakistan Uruguay South Africa Spain Sri Lanka Thailand United Yugoslavia' States 328 Part 2 (continued) Low- and middle-income Low-income Middle-income High-income Severely Moderately Less Severely Moderately Less Not classified Group indebted indebted indebted indebted indebted indebted by indebtedness OECD non-OECD Gibraltar Andorra Guam Channel Isle of Man Islands Marshall San Marino Not classified Islands Virgin Islands by export Mayotte (US) category Micronesia, Fed. Sts. New Caledoma Puerto Rico No. of economies 201 30 17 7 21 16 47 24 21 18 Refers to the former Czechoslovakia; disaggregated data are not yet available. Other Asian economiesTaiwan, China. Refers to the former Socialist Federal Republic of Yugoslavia; disaggregated data are not yet available. Definitions of groups These tables classify all World Bank member economies, plus all of the two key ratios exceeds 60 percent of, but does not reach, the other economies with populations of more than 30,000. critical levels. For economies that do not report detailed debt Major export category: Major exports are those that account for statistics to the World Bank Debtor Reporting System, 50 percent or more of total exports of goods and services from one present-value calculation is not possible. Instead the following methodology is used to classify the non-DRS economies. Severely category, in the period 1987-89. The categories are: nonfuel indebted means three of four key ratios (averaged over 198 8-90) primary (SITC 0,1,2, 4, plus 68), fuels (SITC 3), manufactures (SITC 5 to 9, less 68), and services (factor and nonfactor service are above critical levels: debt to GNP (50 percent); debt to exports receipts plus workers' remittances). If no single category accounts (275 percent), debt service to exports (30 percent); and interest to for 50 percent or more of total exports, the economy is classified exports (20 percent). Moderately indebted means three of four key as diversified. ratios exceed 60 percent of, but do not reach, the critical levels. All other low- and middle-income economies are classified as Indebtedness: Standard World Bank definitions, of severe and less-indebted. moderate indebtedness, averaged over three years (1989-91) are used to classify economies in this table. Severely indebted means Not classified by indebtedness are the republics of the Former either of the two key ratios is above critical levels: present value Soviet Union and some small economies for which detailed debt data are not available of debt service to GNP (80 percent) and present value of debt service to exports (220 percent). Moderately indebted means either 329 How to order World Development Report CUSTOMERS IN THE UNITED STATES: CUSTOMERS OUTSIDE THE UNITED STATES: Complete this coupon and return to: Contact your local World Bank publications distributor or World Bank Publications branch of Oxford University Press for information on Box 7247-8619 prices in local currency and payment terms. If no Philadelphia, PA 19170-8619, U.S.A. distributor is listed for your country, use this order form To have your order shipped faster, call (202) 473-1155 to and return it to the U.S. address. Orders from countries charge by credit card, or send this completed order coupon with distributors that are sent to the U.S. address will be by facsimile by dialing (202)676-0581. returned to the customer. Order Unit Total number Title Quantity price amount($) 60890 World Development Report 1993: Investing in Health 19.95 60876 World Development Report 1992: Development and the Environment 19.95 60868 World Development Report 1991: The Challenge of Development 19.95 60851 World Development Report 1990: Poverty 19.95 Other language editions are available. Please contact customer service or your local distributor for languages and availabifity. Subtotal US$ * SHIPPING AND HANDLING charges are US$3.50 if paying by check or credit card. If purchase order is used, actual shipping and handling costs will be charged. For air Shipping and handling* US $ mail delivery outside the U.S., include an additional US$6.00 per item. Allow 6-8 weeks for delivery by surface mail outside the U.S. Total US $ CHECK METHOD OF PAYMENT U Enclosed is my check, payable to World Bank Publications. U Charge my U VISA U Mastercard U American Express Credit card number Expiration date Signature U Bill me. (Institutional customers only. Purchase order must be included.) PLEASE TYPE OR PRINT. Legible information ensures prompt and correct delivery. Date Ship to: (Enter if different from purchaser.) Name Name Firm Firm Address Address City State City State Postal code Country Postal code Country Telephone Telephone Thank you for your order WR93 Distributors of World Bank Publications ARGENTINA The Middle East Observer KENYA SOUTH AFRICA, BOTSWANA Carlos Hirsch, SRL 41, Sherif Street Africa Book Service (E.A.) Ltd. For single titles: Caleria Guemes Cairo Quaran House, Mfangano Street Oxford University Press florida 165,4th Floor-Ofc. 453/465 P.O. Box 45245 Southern Africa 1333 Buenos Aires FINLAND Nairobi P.O. Box 1141 Akateeminen Kirjakauppa Cape Town 8000 AUSTRALIA, PAPUA NEW GUINEA, P.O. Box 128 KOREA, REPUBLIC OF FIJI, SOLOMON ISLANDS, SF-00101 Helsinki 10 Pan Korea Book Corporation For sutacription orders: VANUATU, AND WESTERN SAMOA P.O. Box 101, Kwangwhamun International Subscription Service D.A. Information Services FRANCE Seoul P.O. Box 41095 648 Whitehorse Road World Bank Publications Craighall Mitcham 3132 66, avenue d'Ihna MALAYSIA Johannesburg 2024 Victoria 75116 Paris Univ. of Malaya Coop. Bookshop, Ltd. P.O. Box 1127, Jalan Pantai Baru SPAIN AUSTRIA GERMANY 59700 Kuala Lumpur Mundi-Prensa Libros, S.A. Gerold and Co. UNO-Verlag Castello 37 Graben 31 Poppelsdorfer Allee 55 MEXICO 28001 Madrid A-loll Wien 13-5300 Bonn I INFOTEC Apartado Postal fl-860 Librerla lnternacional AEDOS BANGLADESH HONG KONG, MACAO 14060 lialpan, Mexico D.F. Consell de Cent, 391 Micro Industries Development Asia 2000 Ltd. 08009 Barcelona Assistance Society (MIDAS) 46-48 Wyndham Street NETHERLANDS House5, Road 16 Winning Centre Do Undeboom/lnOr-Publikalies SRI LANKA AND THE MALDIVES Dhanmondi R/Area 2nd Floor P.O. Box 202 Lake House Bookshop Dhaka 1209 Central Hong Kong 7480 AE Haaksbergen P.O. Box 244 100, Sir Chittampalam A. Branch offices: INDIA NEW ZEALAND Gardiner Mawatha Pine View, 1st Floor Allied Publishers Private Ltd. EBSCO NZ Ltd. Colombo 2 100 Agrabad Commercial Area 751 Mount Road Private Mall Bag 99914 Chittagong 4100 Madras- 600 002 New Market SWEDEN Auckland For single titles: 76, K.D.A. Avenue Branch offices: Fritzes Fackboksforetaget Kulna 9100 15 J.N. Heredia Marg NIGERIA Regeringsgatan 12, Box 16356 Ballard Estate University Press Limited S-103 27 Stockholm BELGIUM Bombay - 400 038 Three Crowns Building Jericho Jean Do L.annoy Private Mail Bag 5095 For subscription orders: Av. du Roi 202 13/14 Asaf All Road Ibadan Wennergren-Williams AB 1060 Brussels New Delhi - 110 002 P.O. Box 1305 NORWAY S-171 25 Solna CANADA 17 Chittaranjan Avenue Narvesen Information Center Le Diffuseur Calcutta -700 072 Book Department SWITZERLAND C.P. 85, 1501B rue Ampere P.O. Box 6125 Etterstad For single titles: Boucherville, Québec Jayadeva Hostel Building N-0602 Oslo 6 Librairie Payot J4B 506 5th Main Road, Gandhinagar Case postale 3212 Bangalore - 560 009 PAKISTAN CH 1002 Lausanne CHILE Mirza Book Agency Invertec ICT S.A. 3-5-1129 Kachiguda 65, Shahrah-e-Quaid-e-Azam For subscription orders: Americo Vespucio Norte 1165 Cross Road P.O. Box No. 729 Librairie Payot Santiago Hyderabad -500 027 Lahore 54000 Service des Abonnements Case postale 3312 CHINA Prarthana Flats, 2nd Floor PERU CH 1002 Lausanne China Financial & Economic Near Thakore Baug, Navrangpura Editorial Desarrollo SA Publishing House Ahmedabad - 380 009 Apartado 3824 THAILAND 8, Da Fo Si Dong Jie Lima I Central Department Store Beijing Patiala House 306 Silom Road 16-A Ashok Marg PHILIPPINES Bangkok COLOMBIA Lucknow -226001 International Book Center Infoenlace Ltda. Suite 1703, Cityland 10 TRINIDAD & TOBAGO, ANTIGUA Apartado Aereo 34270 Central Bazaar Road Condominium Tower I BARBUDA, BARBADOS, Bogota D.E. 60 Bajaj Nagar Ayala Avenue, H.V. dela DOMINICA, GRENADA, GUYANA, Nagpur 440 010 Costa Extension JAMAICA, MONTSERRAT, ST. COTE D'IVOIRE Makati, Metro Manila KITrS & NEVIS, ST. LUCIA, Centre d'Edition et de Diffusion INDONESIA ST. VINCENT & GRENADINES Africaines (CEDA) Pt. Indira Limited POLAND Systematics Studies Unit 04 B.P. 541 Jalan Borobudur 20 International Publishing Service #9 Watts Street Abidjan 04 Plateau P.O. Box 181 UI. Piekna 31/37 Curepe Jakarta 10320 00-677 Warzawa Trinidad, West Indies CYPRUS Center of Applied Research IRELAND For suecription orders: TURKEY Cyprus College Government Supplies Agency II'S Journals Infotel 6, Diogenes Street, Engomi 4-5 Flarcourt Road UI. Okrezna 3 Narlabahce Sok. No.15 P.O. Box 2006 Dublin 2 02-916 Warszawa Cagaloglu Nicosia Istanbul ISRAEL PORTUGAL DENMARK Yozmot Literature Ltd. Livraria Portugal UNITED KINGDOM SamfundsUtteratur P.O. Box 56055 Rua Do Carmo 70-74 Microinfo Ltd. Rosenoerns Allé 11 Tel Aviv 61560 1200 Lisbon P.O. Box 3 DK-l970 Frederiksberg C Alton, Hampshire GU34 2PC ITALY SAUDI ARABIA, QATAR England DOMINICAN REPUBLIC Licosa Commissionaria Sansoni SPA Jarir Book Store Editora Taller, C. por A. Via Duca Di Calabria, 1/1 P.O. Box 3196 VENEZUELA Reslauración e Isabel la Catblica 309 Caaella Postale 552 Riyadh 11471 Libreria del Este Apartado de Correos 2190 Z-1 50125 Firenze Aptdo. 60.337 Santo Domingo SINGAPORE, TAIWAN, Caracas 1060-A JAPAN MYANMAR,BRUNEI EGYPT, ARAB REPUBLIC OF Eastern Book Service Information PublicationsPrivate, Ltd. Al Ahram Hongo 3-chome, Bunkyo-ku 113 Golden Wheel Building Al Calaa Street Tokyo 41, Kallang Pudding, #04.03 Cairo Singapore 1334 The World Bank Becausegood health increases the economic prpducthrit of individuals and the economic growth rate of countries, investing in health is one means of accelerating development More importat, good health is a goal in itself )uring the past forty years life expectancy u the developing world has risen and child mOrtality has decreased, sometimes dramatically But progress is only one side o the picture The toll from childhood and tropical diseases remains high even as new problei including AIDS and the diseases of aging populationsappear on the scene And all counthes are struggling with the problems of controlling health expenditures and making health care accessible to the broad population This sixteenth annual World Development Report examines the controversial q1ieS tions surrounding health care and health policy its findings are based in large part on innova- tive research, including estimation of the global burden of disease and the cost-effectiveness of interventions These assessments can help in setting priorities for health spending The Report advocates a threefold approach to health policy for governments in devel- oping countries and in the formerly socialist countries Foster an economic environment that will enable households to improve their own health .ro1ciesor econo flic growth that ensure income gains for the poor are essential. So, too, is expand&tmvestment in schooling, particularly for girls Redirect government spending away from specialized care and toward such low- cost and highly effective activities as immunization, programs to combat micronutrient defi- ciencies, and control and treatment of mfectious diseases By adopting the packages of public health measures and essential clinical dare described in the Report, developing countries could reduce their burden of disease by 25 percent Encourage greater diversity and competition in the provision of health services by decentralizing government services,' promoting competitive procurement practices, fostering greater involvement by nongovernmental and other private organizations, and regulating insurance markets. These reforms could translate into longer, healthier, and more productive lives for people around the world, and especially for the more than 1 billion poor As in previous editions, this Report includes the World Development Indicators, whih give comprehensive1 current data on social and economic development in more than 200 countries and territories The Indicatois are also available on diskette for ise with personal conp/uters Special appendices to the Report provide hetlth statistics and estimates of the global burden of disease. 90000 Cover design by Walt Rosenquist 9 780195 208900 ISBN 0-19-520890-0