- - .- - *VE. U 'mi 2Zt- . * m um --U *  § I me i * I 4' * I hr I. lit.  4  . . <- A .   '.,  - E7im Environmentally Sustainable Development Series Proceedings Culture and Development: Proceedings of an International Conference ESD Proceedings Series no. 1 (1994) (Also in French) Valuing the Environment: Proceedings of the First Annual International Conference on Environmentally Sustainable Development ESD Proceedings Series no. 2 (1994) Overcoming Global Hunger: Proceedings of a Conference on Actions to Reduce Hunger Worldwide ESD Proceedings Series no. 3 (1994) Traditional Knowledge and Sustainable Development: Proceedings of a Conference ESD Proceedings Series no. 4 (1995) The Human Face of the Urban Environment: A Report to the Development Community ESD Proceedings Series no. 5 (Forthcoming) The Human Face of the Urban Environment: Proceedings of the Second Annual World Bank Conference on Environmentally Sustainable Development ESD Proceedings Series no. 6 (Forthcoming) The Business of Sustainable Cities -Public-Private Partnerships for Creative Technical and Institutional Solutions ESD Proceedings Series no. 7 (Forthcoming) Enabling Sustainable Community Development ESD Proceedings Series no. 8 (Forthcoming) Studies and Monographs (formerly Occasional Paper) The Contribution of People's Participation: Evidencefrom 121 Rural Water Supply Projects ESD Occasional Paper Series no. 1 (Forthcoming) Making Development Sustainable: From Concepts to Action ESD Occasional Paper Series no. 2 (1994) Sociology, Anthropology, and Development: An Annotated Bibliography of World Bank Publications 1975-1993 ESD Studies and Monographs Series no. 3 (1994) The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community ESD Studies and Monographs Series no. 4 (1995) The World Bank's Strategy for Reducing Poverty and Hunger A Report to the Development Community Prepared by Hans P. Binswanger and Pierre Landell-Mills Foreword by Lewis T. Preston Preface by Ismail Serageldin k & Environmentally Sustainable Development Studies and Monographs Series No. 4 ESD The World Bank, Washington, D.C. C) 1995 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing March 1995 This report has been prepared by the staff of the World Bank. The judgments expressed do not necessarily reflect the views of the Board of Executive Directors or the governments they represent. Library of Congress Cataloging-in-Publication Data The World Bank's strategy for reducing poverty and hunger: a report to the development community / Hans P. Binswanger, Pierre Landell-Mills; foreword by Lewis T. Preston. p. cm - (Environmentally sustainable development studies and monographs series ; no. 4.) ISBN 0-8213-3174-4 1. World Bank-Developing countries. 2. Food supply-Developing countries 3. Food relief-Developing countries. 4. Poor- Developing countries. I. Binswanger, Hans P. II. Landell-Mills, Pierre, 1939- III. Series. HG3881.5.W57W696 1995 332.1'532-dc2O 95-1366 CIP Contents Foreword v Preface vii Acknowledgments ix Abbreviations x Notes on Sources xi Executive Summary 1 Actions to Reduce Hunger 9 PART I Growth, Poverty, and Hunger 11 Causes of Hunger and Elements of Food Security 11 World Bank's Strategy to Reduce Poverty and Hunger 13 Poverty and Hunger during Structural Adjustment 16 Food Production and Consumption 18 Social Sector Expenditures 22 Participation and Decentralization 24 PART 11 Specific Actions and Programs 29 Urgent Low-Cost Actions to Reduce Hunger 29 Micronutrients and Protein-Energy Malnutrition 30 Immunization 31 Mass Treatment for Parasitic Worms 32 Food Supplementation 33 General Food Price Subsidies 34 Rationed Food Subsidies 34 Food Coupons or Food Stamps 35 Targeting Hungry Families through Nutrition and Growth Monitoring 35 iii iv The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community Improving Effectiveness of Food Aid and Dealing with Famines 38 Food Aid 38 Famines 39 Interventions to Improve the Incomes of the Poor and Hungry 41 Education 41 Urban Informal Sector Employment 41 Credit for the Poor 42 Food Security in Resource-Poor Areas 43 Land Reform 47 Agricultural Research and Extension 48 Technology and World Food Supply 48 Access of the Poor to Improved Technology 49 World Bank Actions 50 Notes 53 Selected References 55 Tables 1 Direct and indirect contributions of malnutrition to the global burden of disease, 1990 31 2 Costs and health benefits of the EPI plus cluster in two developing country settings, 1990 32 3 Distribution of the annual income transfer from the general food subsidy, Egypt, 1981-82 35 4 General and targeted subsidies, Jamaica, 1988 38 Figure 1 Changes in GNP and prevalence of underweight children from late 1970s to early-late 1980s. 15 Boxes 1 Diagnosis checklist for poverty assessments 16 2 Importance of economic growth for reducing hunger and poverty 16 3 Adjustment program design checklist 17 4 Does commercialization hurt the poor? 20 5 Redirecting social expenditures in Brazil 23 6 Checklist for evaluating public expenditures 23 7 Ghana rural water supplies-importance of local participation 25 8 Beneficiary assessment 26 9 Rural development administration in India 27 10 Oral rehydration therapy: an effective, low-cost way of reducing deaths 30 11 Tamil Nadu Integrated Nutrition Project: making supplementary feeding work 34 12 Chile's success in combining health and nutrition programs 36 13 Honduras Family Assistance Program 37 14 Impact of food aid in Africa 38 15 Drought relief in Africa: food interventions in Botswana 40 16 Land resource degradation is not inevitable: Machakos, Kenya 44 17 Environmental management project in Burkina Faso 45 18 Land reform options in South Africa 48 19 A food supply crisis is unlikely-at least for now 49 Foreword H Tunger in the midst of plenty is one of * Implementing better agricultural technologies the most difficult development chal- and research-through stronger support for the lenges of our time. Aggregate food pro- Consultative Group on International Agricul- duction continues to increase; yet hunger also tural Research (CGIAR) and for national agri- continues to blight the lives of hundreds of mil- cultural research systems (NARS), for example lions of people. Hunger is sometimes caused by * Focusing on the vital link between environ- drought, disease, or war-and there is always a mental sustainability and increased food need for a strong humanitarian response in production those cases. But the more widespread and deep- * Expanding the participation of the poor in rooted form of hunger is caused by people hav- development through increased access to ing neither the capacity to produce food, nor the credit, land, and services. income to buy it. Fundamentally, hunger is The Bank, working with our partners, is caused by poverty. deeply engaged in supporting these efforts. But If we want to reduce hunger effectively, we much more needs to be done. The discussions have to reduce poverty-and that requires action at the Conference on Overcoming Global Hunger across a broad spectrum: held in Washington, D.C. November 30- * Supporting government policies that encour- December 1, 1993 provided a useful platform age growth and employment for the poor, and -and many good ideas-for strengthening col- removing policies that discriminate against laboration and intensifying action. agriculture and peasant farmers * Investing in people's capacities through edu- Lewis T. Preston cation, health, family planning, and nutrition President, The World Bank v _ _ M Preface O utrage over hunger in the midst of important outcomes. First was a reaffirmation by plenty drove U.S. Congressman Tony the Bank of its commitment to fight global hun- Hall to a hunger strike in April 1993. ger, expressed in the present document. The sec- This action sparked the World Bank, long com- ond outcome was an in-depth exploration of the mitted to the fight against global hunger, to join possibility of the Bank's joining forces with other forces with Hall and to organize the Conference donors to create a new Consultative Group to on Overcoming Global Hunger, which took Assist the Poorest of the Poor (CGAPP), funded place in Washington, D.C., November 30 to with $100 million to promote the replication and December 1, 1993.1 The conference was orga- growth of NGO-managed programs that provide nized in close collaboration with a number of financial services for the poor. Third, multilateral nongovernmental organizations (NGOs), consultations have begun among national gov- including Bread for the World, InterAction, ernments, NGOs, bilateral and international Results, and World Vision. Its participants agencies, and the World Bank on country-spe- included Hall, President Ketumile Masire of cific programs to combat poverty and hunger, Botswana, United Nations Secretary General starting with Mali, Nicaragua, and the Boutros Boutros-Ghali, former U.S. President Philippines. Jimmy Carter, and World Bank President Lewis This report is the result of discussions within T. Preston. The conference's objective was to the Bank as well as dialogues with NGOs and raise awareness about the problems of hunger other international agencies concerned with associated with extreme poverty as opposed to poverty and hunger. It was prepared by Hans hunger caused by natural disasters such as Binswanger and Pierre Landell-Mills with inputs drought. For the Bank, the conference was from many colleagues. The report summarizes intended to review and reinforce the array of Bank policy on these issues and details a coordi- tools the Bank deploys in the fight against nated program of actions that the Bank, in col- hunger. laboration with other organizations, will The conference and subsequent activities promote in the next few years. This program will have greatly enhanced cooperation between be evaluated periodically and refined based on NGOs and the Bank. Attended by more than new knowledge and further consultations inside 1,200 participants, the conference yielded three and outside the Bank. vii viii The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community These actions stem from a profound commit- has the means to feed its people is unconscion- ment to ending the continued presence of hun- able and unacceptable. We must become the new ger in our midst. In the nineteenth century some abolitionists. people declared that slavery was unconscionable and unacceptable, that it degraded the free as well as the slaves, and that it must be abolished. Ismail Serageldin They were called abolitionists. Today hunger Vice President, The World Bank associated with extreme poverty in a world that Chairman, Conference on Overcoming Global Hunger Acknowledgments T his report was prepared by Hans Alberto Valdes, Isabel Vial, Harry Walters, and Binswanger and Pierre Landell-Mills with Jacob Yaron provided materials and valuable sug- the asssistance of Howard Isenstein, under gestions. Chris Brown edited the report. The the direction of Ismail Serageldin. Harold authors also wish to acknowledge the members of Alderman, Alan Berg, Robert Christiansen, Jean- the Overcoming Global Hunger Conference Steer- Jacques De St. Antoine, Margaret Grosh, Peter ing Committee for their invaluable contributions. Heywood, Norman Hicks, Dean Jamison, In addition, the authors wish to express their Christine Jones, Victor Lavy, Johannes Linn, appreciation to Eveling Bermudez, Alicia William McGreevey, Judith McGuire, Oey Hetzner, Paul Holtz, Dean Housden, and Meesook, Donald Mitchell, Phillip Musgrove, Heather Imboden for bringing this document to Martin Ravallion, Lyn Squire, David Steeds, publication. Tomoko Hirata designed the cover. ix Abbreviations CBO community-based organization CGIAR Consultative Group on International Agricultural Research FAO Food and Agriculture Organization of the United Nations FINCA Foundation for International Community Assistance IFAD International Fund for Agricultural Development IMF International Monetary Fund NGO nongovernmental organization T&V training and visit UN United Nations UNDHA United Nations Department of Humanitarian Affairs UNDP United Nations Development Programme UNICEF United Nations Children's Fund WFP World Food Programme WHO World Health Organization x Notes on Sources his report draws heavily on World Bank (New York: Oxford University Press, 1993); T publications, including Poverty and Hunger: Investing in Nutrition: With World Bank Assistance Issues and Options for Food Security in (Washington, D.C., 1992); and Adjustment in Developing Countries (Washington, D.C., 1986); Africa: Reform, Results, and the Road Ahead World Development Report 1990: Poverty (New (Washington, D.C., 1994). Boxes without attribu- York: Oxford University Press, 1990); Poverty tion were written for this report by the authors Reduction Handbook (Washington D.C., 1992); listed in the Acknowledgments. World Development Report 1993: Investing in Health xi Executive Summary H _ ~unger is the most deplorable manifesta- While research has shown that broad-based tion of poverty.2 It can take three forms, growth is the most important factor in reducing all of which are considered in this report: poverty and hunger, many cost-effective targeted * Starvation, a life-threatening condition caused interventions can substantially accelerate the by insufficient food often associated with reduction of poverty and hunger and should be famines implemented along with broad-based growth * Chronic undernutrition, due to a persistent lack policies. Country-specific conditions and the spe- of calories and protein cial characteristics of the subgroups of the poor * Specific nutrient deficiencies, from inadequate determine the most appropriate interventions. protein and micronutrient intake, often com- Despite the efforts of many international orga- bined with infections and inadequate nutri- nizations, nongovernmental organizations tional knowledge. (NGOs), and donor and recipient countries, A staggering 750 million men, women, and hunger persists. Recognizing this problem, the children go hungry every day. This need not hap- World Bank, in collaboration with NGOs, held a pen. For those of the poor who are unable to pro- Conference on Overcoming Global Hunger in duce their own food, avoiding hunger depends Washington, D.C. on November 30-December 1, primarily on income to buy food: selling cash 1993. The goals of the conference were building crops or obtaining other cash income or income on past initiatives, moving from rhetoric to transfers. Evidence indicates that broad-based, actions to reduce global hunger, determining sustainable economic growth is the best strategy how the Bank can better support such actions, for reducing poverty and hunger. and raising public awareness about the serious- For these reasons, the World Bank strives to ness of global hunger. Drawing on the conference relieve hunger by promoting sustainable agricul- deliberations, this report states the Bank's actions ture and seeking other ways to assist poor people and plans to help countries overcome hunger. increase their incomes. Given its mandate, the The Bank recognizes that it cannot carry out these Bank is well-suited to support medium- and long- actions alone. Defeating hunger requires the range projects to alleviate hunger, leaving famine active participation and collaboration of national relief to better-equipped international organi- governments, international organizations, bilat- zations such as the United Nations Department of eral agencies, NGOs, community-based organi- Humanitarian Affairs (UNDHA), the World Food zations (CBOs) and the empowerment of the Programme (WFP), and the Red Cross. poor. 2 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to tle Development Community Broad-Based Growth such as subsidized credit or parastatal marketing, have favored the better-off. The Bank's poverty Hunger reduction is an integral part of a two- reduction strategy therefore advocates eliminat- pronged approach to sustainable poverty reduc- ing exchange rate and trade policies that result in tion consisting of broad-based economic growth urban bias. Liberalizing agricultural trade, espe- across income groups and improved access to cially trade in food, is often controversial, and education, health care, and other social services. many NGOs believe that smaliholder production The two elements are mutually supporting; one of food should be protected. without the other is insufficient. As the NGOs Eliminating urban bias also involves redirect- emphasized in their submission after the confer- ing public expenditures in rural areas toward pri- ence, "...it is equitable patterns of growth which mary education and health care and focusing are essential to ending poverty and hunger." agricultural support toward the smallholder sec- Broad-based growth can generate sustainable tor. Government farm credit, for example, should income opportunities for the poor. It can enable be targeted to small-scale farmers and to the poor the poor to become more productive, and thereby who would be left out if credit was left entirely to help eliminate hunger. Caloric intake increases the private sector. Such support can enhance the with income and reduces the number of under- competitiveness of small farmers and increase the weight children. Broad-based growth relies on demand for labor, indirectly helping agricultural using the poor and hungry's most abundant workers. Efforts to reduce urban bias in public resource-labor. Countries that have succeeded expenditures have been less successful than reduc- in reducing hunger have sustained rural devel- ing the bias in exchange rate and trade policies, opment and emphasized urban employment. even in some of the strong adjusting countries. This has increased livelihood security, reducing If developing nations were to grow at the rate morbidity and mortality in the process. East Asia did in the mid-1980s, 500 million fewer Policies to achieve broad-based growth and people would be poor in 2000 than currently pro- develop human resources include eliminating jected. Even then, there would still be hundreds antiemployment biases in trade policies, market of millions of hungry people. Cost-effective regulations, the tax structure, labor laws, and actions targeted to improve the health, nutrition, financial sector policies. They focus government and income-earning capacity of the poor are nec- expenditures on basic education, health care, and essary to reduce this suffering, while simultane- other social services for boys and girls, and on ously pursuing broad-based growth and human growth-enhancing public investments and pro- resource development. grams. They promote investment in safe drinking Poverty assessments are the major analytical water and sanitation. They provide education, tool for adapting the Bank's poverty strategy to credit, productive resources, and employment specific country conditions. Understanding the opportunities to women, who play an especially relationship between hunger and other economic important role in reducing hunger. They ensure factors in a country requires a thorough treatment sustainable resource use through cost-effective of hunger in the Bank's country poverty assess- environmental policies and programs, and avoid ments. The accuracy and depth of poverty assess- the despoliation of natural resources. ments can be improved with NGO and CBO In many countries the largest number of the participation. poor and hungry live in rural areas, and even in other countries hunger is often more pronounced Poverty and Hunger in rural areas than in cities. Often the explanation during Structural Adjustment lies in persistent urban bias. Agricultural sectors have been taxed by a combination of overvalued Many countries have been hit by external shocks exchange rates, industrial protection policies, and have postponed adjustment to lower exter- and sometimes export taxes. Policies and pro- nal earnings. Others have veered off the road to grams to offset this taxation have not been able to broad-based growth by creating unsound eco- compensate the rural areas. In addition, many of nomic structures and neglecting human resource the programs intended to compensate farmers, development. Symptoms of inappropriate poli- Executive Summary 3 cies include unsustainable government expendi- who are able to sell some of their food production tures and balance of payment deficits. Other gain. Any change in producer prices of food, symptoms include large social spending imbal- whether up or down, therefore has an immediate ances in favor of the middle and upper classes adverse impact on some of the rural poor. and urban biases in exchange rate and trade poli- In the longer run increases in food prices may cies and in the allocation of public expenditures. not affect net rural buyers adversely if structural These countries must undergo structural adjust- adjustment also leads to increases in the demand ment to return to broadly shared growth and for their labor and in higher rural wage rates. poverty reduction. Delay is particularly costly for Such wage increases are often delayed, leading to the poor, who are hit hardest by economic sequencing and safety net issues. Evaluating the decline. Structural adjustment uses a country's impact of structural adjustment on the poor must resources more efficiently, including the labor of look at the joint net impact of structural adjust- the poor and the agricultural and human ment on food production, labor demand, food resources of rural areas. The process is often prices, and wage rates of different subgroups painfully sluggish and full of political conflicts. among the poor. Structural adjustment provides many oppor- Structural adjustment usually leads to tunities to reform policies and programs so that increases in the price of tradable agricultural they enhance the income and livelihood security commodities. The question becomes whether the of the poor. Examples include focusing agricul- poor can participate in the production and tural extension on household food security of income gains that become available in export sec- small and marginal farmers (especially women), tors. The worry is that women engaged in non- reorienting health systems to primary care, and tradable food production will not benefit, and ensuring that the ultra-poor have access to these that smallholders in general may not be able to services. The capacity of countries to respond to respond to the new opportunities. Even worse, international food price increases and weather these vulnerable groups may be squeezed out by shocks, and to monitor the nutritional status of more powerful producers who become interested the poor, can also improve during reform. in the production of profitable exportables. Structural adjustment often has immediate Policymakers need to focus on supporting small- positive impacts on rural incomes. But it also may holder development and preventing anticompet- be associated with a transitional recession and itive behavior in output markets and distortions lay-offs brought about by reductions in govern- in the input and land markets, which constrain ment spending and the elimination of protection the poor's participation in adjustment gains. of inefficient industries. These effects hurt certain Opening of the trade regime sometimes vulnerable groups unless actions are taken to mit- reduces the price of domestically produced igate the impact on these groups. goods, either because they were heavily pro- The data for Africa, Latin America, and Asia tected prior to adjustment, or because it takes indicate no major change in the share of social time for private markets to emerge after the with- spending during adjustment. But the data sug- drawal of a parastatal. Sometimes producer gest little progress in the reallocation of social prices have fallen because international prices fell spending from the better-off to the poor. Reform- just when a country was trying to liberalize inter- ing public sector spending to protect the hungry national and domestic agricultural trade. In sev- and poor and to improve their social and eco- eral African and Latin American countries all nomic development remains a major unfinished three factors have recently coincided, limiting task, even where other components of structural rural gains from structural adjustment and adjustment have already been implemented. imposing losses on some producers. If these dif- Evaluating the impact of structural adjust- ficulties are transitory it becomes appropriate to ment on incomes, food production, and food con- temporarily restrain imports and help poor pro- sumption of the poor is difficult. If food prices ducers over the transitory difficulties in market- rise as a consequence of adjustment, the urban ing and storing their crops. poor lose. In the rural areas the net buyers of food If a country has little comparative advantage lose, at least in the short run, while the rural poor in a specific food crop, the benefits of importing 4 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community the food for poor buyers may outweigh the losses this way can local communities take control of of the poor engaged in its production. This is one their lives and program outcomes. Greater of the reasons why free trade in food is so con- responsibility must be accompanied by greater troversial. Special assistance to help the poor accountability through reformed local govern- engaged in the production diversify or find nona- ment. This is essential to reducing the risk that gricultural livelihoods is then needed. local elites will appropriate resources placed in Specific targeted measures can counter the their stewardship to the detriment of the vulner- adverse effects vulnerable groups suffer in the able members of the community. short- to medium-run. Protecting or increasing public expenditures that benefit the health and Specific Interventions nutrition of the poor is especially important. Where food prices are likely to rise, safety net While accelerating and broadening economic actions such as social funds and other employ- growth is the best way to reduce hunger, specific ment-oriented programs must be accelerated. A targeted interventions can greatly accelerate cautious goal is ensuring that the most vulnera- hunger reduction and the associated suffering ble do not suffer a further decline in their welfare. and mortality. Cost-effective targeted interven- A more ambitious objective is creating conditions tions can reduce hunger directly or indirectly by that will start them on the road to more secure increasing incomes of the poor. The specific mix livelihoods. Careful analysis is required to iden- of actions depends on the circumstances of each tify the vulnerable groups most likely to lose at country and on the special characteristics of the the beginning of adjustment, to strengthen exist- diverse subgroups of the poor. The characteristics ing safeguards, and to implement new protec- of the poor and hungry must therefore be ana- tions where existing programs are inadequate. lyzed in each country's poverty profile and CBOs and NGOs often can help identify the likely resulting poverty strategy. The interventions victims and help implement the safety net and described below come from best practices dis- targeted development measures. tilled from worldwide experience. They are not exhaustive, but have been successful in many Participation and Decentralization cases. The actions are ordered roughly by the immediacy of their effectiveness in reducing Development programs are far more likely to suc- hunger, and by the cost and complexity of their ceed when governments, communities, and indi- implementation. Priority setting cannot be done viduals are active partners in the conception, globally. It must be tailored to local or country design, and implementation of the programs. conditions. It is often unnecessary for govern- Participation makes development more client- ments to execute these programs. They can be oriented, building local ownership of programs. implemented by CBOs, NGOs, or the private sec- This promise has too often been neglected. tor. But they all require partial, if not full, govern- Beneficiary assessment is a highly useful tool ment financing. for collecting information on the sociocultural, Many of these interventions also will improve demographic, and gender dimensions of a pro- efficiency and productivity at all levels-individ- gram. Assessing the beneficiaries or a program ual, regional, and national. These productivity- strengthens program design and monitors pro- enhancing actions are not usually done by the gress. It is only one step from a top-down market; on the contrary, they usually reflect mar- approach to one that fully involves the various ket or government policy failure. For example, stakeholders. Consultation is necessary but the poor may not be able to buy land in the mar- insufficient by itself. ket because of a lack of savings, nonexistent mort- Decentralization of decisionmaking and gage banking, or distorted land policies and resources is another key ingredient in creating prices. Mass immunization and deworming are local ownership of poverty and hunger reduction public goods whose benefits cannot be fully cap- programs. The intended beneficiaries must play tured by the private sector. a part in the decisionmaking process through When the primary objective is wealth redistri- local assemblies and CBOs and NGOs. Only in bution or poverty and hunger reduction, the Executive Summary 5 choice of intervention must attempt to transfer with nutrition and health education to change resources to the poor in the most cost-effective behavior. Programs in Chile, Honduras, and way. For example, general food price subsidies Mexico show that health clinics are the most typically transfer food resources to the poor at a effective vehicle for targeting food to malnour- high cost. Targeted nutritional monitoring and ished children and their families. Health clinics feeding programs may achieve the same goal are well-placed to take a holistic approach to much more cheaply. Relative cost-effectiveness is hunger, providing medical treatment, immuniza- thus a key factor in choosing how best to inter- tion, health and nutrition education, and food vene on behalf of the poor. supplementation. Clinics can monitor hungry children and their parents and provide food to Urgent Low-Cost Actions other family members, who are also likely to be hungry. The worldwide oral rehydration therapy cam- On-site feeding is most appropriate when the paign led by United Nations Children's Fund specific vulnerable child or mother is the target. (UNICEF) has saved millions of children who Rather than physically distributing take-home would otherwise have died from dehydration food rations, food entitlements should be pro- caused by diarrhea. It was agreed at the 1990 vided through on-site feeding, food stamps, or World Summit for Children that priority should free ration cards. Whenever possible, food enti- be given to a number of other simple, low-cost tlements should cover the food deficits of the actions. This conclusion has since been reaf- entire family, not just the affected child or mother. firmed repeatedly. These programs can be imple- The use of coupons and ration cards avoids the mented through community-led delivery and costs and complexities of physically handling would usually include nutrition education aimed food. at behavioral change. The measures include treatment of intestinal Food Aid parasites, immunizing against measles and other childhood diseases, and providing vitamin A, Food aid is an important resource which, prop- iodine, and iron supplements. These programs erly channeled, distributed, or sold, will not are affordable and easily implemented in even undermine domestic food production. Making the poorest countries with assistance from inter- sure that food from an international organization national organizations such as UNICEF and the or a donor country effectively benefits those who World Health Organization (WHO). A large body need it is a major concern. Transporting donated of evidence indicates that diseases, parasites, and food long distances in a country is wasteful and other health problems can suppress appetites and costly Monetizing food aid and using the money rob the hungry of nutrients. The world's hungry to assist the hungry wherever they are is more should no longer have to wait for these simple, efficient. The proceeds can finance employment low-cost measures to be taken. generation or targeted food supplementation. Food Supplementation Famines Many countries already have schemes in place to Famines, as opposed to chronic undernutrition, subsidize food consumption and nutrition. These are specific, time-bound events. They often occur programs have benefited many people. However, without a decline in food availability at the some methods of providing the hungry with food national level. Moreover, sharp declines in food are more effective and less costly than others. availability do not always result in famine. The Shifting budget resources to more effective food poor suffer most in famines. Effective action subsidy and nutrition programs can often drasti- requires understanding the way markets and cally reduce hunger at constant or even reduced governments work in a crisis, and the capacity of cost. The most effective targeting mechanism is people to protect themselves. Alleviating famines nutrition monitoring in health posts, schools, and requires early warning mechanisms, food supply other social programs. This should be combined management, food supplementation, employ- 6 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community ment generation, and agricultural assistance. Such growth requires a crime-free environment, Properly implemented, these strategies can be improved land rights for men and women, and highly effective. reduced regulation. The main actors in famine relief are the Providing the poor with access to credit has UNDHA, the WHO, the Food and Agricultural been difficult. The Bank has had little success in Organization (FAO), and UNICEF, along with developing and funding credit for the very poor. many CBOs and NGOs. The International Done successfully, credit helps the poor partici- Monetary Fund (IMF) has a special facility to pate in economic growth. The Grameen Bank of finance food imports. The World Bank's role is Bangladesh is an example of a highly successful assisting countries in overcoming poverty and program that has helped increase the incomes, the institutional weaknesses that are most often and consequently the calorie intake, of the rural the root cause of famine. For this, the Bank should poor. It lends primarily to poor women, with coordinate closely with United Nations (UN) dis- excellent credit recovery. This approach is now aster relief, show flexibility in lending for the for- being extended to other developing nations. The eign exchange requirements of food emergencies, Bank has allocated $2 million to the Grameen and help countries recover from such crises and Trust to assist similar programs in other countries. return to normal development. In addition, the Bank is exploring mechanisms to Famine from desertification and drought promote other promising nongovernment pro- develops slowly. It can be greatly mitigated by grams to assist the very poor. advance warning and prompt countermeasures. The Foundation for International Community Governments can stabilize prices and food sup- Assistance (FINCA) has helped establish a net- plies by building up food reserves or the financial work of more than 1,800 village banks in Latin reserves for the purchase of food imports. America and Africa which provide small loans to Drought-prone countries can strengthen their self-employed borrowers, many of whom are capacity to mitigate drought. The Bank is inte- women. In Bolivia, Thailand, and West Africa, the grating the impact of drought into country assis- Freedom from Hunger organizations are pio- tance strategies and project design. neering credit with education, again using village Another major source of vulnerability is banking. In Pakistan the Aga Khan Rural Support volatile international food prices, such as the Program also combines credit with training to world food crises of the early 1970s. The Bank, in reach women in the poorest villages. In Indonesia conjunction with the IMF Food Facility, the WFP, the International Fund for Agricultural Develop- and others, can help countries design cost-effec- ment (IFAD) funds a rural credit program to tive strategies and financing arrangements for assist poor farmers and landless workers in 2,000 such emergencies. villages, based on self-management principles. Group-based lending for microenterprises is Income Generation Programs most appropriate where there is high population density, landlessness, and a thriving urban or Specific targeted interventions to improve the rural nonfarm economy. In some countries it is incomes of the poor and hungry can powerfully being tested in urban settings. Such credit, how- complement broad-based growth policies. Pri- ever, is not necessarily a powerful or cost-effec- mary education, especially for women, not only tive tool to assist the very poor in sparsely reduces fertility and improves child welfare and populated and poor agroclimatic zones, such as nutrition, it also greatly enhances the income- the semi-arid tropics. earning capacity of the poor. Where soil, water, and other conditions are Informal sector employment reduces poverty poor and growth is slow, special programs can directly and provides safety nets in times of eco- improve the management and income- nomic crisis. Cities should encourage the growth generating potential of natural resources in a of the informal sector. Hawkers, microenter- cost-effective and sustainable manner. Such pro- prises, artisans, and the like not only generate sig- grams should usually be community based. nificant income, they also are efficient channels of Where this does not create sufficient income, gov- marketing, distribution, and waste recycling. ernments must establish policies that help the Executive Summary 7 hungry migrate to cities. Where migration is small farmers continue to be effective food pro- insufficient or unfeasible, men's and women's ducers and competitive exporters. incomes need to be supported by special employ- ment schemes. Successful examples are the World Bank Actions Maharashtra Employment Guarantee program in India and employment programs in Botswana. Poverty reduction is the cornerstone of the World The Bank first assisted land reform in the Bank's mission. This goal encompasses the more 1960s, in Kenya. The 1975 Land Reform policy specific objective of reducing hunger in its three paper emphasized the Bank's readiness to sup- forms. In addition to raising the incomes of the port land reform and other land policy options. poor so they can purchase the food they need for Until recently, however, few of the 1975 propos- an adequate diet, a sound poverty and hunger als had been implemented. The end of the Cold strategy includes other more specific measures to War and the disappearance of farming collectives overcome poor nutrition. The Bank's country and apartheid provide new opportunities to poverty assessments provide an integrated transfer land to the poor. framework for country assistance strategies to The Bank encourages land reform where land raise the incomes of the poor and improve nutri- is concentrated in the hands of a few. Giving the tion and food security. poor secure rights to land allows them to increase The Bank's basic strategy for poverty reduc- their income and access credit, and is an impor- tion was set out in World Development Report tant security factor in emergencies. Moreover, 1990. The Bank's 1992 Poverty Reduction Hand- smaller farmers in developing countries are more book provides detailed guidance for poverty efficient than larger ones and use more labor. reduction operations to the Bank's country Land rights of women are often especially weak teams. In addition, recent work in Africa has and must be strengthened. In South Africa mar- shown how assistance programs can include ket approaches to land reform involving CBOs drought relief and famine avoidance to antici- are being encouraged rather than expropriation pate and substantially mitigate slow-onset by land reform agencies. disasters. International experience with strengthening the economic opportunities of the Agricultural Research poor is now being widely disseminated and inte- grated into various sectoral projects. And collab- Efforts are needed to improve the technology oration with development NGOs and CBOs in available to small farmers. The Malthusian night- poverty projects is expanding. mare of widespread starvation has so far been Recognizing the importance of a participa- postponed by rapid advances in agricultural tory approach, the Bank is preparing a participa- technology such as the green revolution of the tion sourcebook to provide guidance on 1960s and 1970s. But hundreds of millions of peo- participatory approaches. This sourcebook will ple continue to go hungry, and the global popu- complement social assessment guidelines, and lation continues to expand at an alarming rate. an intensive effort will be made to train staff in International research must be bolstered to these techniques. ensure the Malthus prophesies do not come to The Conference on Overcoming Global pass. Intensified support for the CGIAR is Hunger initiated a valuable dialogue with the required for commodities important in the con- NGO community. There is a shared desire to see sumption and sustainable production of the poor. hunger visibly reduced. The Bank is ready to par- More research at the national level in developing ticipate in pilot tripartite in-country consulta- countries can help adapt the findings of interna- tions to be initiated locally as a forum for tional agricultural research to local conditions- exchanging views among government represen- especially farming systems of the poor. The Bank tatives, NGOs, and the Bank on country-specific will continue emphasizing agricultural extension hunger reduction strategies. for small farmers-both men and women- Recent research, World Development Report through the training and visit (T&V) system and 1993, reviews of structural adjustment efforts and other approaches. Only through these actions can public expenditure patterns, and the discussions 8 The World Bank's Strategy for Reducing Poverty and Hunger: A Report to the Development Community at the conference have identified those actions require dissemination inside and outside the that are particularly important in reducing hun- Bank. The specific actions are listed in the table ger, and which should be more systematically below. Bank actions in support of the program are included in Bank-supported programs. They will discussed in the last section of this report. ~~~~____ M ___ .ov,:xo,EzclsgBs- f . BsStE:::.t'A ..4,,c.:o,:4,£,: a.,:8.:*:a, tdl5.Ev Actions to Reduce Hunger Other Agencies Active Actions to Reduce Hunger: World Bank Support to Member Countries in This Area PART 1: Growth, Poverty, and Hunger 1. ASSISTANCE STRATEGY TO REDUCE POVERTY NGOs, IMF, regional develop- The Bank will promote broad-based, employment-intensive growth and improved access of the poor ment banks to education, health care, and other social services (as discussed in World Development Report 1990). Specifically, the Bank will continue to support measures that: * Encourage macroeconomic stability and openness of the trade regime. * Reduce direct and indirect taxation of farm sectors to encourage domestic food production and rural income and employment growth. * Eliminate antiemployment taxes and credit subsidies. * Reduce direct and indirect labor taxation and increase cost-effective, labor-intensive infrastructure construction. * Accelerate redirection of government spendingto safe drinking waterand sanitation, basic education, basic health care, and other social services for the poor, especially women. 2. PROTECTING THE POOR AND HUNGRY IN STRUCTURAL AND SECTORAL ADIUSTMENT PROGRAMS IMF, regional development * Seize opportunities in program and policy reforms to increase the income-earning potential and banks, NGOs livelihood security of the poor. * Encourage compensatory expenditures for vulnerable groups through safety net actions including social funds and employment-oriented programs, especially where basic food prces are likely to rise. *Assist in better targeting food subsidies to the poor and vulnerable groups including through such programs as items 4, 5, and 6 in this list. * Ensure that funding levels for synergistic primary health, nutrition, and targeted food assistance programs remain sufficient to maintain or expand existing levels of coverage. 3. PARTICIPATION AND DECENTRALIZATION IN INVESTMENT LENDING NGOs, CBOs * Integrate the lessons from the learning groups on participation in the Bank's operations. * Ensure greater beneficiary participation by decentralizing functions and fiscal resources to local governments and community-based organizations (CBOs)3 and by strengthening accountability of local leaders to ultimate beneficiaries. PART fI: Specific Actions and Programs 4. SIMPLE, Low-COST ACTIONS UNICEF, WHO, bilateral In follow-up to conclusions of World Development Report 1993 and the goals of the 1990 World Summit donors, NGOs for Children and the 1992 International Conference on Nutrition,4 the Bank stands ready to collaborate with UNICEF in assisting borrowing countries to prepare and finance actions for implementing synergistic, low-cost, hunger-reducing health and nutrition interventions to achieve the following actions and goals: 9 10 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community Other Agencies Active Actions to Reduce Hunger World Bank Support to Member Countries in This Area * Reduce vitamin A, iodine, and iron deficiencies through fortification, supplementation, and dietary modification. * Expand childhood immunization coverage from 80 percent to 90 percent. * Implement a low-cost program for control of parasitic infections that cause anemia and malnutrition. These simple low-cost actions would be implemented through community-level delivery with nutrition education for behavioral change. 5. TARGETED FOOD SUPPLEMENTATION UNICEF, CBOs, NGOs Encourage countries to shift food subsidies to targeted food assistance programs that provide food entitlements to children and their families where children or mothers have been identified as being at risk of becoming malnourished or having low birthweight babies in health posts, schools,or other social programs. Food entitlement may involve on-site feeding, food stamps, or free ration cards rather than take-home food distribution. 6. MONETIZED FOOD AD WFFP bilateral food aid, CBOs, Assist borrowing countries and implementing agencies in monetizing food aid and making counterpart NGOs funds available to finance the above food supplementation programs and the synergistic interventions provided in these programs for nutrition educabon, micronutrient supplementation, and health. Monetization should be done without depressing domestic producer incentives. 7. PROMPT RESPONSE TO DROUGHT AND FAMINES UNDHA, FAO, WFP, UNDP * Integrate the impact of drought into the country assistance strategies and project designs of drought- IME CBOs, NGOs prone countnes or regions. * Strengthen the counthes' capacity to prepare for and mitigate drought. This will reduce the need for emergency relief and smooth the transibon from drought to recovery. * Build on recent experience with Southern African drought relief and coordinate the Bank's lending more closely with the UN's disaster relief operations and subsequent recovery requirements. * Assist countries in designing cost-effective strategies and financing arrangements to stabilize the prices of the most essential staple foods in case of large, temporary, international price shocks, drought, or other calamities. * Improve lending instruments to provide greater flexibility for foreign exchange requirements associated with drought and famines. 8. INCOME GENERATION PROGRAMS IFAD, CBOs, NGOs Within the context of country-specific poverty strategies, evaluate and provide support for targeted income generation programs to: * Improve access to and quality of preprimary and primary education for disadvantaged groups, especially girls. D Encourage countries to provide crime-free space and improved land rights for men and women, and to remove excessive regulation of informal sector activities. *Within the context of overall financial sector reform, support policies and programs that make financial services available to the poor, particularly women, and address the credit and savings needs of the self-employed poor. *Assist communities in semi-arid areas, mountain zones, and other difficult agroclimate zones in improving their management of key natural resource (pasture, forests, watersheds) as, for example, in the Burkina Faso environmental management project; where income generation potential is poor, help finance employment generation programs along the lines of the largely self-targeting Maharashtra Employment Guarantee Scheme and Botswana employment programs. * Encourage countries with highly unequal land distribution to design market-assisted land reform programs as, for example, is being initiated in South Africa. 9. AGRICULTURAL RESEARCH AND EXTENSION IFAD, FAO, CBOs, NGOs, To ensure long-term food supplies and improve rural incomes: ILO, WFP * Emphasize small farmer extension, as in T&V and other similar programs. * Strengthen national agricultural research systems, with special emphasis on the commodities and farming systems characteristic of the poor. * Mobilize a continued high level of support for the CGIAR and encourage research on commodities that are particularly important in the consumption and production of the poor, and on policies and programs aimed at raising the purchasing power of the poor 1 0. FOLLOW-UP TO THE CONFERENCE ON OVERCOMING GLOBAL HUNGER FAD, FAO, UNDP CGIAR, Participate in locally organized tripartite pilot in-country consultabons in countries where representatives CBOs, NGOs of the government, CBOs, NGOs, and the Bank can explore improvement in specific policies or actions to reduce hunger. PART I Growth, Poverty, and Hunger he first part of this report focuses on the insufficient to prevent starvation or alleviate mal- causes of hunger and on the World Bank's nutrition. Only if poor households have the strategy for reducing poverty and hunger income to buy food, or the land and water by promoting broad-based, employment-inten- resources to produce it, can they be sure they will sive growth and by providing basic social services not suffer from hunger. This is why recent to the poor. It also emphasizes the importance of debates have focused so much on household food investing in people and the poor's role in their security. own development. Part H details specific actions As Shlomo Reutlinger stated in 1977, when and programs that have proven successful. great attention was focused on food supply: Causes of Hunger and Elements Malnutrition can be wholly or partly of Food Security caused by poor quality of nutrition or poor health status of the afflicted popula- Hunger is the most deplorable manifestation of tion. However, in recent years nutrition poverty. It can take three forms, all of which are scientists have reached some measure of considered in this report: agreement that many millions of people, * Starvation, a life-threatening condition caused adults and children in developing coun- by insufficient food often associated with tries suffer from malnutrition ... because famines they do not have the means to obtain * Chronic undernutrition, due to a persistent lack enough of their accustomed diet of calories and protein (Reutlinger 1977). * Specific nutrient deficiencies, from inadequate protein and micronutrient intake, often com- Amartya Kuma Sen, in Poverty and Famines: bined with infections and inadequate nutri- An Essay on Entitlement and Deprivation, found tional knowledge. that during the Ethiopian famine of 1972-74 there All three forms are often considered to be pri- was "no abnormal reduction in food output, and marily a problem of food supply or availability. [national] consumption of food per head at the Indeed, it is true that food availability in a given height of the famine in 1973 was fairly normal for area is necessary to avoid famine. Availability can Ethiopia as a whole" (Sen 1981). But 50,000 to be ensured either by sustainable production, stor- 200,000 people died. Most were nomadic and age, or imports. But adequate food supply is seminomadic herders who saw their incomes 1 1 12 The World Bank's Strategy for Reducing Poverty and Hunger: A Report to the Development Community plummet when livestock prices fell and they and prices and assisting vulnerable groups could not afford to buy food. with emergency employment programs, These findings startled some in the develop- income transfers, or food. ment community and have yet to penetrate the * Within countries thefood-insecure poor comprise consciousness of many outside it. Even within the different subgroups. These subgroups are dif- development community, some groups continue ferentiated by location, occupational patterns, to remain primarily concerned with providing in- asset ownership, race, ethnicity, age, and gen- kind food aid, improving production, and other- der. Cost-effective programs to improve food wise trying to assure that food is available to poor security must be tailored to the needs and cir- people in developing countries. cumstances of each group (World Bank 1986). Development organizations used findings Food security does not involve whether food from Reutlinger, Sen, and others to create a new is imported or not, merely whether it can be framework for reducing hunger. This framework, accessed. Of course, producing your own food articulated in a 1986 World Bank policy study, provides both domestic availability and pur- Poverty and Hunger, established these key princi- chasing power. In inaccessible, remote areas, and ples of food security: for subsistence farmers, food availability and * Food security is access by all people at all times to access are essentially the same. Elsewhere, food enough food for an active and healthy life. Food security can also be met by purchases and insecurity, by contrast, is lack of access to imports, especially where domestic agricultural enough food. Countries with widespread endowments are poor and can produce food only undernutrition have a national food security at a high cost. Older schools of thought empha- problem. sized domestic production and self-sufficiency • Food security requires meeting two conditions. based on the assumption that imports are unreli- One is ensuring that there are adequate food able and importing countries could easily be supplies available through domestic produc- blackmailed by withholding of food exports. tion or imports. The other is ensuring Such fears were clearly exaggerated. Even during households whose members suffer from mal- the worst food crises, as in the early 1970s, there nutrition can acquire food, either because they have always been alternative suppliers, even to produce it themselves or because they have unpopular governments. No studies indicate the income to acquire it. hunger resulted from import suspension. If a * Food insecurity is either chronic or transitory. country needs to import and has insufficient for- Chronic food insecurity involves a con- eign exchange reserves to do so, the International tinuously inadequate diet caused by the Monetary Fund (IMF) maintains a special lend- persistent inability to acquire food by any ing facility to help countries buy food, the IMF means-producing, buying, bartering, shar- Food Facility. The facility's infrequent use since ing, foraging, and so on. Transitory food inse- its creation in the 1970s is a reflection of the low curity is a temporary decline in a household's food prices that have prevailed for the past two access to enough food, arising from instability decades, not a lack of potential usefulness. of food prices, food production, or household The Bank's policy framework on poverty and incomes. hunger, reiterated in World Development Report * Policiesfor reducing chronicfood insecurity differ 1990, emphasizes broad-based, employment- from thosefor reducing transitoryfood insecurity. intensive growth and human resource develop- Policy options for reducing chronic food inse- ment as the best way to reduce poverty and curity include increasing the food supply hunger (World Bank 1990). The validity of this (through domestic production, imports, or strategy is continually reaffirmed. Bank country improving market integration), targeting studies consistently demonstrate that increasing development assistance or income transfers to income reduces hunger more than any other the poor, and helping the poor obtain knowl- action. Direct special measures, such as bolster- edge about nutrition and health practices. ing health care and targeted feeding for children Policy options for reducing transitory food and nursing mothers, are also usually necessary insecurity may include stabilizing supplies to reduce malnutrition. Growth, Poverty, and Hunger 13 Nutritional status is strongly affected by As a result child mortality remains unusually behavior, such as breast feeding, sanitation high and primary enrollment unusually low. The habits, or the use of local vegetables. Conse- poor should be better-equipped to take advan- quently, health and nutrition education must be tage of economic opportunities. Some other an integral part of the strategy to reduce malnu- countries, by contrast, have long stressed social trition and specific micronutrient deficiencies. services, but growth has been too low. In Cuba The importance of the direct interventions and Sri Lanka, for example, primary enrollment has been further strengthened over the past few rates and under-5 mortality rates are exception- years by: ally good, but the potential for raising the * New research showing that the consequences incomes of the poor has been wasted for lack of of malnutrition are far more negative than economic opportunity. previously assumed. Progress has been greatest in countries that * Evidence that micronutrient deficiencies, implemented both parts of the strategy. By pro- infectious diseases, intestinal parasites, and moting the productive use of labor, these coun- environmental factors often contribute to mal- tries have furnished opportunities for the poor nutrition as much as calorie deficiencies. and, by investing in health, nutrition, and educa- * Recognition that economic growth alone is tion, they have enabled the poor to take full insufficient to eliminate hunger. Economic advantage of the new possibilities. growth would require more than a generation Policies to achieve broad-based growth and for most countries to achieve significant human resource development reduce or eliminate progress for the lowest income groups. Within antiemployment biases in trade policy, market reg- the overall growth-based poverty strategy, ulation, the tax regime, labor laws, and financial more direct steps are needed to accelerate sector policies. They focus government expendi- hunger reduction. tures on basic education and health care and other * Increasing incomes also are insufficient, even basic social services, and on growth-enhancing among low-income groups. Health and nutri- public investments and programs. They finance tion education improves behavior and nutri- cost-effective infrastructure using labor-intensive tional status. techniques. They provide resources, credit, and i Greater awareness that governments and employment opportunities to women-who play donors, including the Bank, need to consider an especially important role in any hunger reduc- the consequences of adjustment operations on tion strategy. And they ensure sustainable food consumption and nutrition and encour- resource use through cost-effective environmental age compensation for the needy. policies and programs. Broad-based growth best helps the poor World Bank's Strategy to Reduce Poverty because it relies on productively using their most and Hunger abundant resource-labor. In virtually every developing country, poor people depend on The Bank's poverty reduction strategy has two income from work on their own land, from important elements. The first is promoting broad- wages, or from other self-employment. Countries based, employment-intensive growth. This calls that have succeeded in reducing poverty over the for policies that harness market incentives and long term have encouraged broad-based rural build infrastructure, technology, social, and polit- development and urban employment, thereby ical institutions. The second element is providing increasing the returns to small farm production basic social services to the poor. Access of the and wage labor. With increased incomes, the poor poor to clean drinking water, primary education, can afford more food, reducing hunger in the primary health care, family planning, food, and process. Improved nutrition and lower morbidity nutrition are especially important. and mortality follow. The two elements are mutually reinforcing. In The importance of growth in reducing some countries, such as Brazil and Pakistan, poverty and hunger cannot be overstated. In the growth has raised the incomes of the poor, but 1980s the child mortality rate fell more than twice social services have received too little attention. as much in countries where average incomes rose 14 The World Bank's Strategyfor Reducing Poverty and Hun1ger: A Report to the Development Community by more than one percent a year than in countries zations (NGOs) think this is a mistake, as will be where average incomes fell. discussed below. Human resource development improves the The Bank also advocates redirecting govern- education, health, and earnings capacity of the ment support toward agricultural smallholders poor. Moreover, better education of parents, espe- and eliminating credit and other subsidies to cially women, reduces fertility and directly large, commercial farms. Government finance for improves child nutrition and health. Improved credit programs should focus on the small-scale nutrition, especially among pregnant and lactat- sector and the poor. Redirecting support will ing women and the very young, not only enhance the competitiveness of small farmers improves survival chances and quality of life, but and increase the demand for labor, indirectly also increases cognitive test scores, school perfor- helping agricultural workers. Reducing urban mance and, ultimately, productivity and income. bias in public expenditures has been less success- Many of the poor live in rural areas, where ful than reducing the bias in exchange rate and hunger is often most pronounced. A major reason trade policies, even in some of the strong adjust- for their poverty has been persistent urban bias in ing countries such as Mexico. policies and development expenditures. Agricul- The Bank has always favored land reform to tural taxation by developing country govern- promote more efficient small-scale farming. This ments during 1960-80 was often enormous (Shiff policy was spelled out in the 1975 Land Reform and Valdes 1992). The greatest negative impact policy paper, which recognized the political diffi- on agriculture was from the indirect effects of culties associated with land reform (World Bank overvalued exchange rates and industrial protec- 1975). The Bank was rarely able to translate its tion policies, not from direct agricultural policies favorable view of land reform into support of a such as export taxes. The direct policies often pro- country-led reform program. The demise of the vided some protection, especially to food crops Cold War, democratization, and developments such as wheat in Guatemala or Nigeria, and other such as the end of apartheid in South Africa are crops with limited comparative advantage. Still, opening new opportunities. these protections and direct credit and infra- A recent report by the UN's (United Nations) structure support could not compensate for what Administrative Committee on Coordination's was taken out of rural areas by export taxes, Subcommittee on Nutrition clearly demonstrates exchange rate overvaluation, and industrial sec- that countries experiencing broad-based growth tor protection. enjoy sharply declining rates of underweight Direct support through credit and marketing children (figure 1). Indonesia and Malaysia, both programs only reaches the poor in the few coun- successful in reducing the number of under- tries that also tax agriculture less. In most places weight children, needed to adjust because of the these programs instead support the better-off fall in oil prices and other commodities, which commercial farmers. Social sector expenditures- were major sources of foreign exchange. They primary health care, primary education-are also adjusted their fiscal policies, depreciated their usually biased against rural areas. The Bank's currencies, liberalized their trade regimes, and poverty reduction strategy advocates eliminating deregulated their industries. These policies trade rate and exchange policies that result in raised the relative price of agricultural goods, urban bias. helping farmers who make up the bulk of the Structural adjustment during the 1980s and poor in both countries. The incidence of poverty early 1990s has reformed the exchange rate in Indonesia dropped from 28 percent in 1984 to regime and trade policies in many countries, about 16 percent in 1987. especially in Latin America. But taxation of agri- Strong support was expressed at the Con- culture is still severe in many other countries, ference on Overcoming Global Hunger for pro- especially in Africa. This part of structural adjust- moting employment-intensive growth as the ment is often highly controversial. Opening trade most viable approach to reducing poverty in the for agricultural foods favors agricultural exports medium and long run. Within this framework, where many developing countries have compar- the Bank will continue to support measures ative advantage. Many nongovernmental organi- aimed at: Growth, Poverty, and Hunger 15 * Macroeconomic stability, liberalized markets, also required to define appropriate programs and open trade for the different subgroups. * Reducing direct and indirect taxation of farm * Incentives and regulations. The review of coun- sectors to encourage domestic food produc- try policies is intended to identify biases that tion, rural income, and employment growth limit the efficient use of assets owned by the * Eliminating taxes and subsidies that discour- poor or constrain the poor from acquiring age employment human capital, land, or credit. Policies exam- * Reducing direct and indirect labor taxation ined include those affecting the patterns of and increasing cost-effective, labor-intensive growth and labor demand, the incentive infrastructure construction framework governing relative prices, the reg- * Redirecting government spending to safe ulatory framework, and short-term macro- drinking water and sanitation, basic educa- economic management. Policies affecting the tion, basic health care, nutrition, and other poor's accumulation of assets-human capi- social services for the poor, especially women tal, land-and interactions with the environ- and children. ment also should be considered. A poverty assessment is the Bank's principal * Public expenditures and institutions. A public instrument for providing a factual and analyti- expenditure review determines whether the cal basis for a country poverty strategy. composition and structure of expenditures- Although the focus, topic selection, and phasing and service delivery-support macroeco- of work for the poverty assessment depend on nomic and sectoral policies efficiently and country circumstances, it normally covers the cost-effectively. From the poverty perspective, following: the objective is to identify possible realloca- * Poverty profile. The profile provides data about tions of resources-within and across sec- the extent and nature of poverty and identifies tors-that will expand the access of the poor relevant subgroups of the poor by their dis- to physical and social infrastructure and tar- tinguishing characteristics. These characteris- geted programs. tics include age, gender, location, occupation, * Safety net. An analysis of the extent and nature education, asset ownership, ethnic origin or of poverty and the extent of the poor's vul- race, family structure and dependency ratios, nerability to natural or economic changes and so on. The profile is used to identify should identify the type of safety net poverty trends and guide policy analysis. It is needed-and its scope. This analysis should be combined with an assessment of the extent, Figure 1. Changes in GNP and prevalence of underweight effectiveness, and affordability of existing children from late 1970s to early-late 1980s safety nets, and of the existing institutional Percentage of underweight preschool children capacity to implement any recommended (below -2 standard deviations weight for age) changes. 80 Earlier Later Country poverty strategy. Based on the above \ia survey survey analyses, the strategy should lay out priority 60 . government actions for reducing poverty. \ Indonesia Objectives, institutional arrangements, and a 40 Pakis\t Viet Nam timetable should be specified. The strategy Th ailand should take the country's implementation Malaysia capacity and any political constraints into 20 Lesotho Bolivia account. The strategy should specify options and recommend choices when tradeoffs 5 between objectives are identified-as between 0 500 1,000 1,500 2,000 poverty reduction and economic growth or GNP per capoa (US$) protection of the environment. Note Datagivenforthefollowingsurveyyears: Bolivia, 1981, 1989; India, Box 1 provides a poverty assessment check 1975-79, 1988-90; Indonesia, 1980, 1987; Lesotho, 1976, 1981: Malaysia, 1983, 1986; Pakistan, 1977, 1990; Thailand, 1982, 1990, and Viet Nam, 1986, 1989 list. An evaluation of the poverty assessment Source: Unaed Nabons, Second Report on the World Nutnoon Stuotion, vol. II. 1993 published in the Poverty Reduction Handbook 16 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community (World Bank 1992a) and in the poverty progress growth, requiring specific well-targeted and cost- report (World Bank 1993b) suggests that current effective interventions to sharply accelerate attention to hunger is insufficient. hunger reduction. Several are cheap, simple, Even if the strategies resulting from poverty direct interventions known to save and improve assessments are implemented successfully, there lives. These are discussed in part II of this report. will still be more than 550 million people living on less than $0.75 a day (1993 prices) in 2000, most Poverty and Hunger of them hungry (box 2). Moreover, the world's during Structural Adjustment poor, sick, old, and those who live in resource- poor regions will find it difficult to participate in A key issue raised at the conference was the growth. They will continue to experience severe impact of structural adjustment on hunger. Many deprivation. Many others will suffer temporary countries have been hit by external shocks and setbacks from seasonal variations in income, loss have postponed adjustment to lower external of the family breadwinner, famine, or adverse earnings. Others have veered off the road to macroeconomic shocks. As conference partici- broad-based growth by creating or sustaining pants discussed, a comprehensive approach to unsound economic structures and urban bias and hunger reduction goes beyond broad-based by neglecting human resource development. Box 1. Diagnosis checklist for poverty assessments Growth crops produced by the poor? Safety net * How do macroeconomic and sec- *How are other intermediate *How extensive are existing toral polices affect the demand indicators of poverty affected? safety net programs-including for skilled and unskilled labor? *How much access do the poor traditional and other nongov- * To what extent do labor market have to productive assets and ernrmental arrangements? regulations impede labor mobil- infrastructure? * Are they cost-effective and well- ity and absorption? targeted to the most vulnerable? *To what extent do regulations Human capital *If there is no safety net, are the about land or financial markets *How much access do the poor country's institutions devel- discriminate against the poor? have to basic social services? oped enough to implement *How do regulations affect the *Is the balance in public social safety net programs? growth of urban employment? expenditures appropriate? * How do agricultural taxes affect * How appropriate are the institu- Source: Poverty Reduction Handbook rural employment and incomes? tions, planning, and delivery (World Bank 1992a). * How does policy affect the capacity? How severe are short- unskilled wage or the prices of ages of skilled labor? Box 2. Importance of economic growth for reducing hunger and poverty Research by the Bank has found poverty by 2000. But if the devel- about 977 million would be living that if current economic growth oping world were to grow as fast in poverty by 2000, as indicated in trends continue, there will be and reduce poverty as quickly as the table below: about 1.3 billion people living in East Asia during 1985-90, only Number of poor in developing world under alternative growth patterns 2000, if the developing world had the rate of Millions of people 2000, if the 1985-90 poverty reduction found living on less than: 1985 1990 trend continues in East Asia in 1985-90 $1/day 1,051 1,133 1,317 977 $0.75/day 614 644 708 551 Source: Implementing the World Bank's Strategy to Reduce Poverty (World Bank 1993b). Growth, Poverty, and Hunger 17 Box 3. Adjustment program design checklist The following questions should be for unskilled labor greater? * What are the tradeoffs with fis- asked in the design of adjustment *How do prices of the products cal management? operations: consumed by the poor differ * How adequate is the monitoring * How does the poverty outcome along these paths? system for tracking changes in differ under alternative adjust- *How much domestic demand the welfare of the poor during ment scenarios? contraction is necessary to adjustment? *How do wages and employ- restore external balance? *What additional investments mentof unskilled workers differ *What is the likely distributional are warranted to facilitate the under the scenarios? impact of the contraction? poor's participation in the eco- *How mobile is labor between *What specific risks under the nomic growth to be generated the tradable and nontradable program bear on the poor? by adjustment? sectors? *Which groups are likely to be * How flexible are wages in nom- adversely affected? Source: Poverty Reduction Handbook inal terms? in real terms? * How can the poorest and most (World Bank 1992a). * Along which path is the demand vulnerable groups be protected? These policies are often fiercely defended by attention is focused both on how adjustment poli- entrenched interest groups. Symptoms of inap- cies affect the poor and on specific measures to propriate policies include unsustainable govern- cushion the short-term pain. The Bank now has a ment spending, balance of payment deficits, and checklist of questions to ask when appraising large social spending imbalances favoring the adjustment operations (box 3). middle and upper classes and urban areas. The starting point for adjustment is macro- Structural adjustment is the main instrument to economic disequilibrium. The usual symptoms restore broad-based growth in countries that are an unsustainable current account deficit, have experienced macroeconomic instability, internal financial problems, and slow growth. lack of growth, and increased poverty. Structural Adjustment has two objectives: reducing the adjustment programs seek to restructure the pro- demand for imports and domestic goods to sta- ductive capacities of a country. In practice, this bilize economic conditions, and restructuring the means using monetary, fiscal, trade, and other economy to reach a higher growth path. Since macroeconomic policy tools along the lines dis- restructuring calls for reducing urban bias and cussed in the previous section. more efficient use of labor, it is fully consistent Macroeconomic difficulties were triggered in with broad-based growth shared across income the 1980s as the debt crisis and international groups, which will reduce poverty over time. recession exposed structural weaknesses. When Targeting measures will counter the adverse structural adjustment programs were imple- effects specific vulnerable groups suffer in the mented, little attention was initially paid to their short to medium run. Protecting or increasing effects on the poor. Macroeconomic issues public expenditures on the health and nutrition seemed more pressing. The aim was to achieve a of the poor is especially important. Where food rapid transition to new growth paths. As the prices are likely to rise, safety net actions, includ- decade continued it became clear that macroeco- ing social funds and other employment-oriented nomic recovery and structural change were slow programs, need to be speeded up. A cautious goal to come in many countries. While structural is ensuring that the most vulnerable do not suffer adjustment often increased rural incomes, evi- a further decline in their welfare. A more ambi- dence of declines in other incomes and cutbacks tious objective is creating conditions that will in social services became evident. Many start them on the road to more secure livelihoods. observers called attention to the situation, but it Careful analysis is required to identify the vul- was the United Nations Children's Fund nerable groups most likely to lose at the begin- (UNICEF) that brought the issue onto center ning of the adjustment effort, to strengthen and stage. As a consequence this aspect is now rou- further develop existing safeguard mechanisms, tinely reviewed in all adjustment programs and to develop and initiate new protections financed by the Bank. As UNICEF advocated, where existing programs are inadequate. CBOs 18 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community and NGOs can often not only help identify the dition, it can be used to establish a baseline for likely victims but also can assist in the implemen- monitoring during adjustment. The profile can be tation of the safety net and targeted development used to identify the key indicators that are corre- measures. lated with the incomes of the poor and that can Country circumstances determine when indi- be measured readily during adjustment. A simi- vidual adjustment operations should focus more lar approach can be used to monitor key social specifically on poverty reduction-by addressing indicators and access to public services. Targeted distortions and regulations that especially disad- measures are needed to counter the adverse vantage the poor and by supporting a reorienta- effects that specific vulnerable groups suffer in tion of public expenditures toward infrastructure the short to medium run. and social services for the poor. Within the over- The impact of structural adjustment on the all spending envelope given by the macroeco- poor remains highly controversial. The Bank has nomic framework, special efforts should be made produced several reports on the impact of struc- to safeguard, and increase where appropriate, tural adjustment, the latest of which, Adjustment budgetary allocations for basic health, nutrition, in Africa: Reform, Results, and the Road Ahead, pro- and education. While these measures are subject vides a comprehensive evaluation (World Bank to administrative and fiscal feasibility, all adjust- 1994). This report reviews only those issues ment operations are expected to provide protec- directly related to food production and food con- tion for the most vulnerable. sumption and to targeted expenditures for pri- The short-run effects of adjustment can create mary education, primary health care, and difficulties for two reasons. First, the process of nutrition programs. economic restructuring is often sluggish and uneven. Because existing distortions and privi- Food Production and Consumption leges are defended by politically powerful groups, adjustment programs may not be imple- At the preparatory workshop for the Conference mented fully and rapidly. Where they are, firms on Overcoming Global Hunger, participants and labor markets take time to adjust, and in the questioned whether adjustment improves food meantime economies can suffer higher unem- production, food consumption, or household ployment or underemployment and labor food security. Evaluating the impact of adjust- incomes may decline. Second, demand-reducing ment on incomes, food production, and food con- measures may be unavoidable, and these are sumption of the poor is difficult because of likely to hurt the consumption of both the poor conceptual difficulties and inadequate data. One and the nonpoor. The need for cuts in public conceptual difficulty stems from differences in spending can lead to a particularly sharp short- the impact of food price increases on the urban run conflict with two essential parts of the strat- and rural poor. The urban poor lose when food egy advocated in this report-delivery of social prices rise-at least in the short run. So do those services and provision of transfers and safety nets. rural poor who do not produce their own food or Structural adjustment often has immediate have surpluses to sell. In Mexico, for example, positive impacts on rural incomes. Yet it is usu- about half of the rural poor are net sellers of ally associated with a recession and lay-offs maize, while the other half are net buyers of brought about by reductions in government maize. Any change in maize prices, up or down, spending and elimination of protection of ineffi- has an immediate adverse impact on about half cient industries. Some of the rural and urban poor of the rural poor in Mexico. will suffer unless actions are taken to mitigate or In the longer run, increases in food prices may offset the impact of the recession on the vulnera- not affect net buyers adversely if growth else- ble groups. The poverty profile and short-term where in the economy or the elimination of dis- indicators of poverty can help in the design of tortions against agriculture increases demand for reform programs and systems for monitoring their labor and increases rural wage rates. Such poverty during adjustment. Because the poverty wage increases are often delayed, leading to profile indicates who the poor are, where they sequencing and safety net issues. Ravaillon live, and their economic activities and social con- (1990), for example, estimates that it takes about Growth, Poverty, and Hunger 19 three years for an increase in the price of rice in landless workers engaged in wheat production Bangladesh to result in a corresponding rise in will lose. The region may have difficulty finding the wage rate sufficient to offset the impact of the another crop with comparative advantage. Poor rice price rise on the rural poor. Any evaluation bread consumers all over Guatemala, however, of the impact of structural adjustment on the poor will gain from the opening of the wheat trade. For must examine food production, labor demand, this reason it may be a good thing. This is one of food prices, and wage rates of different sub- the reasons why free trade in food is so contro- groups among the poor. versial. If the policy is implemented, special pro- Structural adjustment usually increases the grams may be needed to help highly specialized price of tradable foods and nonfoods. In rural losers find new agricultural enterprises or non- areas, export production becomes much more farm livelihoods. In Mexico, for example, the profitable relative to domestic production. The Bank supported just such a strategy to deal with concern is whether the rural poor can participate the anticipated decline of maize prices as a con- in the newly available production and income sequence of the North American Free Trade gains. A special concern is that women engaged Agreement. in nontradable food production will not benefit, Assessing the impact of a specific structural and that smallholders in general may not be able adjustment program on incomes and food pro- to respond to the new opportunities. Even worse, duction and consumption is a difficult and data- these two groups may be squeezed out by more intensive task. One approach is to ask whether powerful producers who become interested in the poor are better-off as a result of economic the production of profitable exportables. reforms than they would have been in their Policymakers need to focus on supporting the absence. This approach requires one to construct production efforts of the poor and preventing a counterfactual scenario of how the economy anticompetitive behavior in output markets and would have evolved in the absence of reforms. A distortions in input and land markets, which second approach is to ask whether economic reduce the smallholders' ability to participate reforms have improved the welfare of the poor. fully in the new opportunities. Box 4 elaborates This approach compares the welfare of the poor on these points. at two points in time and thus requires baseline Opening the trade regime has sometimes data at the time prior to the implementation of reduced the price of domestically produced reforms. A third approach asks how the welfare crops, either because they were heavily protected of the poor under the current set of reforms com- prior to adjustment or because it takes time for pares with the welfare of the poor under a more private marketing to emerge after the withdrawal ideal set of reforms. Like the first approach, it of a parastatal. Sometimes prices have fallen requires the construction of a counterfactual. because international prices fell just as a country Each approach requires income data from a tried to liberalize agricultural trade. In several national household survey, preferably disaggre- Latin American countries all three factors coin- gated at the individual level. The data should be cided. If these difficulties are transitory, it supplemented with information on food con- becomes appropriate to temporarily restrain sumption and other welfare measures. Baseline imports and help poor producers market and information on the same set of households or store their crops during the transition. Such tem- information to construct counterfactual scenarios porary assistance should be withdrawn when the is also needed. Household survey data permit- markets and producers have had time to adjust to ting a longitudinal assessment of the impact of the new realities. adjustment on poverty are absent in most coun- Where a country lacks comparative advan- tries. In the absence of better data, one can only tage, adjusting to reductions in prices of previ- compare benefits to the poor from the preadjust- ously protected commodities is difficult. Many ment policies with the likely impact of policy rural areas depend on the sale of a single crop for reforms on their incomes. the domestic market or for export. If wheat prices The impact of structural adjustment has in the wheat-producing highlands of Guatemala been studied most carefully in Africa. In Africa, decline, both the net sellers of wheat and the the economic crises of the late 1970s and early 20 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community 1980s caused large declines in per capita Recession in the early to mid-1980s, precipi- income-and certainly hurt the poor. Most tated by an adverse movement in the terms of African countries fared badly on all fronts in the trade and government strategies in response to early 1980s. Growth was stalled or negative. the crisis, worsened things for the poor. In coun- Policies were strongly biased in favor of urban tries with fixed exchange rates, the initial areas and against agriculture, where most of the response was to tighten foreign exchange poor earn their living. Spending and the provi- rationing and to defend the official exchange rate. sion of rationed goods were heavily skewed to From Ghana to Tanzania, this heightened dual- the better-off. ism in markets. Influential people derived greater Box 4. Does commercialization hurt the poor? Commercialization, like technol- and children increase. In The Coerced production is another ogy, can be the prime impetus for Gambia, Guatemala, Kenya, the threat to successful commercial- poverty-alleviating growth. How- Philippines, and Rwanda, women's ization. Governments and monop- ever, many NGOs are skeptical role in new technologies or com- olists may coerce production in an about export crop production, mercialized crops was greatly effort to shift losses from a poorly believing it exposes the poor to reduced, even where women were designed commercialization unnecessary risks. Recent case important contributors to farm pro- scheme, or to capture excessive studies by the International Food duction before the change. profits. In Rwanda an unsuccess- Policy Research Institute indicate In The Gambia, where rice was ful tea production program technological changes and com- traditionally a women's crop, resulted in smallholders being mercialization generally benefit women had trouble using new coerced to produce unprofitable the poor in regions producing rice irrigation technology because tea. The new processing capacities commercial crops for export mar- they couldn't hire the necessary were not utilized, so the parastatal kets, increasing incomes and job labor. But household income tea factory expanded by expropri- security. increased, increasing caloric ating nearby small farms. Commercialization is usually intake and reducing women's sea- Egyptian cotton and rice schemes introduced to create new markets sonal weight fluctuations. In the and Chinese cereal programs also where demand for traditional Guatemalan export vegetable have suffered from coerced pro- commodities has become inelas- cooperative, household food con- duction because of bad program tic. Commercialization by itself sumption improved despite male design. does not generally bring adverse control of new income. Para- The opposite concern is exclu- consequences. Bad consequences doxically, the expanded employ- sion from new production oppor- usually reflect bad policies-con- ment demand increased child tunities. Colonialism frequently straints on trade, production coer- labor to the point where school reserved major production oppor- cion, and ill-advised tenancy schedules had to be altered, while tunities for whites. Similar con- laws. income increases enabled new straints on smallholders have In Guatemala opening new investments in improved child re-emerged in the form of regula- markets for vegetable exports welfare. Rather than undermining tions passed for the indigenous boosted production of high-value, the benefits of commercialization elites. Until 1992 in Malawi, small labor-intensive crops. Small farm- and technological change, these farmers were only allowed to pro- ers' incomes grew, and agricul- examples reiterate the need to duce tobacco under contractual tural employment jumped 45 improve women's access to pro- arrangements with large estate percent. This success reflected ductive resources. owners, for whom tobacco produc- good farming conditions, roads, Another common concern with tion was reserved by regulation. cooperatives, and farmer know- commercialization is increased Commercialization and tech- how. incentives for landowners to evict nological change must be accom- Commercialization is not with- tenants and move to owner opera- panied by enlightened policy. out its problems. In many countries tion of farms. In the Philippines, an Otherwise increased income and it is difficult for women to take administrative ruling excluded ten- employment may be undermined advantage of new export markets ants from contracts for sugarcane by inequitable distribution. developed through commercializa- growing, leading to landlessness tion and technological advance- and reduced tenant welfare. Such Source: Binswanger and von Braun ment-even though household consequences can be avoided by 1991. incomes and the health of women good policy. Growth, Poverty, and Hunger 21 and greater rents from scarce foreign exchange. producers in Africa still rely on selling their pro- By the early 1980s in Tanzania, for example, the duce for a significant part of their income, but the availability of consumer goods in rural areas share of tradables they sell is low. Improvements plunged. With the collapse of urban labor mar- in the price of tradable crops (as a result of real kets in many countries in the early 1980s, the exchange rate depreciations and marketing urban middle class probably suffered the largest reforms) have a positive, direct impact on the welfare declines. The urban poor, a much smaller incomes of the rural poor in the short run, but the group, also suffered. But they were partially pro- effect is not large given the low share of export- tected by the greater resilience of the informal able tradable goods they produce. sector, where most urban poor obtained their Many governments have reduced taxation of livelihoods. Rural and urban poor women were agriculture (through real exchange rate depreci- more vulnerable, and were probably dispropor- ations and marketing reforms). But only a third tionately hurt by falling incomes, rising prices, of the adjusting African countries increased real and cutbacks in social services. producer prices for export crops because of In countries with flexible exchange rates there major declines in world prices. In the other was generally no rise in rationing of foreign adjusting countries, reforms have not increased exchange or other products. But the failure to real producer prices, but have stemmed the tackle fiscal problems and decline of competi- impact of sharp falls in world prices. So many tiveness led to worsening poverty, especially in rural areas have not seen export booms as a rural areas. Absent exchange rate adjustment, the result of adjustment. impact on poverty was even more severe, as Even where exportable prices rose in real household data from the C6te d'Ivoire Living terms, women probably benefited little, because Standards Measurement Survey clearly showed. they primarily produce food for home consump- GDP declined under an abandoned adjustment tion and lack access to resources to cultivate effort, and the number of people living in extreme export crops. For net sellers of food it is the pro- poverty increased by 57 percent. The decline in ducer prices that count, and data about producer income (as opposed to a redistribution of income prices are sketchy. Evidence suggests there have from the poor to the wealthy) was the biggest not been strong increases in food crop producer influence on poverty during the second half of prices, particularly parallel market prices, the 1980s. though official producer prices have increased in How did countries that implemented adjust- some countries. This is consistent with food pro- ment policies fare? Depreciations of the real duction data which show the rate of growth in exchange rate are strongly associated with food production has increased in African coun- renewed growth in Africa. Economic modeling tries that reduced taxation of agriculture. This is for C6te d'Ivoire (Lambert, Schneider, and Suwa likely to have improved the lot of the net food 1991) and Cameroon, Madagascar, Malawi, and sellers. Niger (Dorosh and Sahn 1993) indicate that real At the same time, real depreciations and agri- exchange rate depreciations reduced both cultural reforms probably had little impact on the poverty and income inequality. Half the adjust- consumption of net buyers of food. First, the rural ing countries effectively implemented basic poor in most African countries spend the most on adjustment reforms and saw GDP per capita food products such as millet and cassava that growth increase. In countries that have under- have little international market value. Domestic taken some reforms and increased GDP per production of nontradable foodstuffs has capita growth rates, the majority of the poor are increased in recent years, holding down real price probably better-off, and almost certainly no increases. The consumption of own-produced worse off. nontradables ranges from 32 percent in The Evidence on income sources and food con- Gambia to 88 percent in the west and south of sumption patterns of the rural poor (90 percent of Madagascar. This high share insulates the poor Africa's poor) also allow some inferences about from increases in imported food prices resulting how adjustment reforms have affected the poor from real exchange rate depreciations and mar- (Dorosh and Sahn 1993). Most poor agricultural keting reforms. 22 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community For the urban poor, it is real consumer prices countries in the 1980s to determine how social ser- that count. Liberalizations of the marketing of vices and social well-being fared during the eco- domestic and imported major staple foods have nomic stringencies of the decade.5 Real per capita not pushed up real open market prices even public social spending on health, education, and when there have been large real depreciations. social security fell during some part of the 1980s Limited supply kept parallel market prices high in every country in the study The share of health in the preadjustment period. Indeed, real con- and education expenditures in total government sumer prices have fallen in some countries, expenditures fell while social security rose. including Tanzania, which had the most severe Numerous efforts to increase the efficiency and rationing. Real consumer prices for maize, rice, equity of social services were unsuccessful, judg- and beans all fell sharply between 1985 and 1987, ing from available data. In Brazil the misallocation when food crop marketing was first liberalized. of social expenditures remains significant (box 5). Even in 1992, real prices for these staples Despite lower funding, and no apparent increases remained well below those of the early 1980s in equity or efficiency, social indicators generally (Mayfield 1992 and van den Brink 1993). Major improved in the 1980s. Possible explanations for reforms of the rice sector in Guinea and Mada- this apparent paradox include measurement gascar have not increased real prices relative to error, time lags, technological changes, the grow- the open market prices before liberalization. ing role of NGOs and the private sector, and cur- Careful attention to the impact of reforms on rent benefits of past investment in water, population subgroups at high nutritional risk is sanitation, and women's education. needed. There are clear instances where specific In Africa, there is no evidence of a trend to poor rural groups have been hurt, or have bene- shift education spending away from the second- fited only marginally. For example, in the ary and tertiary levels and toward the primary Mulanje and Phalombe districts of Malawi, the level. Countries varied greatly in the share of ADMARC depots were closed prematurely, education expenditures allocated to primary before the market or alternative programs could education: from a low of 33 percent in Uganda to take over its food security role. They had to be a high of 86 percent in Ghana. For most countries, reopened. Attention is needed to timing and the share was between 40 and 55 percent, with a sequencing adjustment efforts to avoid such median of 52 percent, roughly in line with the occurrences. Poor urban consumers in Mada- median in Asia in 1985 (Tan and Mingat 1992). gascar benefited somewhat from controlled rice The poorest children do not attend school. prices before the devaluation, but were hurt by The story is much the same for health. Many the removal of subsidies (though they benefited African countries have a strong bias toward sec- from other reforms). Some of the urban sector has ondary and tertiary care. There is no trend been hurt by cutbacks in public sector employ- toward primary health care or basic health ser- ment. But the effects may be limited, as many of vices. In addition, personnel expenditures tend to the retrenched were not poor to begin with and squeeze out expenditures for essential materials, have been able to find work elsewhere, according drugs, and operation and maintenance. Aver- to surveys in two countries. aged across countries, personnel costs absorb about 70 percent of the recurrent health budget. Social Sector Expenditures Box 6 provides a checklist against which to eval- uate public expenditures. Some countries, how- What has happened to social sector spending in ever, are shifting spending to lower-level health adjusting countries? And what have been the facilities (where the most cost-effective health consequences for health and nutritional status? interventions are carried out) and to nonsalary The data do not show major cuts in health and recurrent items. education expenditures, but they do show con- Throughout the 1980s, Zimbabwe directed tinuing social spending misallocations that work new investment to completing the basic health against the poor. infrastructure-doubling the number of rural A recent study on Latin America (Grosh 1990) health centers from about 500 to more than 1,000 traced public social sector expenditures for nine during 1980-90. Senegal has set annual targets for Growth, Poverty, and Hunger 23 increasing its spending for drugs, transport, and study found that neither was happening, and that maintenance. infectious diseases continue to exact a heavy toll A Bank study recently examined whether (Griffin 1992). Government programs need to be government health expenditures in Asia were reoriented to give children a healthy start, ensur- targeted at the poor and high mortality areas. The ing environments free from preventable deadly Box 5. Redirecting social expenditures in Brazil Level and composition more likely to improve health and local accountability, new approach- Brazil spends a higher share of potential quality of life. es to financing, and expansion of gross domestic product (GDP) on the federal role in quality control social services than other middle- Delivery of services and consumer protection. Target- income countries. Total social Resources are poorly managed in ing could be improved by shifting spending in 1986 was an estimated agencies and programs, with waste- emphasis within certain sectors 25 percent of GDP, comprising fed- fully high administrative and per- toward programs that help the eral government programs (9 per- sonnel costs. poor and away from programs that cent of GDP), state and local subsidize the middle class and the government social expenditures (9 Financing and sustainability rich. Mechanisms to improve tar- percent), and private household Financing of social programs con- geting include user charges, in- (nonreimbursed) spending on tributes to mistargeting and ineffi- creased consumer information, and health and education (7 percent). ciency. The large share of revenues deregulation of prices for private earmarked to particular federal social services. Management and Incidence agencies or activities has built efficiency of some federal pro- The poorest 19 percent of the popu- empires and impeded the rational grams could be improved through lation receive only about 6 percent allocation of resources. Heavy more rapid decentralization of of social benefits. A large share of reliance on payroll taxes-which responsibility to the states. Financ- social expenditure benefits higher- are regressive and payroll-based ing of social programs could be income groups. Regional inequali- social contributions have distorted improved by reducing reliance on ties are also severe. the economy. The use of ad hoc payroll taxes and increasing agreements (convenios) for revenue reliance on income, value-added, Economic efficiency transfers creates inefficiencies by and corporate profit taxes, and Inefficient spending is indicated causing stop-and-start spending greater local cost recovery. Options by strikingly low social indicators, patterns and arbitrary allocation to strengthen the federal govern- particularly in health and educa- of funds. Direct cost recovery of ment's role in quality control tion. A large share of social spend- social services is inadequate. include preparing a social budget, ing pays for private rather than reviewing social welfare indicators public goods. For example, an esti- Safety net annually, and establishing a pro- mated 78 percent of all public No safety net is provided for the gram to evaluate public and private health sector spending is devoted poor. social services. The government to high-cost curative hospital ser- has recently begun to address some vices; only 22 percent goes to basic Recommendations of these issues. preventive health care-despite Improved services for the poor can the fact that preventive care is be achieved by better targeting, Source: Poverty Reduction Handbook more cost-effective and is much decentralization with increased (World Bank 1992a). Box 6. Checklist for evaluating public expenditures In evaluating the overall public * On whom is it being spent? * How are expenditures being expenditure program and the indi- * How efficient and cost-effective financed? Are they sustainable? vidual sector programs, the fol- are these expenditures? * How effective is the safety net? lowing specific issues need to be *How well are services being addressed in establishing the delivered? Source: Poverty Reduction Handbook information base for the analysis: *Is there a role for the private (World Bank 1992a). *How much is being spent, and sector or nongovernmental on what? organizations? 24 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community diseases. The poorer countries are, the less they seize the opportunities provided in adjustment can spend on health. Yet the poorer their health programs to redirect policies and programs in outcomes, the more the emphasis should be on favor of the poor; better target social expen- public health. ditures and food subsidies to the poor and vul- Continuing misallocations of social spending nerable, partly through the programs discussed clearly work against the poor, but trends in aggre- in part II of this report; increase compensatory gate public spending do not tell the full story. First, expenditures for vulnerable groups through the private sector provides many social services. safety net actions including social funds and Second, inferences about the delivery of basic ser- employment-oriented programs; and provide vices from aggregate public spending are mis- funding levels for primary health, nutrition, and leading. For example, in C6te d'Ivoire the poor targeted food assistance programs sufficient to were forced to cut their spending on education and maintain or expand existing levels of coverage. health care in response to declining incomes even The Bank's lending to reduce poverty and though public spending for health was protected. hunger complements its structural adjustment Girls particularly suffered. Third, there could be efforts. Poverty reduction is the central theme of regional differences in access within countries that the Bank's lending programs. To this end, Bank- do not show up in national data. The remote poor supported investment operations seek to may have little access to public services. Finally, improve the targeting and quality of basic health there has been little work on the direct impact of and education services, provide a social safety changes in social spending on the health and nutri- net, improve access to safe water, agricultural tion of the poor. Disaggregated budget data would extension, and credit, increase food security, and enable a detailed analysis of benefits to the poor enable the poor to acquire land and other assets. from public spending. A relatively small amount of well-targeted Participation and Decentralization funds can significantly impact hunger. A review of food programs in Latin America concluded A recurrent theme during the conference was the that, with sensible targeting, the total nutritional need to involve the beneficiaries and local orga- needs of Latin America could be met several nizations with the identification, design, moni- times over with current expenditure levels toring, and evaluation of all hunger related (Musgrove 1991). Targeted programs may not programs. As the NGOs stressed at the con- stand out in the context of aggregate health and ference, projects and programs that aim to pro- education expenditures. The lack of progress on vide social or economic services to a large overall expenditure reallocation hides signifi- number of people will be more effective if they cantly better targeting of social programs at the have strong beneficiary participation. When poor. This is especially true in Latin America, as communities and individuals take full owner- Honduras' health and nutrition program shows. ship of projects, they are far more likely to suc- Broad review of Latin American food programs ceed (box 7). Irrigation, forestry, livestock, water, finds sensible targeting criteria included in most. health, and other projects undertaken by the These steps toward better targeting of relatively Bank with local participation have performed small programs are significant advances. much better than those without beneficiary Participants in the preparatory workshop for involvement. At the conference Bank officials the Conference on Overcoming Global Hunger and NGO representatives acknowledged that discussed the poor progress on better-targeted community participation is too often neglected, social spending. The poor and hungry need to be and that greater effort is needed to promote par- heard. Borrowing governments ignore these ticipatory development. To achieve true partici- groups when deciding how to target social pation, new attitudes and new skills are needed. expenditures and overall budgets. This continues Introducing this new approach will require care- despite consistent Bank efforts to persuade gov- ful planning and management, particularly ernments to better help the neediest. where there is little experience with beneficiary Where it supports structural adjustment, the participation in public sector projects and service Bank is committed to encouraging countries to delivery. Growth, Poverty, and Hunger 25 With a grant from the Swedish International must respect practices and food preferences Development Authority, the Bank created a which contribute to cultural identity. Many learning group to understand the role of local NGOs and CBOs have an innovative, entrepre- participation in development projects. This neurial approach to development. They have cre- group has studied twenty projects that included ated programs-such as the Grameen Bank in a high degree of local participation. It is incor- Bangladesh-which have clearly reduced porating the lessons learned in a Participation poverty and hunger. The Bank is increasingly Sourcebook to help guide governments and Bank involving NGOs and CBOs in all phases of pro- staff in adopting participatory approaches in pro- ject and program development and execution. At ject work. the conference the Bank underlined its commit- The institutional design of poverty-reduction ment to further developing mechanisms for projects.is also critical. Local NGOs and CBOs Bank-NGO collaboration. should be encouraged to provide leadership in Participation by poor and hungry groups in such programs. Special care is needed to ensure program administration often requires devolving that local vested interests do not obstruct the central government authority and fiscal proper administration of programs. Social analy- resources from hierarchial bureaucracies to local ses can help clarify the gender, sociocultural, and governments and communities. As the NGO demographic dimensions that may significantly statement to the conference emphasized, true influence the outcome of programs. Beneficiary empowerment of the poor depends on increasing assessment is one promising method of collecting their access to productive resources-land, capi- information on these factors. It has been used to tal, technology, and skills. This empowers local strengthen program design in a number of Bank groups to participate in their own development. projects (box 8). The best-intentioned central bureaucracies do not NGOs and CBOs can provide valuable sufficiently consider local factors as a municipal- insights about a country's culture, practices, and ity or CBO would. A review of sector reports attitudes to project design and implementation. completed in more than ten countries reveals that NGOs rightly stress that effective participation development projects would have been more requires full recognition, respect, and support of successful had the borrower governments community rights and social processes. Programs devoted greater fiscal resources and responsibili- Box 7. Ghana rural water supplies-importance of local participation Between 1973 and 1981 the Water health component, using public ject will expand the number of Utilization Project in Ghana radio broadcasts, were incorpo- wells and handpumps to include installed 2,700 boreholes and rated. An estimated 75 percent of all villages in the region and handpumps in 1,000 poor villages the rural population in this area replace the pumps with a more with a total population of 600,000 now has access to safe water. appropriate model manufactured to 700,000 people. It soon became About 90 percent of the pumps are in Kenya. There is also pressure for evident that residents used the working and health standards this project to expand into more wells only when other water was have improved; the incidence of activities. Others argue, however, not available-wellwater did not guinea worm and diarrhea has that the project should remain taste the same-and as long as the declined and women use less focused on water supply and that pump did not break down. effort to collect water. the next stage should be to shift Villagers repaired the pumps only This project demonstrated that responsibility to Ghanaians-and reluctantly because they viewed the learning period can be long and encourage the government to them as Canadian or government that projects must be allowed- adopt similar approaches in other pumps-not their own. indeed, helped-to evolve. The parts of the country. The project began to improve project did not succeed until the in the early 1980s when the project beneficiaries were active partners Source: The Water Utilization Pro- focus was shifted from technology and participated fully. The project ject: A Case Study otn a Water and to community development. The today isvastlydifferentfrom when Hlealth Education Project (CIDA participation of the villagers, par- it was first established. 1990). ticularly women, and a public The current phase of the pro- 26 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community ties to local authorities. In Mexico, for example, The Municipal Funds of Oaxaca (one of the poor- municipalities successfully implemented tens of est states in Mexico) directly transferred project thousands of small-scale development projects resources to small rural communities. For the first once they were given small sums and sufficient time, these communities had the fiscal resources authority under the Bank-assisted Municipal to carry out small local projects. These resources Funds Programs. helped to mobilize considerable contributions How well does decentralization work in serv- from beneficiaries. ing the poor? Much depends on local political Decentralization of fiscal resources must be and social structures. For example, a recent report accompanied by mechanisms for ensuring greater on the results of the decentralization of rural accountability of local governments to their pop- development administration in Karnataka, India, ulations, as has occurred in India with the recent suggests that local governments have speeded up reintroduction of an elected local government the execution of development projects and pro- (box 9). The Bank is progressively strengthening grams, and made them correspond more closely Bank-assisted operations by integrating the to the preferences of local communities. Never- lessons from the Learning Group on Participation theless, they have not increased those program into its activities. This includes encouraging activities which directly benefit the poorest or greater beneficiary participation through greater better dealt with environmental externalities. decentralization of functions and fiscal resources Central support for such programs is necessary to local governments and CBOs and strengthen- even if they are locally planned and executed. ing local accountability to beneficiaries. Box 8. Beneficiary assessment Beneficiary assessment is a largely Findings from beneficiary The residents preferred paying qualitative method of inquiry that assessments conducted to date rent or the higher rates for seeks to sensitize decisionmakers include: trucked water. Loan terms were to the points of view of key actors *Renters in a squatter upgrading restructured to better accom- in development-beneficiaries, project in La Paz, Bolivia who modate beneficiaries' financial service providers, and managers. had not been provided water situations. Since 1981 beneficiary assessments and sewer connections contin- *Parents in rural Mali, where have been conducted for twenty- ued using traditional ways. This only 15 percent of primary six World Bank-financed projects negated the value of the school-age children were in in seventeen countries. Assess- improvements for the entire school, were found to value for- ments have so far been conducted area. Subsequent upgrading mal education mostly as a step- in agricultural, urban, family plan- covered all residents of a site- ping stone to public sector ning, health, educational, indus- homeowners and renters. employment, for which there trial, and energy projects by local *Residents in diverse urban has been a rapidly diminishing personnel, with orientation pro- areas-in Burundi, Ecuador, demand. An informational cam- vided by Bank staff or consultants. and Thailand-rejected long- paign is being conducted about Beneficiary assessment in- term loans for piped water and the benefits of literacy and volves interviews with a sample of home improvements. Techni- mathematical skills to the fami- beneficiaries followed by one to cians had estimated the ability lies' agricultural production in three months resident observation of these people to pay for bet- order to encourage parents to of project beneficiaries. Project terment on the basis of average send their children to school. managers or policy designers par- monthly income, but the people ticipate in the assessment from the considered such long-term Source: Poverty Reduction Handbook outset. Interviewers are trained to commitments too risky. They (World Bank 1992a). See also be precise, based on a preformu- feared that, during unemploy- Salmen (1987 and 1992). lated guide, to facilitate quantifica- ment, they would not be able to tion of findings. afford the monthly payment. Growth, Poverty, and Hunger 27 Box 9. Rural development administration in India The Indian constitution developed block, and district level became After forty-five years of inde- after independence (1947) gave the mandatory. The state's own cen- pendence, India recently approved states strong powers to administer tralized development agencies a constitutional amendment that their own development. Agricul- were subordinated to the elected makes similar changes mandatory ture and irrigation, for example, council, which received substan- for all the states that have ratified are state matters. While it man- tial authority to decide develop- the constitutional amendments. dated elected local governments at ment spending. These measures Apart from the mandate of direct village and district levels-the were opposed by the state level elections to the village, block, and Panchayati Raj-the constitution bureaucracies and specialized district councils, the amendment left the assignment of local func- development agencies; however, mandates an independent elec- tions and resources to each state. the result has been accelerated toral commission at the state level Most states starved local govern- rural development, much more to supervise the elections, and a ments of resources and let the elec- closely matched to the desires and finance commission with statutory toral process at the village and needs of local populations, and a representation of both the state and district level decay. reduction in misappropriation of the districts. The commission A revival of the Panchayati Raj funds and corruption. The local assigns responsibilities to the dif- system began in the 1980s in the bodies in Karnataka still do not ferent levels of govemment and states of West Bengal and fully take the interests of the poor redistributes the fiscal resources of Kamataka, where district elec- into account, but service delivery the state among the state, the dis- tions to the councils at village, has improved. tricts, and the villages. PART II Specific Actions and Programs Dart I of this report discussed ways to reduce precise priority depends on country considera- poverty and hunger with general economic tions and on the specific characteristics of the dif- policies and investments in human capital, ferent subgroups of the poor. and how to adjust economies that have veered off the path of broad-based growth while protecting Urgent Low-Cost Actions to Reduce Hunger the welfare of the poor. The report also discussed general policies to increase effective participation Increasing the income, education, and access of of the poor in their own development. In part II poor people in developing countries to basic the report deals with direct interventions that can health care, clean water, and sanitation is the cost-effectively reduce hunger or increase the surest way to reduce hunger. However, experi- incomes of the poor. Participants at the Con- ence has shown that remarkable short-term ference on Overcoming Global Hunger empha- results can be achieved-even without growth- sized the need for greater support for such by implementing a variety of low-cost direct interventions. interventions. These actions can be undertaken The most appropriate mix of actions depends by governments, NGOs, or community groups on the economic circumstances of each country with support from donors and the specialized and on the characteristics of the poor and hungry. agencies and organizations of the UN, such as the These factors should be analyzed in the poverty World Health Organization (WHO), UNICEF, profiles and strategies for each country. This sec- and the World Food Programme (WFP). Several tion emphasizes actions likely to be cost-effective of these interventions are simple and cheap, and in most countries. The actions represent best can therefore be successful even where the gov- practice distilled from the worldwide experience ernment is weak. There is no excuse for delaying of countries and international organizations. The implementation of these actions. The poor and interventions will not likely result from private hungry cannot wait for macroeconomic stabiliza- sector initiative. They can be implemented by tion and structural reform. NGOs and community groups, the government The best recent example of a simple, low-cost or the private sector, but they require partial or action (implemented by countries with interna- full government financing. The actions are tional assistance from UNICEF, the U.S. Agency ordered roughly in terms of the immediacy of for International Development, and others) is the their impact on the reduction of hunger, and the campaign to make oral rehydration therapy cost and complexity of their implementation. The widely available in developing countries (box 29 30 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community 10). It has saved and improved millions of lives, mary cause of acquired blindness in children. even though the therapy does not cure diarrhea The WHO calculates that 13.8 million children or its underlying causes-inappropriate behav- have some degree of eye damage because of vit- ior combined with lack of clean water, sanitation, amin A deficiency; of these, 250,000 to 500,000 go and health services. blind every year, and two-thirds of the blinded As discussed in World Development Report 1993 children die. (World Bank 1993c) and during the Conference Iodine deficiency causes mental retardation, on Overcoming Global Hunger, there are at least delayed motor development, and stunting, as three other sets of actions that can cheaply and well as neuro-muscular speech and hearing dis- quickly reduce malnutrition: orders. Iodine deficiency is the leading prevent- * Measures to overcome micronutrient defi- able cause of intellectual impairment in the ciencies world. Iodine deficiency causes cretinism in * Expanded childhood immunization about 5.7 million people and mental retardation * Control of parasitic infections that cause ane- in about 20 million people. mia and malnutrition. Iron deficiency is the most common micro- nutrient disorder. It reduces physical productiv- Micronutrients and Protein-Energy Malnutrition ity and the capacity to learn. By reducing appetite, it may diminish a child's food intake Micronutrient deficiencies seriously undermine and growth. Women especially suffer from iron the health and productivity of the poor-espe- deficiency because menstruation and childbear- cially deficiencies of vitamin A, iodine, and iron. ing raise their need for iron. Anemia-a shortage Both vitamin A and iodine deficiency are partic- of iron in the blood-increases the risk of death ularly common in Asia and Sub-Saharan Africa. from hemorrhage during childbirth. The problem Vitamin A deficiency increases the severity and is most active in India, where 88 percent of preg- mortality of infections, especially measles and nant women are anemic. Almost 60 percent of diarrhea. In many countries one quarter of early women are anemic in other parts of Asia. childhood deaths could be prevented by ensur- Diets must contain both energy and protein. ing children receive adequate vitamin A. Vitamin Because little is known about the relative impor- A deficiency also causes vision loss and is the pri- tance of adding energy or protein to an initially Box 10. Oral rehydration therapy: an effective, low-cost way of reducing deaths One of the most successful exam- for less than $0.10 a packet. persistent diarrhea, which require ples of a simple, low-cost action is UNICEF, USAID, and others then appropriate antibiotic treatment. oral rehydration therapy (ORT) to aggressively promoted the Developing countries and aid reduce death caused by diarrhea. packet's use in the developing organizations therefore must also Ten years ago, diarrhea was one of world through education pro- increase efforts in other areas. the major killers of the world's grams, including contests and Some of the areas include improv- children, claiming nearly 4 mil- public demonstrations, and by ing water cleanliness and sanita- lion lives each year. Most of the tapping the support of civic orga- tion and educating parents about victims died of dehydration, a nizations, religious groups, and preventing diarrheal disease, symptom of larger, complex prob- heads of state. As a result, ORT is such as the importance of breast- lems, including lack of adequate used by one family in three in the feeding, giving children extra education, nutrition, and water developing world, saving about meals after the illness is over, and and sanitation. one million livesa year. Diarrhea is washing hands before touching In the mid-1970s, however, doc- no longer the leading cause of food. tors found that an oral solution of childhood death. Nevertheless, low-cost, highly sugar, salt, and water could pre- Still, diarrhea is by no means a effective programs such as ORT vent dehydration better and conquered foe. It continues to cannot wait and need to be cheaper than an intravenous injec- cause about 1.5 million deaths a expanded rapidly. Such programs tion of glucose, which had been year. One major problem is that are well within the capacities and the standard until then. The solu- ORT cannot address two leading financial resources of even the tion was produced and packaged cause of diarrhea: dysentery and poorest countries. Specific Actions and Programs 31 poor diet, the effects of deficiencies in either or overconsumption of energy, fat, salt, and sugar) both components is combined under the term are risk factors for illness and death. Studies in "protein-energy malnutrition." Protein-energy Asia and Africa consistently show that mild-to- malnutrition raises the risk of death and may moderate stunting or underweight increases the reduce physical and mental capacity. Worldwide risk of premature death and contributes to 25 to about 780 million people are estimated to be 50 percent of childhood mortality. The greatest energy-deficient according to WHO standards. It risk occurs for children in their second year, after is not known how many of them are also protein- they are weaned. Malnourished children die deficient, or how many people who get enough principally from measles, diarrhea, respiratory energy may still suffer from a protein shortage. disease, tuberculosis, pertussis, and malaria. According to World Development Report 1993, Child deaths from these diseases cost 231 million in 1990 dietary deficiencies caused a direct loss of DALYs, making the total burden attributable to almost 46 million disability-adjusted life years malnutrition at least one-quarter that amount, or (DALYs) and represent 3.4 percent of the global 60 million DALYs. burden of disease (table 1). (DALYs measure the burden of disease; they combine healthy life Immunization years lost because of premature death with those lost as a result of disability.) Anemia accounts for One way to tackle such problems is through pop- 1.3 percent of the total female disease burden, and ulation-based health interventions, including the 24 percent among women of reproductive age. Expanded Program on Immunization (EPI), Children under five are the principal victims of which relies on personnel with limited training to vitamin A deficiency, iodine deficiency, and pro- provide drugs, vaccines, or specific health ser- tein-energy malnutrition. The nutritional disease vices directly to target populations-in schools, burden for young children is 32 million DALYs, at work sites, or in households. Government or 6 percent of their total burden of illness. finance for such community-based programs is Malnutrition's total impact on health is much justified because of their high cost-effectiveness. larger. Mild or moderate protein-energy malnu- The objective is usually to provide services to all trition and micronutrient deficiencies (as well as members of a community, because the services Table 1. Direct and indirect contributions of malnutrition to the global burden of disease, 1990 (millions of DALYs. except as specified) Latin America Middle Formilly Established Sub-Saharan OtherAsia and the Eastem socialist market Type of malnutntion 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 ,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 undervveight' 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 defciencyb 13.4 14.0 1.0 7.0 1.8 2.0 0.0 0.0 39.1 a. Based on the global burden of disease attributable to deaths from tuberculosis, measles, pertussis, malaria, and diarrheal and respiratory diseases in children under age frve; in developing countries 25 percent of those deaths are attributed to mild or moderate underweight b. Based on estimated deaths attributable to vitamin A deticiency In age groups 6-Il months and 1-4 years. These account for, respectively, I0 and 30 percent of all such deaths in high-risk countries and for 3 and IO percent of all such deaths in other countnes. Thirty lost DALYs are attributed to each child death, losses are redistributed to the regional classification used in this report. Source World Development Report 1993 (World Bank I 993c). 32 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community create externalities or indirect benefits, and some other vehicle must be found for reaching because the diseases they typically combat are very young children. Adding these two vaccines particular problems for the poor. Health inter- and two micronutrients to the EPI (EPI Plus) ventions are also particularly important for would improve health substantially, particularly hunger reduction, including immunization and in the poorest households, for a modest increase mass treatment for worms. The EPI now reaches of about 15 percent in the cost of reaching each about 80 percent of children in developing coun- child with complete services (vaccines and tries and averts an estimated 3.2 million deaths a micronutrients). The estimated costs and health year at a cost of $1.4 billion a year. benefits of the EPI Plus cluster in two different set- Vaccines to prevent tuberculosis, measles, tings are shown in table 2. Total annual costs range diphtheria, pertussis, tetanus, and polio have from between $2.2 billion and $2.4 billion for EPI revolutionized preventive medicine over the past Plus-less than 2 percent of the public health two decades. Costs are less than $10 per DALY expenditure of developing countries. Expanding gained for measles immunization, and less than coverage from 80 to 95 percent would probably $25 for a combination of polio and DPT (diph- increase annual costs by between $500 million and theria, pertussis, and tetanus) immunizations. $750 million. In low-income countries the These vaccines, together with BCG immuniza- increased coverage would reduce the global dis- tion against tuberculosis and leprosy and immu- ease burden by about 6 percent. nization of pregnant women against tetanus, form the basis of the EPI. As a result of the EPI, Mass Treatmentfor Parasitic Worms the share of children immunized rose from less than 5 percent in 1977 to 20 to 30 percent by 1983. The most common intestinal worms-round- By 1990 polio, DPT, and measles vaccines had worms, hookworms, and whipworms-each reached approximately 80 percent of all children, infect between 170 million and 400 million and about 35 percent of pregnant women were school-age children annually. Schistosomiasis receiving tetanus toxoid. infection, also caused by parasitic worms, affects Had vaccination coverage remained at the almost 100 million school-age children annually. low levels of the 1970s, as many as 120 million The immediate effects of infection-including DALYs a year (the equivalent of 23 percent of the failure to thrive, anemia, and impaired cogni- global burden of disease among children under tion-can now be rapidly reversed by low-cost, age five in 1990) would be lost to the diseases pre- single-dose, oral therapy. Studies of single-course vented by the EPI. At current levels of vaccination treatment of school children with hookworm or coverage these diseases cause a loss of 55 million schistosomiasis in Kenya, with worm-induced DALYs, or 10 percent of the disease burden disease in India, and with trichuriasis in the West among children under age five. The cost of fully Indies showed remarkable spurts in growth and immunizing a child in low-income countries is about $15, with a range of $6 to more than $20 Table 2. Costs and health benefits of the EPI plus cluster in the picesof lbor nd oher ocal twvo developing country settings, 1990 depending on the prces of labor and other local (U.S. dollars except as specified) inputs. Low-income countries Middle-income Two extensions of the EPI appear to be justi- (high mortality countries (low mortality fied. First, coverage should be extended, probably Costs and benefits and fertility) and medium fertility) to 90 percent of all children. The costs of expand- Cost per capita 0.5 0.8 ing coverage are relatively high, but so are the Cost per fully immunized child 14.6 28.6 gains. Those not covered often lack any health ser- Cost per DALY gained 12-17 25-30 vices and are disproportionately vulnerable to the as a percentage of diseases. Second, it makes sense to include addi- income per capita' 0.14 0.03 tional items in the package: hepatitis B and yellow Potential health gais fever vaccines for selected countries and vitamin global burden of disease 6.0 1.0 A and iodine supplements in regions where defi- Note: Figures are based on 95 percent coverage. ciency of these nutrients is highly prevalent. If a. Income per capita in 1990 was assumed to be $350 for low-income countries and $2,500 for middle-,ncome countries. micronutrients are not delivered through EPI, Source: World Development Report 1993 (World Bank I 993c). Specific Actions and Programs 33 development in all the populations studied, try to prepare and finance actions to implement including large percentages of children with the three hunger-reducing health and nutrition asymptomatic infections. Treatment also interventions to achieve the following goals improved cognitive development. through community-level delivery: Because it is not necessary to determine which * Reduce vitamin A, iodine, and iron deficien- species are present, curing worm infections is cies through fortification, supplementation, simple with inexpensive modem drugs such as and dietary modification. Albendozole and Praziquantel. The safety of * Expand childhood immunization coverage these drugs has led the WHO to develop proto- from 80 percent to 90 percent. cols for mass use (where a high prevalence of * Implement a low-cost program for controlling infection exists) and by providers who are not parasitic infections that cause anemia and medically trained. That makes for high cost-effec- other forms of malnutrition. tiveness. Treatment usually cures the current infection, but in endemic areas, children will Food Supplementation inevitably become reinfected. A return to pre- treatment levels of infection typically takes about Direct, low-cost actions to reduce hunger are twelve months for roundworm and whipworm needed. They will not solve chronic malnutrition and twenty-four months or more for hookworm. alone. That will require greater financial and Rates of reinfection can be reduced by environ- institutional resources. Nevertheless, such mental improvements, especially sanitation, but actions have been profoundly successful and where this is impractical or unaffordable, it is highly cost-effective. National health services, cost-effective to repeat the therapy at regular NGOs, and other groups are able to provide basic intervals. health care services to poor groups at affordable The benefits of individual treatment can be costs. Programs to care for pregnant women and significantly enhanced by communitywide treat- children under 3, monitor growth and vaccinate ment which, by lowering the overall levels of children, provide supplemental feeding, and environmental contamination, slows the rate of educate families about proper nutrition and reinfection. Treatment programs targeted at the health have been implemented around the world. most heavily infected group (school-age chil- Growth and nutrition monitoring can also take dren) reduce infection immediately both among place in school feeding programs and other social those treated and in the rest of the population. or community-based programs. Many programs Treatment through schools also allows delivery at have achieved excellent results. Their expansion a relatively low cost. A program in Montserrat must therefore be a high priority. cost less than $1.50 a person for a cycle of eight Programs that provide food are not easy to treatments. A program managed by an NGO in implement effectively. Inadequate targeting and Jakarta initially cost $0.74 a person a year. After lack of attention to other causes of malnutrition expansion to almost 1,000 schools, costs fell to often mean that food is wasted. With proper tar- $0.26. Such programs are extremely cost-effec- geting and attention to changing behavior, how- tive, at $15 to $30 per DALY gained. In light of this ever, some supplementation programs have cost-effectiveness and the burden of disease worked. For example, a program in Tamil Nadu, addressed, the Rockefeller Foundation and the India achieved remarkable gains by distributing United Nations Development Programme food only where children's growth faltered, (UNDP) are initiating a major program to docu- while providing information to mothers through ment and explore the potential of school-based highly motivated community nutrition workers health interventions that focus on deworming, (box 11). This success came despite economic provision of micronutrient supplements, and stagnation. There was no improvement in dis- health education. tricts not participating in the program. World Bank officials recognize that the main Most developing countries have also created actors in these areas are countries, NGOs, programs to assist those who have insufficient UNICEF, the WHO, and bilateral donors. The resources to eat, often costly programs handing Bank stands ready to assist each borrowing coun- out food staples, subsidizing food prices, or giv- 34 The World Bank's Strategy for Reducing Poverty and Hunger: A Report to the Development Community ing food rations or coupons. Many of these pro- percent in 1981 and 10.8 percent in 1982. A com- grams are poorly targeted and wasteful. By redi- mon feature of these schemes is that the rich recting the resources spent in wasteful programs, receive a greater allocation per person than the hunger reduction could be achieved at reduced poor, although this amount often accounts for a cost in many countries. The best programs raise smaller share of income. In Egypt urban house- the real incomes of the direct beneficiaries and holds in the top quartile received 18.1 Egyptian provide a safety net to protect a wider group of pounds (equivalent to 3.4 percent of household the poor against collapses in real incomes. expenditures), compared with 15.4 Egyptian pounds (8.7 percent of household expenditures) General Food Price Subsidies for urban households in the lowest quartile (table 3). All benefits going to the nonpoor add to the General food price subsidy schemes supply costs of reducing poverty. Indeed, if general sub- unlimited amounts of subsidized food to anyone sidies are to provide reasonable transfers to the who wishes to buy it. Brazil, China, Colombia, poor, they become very expensive. In Egypt, only Egypt, Mexico, Morocco, Pakistan, Sudan, about twenty cents of each dollar spent reached Thailand, Tunisia, and, before 1972, Sri Lanka those in the lowest quartile. have all operated schemes of this kind, some A better way to reach the poor is to limit sub- national and some regional. Costs ranged from sidies to commodities consumed mainly by the less than 1 percent of total public expenditures in poor. In Egypt benefits from subsidies on coarse Colombia in 1978 to 1980 to 10 to 17 percent in flour accrue mainly to low-income groups. The Egypt between the mid-1970s and 1984. Egypt's urban poor gain more from the bread subsidy experience illustrates the advantages and draw- than do the rural poor, but the reverse is true for backs of these schemes. the wheat subsidy. Egypt's marketwide food subsidy program has been costly, but it has succeeded in reaching Rationed Food Subsidies the poor. During 1981 and 1982 the transfer value represented a sizable portion of the total expen- An alternative to a general subsidy is to provide diture of the poorest urban and rural groups-8.7 a quota, or ration, of subsidized food to house- Box 11. Tamil Nadu Integrated Nutrition Project: making supplementary feeding work In the late 1970s the government of vided immediately to those who malnourished. Of those receiving the state of Tamil Nadu in South were severely malnourished, and food supplementation, 67 percent India was operating twenty-five feeding for children with faltering gained enough weight to graduate different supplementary feeding growth was provided after one in ninety days; all except the programs. Evaluation showed month (for children aged 6-12 severely malnourished graduated these programs to be ineffective months) or three months (for chil- within 150 days. Because partici- and identified several reasons. dren aged 12-35 months). The chil- pants were fed only when The Tamnil Nadu Integrated dren selected were fed for at least required, food was only 13 percent Nutrition Project, the first phase of ninety days. If they failed to gain at of the project's total cost, much less which ran from 1980 to 1989, was least 500 grams, they were referred than is typical in supplementation designed to target services more to health care, and feeding was programs. When the program effectively, to improve family continued for up to 180 days. began in 1980, 45 to 50 percent of nutrition and health practices, and Intensive nutrition education was the children required feeding. By to improve maternal and child directed at mothers of at-risk chil- 1988 the project had brought the health services. dren. Food supplementation was share down to 24 percent. The Children aged 6-36 months were also offered to women whose chil- decline was achieved by improv- weighed each month. Of every 100 dren were being fed, to those who ing access to health institution ser- children selected for feeding, 44 had numerous children, and to vice and raising demand for such were normal in weight but falter- those who were nursing while services and access to potable ing in growth, 34 were moderately pregnant. water and sewerage. malnourished and faltering, and The project cut severe malnutri- 22 were severely malnourished. tion in half and prevented many Source: World Development Report Supplementary feeding was pro- at-risk children from becoming 1993 (World Bank 1993d). Specific Actions and Programs 35 Table 3. Distribution of the annual income transfer transfer was lower, and administrative efficien- from the general food subsidy, Egypt, 198182 cies led to substantial savings. Stamps for food Type of household Amount of transfer Tronsfer as a percentoge and kerosene were targeted to families with self- and income level (Egyptian pounds) of household expenditure reported incomes of less than 300 rupees a Urban month-about half the population. The govern- Poorest quartile 15.4 8.6 ment removed all other subsidies on food by Richest quartile 18.1 3.4 Rural 1980. Food subsidies'share of GNP dropped from Poorest quartile 11.9 10.8 5.0 to 1.3 percent between the mid-1970s and Richest quartile 15.2 2.7 1984. And their share of total government expen- Source. Alderman and von Braun 1 984. diture fell from 15 to 3 percent during the same period. But the real value of food stamps eroded holds, while permitting unlimited sales on the in the 1980s from 83 percent of the benefits of the open market. Schemes of this type have operated general subsidy at the time of introduction to 43 in Egypt and in South Asia (Bangladesh, India, percent in 1981-82. Similarly, Mexico has Pakistan, and Sri Lanka). The absolute transfer replaced its general urban maize subsidy pro- under a general ration is similar for all income gram with the Tortivale program. The program groups. Thus, the rations tend to be more pro- successfully targets one kilogram of tortillas per gressive than general food subsidies. For exam- day to urban households earning less than two ple, the absolute transfer to the richest quartile in minimum wages. urban areas under Egypt's general food subsidy was 20 percent greater than that received by the Targeting Hungry Families through Nutrition poorest quartile. With the general ration, the rich and Growth Monitoring received 5 percent less than the poor. Ration programs have often been established The most precise way to target food and nutrition for political purposes and have rarely been is through monitoring in health posts and schools targeted to the neediest. But it is possible to intro- and through special programs. Monitoring pro- duce better targeting and a degree of self-selec- grams often combine health, education, and tion into rationing schemes without costly means nutrition with subsidized or free food. Many testing. In northeast Brazil, small amounts of sub- countries currently have primary health care sys- sidized basic foods were sold through shops in tems to help the sick that are completely separate poor neighborhoods that the well-to-do regarded from food distribution systems that help the hun- as unsafe. The inconvenience of the locations and gry. Few have integrated systems that success- the limit on purchase size further discouraged fully address the relationship between hunger affluent nonresidents from participating. India is and health-a child may be taking in enough now targeting 1,752 poor rural blocks for priority calories, but could also be carrying a parasite. attention in its public food distribution program. Chile has achieved this integration over the past It is also expanding the employment guarantee thirty years and shows the long-term payoffs of scheme to the same blocks, so that the poor will this strategy (box 12). A recent United Nations have the income to buy the food in the expanded (UN) study found that "the only effective way to network of ration shops. improve growth is to reduce the burden of infec- tion at the same time as improving dietary intake. Food Coupons or Food Stamps Conversely, health programs that do not attend to increasing food availability at the family level Food stamps are similar to ration schemes, except and improving dietary intake by sick individuals that the quota is measured in nominal currency will have limited impact on growth and morbid- units rather than commodity weight. Food ity" (Tomkins and Watson 1989, p.42). stamps are usually better-targeted to needy Nutrition monitoring in health or family wel- groups than rations. fare programs or in schools can identify under- Facing fiscal crisis, Sri Lanka in 1978 replaced nourished children or their mothers. Through the its forty-year-old general food subsidy and ration children or the mothers the programs can iden- scheme with a food stamp program. The total tify and target other poor people-hungry fami- 36 The World Bank's Strategyfor Reducing Poverty and Htiuger: A Report to tle Dev>elopment Community lies. The hungry child or mother provides reliable children under five through registration at pri- information on the nutritional status of the rest of mary health care clinics. The system encourages the family. While providing the food or purchas- preventive health care and successfully screens ing power for the whole family of the hungry out wealthier households, which tend to use pri- individual is costly, it is usually much cheaper vate facilities. Coverage is high among the than food subsidy schemes, which rely on impre- intended beneficiaries. This part of the program is cise targeting. There may even be a net savings if markedly progressive. Poor, aged, and handi- poorly targeted programs are phased out. Such capped people already on welfare programs and programs can also educate families and monitor households with annual income under 2,600 their progress. Jamaican dollars are also eligible for food stamps. Providing take-home rations for families This part of the program is less well-targeted; would overburden the health system or school- there is both incomplete coverage of eligible based monitoring program. Instead, food stamps households and leakage to nontarget households. or vouchers can enable the family to obtain free Jamaica reinstated a general food subsidy in food in ration shops or other outlets. On-site feed- 1986. Although coverage is wider under the gen- ing is preferred only where the special nutritional eral subsidy scheme, targeting food stamps needs of the monitored child or mother are the through health posts has a larger impact on the primary objective. incomes of the poor, at half the cost (table 4). Providing food entitlement through health- Administrative costs (which are not reflected in based programs is particularly attractive. The the table) are kept low by relying on existing gov- health post can diagnose and treat disease and ernment networks. The value of the stamps has nutritional deficiencies with drugs or special eroded over time, though adjustments were foods. It can also provide vaccinations and health made in 1988, 1989, and 1990. and nutrition education to improve behavior. Chile's programs are based on this approach. The Mexico. In Mexico, a program similar to Jamaican, Honduran (box 13), and Mexican expe- Jamaica's and Honduras' was tested through the riences are also instructive. health centers. While Honduras' PRAF is a coun- trywide program focused on poor departments Jamaica. Jamaica's food stamp scheme was (see box 13), the Mexican program is exclusively introduced in 1984 to protect vulnerable groups rural, and distributes food rations rather than from the full impact of exchange rate movements coupons. Based on the initial positive results of and reduced public spending. Stamps are tar- this pilot, the government is considering a major geted to pregnant and lactating women and to expansion to the poor in selected rural areas. Box 12. Chile's success in combining health and nutrition programs Chile has a long history of com- free food but was also used as a tality rate fell from 119.5 to 16 per bined health and food distribution device to attract beneficiaries to thousand live births; the child programs, the most important ones demand health and nutrition ser- mortality rate dropped from 9.1 to being the Complementary Feeding vices (education and food distribu- 0.80 per thousand; low birth Program, the School Feeding tion). This was accomplished by weight declined to about 6 per- Program, and the Day-Care Center channelling the distribution of milk cent in 1992, a decrease of 45 per- Food and Education Program. The through the infrastructure of the cent since 1976; and life key objectives of the primary health care system, with expectancy at birth increased by Complementary Feeding Program health clinics and rural posts serv- more than 13 years, a world- are preventing malnutrition among ing the population nationwide. record gain. The greatest reduc- the most vulnerable groups and Since 1983 additional food and tion in infant maternal mortality promoting health through periodic more frequent health controls are has taken place in poorer rural medical visits, immunization, and provided to low-weight pregnant areas, where mortality rates were education on the use and demand women, undernourished children, highest. The decline was achieved of health services among low- and those nutritionally at risk. by improving access to health income families. The distribution of Results have been startling. institution service and access to food not only involved provision of From 1960 to 1992 the infant mor- potable water and sewerage. Specific Actions and Programs 37 Box 13. Honduras Family Assistance Program The Family Assistance Program show that about 88 percent of ulation at risk of malnutrition. It is (PRAF) was created in July 1990, funds are spent on food, and the operated in parallel with another and a pilot program was under- balance on school supplies, chil- successful food coupon program taken in 1991 to test the delivery of dren's shoes, and medicine. operated through the primary food coupons through the basic An evaluation of the pilot pro- school network thatbenefits about health network. The coupons are gram found that the food coupon 200,000 children. targeted to low-income children scheme was cost-effective in alle- As a result of the PRAF food under 5 and pregnant or nursing viating poverty and was more effi- coupon program, the government mothers. The intervention is con- cient than other nutrition has decided to reevaluate its long- centrated on the earliest stages of assistance programs. It also pro- term nutrition strategy. The new infancy and childhood to improve vided incentives for using primary government policy, expected to be the chances of reaching children health care services. In fact, no issued in 1994, would favor the before malnutrition causes perma- sooner was the program started food coupon approach over dona- nent damage. To maintain eligibil- than it became apparent that the tions in-kind, which are more ity, beneficiaries must meet health existing health system would not costly to administer. Eventually, surveillance requirements. The be able to cope with the increased most external food aid is expected program is self-targeted since bet- demand for health services which to be monetized and channeled ter-off women usually use private accompanied the introduction of through the food coupon program. clinics rather than public health the food coupon. During the first Breastfeeding and nutrition edu- facilities. year consultations had increased cation programs would be Beneficiaries receive the equiva- by 131 percent at health centers strengthened in the poorest lent of $45 a year in monthly and posts. To cope, a nutrition and departments, and school feeding coupons. The coupons are health project was designed to programs would be improved. redeemed in private stores, which support the expansion of the food One weak spot of the PRAF is the cash the coupons in at private coupon program and help the gov- lack of a comprehensive analysis banks, and the banks rediscount emient strengthen the delivery of the impact of the program on the coupons at the Central Bank. capacity of the primary health care mothers and children's nutritional Thus the PRAF administers the network. As a result of conversa- and health status. There also is still program, but uses the private sec- tions with the Bank and the gov- considerable scope to expand the tor and existing government insti- emient, the World Food program to cover a higher per- tutions. As a result, it has a Programme decided to monetize centage of the population at risk of reasonableadministrativecostof7 its food aid to Honduras and use malnutrition. There is thus need percent of total outlays. To sim- the available funds to participate for additional donor support. plify the administration of the pro- in the food coupon program. Because of its stronger impact on gram, PRAF does not control how The PRAF program currently nutrition at an early age, the food beneficiaries spend the coupon. benefits about 100,000 mothers coupon program distributed Yet, surveys of beneficiaries and and young children. This repre- through the health centers should spot checks of participating stores sents about 25 percent of the pop- receive priority. These examples prove that nutrition and subsidies. Increased access of health posts to food growth monitoring is a highly effective mecha- resources is therefore a high priority. Primary nism to target malnourished families. Targeting health services can be empowered to write "pre- can be done in schools or other social or commu- scriptions" for food for entire families once they nity-based programs, but targeting through the have identified the hungry child or mother. As health system has additional advantages. The the Honduran and Mexican examples show, such health post can identify whether malnutrition is capacity makes the primary health care system due to disease, parasites, or lack of micronutri- more attractive to users. Vaccinations, other pub- ents or food. It can initiate treatment of the dis- lic health inputs, and health and nutrition educa- ease or parasites and remedy nutrient tion can be provided to those most in need. These deficiencies with foods or supplements. Infor- synergies should be fully exploited. mation on malnutrition accrues at low cost in the At the conference the Bank committed itself to routine operation of health posts; this informa- encouraging countries to shift food subsidies to tion should be used to better target food aid and targeted food assistance programs which provide 38 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community food entitlements to families whose children or Table 4. General and targeted subsidies, Jamaica, 1988 mothers have been identified as being at risk of (Percent) malnourishment or low-weight births. Food enti- Targeted subsidy tlement should involve on-site feeding, food Item General subsidy (food stamps) stamps, or free ration cards rather than take home Cost as a share of food distribution. government expenditure' 3.0 1.6 Share of transfer going to Poorest quintile 14.0 31.0 Improving Effectiveness of Food Aid Richest quintile 26.0 8.0 and Dealing with Famines Transfer as a share of expendiure per recipient Poorest quintile 2.3 9.5 For many years in the development community, Richest quintle 0.1 1.0 providing food aid for emergencies and overall Share of households covered food availability has been controversial. Some Poorest quintile 100.0 51.0 . . , ................. ., . Rchestquintle ........... 00.0 6.0 analysts argue that food aid destroys farm incen- Richest quintile tives by increasing supply. Others have stressed a. Does not nclude admirnistrative costs. its unreliability. However, recent evidence sug- Sourc:e: larmaica Statistical Institute and Wotid Bank 1988 and 1989 its unreliability. However, recent evidence sug- gests that, appropriately used, food aid does not damage farm incentives as much as has been Food Aid assumed (box 14). On the other hand, food aid is unreliable and tends to be cut during food crises, Food aid can be a valuable resource for poor as in 1973-75. Industrialized countries no longer countries. There is a large political constituency have large surpluses they need to dispose of. for food aid, which includes farmer groups in They are likely to hold on to more of their stocks, donor countries, individuals, and organizations leaving less for food aid. At the same time, inter- concerned about hunger. In 1989 food aid to national food crises reduce countries' purchasing developing countries amounted to about $2.5 bil- power. Countries should therefore avoid exces- lion. Food aid represented about 18 percent of sive dependence on food aid. Food aid should not total U.S. assistance. be seen as a mechanism to directly augment the What is the best way to use food aid to reduce food supply. It provides the best supplemental hunger? Individuals, NGOs, and donor countries resource if it replaces commercial imports with- want to make sure that donated food actually out depressing prices and reducing the incentives gets dispensed to those who need it. They are for domestic food production. worried that schemes other than direct distribu- Box 14. Impact of food aid in Africa Though food aid averages only 10 *Food aid displaces commercial total net increase in food supply percent of total financial aid to imports and does not add to following an increase in food aid developing countries, it represents domestic food supplies. If there is, however, of lower magnitude more than half the food available is full displacement, prices than expected-because food aid for consumption in Botswana, should not change and there tends to replace almost an equiva- Cape Verde, Mauritania, and will be no effect on incentives. lent amount of regular food Mauritius. * Food aid is determined to some imports. What is the relationship of food extent by local food production. Food aid is more likely to have a aid to food production and com- Butinthemedium run itcangen- positive effect in countries that use mercial imports? Three main erate a positive supply effect that fertilizer intensively One possible hypotheses have been advanced: increases the level of production. explanation for this is that countries * Food aid is an addition to local Lavy (1990) used Sub-Saharan that enjoy a relative abundance of food supplies that ultimately Africa to test these hypotheses. He regular food aid can use the lowers prices and acts as a dis- found that food aid has a signifi- resources made available through incentive to local producers. The cant positive effect on food pro- reduced food imports to invest immediate effects may be small, duction. Any disincentive induced more in the agricultural sector. but a lagged response can be by the additional supply of food is generated. offset by the positive effects. The Source: Lavy 1990. Specific Actions and Programs 39 tion of food simply will not achieve this. Donor governments, NGOs, and international Therefore they often advocate providing donated organizations should assist beneficiary organiza- food directly to the hungry, as donations or wage tions to monetize food aid. Where possible, donor payments. Unfortunately, experience shows such governments should amend laws or regulations direct provision is often costly and inefficient. that obstruct monetization. Monetization will not People who are not consuming enough food depress producer incentives if food aid is sold at do not need to be given food. They instead need import-parity prices, and if it replaces commercial money, food stamps, or ration cards to purchase imports. The proceeds should be earmarked for food. Evidence suggests that poor people spend more effective low-cost food and nutrition pro- up to 85 percent of their earnings on food. They grams, or for employment generation. This will are not wasting money. People will meet their require bilateral donors, the WFP, and other inter- nutrition needs more cost-effectively if they national agencies to carefully coordinate with receive the money equivalent of the commodities ministries of health, agriculture, or food supply, supplied from food aid. Purchasing power pro- and with finance services, the private sector, and vides recipients a wider range of options to deal NGOs in the countries receiving the food aid. with challenges to people's livelihood and food deprivation. In India in the mid-1960s, for exam- Famines ple, sales of food aid and associated foreign exchange savings accounted for 30 to 40 percent Famines often occur without a decline in aggre- of government revenue. Some of that revenue gate food availability, and sharp declines in food paid for the infrastructure for the green revolu- availability do not always result in famine. tion, which in turn led to considerable increases Famine is typically more complex. Understand- in food production. Funds can also finance rural ing its causes and taking effective action require employment. understanding the way markets and govern- Two-thirds of nonemergency food aid is cur- ments work in a crisis and the capacity of peo- rently sold in the markets of developing coun- ple to protect themselves. Alleviating famines tries. Food aid commodities are often sold below requires addressing food supply management, full market value, depressing producer incen- health care, employment generation, and agri- tives. The proceeds from the sales are often used cultural assistance. When done right, such stra- for general government spending. The revenues tegies can be highly effective. Botswana is a from these sales must be better directed to pro- good example (box 15). To avoid famines we grams that give high priority to hunger allevia- must fight poverty. It is typically the poor who tion. The other third of nonemergency food aid is suffer in famines. Their vulnerability depends directly distributed to people in food-for-work on the history of poverty prior to the aggregate projects and other programs. This aid contributes shock. to hunger alleviation, but would be more cost- The main actors in famine relief are the United effective in the form of purchasing power. Nations Department of Humanitarian Affairs By selling the food in the target country cities, (UNDHA), the WFP, the Food and Agricultural the money can finance food coupons and the Organization (FAO), UNICEF, and many com- overhead costs of combined health and nutrition munity groups and NGOs. The IMF has a special programs. A good example is Honduras, where food facility to finance imports of food. the WFP and bilateral donors have used proceeds Famine from drought and desertification is a from the sale of donated food to finance health- slow-onset disaster which can be greatly miti- based food coupons. gated by advance warning and prompt counter- Using food aid in this way will meet the con- measures. Governments can stabilize prices and cerns of NGOs, community groups, policymak- food supplies by building up financial or physi- ers, and others that food aid get to the needy. cal reserves, or through imports. The Bank has Farmers in industrialized countries wishing to sell never sought to be a relief agency, and it is not their excess harvests should also be satisfied. And equipped to become one. The Bank aims to sup- the poor in the target country will have a cost- port countries and other organizations so they effective way of receiving the benefits of food aid. can eliminate particular famines. The Bank also 40 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community has a role to play in fighting poverty and institu- * Move from emergency relief to recovery and tional weaknesses-the root causes of vulnera- development bility to famine. The Bank can help countries * Better coordinate Bank activities with UN dis- recover from such crises, and return to a normal aster relief operations development path. In famine-prone settings, * Design projects better able to cope with and relief and development are two sides of the same adapt to the onset of droughts, building on the coin. The Bank has helped by approving Bank's recent experience in Southern Africa Emergency (Drought) Recovery Projects in * Use more flexible lending instruments to deal Sudan in 1991, an emergency food-related grant with the multiple needs for foreign exchange to UNICEF for use in Somalia in 1992, a major as droughts materialize. drought recovery operation in Southern Africa The Bank will also encourage governments to amounting to nearly $400 million in 1992, and an adopt policies to stabilize prices and food sup- Emergency (Drought) Recovery Project in Kenya plies by building up financial or physical in 1993. reserves, or through imports. By stabilizing The Bank is analyzing the impact of drought prices, governments can reduce speculation and and famine on Africa's human resource and keep food prices within the reach of the poor. development potential. This analysis, scheduled Kenya did this in 1984 when it imported food just for completion in 1994, hopes to substantially before domestic supplies were exhausted. By improve the Bank's response to drought in a contrast, uncertainty about future food supply number of important ways. Among them are (both harvests and imports) in Bangladesh dur- actions to: ing the 1974 famine led to sky-rocketing rice * Integrate the impact of drought into the coun- prices and a collapse in the food purchasing try assistance strategies and project designs of power of poor people. Mass starvation resulted. drought-prone countries The worst period of starvation actually preceded * Strengthen the country's capacity to prepare the threatened decline in aggregate food avail- for and mitigate drought ability. On the other hand, Botswana was able to Box 15. Drought relief in Africa: food interventions in Botswana Like many other African coun- gram. Botswana received large provided for malnourished chil- tries, Botswana had to cope with amounts of food aid in 1982-87, dren at health facilities, and a feed- episodes of severe drought in the but its relief efforts did not depend ing program for primary school 1980s. Unlike some other countries on the timely arrival of these sup- children was maintained through- (Ethiopia, Mozambique, Sudan), it plies. Large-scale imports of food out. Health measures, including has succeeded, through appropri- from South Africa were combined efforts to provide a clean and ate policies, in avoiding the worst with support for rural incomes dependable supply of water, were effects of famine. through public works and grants. also important. During the 1979-80 drought, relief Private traders and retailers were Despite large decreases in in Botswana relied almost entirely used to distribute food. Food domestic food production, the on transporting food aid into prices remained broadly uniform program was effective: the per- famine-affected areas and distribut- across the country during the centage of children who were ing it to the destitute. Because of drought period. undernourished had fallen in 1986 logistical difficulties, this approach "Take-home" rations were dis- to less than predrought levels. was unsuccessful. Beginning in tributed to all households that con- There were no deaths from starva- 1981 Botswana implemented a tained children under 10, pregnant tion. During the 1991-93 Southern broad, integrated program for and lactating women, and desti- African drought Botswana used drought relief and recovery that tutes. The government set up a similar tools to prevent the combines food supply manage- special trucking operation to sup- drought from creating a famine. ment, employment generation, and ply maize to remote areas. Over agricultural assistance. one-third of the population Source: World Development Report Effective food supply manage- received free rations during the 1993 (World Bank 1993c). ment lies at the core of the pro- drought. Intensive feeding was Specific Actions and Programs 41 avert tragedy by importing food from South ulation is increasing, policies that facilitate out- Africa and supporting rural incomes through migration are essential. But in many of these public works and grants (see box 15). regions additional programs requiring govern- A major source of vulnerability is the volatil- ment subsidies will still be necessary to meet ity of international food prices, as with the world basic needs, maintain yields, preserve natural food crises of the early 1970s. The Bank, in con- resources, and generate employment. junction with the IMF Food Facility, the WFP, and others, can assist countries in designing cost- Education effective strategies and financing arrangements for food emergencies caused by international The role of education for both boys and girls has price increases and droughts. been stressed several times in this report. Edu- cation accelerates growth and earning potential. Interventions to Improve the Incomes It has direct impact on hunger since it imparts of the Poor and Hungry health and nutrition information. It reduces fer- tility rates and therefore lowers the high depen- Achieving broad-based growth, regardless of the dency levels common among poor households. level of structural adjustment involved, requires Yet it remains insufficiently targeted to the poor, specific policies and programs attuned to two especially poor girls, in many countries. The broad tasks. First, economywide and sectoral main problem remains access to preprimary and policies must encourage employment-intensive primary education, which increase cognitive rural development and urban employment. skills and earning capacities more cost-effectively Experience indicates that this requires moderate that secondary and higher education. taxation of agriculture and relatively undistorted product and factor markets. Urban Informal Sector Employment Second, specific policies and targeted pro- grams must improve the poor's participation in Cities in developing countries are growing growth and reduce hunger. The poor need rapidly. The influx of rural residents in search of increased access to primary education, credit, employment and improved living conditions land, public infrastructure, and services. continues. Even where economic growth is Technical change must be accessible to small strong, many cities simply cannot provide formal farmers and the urban poor. The specific pro- sector employment for all the migrants, and grams should emerge from country poverty urban unemployment is unusually high. One strategies. Areas where the Bank has been suc- often overlooked solution is to encourage infor- cessful in the past have taken place in a wide vari- mal sector employment growth. Informal ety of settings. Land transfers can reduce poverty, employment includes street vending, shoeshin- especially where land distribution is highly ing, lottery ticket sales, street-corner repairs, and unequal. Other policies to increase and secure small-scale retailing and manufacturing. access to land, especially for women, can also These activities are sometimes held in low reduce poverty. Subsidized credit programs have regard by policymakers in developing countries. failed to reach the poor, but approaches such as Many policymakers believe the informal sector is group lending offer promising alternatives. unproductive. Local business elites often oppose Flexible programs that involve the intended ben- informal sector activities because they compete eficiaries, build institutions, employ NGOs and directly with their own businesses. Cities often local groups, and respond to local needs are the outlaw vending near churches, government best way to mold infrastructure, services, and buildings, and central squares. They also require technology to the needs of local communities and excessive licenses and tax payments, as in Cali, the poor. Colombia. As in most Latin American cities, the Resource-poor regions, where poverty and street traders of Cali are closely regulated by environmental degradation are interrelated, municipal and departmental authorities. Several require a different approach. Since the potential hundred pages of municipal and police regula- for growth in these regions is limited and the pop- tions specify the conditions for street trading. 42 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community Street traders must carry identification cards and show. Bank-supported agricultural credit pro- other official documents. Most also must obtain grams have usually failed to reach those groups, municipal trading licenses and health permits. despite considerable effort. Many borrowers are Informal sector activities are not negative. The not creditworthy and do not have profitable pro- informal sector can contribute to increases in jects. Instead, successful programs work with income and food intake in several ways. Informal individuals or groups on a long-term basis. sector workers are important distribution chan- Successful programs emphasize savings, credit nels for goods. They play a key role in solid waste discipline, and trust. Many of the successful pro- recycling. Informal sector workers build skills grams have found that women-the borrowers that may lead them into the formal sector. most often discriminated against-are capable Children, for example, contribute to household entrepreneurs with profitable, small-scale pro- eamings through jobs, and with age may become jects and surprisingly low loan delinquency rates. employees of construction firms. For poor Small-scale credit is therefore a particularly effec- women, the informal sector can provide alterna- tive way to reach women and help them to tives to prostitution. increase income and their children's calorie Informal sector jobs should not be scoffed at. intake. In many cases, workers in the informal sector do Group-based lending for microenterprises is as well as or better than those with formal jobs. most appropriate in areas with high population Vendors in Lima, Peru, enjoy incomes 38 percent density, landlessness, and a thriving semi-urban higher than the legal minimum wage. Food ven- or rural nonfarm economy. In some countries dors in Bangkok made three times the minimum group-based lending is being tested in urban set- wage. Average incomes in the informal sector are tings. Such credit is not effective in sparsely pop- often depressed by disproportionate numbers of ulated or risky agroclimate zones, such as the apprentices and other trainees. Average wage semi-arid tropics. comparisons should adjust for this, but generally In recent years, few local initiatives have suc- do not. ceeded as well as the Grameen Bank of The Bank encourages developing countries to Bangladesh. Beginning in one village in 1976 as strengthen the informal sector as a means of pro- an action-research project conducted by moting growth and reducing unemployment and Professor Mohammed Yunus, Grameen Bank has hunger. Restrictions on the informal sector grown to more than 1.4 million borrowers. It should be avoided, if not eliminated. Govern- extends over $14 million in loans every month, ments should provide crime-free space, improve with an average size of $75. It had an outstanding land rights, and reduce public administration, portfolio of $113 million at the end of 1992, while regulation, and taxation constraints on the infor- enjoying a repayment rate of 97 percent. Ninety mal sector. Education and programs to help four percent of Grameen Bank's borrowers are entrepreneurs can also be useful. poor, rural women. Grameen Bank lends to five to eight self- Creditfor the Poor selected persons who have agreed to form a group in which they guarantee to monitor each The poor, especially poor women, are generally other. Such group lending helps screen out risky excluded from formal finance, both as borrowers borrowers, and creates peer pressure to monitor and savers. Private banks exclude the poor and enforce the terms of the loan. Group enforce- because small transactions are unprofitable. What ment substitutes for collateral, which most poor counts for the poor more than low interest rates is people lack. The group is organized and trained access to interest-bearing savings and timely by Grameen Bank workers. small loans. Even if they have to pay interest rates Independent research (Khandker, Khalily, and above the market rates faced by larger borrowers, Khan 1993) has documented Grameen Bank's sig- the poor are often much better-off compared with nificant impact on its members. Compared to borrowing in the informal market. nonparticipant control groups, Grameen Bank Providing financial services to the poor and borrower incomes rose by more than 50 percent, women is not easy, as numerous failed programs and per capita food and nutrient intake by 9 per- Specific Actions and Programs 43 cent. Expenditures on clothing, health, and edu- community group efforts to spread such pro- cation rose by 18 percent. Employment, particu- grams. The Bank announced at the conference larly of women, rose by 36 percent, compared that it would join donors and other partners in with only 18 percent in villages without Grameen exploring programs that address the credit and Bank branches. savings needs of the self-employed poor. As a The Grameen Bank model has been success- first step, the Bank will begin consultations with fully replicated in pilot programs in other devel- other donors regarding long-term support and oping countries. A Grameen Trust was recently funding for microcredit programs. The Bank also established in Dhaka with the objective of pro- will commit $2 million in 1994 to assist Grameen moting Grameen replicators outside Bangladesh. Bank replicators in countries other than The oldest replicator is Amanah Ikhtiar Malaysia Bangladesh. (AIM) in Malaysia, which currently has more Credit extended by the Grameen Bank and than 19,000 borrowers. AIM is adding 1,000 new similar programs works well in many places. borrowers a month, while maintaining a repay- However, there are many regions where it cannot ment rate of more than 99 percent. Some of the make much of a dent. Credit presupposes eco- AIM's staff were trained at Grameen Bank in nomic growth. If an economy is growing, people Bangladesh. Other similar programs include borrow money and are able to pay off the princi- Project Dungganon (Philippines), Savecred (Sri pal and interest with increased income. With Lanka), and the Mudzi Fund (Malawi). increased income, they are able to save, purchase The family of Grameen Bank replicators is by equipment, and generally reinvest their profits. If no means the only initiative of this type. For there is no economic growth, however, the only example, the Foundation for International way to increase income and pay off the loan is by Community Assistance (FINCA) has helped taking business away from competitors. establish a network of more than 1,800 village Established small entrepreneurs may be hit hard banks in Latin America and Africa that provide as a result, and hunger and poverty may simply small loans to self-employed borrowers, includ- be shifted around. Banks are likely to lend only to ing many women. Last year these banks the most financially healthy farmers and enter- processed 2 million loan payments, with an aver- prises in poor economic environments. age repayment rate of 97 percent. In Bolivia, Thailand, and West Africa the Food Security in Resource-Poor Areas Freedom from Hunger organizations are pioneer- ing a system of credit with education, also using An increasing number of poor people live in envi- the village banking approach. Loans are pro- ronmentally fragile areas with little agroclimatic vided to small, self-managed associations, with potential, such as the Loess Plateau in China, the field staff providing advice on health, nutrition, highlands of Bolivia and Nepal, the African and microenterprise management. In Pakistan Sahel, and the semi-arid tropics. These regions the Aga Khan Rural Support Program also com- need a special development strategy for three rea- bines credit with training to reach women in the sons. First, their growth potential is limited, and poorest villages. In Indonesia the International approaches based on intensive agriculture or Fund for Agricultural Development (IFAD) has credit are unlikely to succeed. Second, they are funded a rural credit program to assist poor farm- increasingly occupied by poor people with the ers and landless workers in 2,000 villages, also fewest skills and the least access to infrastructure based on self-management principles. and supplies. Third, environmental degradation Within the context of overall financial sector in these regions adversely affects both the imme- reform, the Bank will provide increased attention diate area and regions downstream or downhill. to programs that make formal financial services The causes of these growing pressures on nat- available to the rural and urban poor, particularly ural resources are complex and interconnected. women. Since Bank-assisted programs have In many countries poor farmers are being rarely been as successful as programs designed marginalized and pushed to frontier areas. In and managed by NGOs and community groups, addition, population growth and the commer- the Bank is willing to directly support NGO and cialization of agriculture have forced farmers 44 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community who once relied on environmentally sustainable recent research on the Machakos district shows cultivation to use their land more intensively (box 16). Where land and climate are good, this poses little Providing massive infrastructure to develop threat. But the intensification of traditional farm- resource-poor regions may be neither cost-effec- ing methods, such as slash-and-burn agriculture, tive nor viable. A better strategy would start with has damaged the productivity of many marginal investment in education and training to spur out- soils and areas. Overgrazing, mismanaged irriga- migration to areas of better potential. Spending to tion, and an ever-widening search for fuelwood meet basic needs such as health care and drink- often accelerate the decline. Policies that discrim- ing water also will be required. Outmigration inate against smallholders in granting access to would be most effective in countries where labor land and forests make matters even worse. For demand is growing strongly in other regions. example, land policies have directed population Experience, however, shows that migration is movement away from the most productive land only a partial solution. Growth in areas of greater by giving a few large estates preferential access, potential is often not high enough; many as in Southern Africa, or by limiting migration, as resource-poor regions have rising populations in Tanzania. Resource degradation is not the only despite outmigration. Additional programs will possible outcome. When policies and market therefore be necessary. These programs will opportunities for smallholders are less discrimi- involve training poor farmers in better techniques natory, the rural population improves the sus- for farming, animal husbandry, and soil and tainability of the resource base over time, as moisture conservation, increasing the opportuni- Box 16. Land resource degradation is not inevitable: Machakos, Kenya Increasing population growth and by strong links between a local soci- has been a corresponding reduc- agricultural intensification in a ety with a broad, development-ori- tion in the area of general grazing, low-income area do not have to be ented leadership structure; and bush, and scrub. Much of the land accompanied by land resource nonagricultural and urban growth used is now under continuous cul- degradation, and land resource support the above processes by tivation, and almost all of the cul- management cannot be divorced reducing the share of the society tivated area is in some form of from the overall development that relies on the land resource for terracing. The rate of erosion has process. These are two of the prin- its livelihood and increasing the been sharply reduced, though it cipal conclusions of a recently con- options open to those who choose does still occur, and there is no evi- cluded study of land resource to do so, by creating demand for the dence to suggest that the quality of management in the Machakos dis- products of land. soils is declining under current trict of Kenya covering a sixty-year Agricultural conditions are poor practices. period from 1930 to 1990. in the Machakos district. The area The farmers of Machakos have Other conclusions of the study, is semi-arid, with annual rainfall made a large investment in their conducted by a team led by the of 600-1,000 millimeters, divided land resources. This has been Overseas Development Institute into two wet seasons. Few of the achieved without significant use (ODI) of London under the super- soils are classified as good. In the of credit in any formal sense. Cash vision of the Bank's Environment late 1930s the district was consid- earned from off-farm work and Department, include: market out- ered by the colonial administra- from the sales of farm products lets are important to providing a tion to be degrading alarmingly appears to have supported the use wide range of economically and and to be rapidly approaching, if of families' own labor (or labor technically viable land-use options not exceeding, its capacity to sup- mobilized through groups). Since which permit the user to devise an port its inhabitants and their live- the 1950s an ongoing process of overall sustainable system and to stock. Today the area has a agricultural innovation and modify it under changing eco- population five times as great and change has also taken place in the nomic and social conditions; this the value of agricultural output district. Major changes have process is an integral part of the per capita (at constant prices) is occurred in the structure of the process of agricultural innovation; estimated to be three times larger society. Women have taken a more land resource management innova- than it was then. active role and this has influenced tions come from multiple sources The cultivated area has family roles, including traditional and their introduction is enhanced expanded by four times and there agricultural tasks. Specific Actions and Programs 45 Box 17. Environmental management project in Burkina Faso The natural resource endowment prises alone. To reap the full bene- ated uncertainties that rendered of Burkina Faso is unusually fits, collaboration between enter- land improvement works risky. harsh. With population increasing prises was necessary. The project is the first phase of a at an annual rate of 2.6 percent in The Environmental Manage- long-term program to reverse nat- the 1980s, the poor's resources are ment Project was designed on the ural resource degradation through coming under increasing stress. basis of about twenty pilot opera- participatory and holistic commu- In the late 1970s the better- tions undertaken in the preceding nity land management planning endowed but disease-prone south- five years to test ways of enabling and implementation. The first west was opened up to rapid and obtaining collaboration such plans are now being final- settlement by the eradication of between enterprises. The out- ized. The plans include specific river blindness. At the same time, come was l'approche terroir-com- land improvement works, land emigration to the better-endowed munity land management husbandry practices to be C6te d'lvoire was an attractive planning and implementation adopted, and land-use rules gov- option available to and exercised based on a social unit (the com- erning access to common assets by many young Burkinabe. As munity's terroir) rather than a such as pasture, forests, and water. these options began to close in the physical unit such as a watershed. Works such as check dams or dikes mid-1980s, labor became increas- The project, cofinanced with and shelter belts will be funded by ingly abundant relative to land. France, Germany ,and Norway, the project as a grant, with the Land improvement practices were became effective in February 1992 community contributing the rapidly adopted by individual after a ten-month delay occa- unskilled labor. Formulation of farm enterprises. This process was sioned by the proviso that the these plans is an unavoidably slow aided by effective research and land tenure law of 1984 be process: blueprints are of no use extension. But it became apparent amended. That law had national- since site-specificity is an essential that some practices could not be ized rural land and, though never ingredient of ownership. implemented by individual enter- effectively implemented, had cre- ties for diversification and off-farm employment, control and agricultural production in the Loess and providing local user groups (such as the vil- Plateau, a 630,000 square kilometer area in the lage forestry associations in Korea and the range- northwest. By constructing new terraces, elimi- land management associations in Botswana) with nating uncontrolled grazing, and constructing rights to manage degraded communal lands. A soil dams to create cultivable flatlands, China recent example is the community-based resource increased per capita grain production more than management project in Burkina Faso (box 17). 30 percent. In many of these areas, farmers can adopt Special attention also needs to be given to low-cost, low-input technologies that would community-based forestry management in increase and stabilize yields, diversify produc- resource-poor areas, both to provide a source of tion, and maintain the resource base. One such income and to ensure the sustainability of forests. technology, contour cultivation, has raised The Bank supports investments in social and par- yields substantially-by 6 percent to 66 per- ticipatory forestry for three reasons. First, gov- cent-on slopes of up to 32 percent. When con- emient policing and centralized control of tour cultivation is supported by a vegetative forests have been ineffective in reducing degra- barrier-vetiver grass, for example-the bene- dation. Second, forest dwellers are poor, and fits are even greater. Vetiver grass has been used forestry programs must address their needs. for many years in the Caribbean, Kenya, and Finally, people are more likely to manage forests South Asia. It holds soil while allowing for water when their interests in the forests (income, filtration. It is cheap to establish, manage, and tenure, culture) are protected. Bank investment in maintain. In some regions more substantial social forestry programs increased from about 15 interventions will be required. Programs that percent of total forestry lending during 1967-80 target only a few households or villages cannot to more than 65 percent during 1980-92. prevent soil erosion. In China, for example, the Project evaluations indicate that Bank- government started research to improve erosion supported investments in such programs are 46 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community economically viable and environmentally sus- populations. Outmigration to towns and other tainable (the number of trees planted and main- rural areas is therefore necessary, as are programs tained were higher than in reforestation sites in to assist emigrants. Where migration is limited, more than half of thirty-five projects reviewed employment programs should be created. since 1985). While there are no direct measure- Employment problems are particularly acute for ments of changes in the income of resident pop- women during the off-season, since they can ulations derived from forestry, these programs rarely migrate in search of jobs. Employment provide effective mechanisms to distribute bene- problems are also severe during drought years. fits from the forestry project. Three types of built- Employment schemes raise income and improve in mechanisms for income flows to local residents nutrition. They also maintain and create rural have been used since 1985. The first income-shar- assets such as roads, irrigation, and soil conser- ing mechanism is investment bonuses in the form vation and forestation programs. of cash incentives for private tree planting in Since poor people are willing to work for low smallholders' woodlots. In Malawi (1986) these wages, public employment programs can offer bonuses included credit for fertilizers and other wages that screen out the nonpoor so that inputs. More farms planted crops underneath resources can be used more effectively In two tree shades, thereby expanding household large rural schemes in South Asia-the income. Maharashtra Employment Guarantee Scheme in Collective benefits are a second type of India and the Food for Work Program in income-sharing program. In these programs, dis- Bangladesh-the share of participants with trict and community groups retain a percentage incomes below the poverty line was at least 90 of income from government-financed forest plan- percent in the early 1980s. These schemes have tations and reforestation areas. These proved attracted people who are often excluded from worthwhile in China (1990) and Mexico (1990), other programs, such as women and members of where forests were maintained by local groups. scheduled castes in Maharashtra. Many countries Incomes also improved from sale of nontimber in Latin America and Sub-Saharan Africa admin- forest products. ister similar programs. The third mechanism, agroforestry zoning, The Employment Guarantee Scheme began in supports farm forestry around production and the early 1970s. Its provides employment on protection forests (such as in Algeria in 1992, request, within fifteen days, no more than five C6te d'Ivoire in 1990, India in 1985, Nepal in kilometers from the participant's home village. 1991, and Nigeria in 1987). Beneficiaries are An unemployment allowance is paid when this is given access to the forests. The sale of forest not possible. About three-quarters of the pro- products supplements agroforestry incomes. In gram's budget is spent on wages. The current C6te d'Ivoire (1990) agroforestry zones also scheme provides guaranteed employment year- serve as buffers for nearby large-scale forest round. Political commitment to the program is plantations. Local residents earn additional firm, and the effect on employment is significant. income as hired laborers in the plantation. A sim- In 1984-94, 180 million person-days of employ- ilar approach was used in contractual watershed ment were provided, representing 3 percent of management in Algeria (1992). Assemblies of total rural employment. The Government of local groups and municipal officials are paid by India is generalizing the approach to 1,752 poor the government to protect the area from migrant rural blocks. encroachment. This protection fee is collectively Employment programs have also been effec- managed by assemblies that finance marketing tive in Sub-Saharan Africa, as recently discussed of nontimber forest products such as gum, fruits, in an IFPRI (International Food Policy Research and fuelwood. Institute) study by von Braun, Teklu, and Webb Despite these agricultural and forestry (1992). The Bank has intensified assistance to resource management measures, economic activ- communities to improve their critical natural ities in resource-poor areas are typically insuffi- resource management issues. The Bank is also cient to provide employment for growing stepping up assistance in the design and finance Specific Actions and Programs 47 of employment programs, where income genera- countries. And the Bank has frequently been dis- tion potential is poor. couraged by the political sensitivities of various governments from involvement in land reform. Land Reform Market-assisted land reform is a more promis- ing approach than using land reform agencies The specific programs discussed so far help the and relying on expropriation. Small farmers or poor gain income and obtain food. They do not, farmer groups would be provided with a partial however, address the poor's lack of productive grant and credit to buy the land of their choosing. assets. The rural poor comprise the largest seg- Since land prices reflect nonfarming benefits or ment of the poor. The key for most of the rural specific privileges, such as tax advantages or poor is stabilizing and sustaining income by credit subsidies commonly acquired by large owning land (under community-based or landowners, the price of land usually exceeds the freehold tenure). Land allows the poor to pro- capitalized value of (unsubsidized) agricultural duce food directly, or gives them the rental profits. Beneficiaries are therefore unable to income to buy it. It also improves their access to finance land purchases with credit alone, and a credit and integrates them into broader social substantial grant is necessary. To transfer land to networks-crucial during droughts and other poor beneficiaries in a market-assisted land emergencies. Transferring land from large hold- reform of commercial farms, four conditions ings to small farmers usually increases produc- have to be met: tion and rural growth. In low-wage countries, 1. The privileges and distortions that drive the large farms tend to have lower productivity than price of land over the capitalized value of small farms and to employ proportionally fewer farm profits must be curtailed to reduce the workers. Land reform can improve the efficiency cost of land reform and to prevent subsequent of agricultural production and help employ reaggregation of land. more people. 2. Severe macroeconomic instability must be In economies like Iran, Korea, and Taiwan eliminated so that land will no longer have (China), land reform has substantially improved value as an inflation hedge. A long-term mort- agricultural growth and poverty alleviation. gage market must emerge in which beneficia- Successful land reforms were precipitated by ries can finance a part of the land purchase. social revolutions or other major political 3. The poor must be given a partial grant so they upheavals, such as decolonization. Failure to can pay for any land price premium over the reform highly dualistic systems of land owner- capitalized value of realizable farm profits ship has often resulted in peasant uprisings and and still have a sufficiently low debt-equity long civil wars, as in El Salvador, Guatemala, ratio to engage in own-account farming with- Mozambique, and Nicaragua. In many countries out threat of bankruptcy. that have attempted land reform, however, the 4. Beneficiaries must include more than the few attempts have failed or been very costly. Those remaining commercial farm workers to attempts were government-administered, with achieve a sufficient density of family labor on public agencies buying and redistributing the which the superiority of family farms is based. land. Such programs encountered formidable These preconditions for successful land redis- administrative and other problems. tribution are formidable, and warrant caution in For many years the Bank has recognized the planning new programs. In countries with highly benefits of land reform. As discussed in the 1975 uneven land distribution, however, the potential Land Reform sector policy paper, the Bank has benefits in terms of employment-intensive agri- always been ready to help countries implement cultural growth, social stability, and an improved land reform (World Bank 1975). Bank assistance climate for private investment may be so power- to Kenyan land reform in the early 1960s was the ful that market-assisted land reform should be largest such effort. But the Bank has been reluc- attempted. The Bank has recently reemphasized tant to become more involved because of the poor its commitment to land reform, and will actively performance of land reform agencies in many encourage countries with highly unequal land 48 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community distribution to design market-assisted programs. Technology and World Food Supply South Africa is a country where land reform is urgently needed (box 18). The diffusion of agricultural technology has resulted in steadily falling food prices and Agricultural Research and Extension increasing supply, reducing the possibility of Malthus's prediction of mass starvation. (Thomas The Bank has always emphasized technology as Malthus predicted in the 19th century that the a primary means for agricultural growth. The growth in food supply would be unable to match Bank advocates extension for smallholders of the number of people needing it, and the poor land, supports national agricultural research, and would be unable to afford food at the sharply ris- has been residual grantor to the Consultative ing prices. So far Malthus's theory has not come Group for International Agricultural Research true because technology has improved faster (CGIAR) for more than twenty years. than food demand.) Even in developing coun- Box 18. Land reform options in South Africa In addition to being an emotive reform (legal issues, agricultural forces participants to use some political issue in South Africa, the policy, technical support systems). of their own resources in order skewed distribution of land-86 A politically representative advi- to gain access to land will help in percent of agricultural land is sory group was formed to provide self-selection of participants and owned by large-scale, white farm- advice on the work program in the encourage productive land use. ers-impairs the efficiency of agri- hope that the group would eventu- * The grant elements of the pro- cultural production. A successful ally accept ownership of the pro- gram are essential to redistribut- land reform program is essential to gram. When the South African ing assets and ensuring that increased growth and employment teams completed their reports, a beneficiaries emerge from the and a smooth transition to majority Bank team drafted a synthesis program with a net increase in ruleinSouthAfrica.ToassistSouth report. This report-along with their asset position and a low Africans in a meaningful analysis each of the reports prepared by the debt-asset ratio as a means of of land reform options, the Bank South African teams-was pre- ensuring the viability and sus- summarized international experi- sented at a workshop on land tainability of their enterprises. ence with land reform. This sum- reform options in South Africa. * In addition to addressing the mary complemented an analysis of The main themes that emerged fundamental issue of social jus- local circumstances by South from this workshop were: tice, the program is likely to sig- African experts aimed at develop- *A market-assisted land redis- nificantly increase net rural ing policy options. tribution program is likely to employment. The process began with a work- perform better than one admin- *A redistribution program will shop for South Africans that iden- istered and operated by the not be able to provide land for tified the lessons of international public sector. everyone-the program will experience in several areas of agri- *The program must be under- need to be complemented by a culture (land reform, resettlement, taken in a reformed legal frame- rural safety net and by programs pricing, and marketing) but work in which rights of for urban groups. refrained from applying these individuals, of groups, and of *The program will also be a lessons to the South African case. women are fully protected. vehicle for supporting a wide From the workshop emerged an *The role of the public sector is range of land-use activities, interest in exploring the implica- ensuring adequate supplies of including trading and small- tions of these lessons for South land in the market and monitor- scale enterprises. Africa. Given the difficulty of iden- ing the overall operation of the The study team estimated that tifying a local institution that program. the program would create more would be widely regarded as neu- *Criteria for participation are than one million rural livelihoods tral or objective, the Bank agreed to necessary and must be dis- at relatively low unit cost and, at design a work program for devel- cussed and agreed in advance. the same time, address issues of oping a set of land reform options. * Welfare objectives can be met by social justice. The work program relied on teams including a grant component in of South African experts to prepare the program. Source: Land and Agriculture reports on various aspects of land *A matching grant scheme that Policy Center 1993. Specific Actions and Programs 49 Box 19. A food supply crisis is unlikely-at least for now Modern day Malthusians warn Since diets have improved tries was nearly double that of that Malthus will ultimately be greatly in many countries, total Westem European consumers right: the world will run out of food demand is growing more despite much lower incomes. This food. But for now there is little slowly than in the past. The is changing, and consumption lev- evidence to support this warning. world population growth rate is els for cereals are expected to Since 1980 world cereal yields also projected to slow during the decline by 30 to 40 percent. This have grown by 2.25 percent a next several decades-to 1.0 per- could free substantial supplies for year; world population has cent a year by 2025. This would export. Economic reforms could grown by 1.75 percent a year. further reduce the growth of food also stimulate production, causing Cereals account for 60 percent of demand and weaken the even larger exports. These changes total calories in developing coun- Malthus's warning. could cause the region to shift tries and about half of total crop- Other factors are also changing, from importing 20 percent of the land used for agriculture. The especially in the formerly centrally world's cereals trade to exporting rapid growth in crop yields has planned countries of Eastern an equal amount in span of twenty caused cereals prices to fall and Europe and the former Soviet years. area planted to cereals to decline. Union. These countries were large These calculations do not mean A number of factors suggest that food importers during the 1970s that agricultural technology can be current surplus production and 1980s, but they may become neglected. Only rapid technical capacity will not be needed soon. food exporters within a decade. change and sustainable resource One of the most important The big change will likely be in use can ensure the needed longer- changes in the world food situa- consumption. Because of highly term food supply, and continued, tion is the changing patterns of subsidized prices, per capita con- if not enhanced, research effort is demand in developing countries. sumption of cereals in these coun- essential. tries, the green revolution in the 1960s and 1970s system as a key element in a long-term world- led to high growth rates of food production. At wide food supply strategy. the same time, population growth worldwide has The Bank itself supports international agri- slowed markedly (with notable exceptions). A cultural research with $40 million a year in grants recent Bank study calculates that, with current and chairs the CGIAR. Even though these grants food demand and technology trends, a world- are an important complement to those made by wide food supply crisis is unlikely in the next two other donors, the total amounts are too small rel- decades (box 19). Even with the catastrophe post- ative to the role of the CGIAR system in preserv- poned for two decades, there are hungry people ing germplasm resources, researching ways to all over the world. Moreover, short-term famines increase yields, addressing environmental remain a problem for many countries. This rein- resource management issues, and promoting forces the view that individuals'lack of income to modern biological research for the benefit of purchase food or resources to produce food are at developing countries. Consequently, the CGLAR the root of the hunger problem, not aggregate system is facing a funding crisis. This crisis will supplies. require major attention by the Bank and other Favorable current trends in worldwide food donors. supply do not warrant reduced support for gen- eration and dissemination of technology. Opti- Access of the Poor to Improved Technology mism is only warranted because technology will likely increase at a fast pace. To slow down tech- Technology affects both worldwide aggregate nology development would sharply increase the food supply and the world's poor populations, risk of food supply problems fifteen years from who are mainly rural. Improved technology is now. The basic breakthroughs required to achieve critical to poor countries. Their ability to derive food production growth generally take one or income from their agricultural resources depends two decades to move from discovery to wide- on their relative international competitiveness in spread use. For this reason alone the Bank advo- agricultural production. Competitiveness in agri- cates greater donor support for the CGIAR culture generates export revenues, which can be 50 77e World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community used to buy food imports. Trading agricultural cultural extension programs can successfully tar- commodities for food generally helps the poor by get smaller farmers, many of whom are women, enhancing rural incomes and the ability to buy enabling them to remain competitive and partic- food. An example is the Dominican Republic, ipate in agricultural growth. Much of Chinese which exports high-value tropical fruits and veg- agriculture has grown at 7 to 10 percent a year etables and imports staple foods. over the past 15 years, when average farm sizes About 80 percent of aggregate agricultural were less than half a hectare. This is partly due to research expenditure in developing countries is excellent extension services provided by former concentrated in twenty relatively large Asian and communes and partly due to excellent technical Latin American countries (50 percent in just four and human resource bases. countries). Developing countries more than dou- The Bank is the largest donor for agricultural bled their real aggregate expenditure on agricul- extension in developing countries. During the tural research between 1970 and 1980, but past three decades the Bank lent about $2.5 billion expenditure levels have stagnated since then. The for extension, more than all other donors com- World Bank is lending to many countries in sup- bined. The Bank supports extension through port of their national agricultural research sys- investment projects, either free-standing or as tems (NARS). The performance of many of these extension components in agricultural projects. systems has deteriorated in recent years. This is Except for a few commodity-based extension pro- partly because they have suffered from inade- jects, virtually all the Bank's support for free- quate compensation to retain their best staff and standing extension projects uses the training and partly because they have cut operating budgets. visit (T&V) model. This model emphasizes regu- Even with the limitations of existing support sys- lar, face-to-face contact between well-trained agri- tems, public national agricultural research pro- cultural professionals and farmers. grams can generate high economic payoffs. This Such programs have been highly successful. suggests many countries are underinvesting in For example, an evaluation of a program in agricultural research. Burkina Faso found that the introduction of T&V To be effective, agricultural research must be programs has increased the adoption of complemented with effective agricultural exten- improved practices. Farmers belonging to T&V sion and efficient marketing and input supply contact groups have seen their crop yields services. In most countries research and exten- increase by 25 to 30 percent. The rates of return sion systems need to be better-integrated, includ- estimated for investments in the expansion of the ing farming system research and research current T&V extension system range from 86 to outreach using participatory techniques. In some 187 percent in Burkina Faso. The evaluation also countries there are now private practitioners of showed that the average annual expenditure on agricultural extension. Beyond salesmen of extension per farm family declined by almost 30 seeds, fertilizers, chemicals, and machinery, these percent after T&V programs were adopted at the practitioners are active in the broader field of national level. The Bank will continue to provide farm management, often in intensive livestock or strong support to small farmer extension. horticultural activities. Public extension should not duplicate the work of such private practi- World Bank Actions tioners. Instead, public extension should focus on small-farmer development and natural resources The World Bank's strategy to reduce poverty and management. Priority should be given to poor hunger has been developed and refined over farmers. Where regular, face-to-face contact is the many years. This strategy forms the core of the delivery method, sound managerial principles Bank's development mission. Country poverty must be applied to effectively train and supervise assessments provide an integrating framework to a scattered workforce and make it responsive to plan and implement Bank support to reduce the farming clientele. poverty and hunger in each country. Programs in While rich and middle-income farmers will support of human resource development are usually benefit first from new technology, agri- becoming ever more important in the Bank's Specific Actions and Programs 51 lending portfolio. Moreover, food security pro- the malnourished through nutrition monitoring jects and operations focusing directly on nutri- and provide synergistic primary health care, food tion also are increasing. Experience on making supplementation, and health and nutrition edu- Bank-supported programs complementary with cation. Such programs greatly reduce infant mor- drought avoidance and famine prevention is tality and hunger and improve the health and accumulating, especially in Sub-Saharan Africa. nutrition of poor populations. Shifting fiscal Experience on strengthening the income-earning resources from poorly-targeted food subsidy pro- opportunities of the poor is being widely dis- grams to the provision of food entitlements for seminated and integrated into various sectoral the entire families of the children or mothers projects. Collaboration with development NGOs identified as malnourished in such programs is a and community groups in poverty projects is high priority. These findings require greater dis- expanding. semination inside and outside the Bank. The While many of the social and human financing of food for primary health care and the resource development programs assisted by the expansion of services that allow more wide- Bank have become better-targeted to the poor, spread food availability can come from the real- there has been little progress over the past location of food subsidies to these combined decade in reorienting social and human resource programs. In the Bank the operational responsi- development expenditures in the developing bility for food production and subsidy issues is world from the better-off to the poor and hun- usually in the agriculture divisions, while that for gry. This is despite the considerable emphasis health and nutrition is usually in the human placed on this issue by the Bank in poverty pro- resource divisions. Close cooperation between files, public expenditure reviews, policy dia- the divisions and disciplines is required to logue, and loan negotiations. Progress will achieve the integration of these programs in a depend partly on greater public awareness of cost-effective and fiscally responsible way. options available to governments in the alloca- The Bank's own efforts to assist governments tion of public expenditure. in the provision of microcredit have been less suc- The poverty profiles prepared by Bank staff, cessful than some NGO or community-group- with local participation where possible, provide managed programs such as the Grameen Bank. an important opportunity to analyze hunger in its The Bank needs to learn from the successful pro- three forms, and to consider the policies and pro- grams which approaches can be effectively grams required to overcome hunger. Training exploited in Bank-assisted projects, and explore task managers, staff, and consultants involved in ways to extend wholesale credit in collaboration poverty assessment work will help improve the with such intermediaries. quality of their work. The pursuit of market-assisted land reform in World Development Report 1993: Investing in countries with highly unequal land distribution Health, identified a number of effective, low-cost is both an opportunity and a challenge. Land actions which in many countries could sharply reform is never easy. Using a market-based reduce hunger even in the absence of rapid approach presents issues which must be growth. The challenge now is for the Bank to addressed according to the specific circum- work with countries and other agencies to imple- stances of each country ment these actions as fully and as promptly as The other approaches to strengthening the possible. Efforts will continue to disseminate the income-earnings capacities of the poor have been report's findings on micronutrient supplementa- discussed and widely disseminated in World tion, vaccination, and prevention of intestinal Development Reports, the Poverty Reduction parasites both inside and outside the Bank. Handbook, and other Bank plans. They are associ- Implementation of these programs, and possible ated with specific sectors and best addressed financing within Bank projects in member coun- within sector lending operations. tries, merit high priority. The funding pressures on the CGIAR are Experience inside and outside the Bank has increasing; indeed they are assuming crisis pro- shown the effectiveness of programs that identify portions as some major donors have reduced 52 The World Bank's Strategyfor Reducing Poverty and Hunger: A Report to the Development Community support. The Bank will take the lead in restruc- participating in a few pilot tripartite in-country turing the system and devising a new funding consultations as a forum to exchange views strategy. among governments, NGOs, and the Bank on Finally, there is a shared desire to visibly specific country policies and actions to reduce reduce hunger in poor countries. The Bank is hunger. Notes 1. See Overcoming Global Hunger: Proceedings of a Con- 4. Goals were to achieve, by 2000, virtual elimination ference on Actions to Reduce Hunger Worldwide. ESD Pro- of vitamin A and iodine deficiencies, a reduction of ane- ceedings Series no. 3. Ismail Serageldin and Pierre mia among women of reproductive age by one-third of Landell-Mills, eds. (Washington, D.C.: World Bank, 1994). 1990 levels, and a reduction of severe and moderate 2. Terminological confusions abound. "Hunger" as malnutrition among children under age 5 by one-half of used in this report is sometimes called "malnutrition," 1990 levels. but the latter term is often limited to chronic undernu- 5. The countries are Argentina, Bolivia, Brazil, Chile, trition and specific nutrient deficiencies. Costa Rica, Dominican Republic, El Salvador, Jamaica, 3. Community groups include neighborhood commit- and Venezuela. tees, women's groups, farmers associations, and the like. 53 Selected References More complete reference to primary sources can be Grosh, Margaret. 1990. Social Spending in Latin America: found in the World Bank publications, Poverty and Hun- The Story of the 1980s. World Bank Discussion Paper ger: Issues and Options for Food Security in Developing no. 106. Washington, D.C. Countries and World Development Report 1990 and 1993. Husain, Ishrat. 1993. "Poverty and Structural Adjust- ment: The African Case." Working Paper 9. Human Alderman, Harold, and Joachim von Braun. 1984. The Resources Development and Operations Policy Depart- Effects of the Egyptian Food and Subsidy System on Income ment, World Bank, Washington, D.C. Redistribution and Consumption. Research Report 45. Hyden, Goran, and Shlomo Reutlinger. 1992. "Foreign Washington, D.C.: International Food Policy Research Aid in a Period of Democratization: The Case of Poli- Institute. tically Autonomous Food Funds." World Development Alexander, Nancy, and others. 1993. "Perspectives and 20(9): 1253-60. Proposed Actions in Follow-up to the World Bank Jamaica Statistical Institute and World Bank. 1988. "Liv- Overcoming Global Hunger Conference." Washington, ing Conditions Survey, Jamaica." Kingston. Processed. D.C.: Bread for the World. . 1989. "Survey of Living Conditions, Jamaica." Berg, Alan. 1987. Malnutrition: What Can Be Done? Les- Kingston. Processed. sonsfrom World Bank Experience. Baltimore, Md.: Johns International Food Policy Research Institute (IFPRI). Hopkins University Press. 1993. "Ending Hunger Soon: Concepts for Policy Binswanger, Hans B., and Joachim von Braun. 1991. Actions." Background Paper. Draft. "Technological Change and Commercialization in Agri- Kahnert, Fritz. 1992. Improving Urban Employment and culture: The Effect on the Poor." World Bank Research Labor Productivity. World Bank Discussion Paper no. Observer 6 Uanuary): 57-80. 10, Washington, D.C. Canadian International Development Agency (CIDA). Khandker, Shahidur R., Baqui Khalily, and Zahed Khan. 1990. The Water Utilization Project: A Case Study on a 1993. "Grameen Bank: What Do We Know?" World Water and Health Education Project. Ottawa. Bank, Washington, D.C. Draft. Del Rosso, Joy. 1992. Investing in Nutrition with World Lambert, Sylvie, Hartmut Schneider, and Akiko Suwa. Bank Assistance. Washington, D.C.: World Bank. 1991. "Adjustment and Equity in C6te d'lvoire: De Soto, Hernando. 1989. The Other Path: The Invisible 1980-1986." World Development 19 (11): 1563-76. Revolution in the Third World. New York: Harper and Land and Agricultural Policy Centre. 1993. Options for Row. Reform and Rural Restructuring in South Africa. Rev. Dorosh, Paul A., and David E. Sahn. 1993. "A General version of paper presented at Land Redistribution Equilibrium Analysis of the Effect of Macroeconomic Options Conference on October 12-15, 1993, Adjustment on Poverty in Africa." Cornell Food and Johannesburg, South Africa. Nutrition Policy Program Working Paper 39. Ithaca, Lavy, Victor. 1990. "Does Food Aid Depress Food Pro- N.Y.: Cornell University. duction? The Disincentive Dilemma in the African Griffin, Charles. 1992. Health Care in Asia: A Comparative Context." Policy Research Working Paper 460. Policy Study of Cost and Financing. World Bank Regional and Research Advisory Staff, Policy Research Dissemina- Sectoral Studies. Washington, D.C. tion Center, World Bank, Washington, D.C. 55 56 References Linn, Johannes. 1983. Cities in the Developing World: Teklu, Tesfaye. 1993. Labor-Intensive Public Works: The Policies for Their Equitable and Efficient Growth. New Experience of Botswana and Tanzania. Washington, D.C.: York: Oxford University Press. International Food Policy Research Institute. Mayfield, Malcolm R. 1992. "The Effects of Policy Reforms Tomkins, Andrew, and Fiona Watson. 1989. Malnutrition on the Performance of Agriculture in Tanzania." and Infection. Administrative Committee on Coordina- Consultant's report. Agriculture and Environment tion/Subcommittee on Nutrition. Nutrition Policy Division, Eastern Africa Department, World Bank, Discussion Paper 5. New York: United Nations. Washington, D.C. United Nations, Administrative Committee on Coordi- Mitchell, Donald, and Merlinda Ingco. Forthcoming nation/Subcommittee on Nutrition. 1992. Second 1995. Waiting for Malthus: The World Food Outlook. Report on the World Nutrition Situation, vol. 1, "Global London: Cambridge University Press. and Regional Results." Musgrove, Philip. 1991. "Feeding Latin America's Chil- . 1993. Second Report on the World Nutrition Situa- dren: An Analytical Survey of Food Programs." World tion, vol. II, "Country Trends Methods and Statistics." Bank, Latin America and the Caribbean Technical van den Brink, Rogier. 1993. "A Review of Agricultural Department, Washington, D.C. Statistics of Mainland Tanzania." Background Paper. NGO Statement. 1993. "The World Bank Must Adopt Agriculture and Environment Division, Eastern Africa Fundamental Changes to Achieve Goals of Reducing Department, World Bank, Washington, D.C. Poverty and Hunger." Available at InterAction, von Braun, Joachim, Tesfaye Teklu, and Patrick Webb. Washington, D.C. 1992. "Labour-Intensive Public Works for Food Secu- Nurse, Keith. 1994. "Comment on the World Bank rity in Africa: Past Experience and Future Potential." Policy Paper on Implementing the World Bank's Stra- Reprinted from International Labor Review 131 (1), tegy for Reducing Poverty and Hunger." Southern Reprint 247. Washington, D.C.: International Food NGO representatives of the Steering Committee on Policy Research Institute. World Hunger and Poverty. Draft. Walters, Harry. 1994. "Overcoming Global Hunger: An Ravaillon, Martin. 1990. "Rural Welfare Effects of Food Issues Paper." In Overcoming Global Hunger: Proceedings Price Changes under Induced Wage Responses: Theory of a Conference on Actions to Reduce Hunger Worldwide. and Evidence for Bangladesh." Oxford Economic Papers ESD Proceedings Series no. 3. Ed. Ismail Serageldin 42: 574-85. and Pierre Landell-Mills. 123-42. Washington, D.C.: Reutlinger, Shlomo. 1977. "Malnutrition: A Poverty or a World Bank. Food Problem?" World Development 5(8): 715-24. World Bank. 1975. Land Reform. Sector Policy Paper. Reutlinger, Shlomo, and Anne Marie del Castillo. 1994. Washington, D.C. "Addressing Hunger: A Historical Perspective of . 1986. Poverty and Hunger: Issues and Options for International Initiatives." In Overcoming Global Hunger: Food Security in Developing Countries. A World Bank Proceedings of a Conference on Actions to Reduce Hunger Policy Study. Washington, D.C. Worldwide. ESD Proceedings Series no. 3. Ed. Ismail . 1988. The Challenge of Hunger in Africa: A Call to Serageldin and Pierre Landell-Mills. 143-158. Action. Washington, D.C. Washington, D.C.: World Bank. . 1990. World Development Report 1990: Poverty. Salmen, Lawrence. 1987. Listen to the People: Participant- New York: Oxford University Press Observer Evaluation of Development Projects. New York: . 1992a. Poverty Reduction Handbook. Washington, Oxford University Press. D.C. . 1992. "Beneficiary Assessment: An Approach . 1992b. World Bank Structural and Sectoral Adjust- Described." Working Paper 1. Africa Technical Depart- ment Operations: The Second OED Review. Operations ment, World Bank, Washington, D.C. Evaluations Department. Washington, D.C. Schiff, Maurice, and Alberto Valdes. 1992. The Plunder- . 1993a. "Agricultural Sector Review Paper." Agri- ing of Agriculture in Developing Countries. Washington, culture and Natural Resources Department. D.C.: World Bank. Washington, D.C. Sen, Amartya. 1981. Poverty and Famines: An Essay on . 1993b. Implementing the World Bank's Strategy to Entitlement and Deprivation. Oxford: Clarendon Press. Reduce Poverty: Progress and Challenges. Washington, Serageldin, Ismail, and Pierre Landell-Mills. 1994. Over- D.C. coming Global Hunger: Proceedings of a Conference on . 1993c. World Development Report 1993: Investing Actions to Reduce Hunger Worldwide. Environmentally in Health. New York: Oxford University Press. Sustainable Development Proceedings Series no. 3. . 1994. Adjustment in Africa: Reform, Results, and the Washington, D.C.: World Bank. Road Ahead. Policy Research Report 2. Washington, D.C. Tan, Jee-Peng, and Alain Mingat. 1992. Education in Zeller, Manfred, Gertrud Schrieder, Joachim von Braun, Asia: A Comparative Study of Cost and Financing. World and Franz Heidues. 1993. "Credit for the Rural Poor Bank Regional and Sectoral Studies, Washington, in Sub-Saharan Africa." International Food Policy D.C. Research Institute, Washington, D.C. '|'I ' V 'I P. ii \\V 'ij ,~ ~~~n /. I-@,,,- ij* i '' '-''1' *,,l-, cm4i (.,1z.',,s' ,:-, "~~~~~jl :l'r . i- ' ,,,,c:;,,7., I=' . i 1 : -§( .. ,i 1 *'. i.- !' S 'r *, l 1 qs 1 > .- i - . i; .13174 DEVf * 9 -1-<^1/;-9-; 1-l - { I I >, !two (-821.-31jl, .. { t! t I, i =_ 1^; 1 | r. . ^ ! .- i : +RL BANK'SSTRATEG 0 , ,, . 0 I (,.Sk ('4 siS ? 1 0 ~~~~~1111111111 iC,.>,V1-r~ ~ ~~~~~~~~~ 4000163 I *i!, i , 'nfl-5,