1 4030 1403o Accelerated Development in Sub-Saharan Africa An Agenda for Action The World Bank WASHINGTON, D.C. Copyright ©) 1981 by 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 This report was written by the African Strategy Review Group, consisting of Elliot Berg, coordinator, K. Y. Amoako, Rolf Gusten, Jacob Meerman, and Gene Tidrick, with the assistance of many staff members of the World Bank. The Report draws heavily on the ideas of many colleagues inside and outside the World Bank, and also reflects the views expressed by scholars and officials in Africa and in donor countries. The judgments expressed do not neces- sarily reflect the views of the Bank's Board of Executive Directors or of the Governments that they represent. Library of Congress Cataloging in Publication Data Main entry under title: Accelerated development in sub-Saharan Africa. Bibliography: p. 1. Africa, Sub-Saharan-Economic conditions- 1960- 2. Economic assistance-Africa, Sub-Saharan. I. World Bank. HC800.A54 338.967 81-16828 AACR2 Accelerated Development in Sub-Saharan Africa IBRD 16041 200 200 40 SEPTEMBER 1981 NVorth 0ceo an 400 40' FOR.EROEEYT TROPIC OP 200 ~ ~ ~ ~ 1000R2000IMl 20 200 South A f/on tA U Ico CAP'~~~~~~~~~~~~~~~~~~~~~~ AFRICA I I I I I 0 1000 2000 3000 KM. This mnap has bcean prepared bY the World Bank's staff eaclusivreIy for the con lenience of the readers of the report to r.'hjch it is atUached. The denominations used and the boundaries shosem on this mnap do not innply, on the part of the World Bank and its affiiates, any judgment on the legal status of any terrio.y or any endorsemnent or acceptance of such boundaries. 200 0° 20° 40° Sot ta tc co AA This report has been prepared by the staff of is not a recommendation which derives from the World Bank. It highlights the severitv and any preconceived philosophy of ownership. complexity of the problems facing manv of It derives from considerations of efficiency, the couLntries of Sub-Saharan Africa in their which suggest that governments can more efforts to raise the living standards of their effectively achieve their social and develop- people. ment goals by reducing the widespread ad- The report accepts the long-term objectives ministrative overcommitment of the public of Africaln development as expressed bv the sector and by developing and relying more i [eads of State of the Organization of African on the managerial capacities of private indi- Unitv in the LLu,'os Plan of Actiotn. It emphasizes viduals and firms, which can respond to local that if thee objectives for the year 2000 are needs and conditions, particularly in small- to be achlieved, actions must be taken to re- scale industry, marketing, and service activ- verse the stagnation and possible decline of ities. per capita incomes which are projected for However, the report offers no general pre- the l98Qs. scriptions. The countries of Africa are too The central theme of the report is that more diverse-politicallv, culturally, philosophi- efficienit use of scarce resources-human and cally-to attempt to define a single strategy. capital, nmanagerial and technical, domestic Programs of action must be formulated bv andi foreign-is essential for improving eco- each countrv, and these programs must in- nomic conditions in most African countries. clude the external financial and technical as- From this flow a number of suggestions for sistance which will be required to support the inmprovement of incentives and institu- them. To encourage the reviews and discus- t0onal supLport for production, particularlv in sions which are prerequisites to the formu- agriculture. The public sector wvill have to lation of these country strategies, we have meet the extenisive needs for infrastructure, decided to publish this report, thereby facil- educaLtion,1 health, and other services. The itating widespread distribution. efficient provision of these services will place We hope that the results of these discus- enormous demands on administrative and sions will be a strengthened commitment to managerial capacity-the scarcest resource in a joint effort by African governments, bi- all counitries. It is in this context that the re- lateral donors, and international institutions potrt suggests that African govemments should to accelerate development in Sub-Saharan niot onl\y examine ways in which the public Africa. sector oro'anizations can be operated more etficienrtly, but should also examine the pos- sibility of placing greater reliance on the pri- vate sector. The report emphasizes that this A. W. Clausen v Definitions viii I Introduction The Setting 1 The Present Economic Crisis 2 Sources of Lagging Growth 4 News Priorities and Adjustments in Policy 4 Long-run Strategy Implications 6 Donor Policies 7 2 Basic Constraints 9 Obstacles to Growth: The Postcolonial Situation 9 Progress since 1960 14 The Persistence of Special Constraints 15 3 External Factors 17 Balan ce-of-Payments Deterioration 17 Terms of Trade 18 Export Growth 19 Prospects for the 1980s 21 4 Policy and Administrative Framework 24 Trade and Exchange-Rate Policy 24 Economic Decisionmaking 31 Organization and Management 35 The Size of Government 40 5 Policies and Priorities in Agriculture 45 Trends in Agricultural Development, 1960-S80 45 Action for Rural Development 49 Improved Incentive Structures 52 Refomi of Price, Marketing, and Input Supply Policies 61 Agricultural Research and Extension 69 irrigation 76 6 Human Resources 81 Education 81 Training 85 Health 87 7 Other Productive Sectors 91 Industry 91 Nonfuel Minerals 97 Energy 99 Transport and Communications 105 8 Longer-Term Issues 112 Population 112 Urban Growth 114 Resource Planning 117 Regionalism 118 9 External Assistance in the 1980s 121 The Need for Increased Aid 121 The Effect of Higher Aid and Policy Reform on Economic Prospects 122 Donor Policies 124 Conclusion 132 Statistical Annex_ 134 Contents 137 Kev 141 Introdlction to the Annex 142 Tables 143-86 Technical Notes 187 Bibliograph. of Data Sources 197 NMap frontispiece Text tables I I Sub-Saharan Africa and the World: Basic Data 3 2. I Expatriate Employment as a Percentage of Total Emplovment of Trained Manpower in Early iManpower Survevs 9 2.2 Annual Average Growvth Rates of Enrollments in Educational InstitLtions by World Developing Regions, 1960-76 14 3. 1 Oil-importing African Countries: Current Account Deficit and its Financing, 1970-78 17 3.2 Tenns of Trade and Export Trends, Selected African Countries, by Export Category 18 3.3 Oil Imiports in relation to Exports and GDP in Eight Oil-importing African Countries 18 3.4 Africa's Share of Nonfuel Exports 19 3.5 Africa's Share of World and Developing Countrv Exports of Selected Commodities 21 3.6 Projected Price and Volume of World Trade in Selected Commodities 22 4.1 Public Sector Emplovment and Growth Rates, Selected Countries 41 4.2 The Growvth of Public Administration and Defense relative to GoT' 41 5.1 Growth of Agricultural Exports 46 5.2 Gromwth of Production of Selected Food Crops 47 5.3 Growth of Imports of Selected Agricultural Commodities, 1961-63 to 1977-79 48 .4 Food Aid and Commercial Imports of Cereals, 1975-79 49 5. Growth Rates of Agricultural Production, 1969-71 to 1977-79 50 5.6 Relative Frequency of Government and Private Sector Control in the Procurement and Distribution of Agricultural Inputs 60 vii 6.1 The Social Returns to Education in Africa 82 6.2 Costs of a Student-vear as a Percentage of GNP per Capita 82 6.3 Relative Teaching Costs and Student-Teacher Ratios, Public Primarv Schools 83 7.1 Proven Reserves of Oil and Oil Production 101 8.1 Population Projections for Eight African Countries and Sub-Saharan Africa 112 9.1 Growth of GNP per Person, 1960-90 122 9.2 Projected Perfomrance of Oil-importing African Countries, 1980-90 122 9.3 Net Disbursements of Official Development Assistance (ODA) to Africa 1980-90 123 9.4 Net Disbursements of ODA to Africa, 1975 (Actual), and 1985 and 1990 (Projected) 124 9.5 Actual and Projected Sub-Saharan African Debt-Service Ratios 129 Text boxes Chial;ter 4 Box A: Agricultural Exports from Tanzania and Ghana 26 Box B: An Example of the Costs of Excess Protection 28 Box C: Industrial Incentives in Zambia 29 Box D: Botswana's Planning and Budgeting System 34 Box E: Macroeconomic Indicators, Economic Management and Growth 36 Box F: Public Enterprise Contracts in Senegal 39 Chlapter D Box A: Kenvan Smallholder Agriculture 51 Box B: Northern Nigeria Agricultural Development Projects 53 Box C: Cotton-based Projects in Mali and Upper Volta 54 Box D: "Taxation" of Export Crops 56 Box E: Nigerian Food Imports 57 Box F: Arabica Coffee Marketing in Cameroon's Western Highlands 63 Box G: Measuring Comparative Advantage 65 Box H: lrivatizing Input Supply Systems: The Bangladesh Experience with Fertilizer Distribution 67 Box I: Controlling the Desert Locust in East Africa 72 Chap;ter 6 Box A: Ethiopia's Campaign against Illiteracy 84 Box B: Provision of Rural Water Supplies in Malawi 90 Ch1a;tcr - Box A: Industrial Growth in Malawi 92 Box B: Public Aid as a Catalyst for Private Capital: Petroleum in Mali 104 Box C: Building Local Capacities 107 Ch1a71ter 8 Box A: Two Approaches to Urban Housing 116 Definitions Ecoiaomiic and Ldeinographlic termns are defined in the technical notes to the Statistical Annex. Billion is 1,000 million. Tonis are metric tons (1,000 kilograms). Growthl rates are in real terms unless otherwise stated. Dollars are United States dollars. Synmbols used in the text tables are as fol- lowss: ..Not available; (.) Less than half the unit shown; and n.a. Not applicable. viii In the fall of 1979, the African Governors of of the Organization of African Unity held in the World Bank addressed a memorandum to April 1980. The Lagos Plan endorses objectives the Bank's President, expressing their alarm for the African states to achieve a more self- at the dim economic prospects for the nations reliant, more economically integrated Africa of Sub-Saharan Africa and asking thatthe Bank by the year 2000. This Report deals with short- prepare "a special paper on the economic to medium-term responses to Africa's current development problems of these countries" and economic difficulties. It focuses on how growth an appropriate program for helping them. This can be accelerated and how the resources to Report is the response to that request. achieve the longer-term objectives set by the The Report discusses the factors that ex- African governments can be generated, with plain slowv economic growth in Africa in the the support of the international community. recent past, analyzes policy changes and pro- Like the Lagos Plan, the Report recognizes that gram orientations needed to promote faster Africa has enormous economic potential which grow th, and concludes with a set of recom- awaits fuller development. mendations to donors, including the recom- mendation that aid to Africa should double in real terms to bring about renewed African Til S c ti71< development and growth in the 1980s. The SbSaharan Africa is a region of great di- Report's agenda for action is general; it in- versity. On the one hand, it encompasses dicates broad policy and program directions, overall priorities for action, and key areas for Nigeria, a country of 80 million people (al- donor attention. It does not-indeed could most one uarter of the total opulation of not-deal with specific programs or problems region), pro. dn soe 44 prnd t of of iniida conris regional output. On the other hand, it in- Theindividual Reportdraws. hvyn eBk cludes a number of small countries such as The Report draws heavily on the Bank's Tg n wzln.Teeaecutisrc experience in Africa and its extensive rela- inTogo and Swaziland. There are countries rich in oil and mineral resources2 as well as coun- tions with its African members as well as the tries that are entirely agricultural; countries work of other multilateral organizations and of the tropical rainforest and those of the bilateral development agencies. In particular, semiarid interior; landlocked countries and the Report builds on the Lagos Plan of Action' those fronting the ocean; six countries where the statement of developmentstrategyadopted agricultural output has grown by over 3 per- bv the African Chiefs of State at the meeting cnt urin the has an seven coun- cent a year during the 197(3s3 and seven coun- 1. Organization of African Unity, The Lagos Plan of Action tries where agricultural output has been for the Implementation of the Monrovia Strategy for the Economic Dcvezoptment of Africa, adopted by the Second Extraordinary 2. Botswana, Gabon, Guinea, Nigeria, Zaire, Zambia, and Assemblv of the OAL Heads of State and Government, De- Zimbabwe, among others. voted to Economic Mviatters (Lagos, Nigeria: April 28-29, 3. Cameroon, Ivory Coast, Kenya, Malawi, Rwanda, 1980). Swaziland. 1 unchanged or has actually declined during The Report is necessarily selective in its this period4; countries with a long-established analysis. Many important questions have been indigenous trading class (most of the nations treated elsewhere: problems of science and of West Africa) and those where such a group technology, for instance, are treated at length is smaller and of recent origin; countries which in the Lagos Plan of Action, while others, such are attempting socialist transformation and as higher rates of domestic saving, are likely those which are following the example of the to follow a resumption of growth rather than market economies. precede it. New investment opportunities and There is, nonetheless, considerable homo- growth are the focus of the Report. geneity within the region. African economies are for the most part small in economic terms, a result of low average incomes and small Ti 'rU--';!ti::>:u1iC Cihs populations. Of the 45 states in the region, 24 have fewer than five million people. Only During the past two decades economic de- Nigeria has a gross domestic product (GDP) velopment has been slow in most of the coun- greater than that of Hong Kong. African tries of Sub-Saharan Africa.5 When, in the economies are open; foreign trade typically mid-1970s, the world economy experienced accounts for about a quarter of GDP. They are inflation and recession, nowhere did the crisis specialized economies, most of them agricul- hit with greater impact than in this region. tural, dependent on the export of two or three The picture is not uniformly bleak. There primary commodities. Even in the mineral- are signs of progress throughout the conti- exporting countries, the bulk of the popula- nent. Vastly more Africans are in schools, and tion-rare1v less than 70 percent-works in most are living longer. Roads, ports, and new agriculture, and subsistence-oriented produc- cities have been built and new industries de- tion still accounts for half or more of total veloped. Technical and managerial positions, agricultural output. Only about 20 percent of formerly occupied by foreigners, are now held the population is urban, and modern wage by Africans. Of the 45 countries in the region, emplovment absorbs a very small proportion nine posted annual growth rates of over 2.5 of the labor force-in most countries less than percent per capita between 1960 and 1979 (see 10 percent. Table SA.1).6 In addition to these similarities of economic But for most African countries, and for a structure, other characteristics are common: majority of the African population, the record the scarcity of educated people, the domi- is grim and it is no exaggeration to talk of nance of land-extensive agricultural systems, crisis. Slow overall economic growth, slug- and an extreme ethnic diversity and conse- gish agricultural performance coupled with quent political fragility. All are new states, rapid rates of population increase, and bal- recently emerged from colonial rule, except ance-of-payments and fiscal crises-these are Liberia and Ethiopia. All are tropical, with the dramatic indicators of economic trouble. exception of Lesotho and Swaziland. In al- Between 1960 and 1979, per capita income most all, fertility is high and population is in 19 countries grew by less than 1 percent growing rapicdly-more rapidly than in any per year, while during the last decade, 15 other region of the world. countries recorded a negative rate of growth Finally, there is an extraordinary degree of of income per capita. And by the end of the similaritv throughout the region in the nature 1970s, economic crises were battering even of the policy problems that have arisen, such 5. All references to Africa in this Report are to Sub-Sa- as in rural development, trade, and indus- haran Africa unless otherwise specifically noted. See the trialization, and in the national responses to Introduction to the Statistical Annex for a listing of the them. countries included. 6. All tables preceded by the letters SA are located in the 4. Angola, Congo, Ghana, Mauritania, Mozambique, Togo, Statistical Annex. All other tables, such as 1.1, 2.1, 2.2, etc., Idganda. are in the text. 2 high-growth countries like Kenya, Malawi, and wiped out by a decline of similar proportions Isvorv Coast-where per capita GNP growth had in the 1970s. Consequently, Africa's share of averaged an annual 2.7 percent between 1960 the world market dwindled. As for food crops, and 1979-compelling them to design pro- while data are uncertain, they leave no doubt grams, supported by the Bank, to restructure about general tendencies. Total food produc- their economies. Output per person rose more tion rose by 1.5 percent per year in the 1970s, slowlv in Sub-Saharan Africa than in any other down from 2 percent in the previous decade. part of the World, particularly in the 1970s, But since population was rising rapidly-by and it rose more slowly in the 1970s than in an annual average of 2.5 percent in the 1960s the 1960s (see Table 1.1). and 2.7 percent in the 1970s-food produc- The tragedy of this slow growth in the Af- tion per person was stagnant in the first de- rican setting is that incomes are so low and cade and actually declined in the next. Imports access to basic services so limited. Per capita of food grains (wheat, rice, and maize) income was $329 in 1979 (excluding Nigeria) soared-by 9 percent per year since the early and 5411 when Nigeria is included. Death 1960s-reinforcingfooddependencv. Foodaid rates are the highest in the world and life also increased substantially (see Tables SA.24 expectancy is the lowest (47 years). Fifteen to and SA.29). Since 70 to 90 percent of the pop- twentv percent of the children die by their ulation earns its income from agriculture, the first birthday, and only 25 percent of the pop- drop in production in this sector spelled a ulation have access to safe water. Of the 30 real income loss for many of the poorest. countries classified by the United Nations The deterioration in agriculture and other Conference on Trade and Development (ULNc- internal and global factors led to widespread TA.\) as the poorest in the world, 20 are Af- balance-of-payments crises in the 1970s. Cur- rican. Of the 36 countries listed in the World rent account deficits in the region as a whole Bank's PVorld Dcz'elopment Report 1981 as "low rose from a modest $1.5 billion in 1970 to $8 income" (a per capita income of less than billion in 1980. External indebtedness climbed S370), almost two thirds are African. from $6 billion to $32 billion between 1970 The economic crisis is especially evident in and 1979, and debt service increased from 6 agriculture, and is reflected in output figures. to 12 percent of export earnings in the same Export crop production stagnated over the period. Foreignexchangereserves, whichwere past two decades. A 20-percent increase in comfortable in 1970, fell sharply. In 1979, re- production registered during the 1960s was serves could cover only two months' imports Table 1.1. Sub-Saharan Africa and the World: Basic Data GNP per capita average Per capita growth 1970-79 Life expect- Death rate of Population annual growth rate (percent) Adult literacy ancy at birth children aged (millions) (percent) Volume of rate (percent) (years) 1-4 (per thou- Countnes mid-1979 1960-70 1970-79 Agriculture exports 7976 1979 sand) 1979 Sub-Saharan Africa 343.9 1.3 0.8 -0.9 -3.5 28 47 25 Low-income 187.1 1.6 -0.3 -1.1 -4.5 26 46 27 Nigeria 82.6 0.1 4.2 -2.8 -2.8 .. 49 22 Other middle- income 74.2 1.9 -0.5 -0.4 -3.5 34 50 22 South Asia' 890.5 1.5 1.5 0.0 0.6 36 52 15 All developing 3,245.2 3.5 2.7b 0.1 -1.5 57 58 11 Low-income 2,260.2 1.8 1.6b 0.1 -3.1 50 57 11 M'Aiddle-income 985.0 3.9 2.8b 0.6 1.9 72 61 10 All industrialized 671.2 4.1 2.5" 0.2 5.2 99 74 1 . . not available. a. Bhutan, Bangladesh, Nepal, Burrma, India, Sri Lanka, and Pakistan. b. 1970-80. Source: World Bank data files. 3 and by 1980 reserves had fallen even lower. Son rce k oE+iq Gru:.cut7! Fiscal pressures also intensified in many countries, as indicated bv declining real bud- Africa's disappointing economic performance getary allocations for supplies and mainte- during the past two decades reflects, in part, nance, growingimbalancesbetweensalaryand internal constraints based on "structural" fac- nonsalarv spending, and difficulties in fi- tors that evolved from historical circumstan- nancing local and recurrent costs of externallv ces or from the physical environment. These funded development projects. include underdeveloped human resources, the The crises that evolved in much of the re- economic disruption that accompanied de- gion are particularly disturbing since, during colonization and postcolonial consolidation, the period from 1960 to 1974, world trade and climatic and geographic factors hostile to de- the world economy in general expanded rap- velopment, and rapidly growing population. idly, and many less-developed countries else- These internal factors are discussed further in where experienced relatively high growth Chapter 2. rates. Now, against a backdrop of global eco- Growth was also affected by a set of exter- nomic recession, the outlook for all less-de- nal factors-notably adverse trends in the in- veloped nations-but especiallv for the Sub- ternational economy, particularly since 1974. Saharan region-is grim. Although cyclical These include "stagflation" in the industrial- factors may push prices of some African ex- ized countries, higher energy prices, the rel- ports up from their low levels of the recent atively slow growth of trade in primary past, mounting energy costs, slow growth in products, and-for copper and iron-ore ex- the industrial countries (which translates into porters-adverse terms of trade. External fac- diminished marketsforthe developingworld), tors are the subject of Chapter 3. and reduced growth of international trade The internal "structural" problems and the (factors that have plagued the global economy external factors impeding African economic for the last half decade) will make renewed growth have been exacerbated by domestic African growth difficult. policy inadequacies, of which three are criti- In sum, past trends in African economic cal. First, trade andexchange-ratepolicieshave performance and continued global recession overprotected industry, held back agricul- together explain the pessimistic projections ture, and absorbed much administrative ca- for African development in the 1980s. The pacity. Second, too little attention has been World Development Report 1981, under its most paid to administrative constraints in mobiliz- optimistic set of assumptions about the ex- ing and managing resources for development; pansion of the world economy, forecasts vir- given the widespread weakness of planning, tually no growth in per capita income for the decisionmaking, and management capacities, continent in this decade7; under less favorable public sectors frequently become overex- assumptions, a negative rate of growth (- 1.0 tended. Third, there has been a consistent percent per year) is projected for the poorest bias against agriculture in price, tax, and ex- nations in the region. change-rate policies. These prospects and their political, social, and economic implications are not acceptable Nczcl Priorities either to the countries concerned or to the and Adjlstments i'l PoliCIJ international community. There is an urgent need to understand what has gone wrong A reordering of postindependence priorities and what must be done-by African govern- is essential if economic growth is to acceler- ments themselves and the concerned inter- ate. During the past two decades most Afri- national community-to assure a better future can governments rightly focused on political for Africa's people. consolidation, on the laying down of basic 7. World Bank, World Development Revort 1982 (New York: infrastructure (much of it tied to the goal of Oxford University Press, 1981), Table 1.1. political integration), and on the development 4 of human resources. Relatively less attention ity. This has resulted in slower growth than was paid to production. Now it is essential might have been achieved with available re- to give production a higher priority-without sources, and accounts in part for the current neglecting these other goals. Without a faster crisis. Without improved performance of pub- rate of production increase, other objectives lic agencies, stepped-up growth will be dif- cannot be achieved, nor can past achieve- ficult to achieve. The organization and ments be sustained. Three major policy ac- management of economic activity need to be tions are central to any grovth-oriented reviewed to determine how the resources and program: (1) more suitable trade and ex- energies of all economic agents can be better change-rate policies; (2) increased efficiency mobilized for development-for example, by of resource use in the public sector; and (3) improving government policymaking insti- improvement in agricultural policies. tutions and procedures; by giving the public Exchange-rate and trade policies, ad- sector's development-related agencies- dressed in Chapter 4, are especially critical "parastatals"-clearer mandates and greater for African economies, which are uncom- management autonomy; by giving wider re- monlv "open." Exports account for a large sponsibilities to the small-scale indigenous share of marketed production, and imports private sector; by allowing greater scope for constitute a significant share of consumption. decentralized cooperatives; and by defining Moreover, Africa has more frontiers per square an appropriate role for larger-scale private kilometer than any other region, and they are capital, domestic and foreign. Many govern- highly permeable. The framework of incen- ments have already acted in this area. In tives created by trade and exchange-rate pol- Guinea-Bissau, Mozambique, Senegal, icies is thus especiallv decisive. With respect Uganda, and Zaire, among others, govern- to agriculture, for example, overvalued ex- ments have decided on efficiency grounds that change rates discourage local production: the scope of private sector activity should be farmers obtain less in their local currencies enlarged. for their export crops, while the price of food In most of Africa, four out of every five imports is reduced. The situation is similar in people work in agriculture. It is the main the industrial sector. Also, direct controls over economic sector, generating in most countries trade (for example, import bans and quotas), 30 to 60 percent of GDP, or even more, if na- which are widelv imposed to deal with bal- tional accounts value it properly. Because of ance-of-payments problems, have proved ex- its importance, agriculture is treated at length tremelv costlv to applv, as they require many in the Report. A strategy for faster agricul- trained people and an enlarged administra- tural growth is set out in Chapter 5. Its main tive apparatus. Moreover, they have fre- elements are: concentration of resources on quently been ineffective. smallholders; reform of incentive structures Chapter 4 also considers policy issues which to ensure better prices, more open and com- bear on the efficiency of resource use in the petitive marketing systems, and greater avail- public sector. When African states won in- ability of consumer goods in some instances; dependence, they inherited unevenly devel- a focus in the medium term on making ex- oped economies with rudimentary isting programs work better and on rehabil- infrastructure. Markets often functioned im- itation of existing infrastructure, small-scale perfectly and foreigners dominated trade and irrigation, and rural roads; a major effort in most modern businesses. To speed up de- research on CTops and livestock; and expan- velopment and make their economies more sion of pest control and related activities to "national," the new governments expanded reduce postharvest losses. The strategy also the public sector. It is now widely evident requires careful evaluation of means toward that the public sector is overextended, given self-sufficiency in food. the present scarcities of financial resources, One of the pervasive themes of the Report skilled manpower, and organizational capac- is the critical role of human resources in the 5 development of Africa. The development of tionale for regional economic integration-the human resources is the subject of Chapter 6: goal of the Lagos Plan-and the desirability of education, training, and health. With respect a positive donor role in support of moves to education, one basic strategy issue is ad- toward closer regional economic cooperation. dressed: how to expand schooling most ef- Policy-related factors receive priority in the fectivelv in the face of severe financial Report, because-for many of the countries constraints. The training discussion covers a involved-the prospects for more rapid and wide range of questions: project-related train- sustained economic development are slight ing; expansion of on-the-job training; the need without appropriate adjustments. Domestic for special attention to management training; policies can be altered, although it is recog- and technical assistance. In the discussion of nized that changes of this sort require time health sector strategies a number of themes and are not easy to achieve. Given a more receive attention, including: experimentation suitable policy framework and adequate ex- with low-cost rural health care strategies and ternal support, the region's substantial eco- their gradual expansion, taking into account nomicpotential could be realized. Agriculture, administrative and financial constraints; bet- long neglected, could recapture the growth ter use of existing resources through im- rhythms of the 1960s if the environment were proved planning, policy analysis, and more supportive. Also, some impact of agri- managemenit; mobilization of private as well cultural research should be felt in the next as public energies; rehabilitation of infra- few years; research now under way can yield structure and consolidation of existing health results by the 1990s. In addition, there is the systems; and improvingaccesstopotablewater promise of the continent's largely unexplored and adequate sanitation. minerals, its offshore resources in fish and Although expansion of agriculture is nec- fuel, and the unharnessed power of its great essarilv the centerpiece of any production- rivers. oriented strategy, opportunities for increas- ing output also exist in other sectors-energy, Long-nlin Str'rtal-<:/ 1 -71&C(9t10HS industry (manufacturing, utilities, construc- tion), nonfuel minerals, and transport. Poli- The agriculture-based and export-oriented cies and programs for these sectors are development strategy suggested for the 1980s considered in Chapter 7. is an essential beginning to a process of long- While the focus of the Report is on re- term transformation, a prelude to industrial- sponses to the economic crisis of the 1980s, ization. It is not a permanent course for any certain longer-term problems cannot be ig- country, but one that in Africa generates re- nored. Perhaps the most critical of these is sources more quickly than any alternative and rapid population growth. The six children born benefits more people. Without these re- on average to an African woman during her sources, the foundations of future develop- childbearing years represent the highest total ment cannot be established. The list of what fertility rate in the world. This high fertility, must be done is formidable: administrative combined with declining mortality, has ef- services have to be extended to the rural areas fects on agriculture, urbanization, and gov- to increase social welfare and contribute to ernment spending, which are explored in the building of a sense of national unity; crit- Chapter 8. This chapter also considers the ically needed social and economic infrastruc- problems of Africa's rapidly growing cities, ture must be developed; roads must be built and the responses necessary nozw if urban re- and maintained; suitable schooling must be sources are to be effectively marshalled and offered to everyone; knowledge about the future urban crises headed off. Problems of economy has to be increased by broader and soil conservation, reforestation, and fuel- deeper research and by pilot experiments on wood supply are briefly analyzed, and the a wide front; and more people must be trained. chapter concludes with an analysis of the ra- Inter-African trade relations have to be de- 6 veloped and greater cooperation encouraged The analysis in the Report suggests that a by mieans of joint programs. This will build doubling of aid in real terms by the end of mutual interests and the habit of common the 1980s, from $4.9 to $9.1 billion (from $4.9 efforts, creating a sure basis for increased re- to $17.8 billion in current prices), combined gional integration, such as is envisaged in the with an appropriate reorientation of domestic Lagos Plan of Action. policy, could lead to a regional average an- A strategy focused on agriculture and ex- nual per capita growth rate of almost 2.5 per- ports is thus open-ended, a necessary begin- cent during the remainder of the decade. On ning. It will help generate the resources Africa the other hand, if the established patterns needs to consolidate its political and admin- continue, the overall per capita growth rate istrativeforces,educateandimprovethehealth will be zero or negative, with alarming pos- of its people, and find out what will work sibilities for even steeper downward spirals and ws hat will not. It will bring forth human in some countries. talenit nowv neglected and uncover physical Additional aid commitments will have to resources not yet imagined. And it will open be made now if disbursements are to reach the wav to a future whose shape we cannot required levels by the mid-1980s. These in- vet see. flows should be targeted to improve effi- ciency of resource use in the short and medium term. Quick-disbursing balance-of-payments assistance is critically needed in some coun- tries to permit fuller operation and mainte- \Vhile it is emphasized in the Report that nance of existing productive capacity and African governments must review their pol- infrastructure. A focus on completion of ex- icies and programs if their development ob- isting projects, on making recently completed jectives are to beachieved, it is also recognized projects work better, on rehabilitation, and that policy reform is difficult and delicate. In on maintenance will lead to quick increases all societies, formnidable obstacles prevent quick in output. Productive projects generally should responses to even the most carefully reasoned have highest priority. calls for change. In some cases, consumers Both donors and African governments will and producers, parastatal managers, civil have to change policies and attitudes if the Nervants, ancd industrialists have an interest large increases in aid recommended here are in maintaining existing policies, however in- to come about, and if they are to have their efficient these might be from a national point desired effects. What is needed is a new kind of view. Further, reform often involves tech- of social compact, an agreement within the nical9uestions fraught withuncertainties. But, world community that the struggle against perhaps most ot all, reform programs always poverty in Africa is a joint concern which take time. For these reasons and others, Af- entails responsibilities for both parties. After ncan governments will need additional out- all, foreign assistance has played a more sub- side assistance, and this matter is discussed stantial role in Africa than in most other de- in Chapter 9. veloping regions, in terms of aid per capita, The first step for the international com- share of total investment, technical assist- munity, if it is to assist African countries ance, and project selection and design. Do- through the present crisis and help the region nors have thus contributed to some extent to realize its potenitial, is a commitment to larger the present crisis. Moreover, African states aid flow.s in the I'9Ss. While per capita aid are among the world's newest and least de- to Africa is already relatively high, the needs veloped. They face special economic prob- are particularlY large and pressing when com- lems handicapped by still-acute scarcities of pared with those of most other developing trained and experienced people, fragile polit- re,ions w, hich are at roughly the same level ical systems, and untested institutions. They of per capita 1ncome, are, rightfully, a special concern of the com- 7 munitv as a whole. On the donor side, there- donors must engage in more systematic policy fore, assistance must not only be greater, but dialogue with their African partners. more effective. It will have to be accompanied On the African side, aid inflows have not by closer attention to project selection and always been used effectively; their develop- design, by more flexibility in aid modalities ment impact has been diluted by inadequa- (more financing of local and/or recurrent costs, cies in the domestic policy environment. for example), by more nonproject lending, African governments, therefore, must be will- and by greater attention to the policy envi- ing to take firm action on internal problems, ronment. All of this also implies greater do- be more open to proposals to revise policies nor collaboration than in the past; no donor in the light of experience, and be willing to wishes to finance the recurrent costs of some- accept the proposition that without policy re- bodv's "unsuitable" project. It means also that form higher aid will be difficult to mobilize. 2. BASIC CONSTRAINTS When the Sub-Saharan states won indepen- but the supply of "high-level" (university- dence some 20 years ago, they faced formi- trained) manpower was even more limited. dable constraints to development. These In most Sub-Saharan countries, over three included underdeveloped human resources, quarters of this cadre were foreign. Senior political fragility, insecurely rooted and ill- executive and technical jobs in govemment suited institutions, a climate and geography were dominated by expatriates Zaire, which hostile to development, and rapid population was left without a single African doctor, law- growth. And while the governments have yer, engineer, or army officer at indepen- scored considerable achievements, the legacy dence was an extreme case, but foreigners of history and the facts of geography continue occupied many positions of skill and respon- to hamper African economic progress. sibility even in the countries with the most advanced education systems; in Nigeria, Af- ricans held fewer than 700 of the 3,000 senior (9LztL'L';;. hL) GrowetrthZ: posts in the civil service until the mid-1950s, =rv! PO't(OtiOUkicvl Sit~~tlaioU and in Senegal, 1,500 French technical per- sonnel occupied almost all of the top jobs in UNDERDEVELOPED HUMAN RESOURCES 1961. And where there were large numbers One of the most critical problems of the past of settlers, even fewer Africans were trained 20 vears has been the scarcity of trained man- in modern skills. In Kenya and Tanzania, for power. Table 2.1 shows the great reliance on example, fewer than 20 percent of high-level expatriates for trained manpower in general, civil service posts were in African hands in the early 1960s. Throughout the region, trade and industry Table 2.1. Expatriate Employment as a were almost entirely owned and managed by Percentage of Total Employment of Trained foreigners. As recently as 1975 there were only Manpower in Early Manpower Surveys 80 African-owned shops in the Mozambican Country Year Percent capital of Maputo, and after sixty years of Botswana 1967 42 colonial rule, African-owned and operated Ivory Coast 1962 45 enterprises withmore thanten employees were Kenya 1964 48 extremely rare, even in the relatively ad- Niaerwa 1964 13 vanced economies of Kenya, Uganda, and Nigeria 1964 13 vne cnme fKna gna n Swaziland 1970 35 Zimbabwe.' In the wage sector, the foreign Tanzania 1965 31 Uganda 1967 21 1. Shankar N. Acharya. "Perspectives and Problems of Zambia 1965 62 Development in Low Income Sub-Saharan Africa" in Shan- Source: Adapted from Richard Jolly and Christopher Col- kar N. Acharya and Bruce Johnston, "Two Studies of De- clough, "African Manpower Plans: An Evaluation," Inter- velopment in Sub-Saharan Africa," World Bank Staff Working national Labor Reviezo, vol. 106, nos. 2 and 3 (August/September Paper, no. 300 (Washington, D.C.: World Bank, October 1972), p. 210. 1978), p. 11. 9 presence extended even to first-level super- glected. In 1960, for example, there was just visory positions: in 1960, for example, 300 of one physician for every 50,000 people in Sub- Ghana's 900 foremen were expatriates. Saharan Africa as compared to one per 12,000 This pattern of underdeveloped human re- in other low-income countries. Rural-based sources is partially explained by the fact that programs were few and preventive care was even by the end of the 1950s, advanced ed- very limited. Thus, for most of the population ucation was still largely unavailable to most (which is located in the rural areas), medical Africans: local facilities did not exist or, where care was provided by indigenous practition- they did, African enrollment was often re- ers. Life expectancy was lower than the av- stricted. Thus, in 1958, less than 10,000 Af- erage for all low-income countries (39 years rican students were attending universities at compared with 42), and child death rates home or abroad (one student per 20,000 pop- (deaths of children from one to four years of ulation), some 6,500 of whom were from age) were substantially higher (39 per thou- Ghana and Nigeria. In fact, very few coun- sand compared with 23). Colonial govern- tries had more than 200 students in university ments made valiant attempts to control training, which translates into less than 1 per- endemic diseases in many parts of the region, cent of the relevant school-age population (see but the majority of rural people were not af- Table SA.38). fected; systematic efforts at malaria control, The number of people educated at the sec- for example, were largely restricted to major ondary level was also limited. In the late 1950s, urban centers. the entire region produced only 8,000 sec- The scarcity of managerial and technical ondary school graduates per year, 40 percent cadres at the time of independence had strong of whom were in Ghana and Nigeria. In fact, adverse effects on public administration, in- only 3 percent of high-school-age students dustrial development, wage levels, and costs. were being educated at the secondary level Furthermore, the lack of education among the in Africa in 1960, compared with over 25 per- population reduced the stimuli for progres- cent in the Philippines, 20 percent in India, sive change generally experienced where ed- and 10 percent in Burma. ucation is more widespread.2 Finally, the The severe shortage of skilled labor and debilitating effects of disease and sickness entrepreneurs was also the result of the im- lowered the productivity of the labor force migration policies in the colonial period. For- and the propensity to innovate. eign workers at all skill levels were sought to meet specific labor shortages and later came POLITICAL FRAGILITY to occupy dominant positions as traders and Over the past two decades, a sizeable portion merchants, building contractors and artisans, of the Sub-Saharan region was the scene of industrial entrepreneurs, and skilled manual political and military conflict. In some coun- and clerical workers. While immigrant com- tries, the violence was sparked by liberation munities made important contributions to struggles, although in general the decoloni- African development, their presence in the zation process was remarkably peaceful. In labor market blocked Africans from acquiring the wake of independence, violent internal skills. Market forces also played a part: it was conflict burst forth in many of the new na- often cheaper to import and train foreign la- tions, stemming from the pluralism of African bor than to recruit locally. Moreover, because societies and the difficulties of postcolonial of colonial social conventions, Africans were political consolidation. Because cultures and rarely allowed to supervise non-Africans. As languages are so diverse (probably more so a result, the local population was prevented in Africa than in any other region), the pro- from moving up the skill ladder or assuming cess of national integration-building new in- entrepreneurial roles. Just as educational and training needs were 2. See World Bank, World Development Report, 1980 (New not being met, so too were health needs ne- York: Oxford University Press, 1980), Chapter 5. 10 stitutions and loyalties-inevitably involved required imposed heavy costs on the newly strife. Also, since the borders that the new independent governments. governments inherited frequently cut across The second and related problem was that ethnic lines, clashes were almost assured. of adapting existing national institutions, Civil and military strife and the political which had been closely patterned on those of fragility which it reflected had several nega- the colonial power, to African needs. Systems tive economic effects. First, it forced the post- of local govemment, general administration, independence leadership to give especially health care, and education had to be restruc- high priority to short-term political objectives. tured; the fashioning of "appropriate" insti- Second, it triggered large-scale displacement tutions proved to be a mammoth undertaking, of people. In the 1970s, the number of refu- one that remains unfinished. gees who had fled across national frontiers in Africa rose from 750,000 to over 5 million, THE ECONOMIC INHERITANCE accounting for about half of all refugees Modern economic growth has a relatively brief worldwide.3 In eight countries, the number history in Sub-Saharan Africa. Colonial of refugees during the peak year of flight administration established itself in most cases numbered at least 3 percent of the total pop- in the last two decades of the 19th century. ulation.4 Third, civil strife induced a diversion Economic expansion came quickly in a few of resources to military spending. While the countries-Ghana, Senegal, Uganda, and share of CNP devoted to military purposes re- Zaire, for example-and spread elsewhere mained fairlvconstantfortheregionasawhole later, with interruptions during World Wars in the 1970s (2.9 percent), the proportionnearly I and II and the depression of the 1930s. How- doubled among the poorest group (the low- ever, general and sustained developmentcame income semiarid countries), rising from 2.3 to only after World War II in most of the coun- 4.3 percent of GNP (see Table SA.43). tries of the region. In part because of this time factor, the Af- INSTITUTIONAL ADAPTION rican economies at independence were un- evenly developed and dualistic, more so than The institutional heritage of the postcolonial most other developing regions. Across the states had to be adapted to new political real- continent there were but few islands of mod- ities and harnessed to meet new needs. Two ern economic development. For example, in sets of problems were particularly relevant. West Africa, where peasant production of First, colonial governments had created many export crops was the primary motor of de- subregional organizations that did not prove velopment, modern economic activity took to be viable after independence. Some were place mainly in the forest and coastal zones functional groups, such as the West African extending 200 kilometers inland from the sea. Cocoa Research Institute, the West African In the vast interior, where most of the pop- Examination Council, the East African Rail- ulation was (and still is), evidence of eco- wav and Harbour Authority, and the East nomic change was barely visible, with the African Airways. Others, more important, exception of groundnut production in Nigeria were supranational groupings-the Federa- and Senegal. In Central, East, and Southern tions of French West and Equatorial Africa, Africa, dualism was even more marked; the the Central African Federation, and the East modern economy consisted largely of Euro- African Common Market. All proved no longer pean-run mining enclaves and islands of set- suitable to the new national realities and weTe tler agricultural activity. In the mineral- disbanded. But the reorganization which this producing countries which had significant settler communities, the "native areas" were 3. U.S. Committee for Refugees, 2981 World Refugee Sur- neglected and usuallv targets of discrimina- vey (New York: Hudson Press, 1981). 4. Angola, Burundi, Chad, Ethiopia, Guinea-Bissau, tion. African farmers, therefore, produced lit- Miozambique. Rwanda. and Zimbabwe. tle for the market. :11 Thus, as the postcolonial period began, most hinterland. Public capital investment had in- Africans were outside the modern economy. itially been limited by the shortage of local According to a 1950s United Nations study, resources, as well as by the doctrine of "co- over 70 percent of the land under cultivation lonial self-sufficiency" that prevailed until was devoted to subsistence crops, while less World War II, its central tenet being that col- than 10 pcrcent was planted for export.5 Af- onies should not be subsidized by metro- rican labor was overwhelmingly concentrated poles. In addition, private capital flows into in subsistence-oriented farming. In 1960, there most African countries were much smaller than were probably no more than 10 million Af- in other developing regions, and that which rican wage earners during any part of the went to Africa was very unevenly distributed. year; only in Southern Africa was as much as According to a classic study, 40 percent of the 10 percent of the population engaged in paid total private foreign investment in Africa south employment in the modem sector. Moreover, of the Sahara between 1880 and 1936 went to "circular" migration (the practice of workers South Africa; Zambia and Zimbabwe together returning to their villages more or less reg- received 18 percent; Zaire 11 percent; and ularly) was still quite common throughout the Kenya and Uganda together received 4 per- continent. cent. Elsewhere, investment was negligible.6 The dominance of subsistence production presented special obstacles to agricultural de- CLIMATE AND GEOGRAPHY velopment. Farmers had to be induced to Africa is "preeminently tropical."7 The equa- produce for the market, adopt new crops, and tor bisects the continent, and all of the coun- undertake new risks. Established farming tries that are the subject of this Report are systems, which had evolved over centuries located in the tropics (with the exception of and were well adapted to the local environ- Swaziland and Lesotho), a fact which creates ment, had to be revamped if production was special obstacles to development. First, the to increase. Little was known about new crops, interaction of climate and geography is such new methods of crop rotation, seed protec- that most African soils are delicate, deficient tion, or more productive farming techniques. in organic materials, and in general only mod- Agricultural research and experimentation erately fertile. Well-watered areas are only were lacking, but so too was most basic in- about one quarter of the total; elsewhere, rains formation about rainfall, river flow, soil qual- are inadequate in volume and highly variable ity, farming systems, and patterns of land in time. Moreover, the absence of frost, "the use. Accordingly, the experimental and in- great executioner of Nature," creates espe- tellectual raw material necessary for progress cially burdensome problems of weed and pest in agriculture was very sparse. The fragility control." Second, the search for minerals is of African soils, the irregularity of rainfall, more difficult in the tropics than in temperate and the ecological diversity that characterizes zones, where rock formations are well ex- even small subregions in this part of the world posed. Until two decades ago, prospecting make location-specific, detailed knowledge technology was largely restricted to surface especially necessary; its absence presented an probing; little was known about how to ex- unusually severe obstacle to effective agri- plore for mineral formations where soil over- cultural development. burdens are substantial. Finally, because the Moreover, basic infrastructure was, in some tropical climate is especially hospitable to bac- areas, almost nonexistent: roads, railroads, terial and parasitic diseases and to endemic ports, buildings, and communications sys- tems were scant and did not penetrate the 6. Sally H. Frankel, Capital Investment in Africa (London: Oxford University Press, 1938). 7. Andrew M. Kamarck, The Tropics and Economic Devel- 5. United Nations Department of Economic Affairs, En- opment (Baltimore, Md.: Johns Hopkins University Press, largerment of the Exchange Economy in Tropical Africa (New York: 1976), p. 19. United Nations, 1954). 8. Ibid., p. 17. 12 diseases such as malaria, schistosomiasis, and in climate and ecology remains open, there onchocerciasis, human energy and productiv- has been a definite increase in the vulnera- ity are adversely affected. bility of the semiarid areas. Drought has also played a role: some of the Another climate-related problem emerged poor performance of the 1970s has certainly following the 1950s and the first half of the been due to bad weather. The Sahel experi- 1960s, when it appears that there was better enced a quick succession of drought years than the long-term average in total rainfall between the late 1960s and 1973-74, with only and distribution of rains. This long period of one or two years of recovery in between. A favorable conditions encouraged herders and period of satisfactory weather in the mid-1970s farmers to adapt their husbandry techniques was then followed by a number of poor years, to a seemingly reliable situation. Then, with starting in 1977-78, in both the Sahel and the abrupt and severe climatic changes of the Northeastern Africa. These occurrences re- 1970s, serious and painful readjustments were sulted in a sharp drop in crop production and necessary, and this period of readjustment sev ere losses in livestock. may not yet have ended. Since rhe drought of the early 1970s, there Geography has also had an impact. Africa's has been extensive discussion as to whether large physical size and dispersed population this indicates long-term changes in the cli- create special transport needs and problems. mate with ensuing changes in the ecology- In addition, fourteen of the world's twenty an "advance of the desert." Present evidence landlocked developing countries are located provides inconclusive support for the hy- on this continent; almost one third of all Sub- pothesis of a secular trend in climatic condi- Saharan countries are landlocked, often more tions.9 Instead, there are indications that in than 1,000 kilometers from the sea by the some locations the natural plant population shortest land route. These factors have ob- has been degraded through overgrazing, and vious implications for road construction: long that the expansion of cleared land areas has trunk routes generate relatively low volumes negatively influenced evaporation and rain- of traffic, and extensive feeder networks are fall. But these were the result of acts of man- required. a relative overpopulation and overgrazing in semiarid areas under the pressure of human POPULATION GROWTH and animal population increases-and not to Most African agricultural systems involve very autonomous changes of climate. extensive use of land: shifting cultivation, long While there may be no discernible long- fallow periods, and limited use of manure term trend in average rainfall levels, there and off-farm inputs. As long as land was may be a rise in their variability, with marked abundant, the use of such technology was above- or below-average accumulation be- compatible with steady yields per capita, even coming more frequent. The fact that through- as rural population grew. But land is no longer out Africa marginal zones are now used much plentiful on most of the continent; in fact, more intensively than a generation ago ex- population pressures have existed for many acerbates the effects of irregular rainfall even decades in parts of East Africa (Burundi, further. So, while the issue of long-term trends Kenya, and Rwanda), Southern Africa (Le- sotho, Swaziland, and Zimbabwe), and West 9. Evidence contradicting a secular trend in climate is Africa (Mauritania and Niger). There are im- presented by twvo French scholars in a recent study of runoff portant pockets of high-density settlement in patterns of the Senegal River. Noting the great variability a few countries, such as in Southeastern Ni- of rainfall since prehistoric timnes, the authors argue that the Sahel has recently gone through a period of low rainfall. geria, the Western Highlands of Cameroon, The' predict that this dry cycle should end in 1985 with a the Mossi plateau of Upper Volta, and Sen- cycle of normal rainfall peaking in 1992; another severe egal's northern Groundnut Basin. And in re- drought is predicted for 2005. See Hughes Faure and Jean- , . Yves Gac, "Will the Sahelian Drought End in 198;?" Natuire, cent decades, very high population growth vol. 291 (June 11, 1981), pp. 475-478. rates for Sub-Saharan Africa-2.5 percent per 13 annum in the 1960s and 2.7 percent annually Pro - i,.rc l9uL) in the 1970s-have discouraged production and economic growth. In the past twenty years, African govern- There is evidence, first of all, that returns ments have made great progress in reducing to labor in agriculture are declining. Fallows the impact of these basic constraints. The are being shortened, undermining the regen- greatest strides have been made in the de- erative power of soils. Cassava, which can be velopment of human resources; in particular, grown on soils too impoverished for other the population is significantly better educated staples, is more widely cultivated at the ex- than it was a generation ago. Since 1960, total pense of other crops. Fuelwood has become school enrollments have grown faster in Af- harder to find, and overgrazing and disputes rica than in any other developing region (see between cultivators and pastoral people are Table 2.2). Student population increased from more frequent."' In addition, more marginal 36 to 63 percent of the age group at the pri- landhasbeenbroughtundercultivation,either mary level, from 3 to 13 percent at the sec- in zones of lower uncertain rainfall or on ondary level, and from virtually zero to 1 slopes, which has led to soil erosion and deg- percent at the university level (see Table radation. It is thus likely that part of the de- SA.38). Tens of thousands now graduate from crease in agricultural output per capita in the the continent's secondary schools each year, 1970s can be explained by population pres- and thousands from its universities.'I In fact, sure on arable land. Moreover, population the African record is unique: nowhere else growth has undoubtedly affected exports, as has a formal education system been created many tradeable goods (groundnuts and cook- on so broad a scale in so short a time. ing oil, for example) are currently being con- The story is similar in medical care. Life sumed locally. expectancy, the most important indicator of Also, rapid population growth has contrib- general health status, increased from 39 to 47 uted to huge migrations from the rural areas years-a 21 percent rise in the Sub-Saharan to the cities, with the result that the urban region since 1960. At the same time, child population grew faster in recent decades than death rates fell from 38 to 25 per thousand on any other continent (by 6 percent a year). and the number of medical and nursing per- Thus, in 1960, only three African cities had sonnel per capita doubled, despite very high populations of 500,000; by 1980, there were rates of population growth. 28. Government attempts to meet demands Table 2.2. Annual Average Growth Rates of for basic services and to assure supplies of Enrollments in Educational Institutions by reasonably priced food in these areas have World Developing Regions, 1960-76 strained already tight budgets and hampered appropriate flexibility in food price and pro- Percent duction policies. Region Primary Secondary Higher Total Lastly, growing populations create special Africa 5.7 9.8 10.5 6.2 needs for the provision of basic services, par- Latin America 4.8 10.0 10.9 5.9 ticularly schooling. The number of children Asia 3.9 5.0 9.1 4.2 to be educated has grown so rapidly that it Norh Africa and 7.6 8.0 54 has~~~~~~~~~~~~~Mdl East 4.9eel 7.6iul 8.0 5.4rmet has been extremely difficult for governments Source: David Davies, "Human Development in Sub-Sa- to increase the proportion of children being haran Africa," World Bank Staff Working Paper, no. 406 educated. (Washington, D.C., July 1980), p. 79. 10. See Bruce Johnston, "Agricultural Production Poten- 11. In Nigeria in the mid-1970s, local universities were tials and Small Farmer Strategies in Sub-Saharan Africa" in producing each year more graduates than those who had Shankar N. Acharya and Bruce Johnston, "Two Studies of graduated from Nigerian and overseas universities during Development in Sub-Saharan Africa," World Bank Staff the whole period of colonial rule. See Paul Beckett and Working Paper, no. 300 (Washington, D.C.: World Bank, James O'Connel, Edtucation and Po.oer ii, Nigcria (London: October 1978), p. 80. Hodder and Stoughton, 1977), pp. 9-13. 14 New infrastructure also was created: ports, same obstacles therefore continue to restrain railwavs, roads, and buildings were con- development. The reach of formal education structed at unparalleled rates. Road-building is still limited: for example, the 1978 primary received special attention, both for its eco- school enrollment ratio was only 63 percent nomic effects and because African leaders of the applicable age group for the region as knewv that more and better roads would help a whole. In about one third of the countries, unify their nations. All-weather road mileage less than 50 percent of primary age children and1 the number of vehicles tripled over the are in school; in only six are more than 20 two decades, which indicates something of a percent of the relevant age groups in second- transport revolution. Postindependence Af- ary school (see Table SA.38). And highly rica was "opened up" with extraordinary trained technicians, professionals, and man- speed; as a result, millions of formerly iso- agers remain in very short supply. lated villagers now have access to cheaper The health gap between Africa and the other transport, which creates new options and op- regions also persists. The population contin- portunities for this population on a wide ues to be more exposed to endemic disease scale. (especially malaria) and to diseases stemming Significaint political innovation has also oc- from poor sanitation, malnutrition, and pov- curred, not only in attempts at social trans- erty (intestinal parasites, gastrointestinal ill- formation, as in Ethiopia, Guinea, nesses, and respiratory infections). Life Mozambique, and Tanzania, but in postco- expectancy in Africa is still 27 years shorter lonial adjustments of various kinds. While than in industrialized countries and less than political violence has indeed scarred Africa's in any other developing region. The proba- recent historv, there have been many exam- bility that a one-year-old child will die before ples ot peaceful change and ethnic accom- his fifth birthday is 25 times greater than in modation as well. In some places, the wounds the developed world; the African child death of civil war were closed with extraordinary rate is 67 percent greater than in South Asia speed and humanieness. In Nigeria and Su- and three times higher than in Latin America. dan, unusual efforts were made to accom- Geographic factors continue to impose spe- modate ethnic diversity, and in Zimbabwe, cial hardships on transport development. The racial differences. existence of long routes with light traffic means Enormous efforts have also been made to that road maintenance is particularly de- adapt organizational and administrative ar- manding and expensive. Minimal mainte- rangements-civil service reforms, decentral- nance costs are estimated to be about three ized administration, planning systems, times as high a proportion of GNP as in other educational innovations, experiments in pri- developing countries, and about twice as high mary health care, and others-to the African as in the industrialized countries.12 And in setting. And the past two decades have been Africa'slandlockednations, transportandroad marked by a mobilization of resources and maintenance costs are greater still. energies not known before. Thousands of Population growth, already very rapid, dedicated people, African and foreign, con- threatens to become even more onerous in tributed to this effort, which lay behind the the future. Between 1960 and 1979, popula- achievements of the 1960-80 period. tion rose by 63 percent in the Sub-Saharan region, to a total of 344 million. During the 1970s, when growth rates declined around Tu7,, .°. -s. S f i: the world, in Africa they rose-to 2.7 percent *" i:'.-.:. te);&',slizlts a year. Moreover, the figure may increase, But, while African governments have ener- 12. The median African requirement has been estimated geticallv addressed many of the fundamental at 0.7 percent of GNl' (range: 0.3 to 1.4 percent). See World problems, twenty vears is not much time; the Bank, The Road Maizteciaiice Probleim, forthcoming. 15 since improved health and nutrition are re- ified, though its exploration has barely got ducing infant and child mortalitv at the same under way. time as extremely high fertility persists. This Only a few vears ago the potential of Sub- has already occurred in Ivory Coast, Kenya, Saharan Africa as a source of crude oil was Liberia, Tanzania, and Zimbabwe, where judged to be poor; now there appear to be population growth exceeds 3 percent a year major fields along the West African littoral, (see Table SA.33). and potential also in the East African Graben, Thus, unless there are significant improve- including Madagascar and the Seychelles. ments in agricultural technology, the popu- Proven reserves in oil-importing Africa are lation explosion will lead to further declines only 1 percent of the estimated potential. Even in per capita farm output. And government discoveries which are small bv international attempts to improve rural incomes, achieve standards will go a long wav to satisfy do- food self-sufficiency, provide basic services, mestic demand in most African countries, and and enlarge and improve infrastructure will even create an export capacity. It is much the become even more difficult than before. same with nonfuel minerals; Niger and Bo- Part of the explanation for Africa's slow tswana have been carried forward by recent economic growth since 1960, then, stems from mineral discoveries, and there is good reason the internal "structural" problems listed in to expect similar events in many other coun- this chapter. Those amenable to direct attack, tries of the continent. such as underdeveloped human resources, are Africa has additional resources that can be being addressed. Nevertheless, all of these counted on to facilitate its economic modern- problems will continue to constrain growth ization. Its coastal fishing grounds are among in the 1980s and beyond, and the region's the world's richest. Its great rivers contain a special needs should call forth special atten- substantial share of the world's unexploited tion from the international community. hydroelectric potential. The processing of its At the same time, it is important to stress raw materials will one dav be a major source that for most of the continent a bright eco- of industrial expansion. nomic future is wvithin reach. The human po- In agriculture, in petroleum, in other non- tential of the region has only begun to be fuel minerals, as well as in the industrial sec- developed. The energies of its people, and tor, African economic potential is therefore especially its farmers, are a major resource, very substantial; for most countries of the re- still imperfectly tapped. The continent's en- gion the long-term future is thus not at all dowment in material resources is also prom- bleak. What the region needs no-w-and what ising; Africa's old image as a storehouse of is asked of the international community here- great natural wealth is being increasingly ver- is help in realizing its economic promise. 16 3. EXTERNAL FACTORS The 1970s was a decade of growing balance- Table 3.1. Oil-importing African Countries: of-pavments deficits throughout Africa and Current Account Deficit and Its Financing, external factors certainly plaved a part in this 1970-78 deterioration: oil prices soared, the growth of (billions of dollars, 1978) world trade in primary commodities slowed Item 1970 1973 1975 1978 1980 considerably, and some countries (mostly Current account deficit- Fi- mineral exporters) suffered a severe terms-of- nanced by: 1.5 1.9 6.4 6.6 8.0 trade loss. Oil-importing African countries 1. Net capital flows (including mineral exporters) were hit espe- assistance ciallv hard at the end of the decade, when a (ODA) 1.6 2.1 3.2 3.2 4.3 second round of oil price hikes combined with Private direct in- the collapse of the nonmineral commodity vestment 0.4 0.4 0.4 0.3 0.3 Commercial loans 0.8 1.1 1.9 1.9 2.1 boom to reduce their terms of trade. 2. Change in reserves and But external forces were not the chief factor short-term borrowingb -1.4 -1.7 1.0 1.1 1.3 behind growing deficits: for the oil-importing Memorandum items: African nations as a group, poor export per- Current account deficit ' ~~~~~~as a percent of GDP 2.4 3.6 9.5 8.8 9.2 formance wvas more significant. During the as a percent of GDP 2.4 3.9.5 4.4 9.2 1970s, the volume of exports declined in two ODAas a percent Of GDP 2.7 3.9 4.7 4.4 5.0 thirds of the African countries for which data a. Excluding from current accounts net official transferS thirds o(grants), which are included in capital flows. are available (see Table SA.7), and Africa's b. A minus sign indicates an increase in reserves. share of world trade dropped accordingly. Source: World Bank data files. the second oil price increase in 1978-80 trans- 3'r. .., : - Q '-2, ,-W its Deteriort tioni formed their balance-of-payments position: by 1980, thev recorded a current account surplus Beginning in 1973, African oil-importing equal to 8.5 percent of GNP. countries experienced a pronounced wors- Although official development assistance ening of their balance of payments (see Table (ODA) to the African oil importers increased, 3.1). Since that time, their current account it offset only part of the growing balance-of- deficits (excluding foreign aid grants) aver- payments deficit. The absolute value of their aged about 9 percent of GDrP-twice the figure ODA receipts in real terms nearly tripled be- for all oil-importing developing countries and tween 1970 and 1980, and the ratio of ODA to conspicuously higher than any other region GDP nearly doubled.' Nevertheless, these of the developing world. Oil-exporting Afri- can countries fared much better, of course, but even thev ran a deficit during most of the 1. ODA is the net disbursement of medium- and long-term official loans and grants. Technical assistance is excluded. decade: in 1978, it reached 7.5 percent Of GNP, Real values of ODA and other items in Table 3.1 are based compared with only 4.4 percent in 1970. But on the use of the OECD North GDP deflator. 17 Table 3.2. Terms of Trade and Export Trends, Selected African Countries, by Export Categorya Average annual rate of grozwtl2 Purchasing Net barter terms power of exports of trade Export v'olunme Category 1961-70 1970-79 1961-70 1970-79 1960-70 1970-79 1. Oil exportersb 7.7 12.6 1.2 14.7 7.2 -2.0 2. Mineral exporters' 11.1 -7.7 6.5 -7.1 4.6 -0.7 3. Other primary exporters' 4.9 1.1 1.0 2.9 4.7 -2.1 Subtotal, oil importers (Categories 2 + 3) 7.6 -2.7 3.4 -1.5 4.7 -1.5 Total Sub-Saharan Africa 7.6 1.0 2.9 2.5 5.3 -1.6 a. Country group averages are weighted by value of country merchandise exports in 1970. b. Angola, Congo, and Nigeria. c. Liberia, Mauritania, Niger, Sierra Leone, Togo, Zaire, and Zambia. d. Benin, Cameroon, C.A.R., Chad, Ethiopia, Ghana, Ivory Coast, Kenya, Madagascar, Malawi, Mali, Rwanda, Senegal, Somalia, Sudan, Tanzania, Uganda, and Upper Volta. Source: UNCTAD, Handbook of International Trade and Development Statistics, 1980. countries were forced to expand commercial chapter examine these trends in more detail, borrowing sharply and deplete foreign ex- analyze their causes, and discuss how the change reserves because ODA contributions, external economic environment is likely to which more than covered the deficit in the evolve in the 1980s. early part of the decade, financed only atout half of it at the end. The balance-of-payments problem has other Terms of Trade aspects. First, foreign exchange reserves fell Three major factors influenced the shifting dangerously low; bv 1979 reserves were equal terms of trade in the 1970s: the two large oil to the amount needed to cover about two price hikes (in 197374 and 197-80) the long months' imports, and by 1980 they had fallen slie hi mineral pie (primaril in cope even lower. Second, the external debt contin- and in oran the boomain in com- ued to swell, despite measures taken to re- mod iroe (ofee, co oaind tea),in 17 -7 strain~ ~ , imot an kee th deii wihi modities (coffee, cocoa, and tea), in 1976-78. strain imports and keep the deficit within (Table SA.15 shows price trends of the most manageable bounds (imports indeed grew at important African exports.) a much slower rate in the 1970s than in the Over the last decade, the real price of oil 1960s). Finally, the debt-service ratio for all increased fivefold. In 1970, net oil imports African oil importers rose from about 6 per- absorbed 4.4 percent of the nonfuel export cent in 1970 to 12.4 percent in 1979; of the earings in eight African oil-importing coun- total of 11 developing countries forced to re- tries and 1.3 percent Of GDP (see Table 3.3). negotiate their multilateral loans in recent . . years, six were in the Sub-Saharan region.2 B The main reason for oil-importing Africa's percent of exports and 3.0 percent Of GDP. And deteriorating balance-of-payments position is that the purchasing power of exports declined Table 3.3. Oil Imports in relation to Exports at an average rate of 2.7 percent a year during and GDP in Eight Oil-importing African the last decade: - 1.5 percent in the terms of Countries' trade and - 1.5 percent in export volume (see (medians) Table 3.2).3 The remaining sections of this Item 1970 1978 1980 Net oil imports as percentage of: 2. The debt-service ratio is interest and amortization pay- Nonfuel exports 4.4 12.5 23.2 ments as a percentage of export earnings. Gross domestic product (GDP) 1.3 3.0 5.9 3. The purchasing power of exports is the export value a. The median ratios for Ethiopia, Ghana, Kenya, Mad- deflated by the import price index. This measure is also agascar, Senegal, Sudan, Tanzania, and Zambia. known as the income terms of trade. Source: World Bank data files. 18 between 1978 and 1980, the increase in real 1970s was not the terms of trade, but the slow oil prices of over 80 percent raised oil imports growth of exports: of the 29 countries for which to 23 percent of export earnings and to almost information is available, 24 recorded a lower 6 percent of GDP. rate of export growth during the 1970s than The effect of oil and other commodity price in the previous decade, and 19 had negative changes on the terms of trade is shown in rates of export growth, compared with only Tables 3.2 and SA.13. Several conclusions can one during the 1960s (see Table SA.7). be drawn from these and other data. First, It must be mentioned that while the infor- the terms of trade of African oil importers as mation available supports these conclusions, a group deteriorated by about 8 percent be- the data on exports contain some uncertain- tween 1978 and 1980. This loss of purchasing ties. One problem is that country coverage is power will probably be permanent. Second, incomplete: Botswana, Guinea, and Zim- mineral exporters experienced a strong down- babwe, all important exporters, are omnitted. ward trend in their terms of trade from 1970- Also, most of the data are pieced together 79 (- 7.1 percent per annum). This more than from various sources, and these are not al- offset a favorable trend in the previous dec- ways readily reconcilable. For example, there ade for the group as a whole, and several are some preliminary indications from the im- countries suffered terms-of-trade losses over port statistics for the industrialized nations the entire 1961-79 period: Liberia and Maur- that Africa's mineral and manufactured ex- itania were the most seriously affected. Third, ports (many of which are diamonds or other most African countries, other than mineral resource-based products) in fact increased at producers, experienced favorable terms of the end of the decade, and that the total vol- trade during the 1970s. Oil exporters, of course, ume of exports may have been marginally made spectacular gains, but other primary higher in 1980 than in 1970. Similarly, na- exporters also showed a strong upward trend. tional statistics from a slightly different group Finally, on average, African oil importers ex- of African oil importers show a positive av- perienced less deterioration in their terms of erage annual growth of exports of 1.2 percent trade in the 1970s (- 1.5 percent per annum) for the period 1970-78. than did most other oil-importing groups. The Nevertheless, the picture which emerges is trend in the terms of trade for all oil-import- one of nearly stagnant or declining export ing developing countries was -1.8 percent volumes for the continent as a whole during per annum, and the trend for industrialized the 1970s. And, for the entire 20-year period, countries was - 1.7 percent per annum dur- Africa's share of nonfuel world trade fell (see ing the same period. Table 3.4). Moreover, its share of developing- In brief, past trends in the terms of trade country nonfuel trade, which is perhaps a cannot explain the slow economic growth of more meaningful indicator of performance, Africa in the 1970s because for most coun- increased slightly during the 1960s but fell by tries-mineral exporters being the mnain ex- more than half in the 1970s. ception-the terms of trade were favorable or What accounts for this poor record? The neutral. All oil-importing countries suffered a strong and perhaps lasting downward shift in the terms of trade at the end of the 1970s, Table 3.4. Africa's Share of Nonfuel Exports however. This compounded the balance-of- Percentage share payments problems thev all faced as they en- of world nonfuel of developing-country tered the 1980s. Year exports nonfuel exports 1960 3.1 18.0 1965 2.7 18.0 1970 2.4 18.6 1978 1.2 9.2 The main cause of rising current account def- Source: United Nations, Yearbook of International Trade Sta- icits and shortages of foreign exchange in the tistics. 19 problem is partly structural. Africa is more restrictions may be imposed on a commodity dependent on exports of primary products that threatens serious disturbance to any eco- than is any other region. Thirty-two major nomic sectors (within the EEC); and (2) rules resource commodities accounted for about 70 of origin limitations, which remove the pref- percent of its nonfuel exports during 1976- erential access to manufactured goods de- 78, compared with 35 percent for all devel- rived from imports from outside the African, oping countries and 10 percent for the world. Caribbean, and Pacific (AcP) and EEC groups. And since world trade in most primary prod- But the structure of African trade and spe- ucts grows more slowly than world trade in cial links with the EEC soften the effects of manufactures, Africa's share of total trade some of these policies and even turn them to tends to fall. Also, its key commodities had African advantage. First, the most important lower than average worldwide growth during restrictions are on temperate agricultural the 1970s: growth of world trade in the prin- products or on manufactured goods; in nei- cipal African nonfuel exports slowed from 4.5 ther of these does Africa have much produc- percent a year in the 1960s to 1.5 percent a tion capacity. Thus, even if most of these year during the 1970s, while world trade in restrictions were removed, such an action a broader group of 33 nonfuel primary prod- would benefit the more advanced or more ucts grew at approximately 2.4 percent a year climatically suited developing countries to a during both decades. greater degree. For example, the Overseas Then there are specific explanations for why Development Council estimated that a 60 per- trade in certain important products lagged. cent cut in agricultural tariff and nontariff The copper industry in Zambia, for example, barriers by the developed countries would have is at a mature and fully developed phase, increased African exports by only $292 million cannot be expanded and, therefore, cannot in 1974. This was only 7 percent of the esti- maintain its share of the world market. mated increase for all developing countries, Drought and civil strife affected the produc- or 1.8 percent of African nonfuel exports in tion of particular crops elsewhere. that year.5 Finally, there are external factors-trade Second, the restrictions do not affect Africa restrictions imposed by the developed coun- as much as they do other developing regions tries. As with terms of trade, it must be noted because the Sub-Saharan nations receive pref- that, while these are important, they are not erential access to the EEC market for about 25 critical in determining export growth. percent of their exports (including aluminum On the one hand, the Sub-Saharan coun- ore, bananas, coffee, cocoa, cocoa products, tries are influenced negatively (as are other groundnut oil, palmoil, tobacco, and wood developing nations) by the industrialized products), and the average preferential mar- world's trade restrictions, such as a tariff gin is about 16 percent. Moreover, most other structure which escalates with the degree of products are admitted free under Most Fa- processing of the product, agricultLral price vored Nation (MFN) or Generalized System of supports for its own products, and nontariff Preferences (GSP) provisions. Thus, on bal- barriers to trade (see Table SA.16).4 Many ex- ance, protectionism by developed countries ports to the European Economic Community had little effect on African growth in the last (EEC), for example, are subject to quotas, min- decade. imum price rules, or variable levies. Also, two The developed countries' policies, then, provisions of the Lome Convention may ef- cannot account for Africa's overall poor ex- fectivelv limit the scope for future expansion port performance in the 1970s, when access of processed or manufactured exports to the to foreign markets actually increased. Other EEC on a preferential basis. These are: (1) an ECopen-aprend feguntiardbasis. whih states that 5. Thomas B. Bimberg, "Trade Reform Options: Eco- open-ended safeguard clause, which states that nomic Effects on Developing and Developed Countries" in William R. Cline (ed.), Policy Alternatives for a New0 Interna- 4. See also World Development Report 1981. Chapter 3. tional Economic Order (New York: Praeger, 1979), pp. 220-1. 20 Table 3.5. Africa's Share of World and Developing Country Exports of Selected Commodities Exports of selected primary commodities of Suib-Saharan countries as a percentage Of exports of those products of all the Of exports of those produicts of the whole developing countries world Comniodity 1960 1970-72 1976-78 1960 1970-72 1976-78 Fuels Petroleum 0.3 7.6 8.6 0.3 5.5 7.7 Minerals and metals Copper 47.3 52.1 38.8 25.5 28.2 19.2 Iron ore 10.8 30.3 19.7 4.3 11.4 8.1 Bauxite 5.7 4.7 31.7 4.8 3.4 27.7 Phosphate rock 0.6 13.3 14.3 0.4 7.1 9.4 Ntanganese ore 22.2 53.1 36.9 15.5 29.8 14.1 Zinc 27.7 25.9 18.7 6.8 5.7 3.6 Tin 11.7 9.7 3.6 10.7 8.3 3.0 Lead 12.9 19.4 6.6 4.8 4.1 2.2 Food and beverages Coffee 19.3 29.3 29.1 18.8 28.3 26.8 Cocod 72.8 80.1 72.3 72.8 80.1 69.3 Sugar 4.6 5.6 11.0 3.4 4.0 4.5 Tea 7.1 15.7 19.4 6.2 13.1 15.7 Groundnuts 87.1 74.8 63.5 76.8 53.5 35.5 Groundnut oil 77.3 72.2 56.8 57.0 55.7 43.2 Beef 4.5 4.0 8.6 1.3 1.2 1.6 Palmoil 65.7 22.6 6.7 63.2 20.8 5.7 Bananas 11.3 7.2 4.9 10.6 6.6 4.6 Maize 4.8 4.4 2.5 1.6 1.0 0.4 Nonfood Timber 44.7 22.8 18.5 36.1 5.7 5.1 Cotton 23.2 28.8 22.4 10.3 16.3 11.3 Tobacco 40.6 25.4 19.4 11.2 6.7 8.9 Rubber 7.4 7.9 4.9 5.9 7.7 4.8 Hides and skins 21.2 33.7 23.7 7.9 7.9 4.4 Sisal 68.5 58.3 52.8 62.9 56.8 51.8 Source: World Bank, Commodity Trade and Price Trends. developing regions were able to surmount Three factors explain the region's poor ex- similar obstacles and expand exports; as noted port performance: (1) a policy bias against above, Africa's share of developing-country both agriculture and exports (examined in nonfuel trade fell by half in the 1970s and its Chapters 4 and 5), which has led to slow share of world trade for most individual com- overall growth in production; (2) rapid pop- modities also fell (see Table 3.5). In fact, if ulation 'growth, which, by increasing con- Africa had maintained its 1970-72 share of sumption, has reduced the exportable surplus world trade, its principal nonfuel commodi- of crops such as oilseeds and maize and raised ties would have earned $2.2 billion more each the proportion of land used for domestic food year during the 1976-78 period (a 20.9 percent production; and (3) the inflexibility of African rise), its annual growth of nonfuel exports economies, which has prevented their di- would have been 3.2 percent higher, and its versification into products with rapidly grow- total export growth 2.2 percent greater.6 ing markets. 6. In the case of a few commodities, such as bauxite, tea, and tobacco, Africa's share of world trade increased. As- suming actual growth rates in these products and constant Prospects for the 1980s market shares in all others, nonfuel commodity exports would have been S2.5 billion higher, the annual growth of The 1980s should be somewhat more favor- these commodity exports 3.8 percent higher, and total mer- chandise export volume per annum 2.7 percentage points able for African oil importers than the recent higher. past. Much depends on the future path of oil 21 Table 3.6. Projected Price and Volume of World Trade in Selected Commodities Percentage share of African merchandise Price index Volume index Value index exports 1990 1990 1990 Commodity 1976-78 (1980=100) (1980=100) (1980-100) Minerals and metals 9.92t 127.7w, 137.8w 176.0w Copper 6.00 137.6 129.5 178.2 Iron ore 1.63 115.8 133.9 155.1 Bauxite 0.71 105.6 199.5 210.7 Phosphate rock 0.53 117.4 164.9 193.6 Manganese ore 0.45 90.2 138.6 125.0 Zinc 0.30 153.3 142.5 218.5 Tin 0.22 95.0 110.6 105.1 Lead 0.08 115.6 149.5 172.8 Food and beverages 23.20t 88.8w 131.7w 116.9w Coffee 10.73 96.7 125.6 121.5 Cocoa 7.11 66.2 143.5 95.0 Sugar 1.63 69.8 128.2 89.5 Tea 0.93 96.1 128.6 123.6 Groundnuts 0.73 142.9 84.0 120.0 Groundnut oil 0.67 134.8 122.5 165.1 Palmoil 0.82 131.0 220.0 288.2 Beef 0.30 107.8 167.7 180.8 Bananas 0.17 90.6 135.6 122.9 MAaize 0.11 158.5 113.8 180.4 Nonfood primary 6.61t 126.7w 128.6w 162.9w Timber 2.57 135.7 138.3 187.7 Cotton 2.46 125.1 110.5 138.2 Tobacco 1.10 116.7 141.3 164.9 Rubber 0.48 109.4 140.3 153.5 Subtotal of 22 nonfuel commodities 39.73t 105.1w 133.4w 140.2w Petroleum 43.47 137.0 123.6 169.3 Total commodities including petroleum 83.20t 121.8w 128.3w 156.3w Memorandum Cereals 142.9 Manufactures 100.0 a. Commoditv group averages are weighted by the percentage share of African merchandise exports, 1976-78. Source: World Bank projections. prices, which is highly uncertain. We have increases in the 1970s, but with the increased assumed, for purposes of this study, that the weight of oil in total imports, even relatively price of oil will increase by 3 percent per year small price increases will have a large impact. in real terms during the 1980s.7 At that rate For the typical oil-importing African country, the relative price of oil would increase by a 3 percent annual increase in the price of oil slightlv over one third during the coming implies a 0.7 percent annual decrease in the decade. This is small in comparison to the purchasing power of exports. Not all coun- tries will be affected; some oil importers, such as Cameroon, Ghana, and Ivory Coast, have 7. This and other projections are based on analysis done aspectsfor Ghang domesticopetrole bv the World Bank and incorporated into the global trade proSpects developing petroleum and commodity projections used in the World Development resources which will at least satisfy their own Report 1981 and in this Report. requirements. 22 According to World Bank projections, world value of world trade in minerals is projected trade in the 22 nonfuel commodities of great- to increase by 5.8 percent per annum, with est importance to Africa is expected to in- most of the increase coming in the latter half crease by 2.9 percent per annum during the of the decade, while world trade in food and 1980s (see Table 3.6), which is substantially beverages will increase by only 1.6 percent higher than the average annual increase ot per annum. 1.5 percent for African nonfuel primary prod- The projected rate of growth of trade in ucts in the 1970s. Moreover, the weighted Africa's main exports is lower than that of price of Africa's nonfuel commodity exports overall world trade. This dependence on ex- is projected to rise slightly so that the average ports of slowly growing primary products is value of world trade in these commodities a disadvantage, but exports can be diversified wvill increase by 3.4 percent per year. Broadlv and Africa's share of world trade in most speaking, projected trends for minerals and commodities could be increased with rela- beverages will be the reverse of the 1970s; the tively small effects on prices. 23 4. POLICY AND ADMINISTRATIVE FRAMEWORK Although internal constraints and changes in and that action to correct this situation is ur- the world economv are heavily implicated in gently called for. Africa's slow economic growth, domestic policy deficiencies and administrative con- straints have also been important-in many Trade and Exchange-Rate Poli'cy cases, decisive-and will continue to block economic progress unless changes are made. Trade and exchange-rate policy is at the heart Four concerns of this kind are discussed in of the failure to provide adequate incentives this chapter. Trade and exchange-rate poli- for agricultural production and for exports in cies, wvhich set the pattern of incentives for muchofAfrica.Tradeandexchange-ratepolicy economic growth, are considered first. This comprises policies on the official exchange rate, is followed bv a look at the planning and import duties, export taxes and subsidies, food decisionmakir;g context in which public policy prices, quantitative restrictions on imports, is determined and resources are allocated, and exchange controls. Although there are Organization and management of the public substantial differences among African coun- sector are considered next, followed by a con- tries with respect to both policy and circum- cluding section on the implications of the size stances affecting the availability of foreign of government on economic expansion. exchange, the discussion here will focus on The focus of the analysis is on the efficiency certain common patterns and on ways in which with which resources are used. Economic policies can be changed to stimulate efficiency growth implies using a country's scarce re- and growth. Ensuring an effective structure sources-labor, capital, natural resources, ad- of incentives is not a sufficient condition for ministrative and managerial capacity-more reversing the downward trend of agriculture. efficientiv. Improving efficiencyrequires, first, Other requirements are discussed in Chapter that a country produce those things which it 5, where, in particular, efficient domestic can best produce as compared with other marketing is shown to be a necessary con- countries and, second, producing them with dition for an associated improvement in ag- the least use of limited resources. While the ricultural production. analvsis which follows will be restricted to these efficiency considerations, it is recog- COMMON PATrERNS nized that policymaking inevitably has to em- The most striking similarity in the trade and body widerpolitical constraintsand objectives. exchange-rate policies of African govern- However, the record of poor growth in most ments has been the tendency to let real official Sub-Saharan African countries suggests that exchange rates become overvalued because of inadequate attention has been given to poli- higher inflation at home than abroad. cies to increase the efficiency of resource use Acccording to a study by the African Centre 24 for Monetary Studies, exchange rates rose in 1970s and the resort to quantitative restric- all but one (Sierra Leone) of 19 Sub-Saharan tions, foreign exchange management and in- African countries between 1963 and 1977-78.1 dustrial policy have interacted to intensify the Another study showed that they rose in 19 bias toward import-substitution industriali- of the 27 African countries examined.2 In three zation and to use quantitative restrictions as countries (Ghana, Uganda, and Zaire), the a protective device. Thus, an increasing num- exchange rate appreciated by over 100 per- ber of African countries have been moving cent. This statistical evidence along with the toward a trade and foreign exchange regime increasing use of trade and payments restric- with the following characteristics: licensing of tions, the large profit opportunities offered most imports; quotas or a complete ban on by smuggling, the wide gap between official imports that compete with local production; and black market exchange rates, and the slow automatic protection for any import-substi- growth and loss of market shares of many tution industry; and priority allocation of es- traditional exports, all indicate that overval- sential imports to capital goods, raw materials uation is widespread. Although not all Afri- for local industry, and food. can countries have overvalued exchange rates, This system has not been used by some foreign exchange problems are sufficiently small countries (Botswana and Malawi), by widespread to suggest that most official ex- CFA franc zone countries (where quantitative change rates do not reflect its scarcity. restrictions are usually applied only outside There has been a common pattern of re- the zone), or by countries with a strong bal- sponse to foreign exchange scarcity in most ance-of-payments position (Nigeria before African countries. Governments have relied 1977). It is, nevertheless, in widespread use increasinglv on import restrictions rather than in countries as diverse as Ghana, Ivory Coast, devaluation to conserve foreign exchange. Kenya, Nigeria (since 1977), Senegal, Sudan, More and more countries have imposed higher Tanzania, Zambia, and Zimbabwe. The policy tariffs, quotas, and bans on "nonessential" is in danger of biasing the incentive system imports. Quantitative restrictions have been against objectives to which governments give the favored means of import restriction. high priority-agriculture, exports, food pro- This dependence on import restriction has duction, and rapid industrial development. been reinforced by the import-substitution in- dustrialization policies pursued by most Af- rican governments. Quite separately from balance-of-payments policy, governments have Bias against Agriculture. A trade and ex- tried to promote industrialization through tar- change-rate system that relies heavily on im- iff protection for local industry against com- port restrictions biases the incentive system peting imports. This policy has been combined against agriculture in several ways. First, it with low tariffs or duty-free imports on cap- forces farmers to purchase high-cost local im- ital equipment and raw materials. With the plements. For example, in Upper Volta there recurring balance-of-payments crises of the is a 66 percent tariff on animal-drawn plows and a 58 percent tariff on engines used for 1. African Centre for Monetary Studies, Baiance of Pay- irrigation pumps. Second, it raises the cost of nitents Problems of African2 Countries and Their Effects on Dcel- consumer goods. Thus, in Kenya, imports of olpment Objectizcs (Dakar, Senegal: August 1979), p. 69. The second-hand clothing have been banned and lrr exchange rate is said to appreciate if a countrv's inflation -ate exceeds the world inflation rate, unless it devalues bv there is a 100 percent tarff on textles. This more than the differential inflation rate. measure has doubled the price of clothing 2. This calculation used 1970 as a base vear and uses the and textiles and reduced real rural incomes special-drawing-right (SDR) exchange rate and the consumer by 10 percent. Third, and most importantly, price index of industrialized countries as the foreign com- parators. The previousl' cited study used the U.S. dollar this trade and exchange-rate policy serves to exchange rate. hold down the prices farmers receive for their 25 export crops. Several African countries now fect of overvalued exchange rates and low find that producers of traditional export crops duties on food imports has encouraged a de- cannot be paid enough to cover the costs of pendence on food imports at the expense of production (e.g., cocoa in Ghana, sisal in Tan- domestic production. In particular, wheat and zania, and coffee in Madagascar), even though rice imports have grown at the expense of these are crops in which the countries have local grains and root crops (a trend consid- a strong comparative advantage. For the ma- ered in more detail in Chapter 5) and in parts jor agricultural exports of seven African coun- of West Africa traditional upcountry supplies tries (Ghana, Kenya, Nigeria, Senegal, Sudan, of beef have been displaced by South Amer- Tanzania, and Zambia), producers frequently ican and Australian imports. receive less than half the real value of their crops (see Box A). Finally, the combined ef- Bias against Exports. A second consequence Box A: Agricultural Exports from Tanzania and Ghana During the last 15 years, the volume of exports in to be financed by domestic bank loans either directly Tanzania has declined dramatically. In 1980, the total to the parastatals or to the Government, since neither exports of the country's major commodities (cotton, have the funds required for such a move. coffee, cloves, sisal, cashews, tobacco, and tea, which The story is the same in Ghana. In the past 15 years, account for two thirds of the nation's export earnings) a dramatic and steady decline in cocoa production- were 28 percent lower than in 1966 and 34 percent from a peak of 566,000 metric tons in 1965 to 249,000 lower than in 1973. As a percentage of GDP, export tons in 1979. Ghana's share in world production has eamings fell from 25 percent in 1966 to only 11 percent shrunk from one third in the 1950s and 1960s to one in 1979. And the drop in volume, when combined with sixth in 1979; from first place in cocoa, Ghana has fallen deteriorating terms of trade, produced a severe bal- to third place, behind Ivory Coast and Brazil. ance-of-payments crisis. The main reason for this disastrous decline is the Why such a poor record? In the case of tobacco and heavy "tax" imposed on farmers by Government cotton, exports sagged because domestic consumption through the Cocoa Marketing Board's price policies. increased. But in general, it was the result of severe Producer prices for cocoa lagged behind other relevant cutbacks in production which, in turn, were sparked prices-far behind in most cases. Thus, the price index by the low official prices paid to producers. These did of all consumer goods rose 22 times between 1963 and not keep pace with world prices or domestic costs. 1979; food prices rose at about the same rate and the Thus, terrns of trade for the cash-crop farmers plum- price of cocoa in neighboring countries rose 36 times. meted by almost one third in the last decade (or by 20 In contrast, Ghana's farmers received only 6 times more percent, if food and cash crops are taken together). for their cocoa. The marked drop in prices was not due to unfavor- As a result, producers have been neglecting their able world market conditions, but rather to heavy cocoa trees, and shifting into other crops-particularly "taxes" on farmers-export duties combined with ever- foodstuffs. They have smuggled cocoa to high-priced increasing marketing and administrative costs accu- markets outside; perhaps 15 percent of production takes mulated by the parastatals handling the products. These this route. bloated costs were important factors leading to a re- Since Ghana's fiscal situation has been difficult for duction in farmers' share in export eamings from cof- many years, the Government has had to give high fee, cotton, tobacco, and cashews to below 50 percent. priority to its revenues, and this has contributed to the As a result, producers diverted some of the crops reluctance to raise producer prices and hence the main- (coffee, for example) into unofficial channels, neglected tenance of heavy taxation on cocoa. and even abandoned some tree and bush crops (coffee, This is where exchange rate policy comes in. The sisal, and cashews), and shifted resources into sub- real effective exchange rate for cocoa, which is the sistence production (in the case of cotton and tobacco). number of units of local currency actually received by To ameliorate the problem, Tanzania devalued its producers for a dollar's worth of cocoa exports, ad- currency by 15 percent (devaluations occurred in Oc- justed for inflation, declined drastically over the past tober 1975 and January 1979); but this still did not fifteen years, which explains the disincentives to grow- reduce the wide disparity between Tanzanian costs ing cocoa. Higher producer prices could have been and intemational prices. In addition, the Government paid without reducing governnent revenues, by ad- increased producer prices somewhat, but the scope for justing exchange rates. Instead, the official rate of ex- such adjustments is limited by the current overvalued change of Ghana's currency has remained unchanged exchange rate. Thus, any further price hikes will have for long periods, despite persistent domestic inflation. 26 of the predominant trade and exchange-rate result, they continue to impose high costs on policy in Africa is that it is biased against consumers. exports. As with agriculture, export indus- In brief, the trade and exchange-rate sys- tries are sometimes burdened with high-cost term of most African countries encourages the domestically produced inputs that reduce their development of high-cost, import-intensive, competitiveness in world markets. In Kenya, and capital-intensive industries. The system for example, a survey showed that the cost is usually quite effective in encouraging sim- of locally produced cans alone was higher ple import substitution industries, but, once than the landed price of canned vegetables these have been established, industrial de- from Asian competitors located in the Ara- velopment faces an impasse. Further growth bian Gulf. through import substitution requires produc- Producers of exports are discouraged by tion of intermediate industrial products that these policies, even when high-cost inputs frequently require larger amounts of capital are not a problem. High protection for local and provide little employment. Economic in- production combined with access to imported tegration can expand the scope for efficient raw materia[s at lower rates of duty give high import substitution, and efforts in this direc- protection to import-substitution production. tion should be supported. However, integra- This makes producing for the local market tion will be a long-term process, and the much more profitable than producing for ex- limitations of overprotection in small domes- port. Thus, firms producing for the local mar- tic markets apply with delayed but equal force ket have an edge when bidding for scarce in larger regional markets. Industrial exports, resources. This raises costs for exporters. Af- which could provide a means of sustaining rican governments should continue to en- industrial growth, are discouraged by this courage efficient import substitution but this system. In extreme cases, countries may even should not be achieved through policies which have trouble maintaining production of ex- actively discriminate against exports. On the isting industries due to foreign exchange contrary, the possibilities for developing ef- shortages that limit imports of essential raw ficient export opportunities should be given materials and parts (see Box B). comparable encouragement. This failure to provide incentives for manufactured exports Reduced Flexibility. Trade and exchange-rate is a serious shortcoming because it cuts Af- policy also reduces the flexibility of the econ- rican countries off from the fastest growing omy. Once all "nonessential" imports have part of world trade and from trade with been eliminated, only "essential" imports of each other. capital goods, spare parts, and raw materials remain to be cut in the event that severe for- Distorted Industrial Development. There are eign exchange shortages require such a move. two other wvavs in which the typical trade and Political pressures have usually favored main- exchange-rate svstem adversely affects the taining existing production or employment prospects for long-term industrial growth. levels, even though this is predominantly in First, it encourages import-intensive industry consumer goods industries. As a conse- and discourages the development of domestic quence, scarce foreign exchange is allocated industries that consume local raw materials to maintaining consumption instead of being and labor. Protection typically favors pack- allocated to high-priority investments. There aging or assembly-type industries, which pro- are numerous instances in which bans or quo- vjide very few benefits to the economy either tas on imports of whisky or electronic equip- in foreign exchange, employment, or skill ment have simply led to local bottling and development. Second, the nearly complete assembly of these same items. To be effective, protection given to industry gives no incen- import controls must be supplemented by an tive for growth in productivity. Infant indus- excise or sales tax that reduces the incentive tries tend, therefore, never to grow up. As a for both consumption and local production. 27 Box B: An Example of the Costs right quantity or quality at the right time. of Excess Protection Controls also deprive the government of rev- enue that could otherwise be captured through In 1979, an African firm producing synthetic yams tariffs. Importers lucky enough to receive an sought an increase in the protective tariff from 50 to import license pay a lower price for imports 100 percent plus continued remission of duty on im- but are able to earn a high scarcity premium ported inputs. An analysis of production costs showed that the firm's foreign exchange costs alone were through resale or use of the cheap import in higher than the price of a comparable imported prod- production of scarce goods-and this pro- uct. By the time the foreign exchange cost of im- vides strong incentives for corruption. Fur- ported raw materials, foreign managers, and capital ther, bureaucratic delays are costly and the charges abroad were paid, it cost $1.01 per kilogram t more foreign exchange to produce the product locally emptation to pay (and accept) a premium to than to import it. Moreover, local production was get to the head of the queue is usually quite losing revenue for the treasury. At the existing 50 irresistible. percent tariff, each kilogram of imports would have yielded $1.44 in revenue. This would have been suf- ficient to emplov five times as many local workers SPECIAL CASES (the local wage bill per kilogram was $0.27) as em- ployed in local production. Finally, the high cost of Not all African countries have the same trade the local product prevented other user firms from and exchange-rate problems, nor is there a exporting. Thus, high-cost local production not only blanket policy prescription that fits all coun- lost foreign exchange, but prevented other firms from tries. A brief discussion of two groups of eaming foreign exchange. Although this sample offers an extreme case, the countries-the mineral exporters and mem- existing trade and exchange-rate system in many bers of the CFA franc zone-serves to illustrate African countries encourages countless other in- these differences. stances of "value subtracted." Breakdown of Production Costs Mineral and Petroleum Exporters. Although of Polyester Texturized Yam mineral- and petroleum-exporting economies (dollars per kilogram) are generally not troubled by balance-of-pay- Total Local Foreign ments problems while exports are expanding, this masks longer-term difficulties. The spe- wages and salares 0.30 0.27 0.03 cial problem that mineral exporters face is that Other costs, including their exchange rates can be valued too highly depreciation 3.43 2.13 1.30 for long-term development. A strong balance- Total costs 6.29 2.40 3.89 of-payments position makes importing too easy C.i.f. import price 2.88 - 2.88 and this blocks the development of import- Excess cost 3.41 2.40 1.01 competing and export activities that are needed to generate employment opportunities and to sustain growth. when minerals are depleted. A more balanced approach emphasizing for- Export diversification is particularly difficult eign exchange earnings as well as foreign ex- in these circumstances (see Box C). The ap- change savings provides greater room for propriate policy here is some form of tax on maneuver to the authorities in a crisis. the principal export that captures the surplus income and channels it toward long-term di- Distortions and Dangers of Direct Con- versification. In doing so, it prevents this sur- trols. Reliance on quantitative restrictions and plus export income from being used to other administrative measures makes lavish maintain consumption, wages, and nonpro- use of administrative capacity, the scarcest ductive employment at levels above those resource many African economies have. In sustainable in the long term. Niger, for ex- addition, import controls slow down deci- ample, has earmarked all uranium tax reve- sions and create bottlenecks in production nues for education and agricultural because critical inputs are not available in the development to avoid these dangers. 28 Box C: Industrial Incentives in Zambia Zambia's economy is characterized by extreme dual- the cost of Zambian labor high relative to most neigh- ism between an urban-oriented, modern sector dom- boring countries and to most other developing coun- inated by copper mining and a rural subsistence sector. tries at a similar stage of development. On the other Copper mining has generally contributed over 90 per- hand, because of low tariffs and interest rates, capital cent of exports, one quarter of cDp and (at least before is relatively cheap. the present economic crisis) about one third of gov- The effect of this structure of incentives has been ernment revenues. A persistent theme in Zambian pronounced. Manufacturing output grew at more than development has been the need to generate sustainable 7 percent per annum from independence to the mid- growth in alternative sectors to diversify the economy 1970s. Even so, manufacturing employment grew less away from dominance of copper, but to date little suc- than half as rapidly and the capital-intensity of man- cess has been achieved in developing agriculture or ufacturing techniques grew consistently over time. manufacturing. Since the world copper price peaked Manufactured exports, which amount to less than 2 in 1974 (never again reaching as much as 60 percent percent of total exports, have stagnated. Manufactur- of its peak level in real terms), Zambia has fallen into ing production has been centered in final consumer a prolonged economic slump characterized by falling goods and some intermediate goods using imported real output, inflation, declining foreign exchange re- inputs. The expansion of manufacture for domestic use serves, and a large amount of arrears in payment for has not been based on the development of Zambia's imports. considerable agricultural potential. Instead growth has Part of Zambia's difficulties in developing manufac- been centered in textiles, rubber, and chemicals, all of turing may be traced to its system of incentives which which are highly dependent upon imported inputs. has affected both the structure of production and pro- The dangers of import-dependent manufacturing duction techniques. Tariffs are high on unessential and development are well illustrated by Zambia's current luxurv consumer goods, low on foodstuffs, essential economic crisis. Because the basis of manufacturing is consumer goods, and capital goods, and zero on most not firmly rooted in domestic resources, the possibil- intermediate goods. A system of import quotas and ities for manufacturing expansion are severelv re- foreign exchange licensing reinforces the incentives stricted in times of foreign exchange scarcitv. As copper generated by tariffs; foreign exchange is rationed to export revenues fell, import volumes were severely essential consumer goods and foodstuffs, intermediate restricted, falling about 50 percent between 1975 and and capital goods, while severely restricting allocations 1980. The resulting shortages of imported raw mate- to less essential and luxury consumer items. Govern- rials and intermediate goods have led to a fall in man- ment policies have also affected the cost of labor and ufacturing output, and most manufacturing subsectors capital. The drive for higher mining wages following are operating at less than two thirds of capacity. independence rapidly spread to other sectors, making Memnbers of theCFAFrancZone. TheCFAfranc available for use. This puts a greater burden zone countries face a rather different set of on other policy instruments for maintaining conditions than other African countries. As balance-of-payments equilibrium, particularly part of a common currencv zone, they have on fiscal, monetary, and wage policy. The benefitted from relativelyfreepaymentsamong stresses of the late 1970s have somewhat members, from pooling of resources, and from changed the economic environment within the abilitv to run a deficit financed by the which the franc zone operates. There are fewer French Government through an account at countries with surpluses and more are seek- the French Treasury-the so-called opera- ing credit at the operations account. More- tions account. They also benefit from asso- over, uncontrolled parastatal borrowing over- ciation with a major convertible currency. As seas threatens future balance within the zone a group, the CFA countries appear to have since individual countries build up heavy debt- benefitted from the discipline imposed by the servicing obligations. need to coordinate policies with partner states. However, the need for coordination also im- POLICY CHANGES posesconstraintson individualcountries; some Improvementintradeandexchange-ratepolicy policies, such as monetary growth, must be is essential for accelerated growth. The 1979- coordinated, while other policy options, such 80 deterioration in the terms of trade and the as exchange-rate changes, have not been prospect of further increases in the cost of 29 energy lend greater urgency to policy change. inflation. Although there is little experience The key changes needed are: correction of with sustained attempts to restructure trade overvalued exchange rates that have emerged and exchange-rate policy in Africa, experi- in most countries; improved price incentives ence in other developing areas suggests that for exports and for agriculture; lower and more such measures can work, that they take time, uniform protection for industry; and reduced and that change is eased and hardship re- use of direct controls. For those countries duced if substantial external assistance is willing and able to use exchange-rate changes, available.3 Responsiveness to changes in the devaluation is a powerful tool for restructur- real exchange rate has enabled several coun- ing relative prices and incentives, though some tries in Asia and Latin America to achieve countries will find it possible to improve in- very high rates of growth of both manufac- centives without changing the exchange rate. tured and total exports.4 The alternatives, however, require a combi- For most African countries, trade and ex- nation of effective policies to keep the rise in change-rate policy should be viewed as an incomes and in costs below the rise in other instrument of long-term structural adjust- countries and budgetary policies to provide ment rather than short-term balance-of-pay- appropriate incentives. Budgetary action ments adjustment. Policy changes to would in particular have to provide for ex- restructure incentives might not always have port subsidies or reduced export taxes together immediate and dramatic effects on economic with changes in both the level and structure performance, but the cumulative, long-term of tariffs. Taken together, these actions are effects will be critical to Africa's effort to raise likelv to have an adverse affect on the overall agricultural and exportgrowth rates. The main revenue situation of governments which are effect of such reform on the import side will already facing major problems in this regard. be to change the composition of imports and Furthermore, the administrative burden of to ensure that imports are allocated more ef- export subsidies is significant especially if ficiently rather than to reduce the total import abuse, such as smuggling for re-export, is to level. Moreover, the supply response to be avoided. changes in incentives will take time. Produc- The fiscal and administrative burden would ers will be reluctant to make necessary in- be less severe and the adjustment process vestments until they have been assured that made easier, however, if the official exchange the relative price changes will be permanent. rate were changed and kept at realistic levels. For all of these reasons, full reform of trade Devaluation permits higher prices to be paid and exchange-rate policy cannot be under- to exporters without subsidies. If tariff re- taken at a stroke. However, in a few countries duction and relaxation of import restrictions the exchange rate is so overvalued that a sub- are accompanied by devaluation, prices for stantial devaluation is a precondition for be- import-substitute production can remain con- stant in local currency, thus easing adjust- 3. A study of 24 devaluations found that, on average, ment for local producers. Devaluation, consumer prices increased by less than half the amount of combined with tariff reduction or relaxation the devaluation in the following year, while manufacturing of import restrictions, enables the full effect wages increased by less than one fourth. See Richard N. Cooper, "Devaluation in Developing Countries" in Gustav of the exchange-rate change to be concen- Ranis (ed.), Government and Economic Development (New Ha- trated on exports. ven: Yale University Press, 1971). Anne 0. Kreuger, in Governments are sometimes skeptical that Liberalization Attempts and Consequences (Cambridge, Massa- changes in trade and exchange-rate policy can chusetts: Ballinger, 1978), found that devaluation tended to 0 ~~~~~~~~~~~~~reduce inflation rates below what they would otherwise affect economic performance. They also worry have been. See also Jagdish N. Bhagwati, Anatomy and Con- that the costs may be prohibitive-that at- sequences of Exchange Control Regimes (Cambridge, Massachu- tempts to restructure relative prices (espe- setts: Ballinger, 1978). 4. Bela Balassa, "Export Incentives and Export Perform- cially if devaluation is involved) may create ance in Developing Countries: A Comparative Analysis," hardship for the poor and generate increased Weltwirtschaftliches Archiv, vol. 114, no. 7 (1978). 30 ginning the process of reform which will get eral years; livestock projects may need infor- the economv back onto the right track. And, mation on labor utilization patterns among in all countries, there is a need to keep trade cattle-using sedentary farmers; for effective ar[d exchange-rate policy under continuous rice projects more may need to be known reviewt to ensure that it contributes to the about the division of labor between men and development of an incentive system which women as well as how yields behave over a assists governments in achieving their eco- period longer than two years; rural self-help nomic management and development objec- projects should be based on close sociological tives. External assistance can play a critical analysis. In short, in many sectors good pro- role in helping countries bridge the difficult jects cannot be developed quickly (in 3 to 9 period between initiation of reform and re- months) by visiting teams of specialists. More alization of substantial benefits. lengthy, serious, and focused studies will be required if projects in these key sectors, es- r. -~ >7pecially agriculture, are to be more successful. Tl. ' """- ' cs'. ?>?g}1t57t7ml7Zg World Bank sector work could play a role in identifying specific needs in data or basic Another requirement for more efficient re- suisrltdt oeta e rjcs source use is strengthened economic deci- sionmaking capacity in the public sector. SCREENING EXPENDITURE PROPOSALS Experience indicates a pressing need to rein- AND POLICY ANALYSIS force procedures and institutions in three key Even more essential than the ability to gen- areas: the generation of development pro- erate good projects is the capacity to sort out jects; the evaluation of expenditure requests, those which are good from those which are including project proposals, by central gov- not so good. This is the vital screening func- ernmnent coordinatin-> agencies (finance and plannt ing ancie f in n an tion, which involves appraising government planoning moinisies);.and the formulation o investment proposals, and is the essence of setting priorities. Two units of government normally have PROIECT GE?NERATION the responsibility for screening proposed in- Theabilitvtogenerategoodprojects-thebasic vestment projects and other development-re- units of development action-is essential for lated expenditure requests: planning agencies efficient use of investment funds. With few and ministries of finance. In practice, plan- exceptions, this capacity is weak in Africa, ning agencies in Africa, as in most developing which explains why manv projects are de- areas, have generally played a minor role in veloped by donor technicians and why do- the vetting of projects. They have been en- nors and African governments have developed gaged mainly in the elaboration of medium- special project-preparation facilities. Notwith- term plan documents. But during the process standing these efforts, the lack of well-pre- of putting them together, planning agencies pared projects constrains the flow of aid to a rarely have the opportunity to give systematic number of countries. appraisal of projects proposed by spending The shortage of good projects is accounted agencies. Frequently, the planners find it nec- for in part by organizational factors: the in- essary to include in the plan most projects ability to bring together available data for submitted to them by the technical ministries, technical and economic analysis, and to pack- even some which are little more than project age the outcomes in a form tailored to donor ideas. In part for this reason, the plan's ex- needs. But much more important is the sparse penditure targets almost always exceed re- knowledge base. Successful project prepara- sources available for investment. This means tion often requires location-specific data gen- that selection of investment proposals for fi- erated over a fairly long period of time: port nancing (i.e., the real setting of priorities) is projects may require tidal flow data for sev- made outside the planning process, usually 31 by the finance ministry during the budget role demanded of them in the 1980s and the process. But ministries of finance typically are years beyond. forced to focus on containment of aggregate expenditures and less on their "quality." Their Planning Ministries' Priorities. Formal devel- budget bureaus are usually fully absorbed in opment plans, and the process by which they the annual production of the budget docu- are elaborated, can be useful in charting na- ment; few have succeeded in effectively re- tional strategy and stimulating constructive viewing new development project proposals. dialogue on development issues. But there is Finally, many technical ministries, eager to virtually universal agreement in assessments press forward with their sectoral programs, of planning experiences in Africa (and in other present finance and planning ministries with poor regions also) that the impact of formal faits accomplis-expenditure commitments planning on actual policymaking and invest- made outside the plan, often in collaboration ment programming is slight.' with foreign aid donors. There are many reasons why formal plan- Thus, in government after government, ning has had such limited impact on decision- many investment projects and related claims making, among them: shortages of planners; on resources pass through the public sector's scarcities of data; bureaucratic weaknesses; administrative machinery without proper exogenous shocks; the tendency for plans to evaluation. Improvement of project screening be overambitious; the lack of well-defined institutions and procedures is indispensable projects; and the difficulties of linking me- to better investment programming. dium-term plans to annual budgets. Under- A related weakness in making public de- lying many of these difficulties is the failure cisions is the very limited use of economic to distinguish sufficiently between writing a analysis in policy formulation. Fundamental plan document and actually trying to influ- policy questions such as price-setting for basic ence decisions, a process which means com- staples, wage changes in the public sector, ing to grips with fundamental political and and the granting of tax concessions to new bureaucratic factors. Plan-writing is what most investors are often decided on the basis of inadequate analysis. In part, this shortage of 5. Recent studies of plans from a large number of African policy and program analysis is explained by countries have found that correlations between planned and lack of staff and data. But it also reflects poor actual macroeconomic variables are very weak. These stud- use of existing analysts. In most of the coun- ies conclude that plans have not actually been implemented, and that, according to one observer, "the numbers exercise tries of the region, planning ministries, major which marks the course of macro planning in tropical Africa users of economists, are so absorbed in the appears to be futile." A study of planning in Nigeria in the elaboration and adjustment of medium-term 1960s concluded that it "played a peripheral role in the Government's decision-making processes" and a recent plans that they can give little attention to pol- evaluation of planning in Kenya came essentially to the icymaking. In finance ministries, similarly, same conclusion. See United Nations Center for Develop- staff units capable of providing analytic in- ment Planning, "Implementation of Development Plans: The Experience of Developing Countries in the First Half puts into policy decisions are generally not of the 1970s," Journal of Development Planning, no. 12 (1977). well developed. See also T.Y. Shen, "Sectoral Development Planning in Tropical Africa," Eastern Africa Economic Reviezw, vol. 7, no. STRENGTHENING OF PRIORITY-SETTING 1 (une 1975) and "Macro Development Planning in Tropical AND POLICYMAKING CAPACITY Africa," Journal of Development Studies, vol. 13, no. 4 (July 1977); Edwin Dean, Plan Implementation in Nigeria, 1962-66 Although no continentwide prescription can (Oxford University Press, 1972), pp. 236-7; Eniola 0. Ad- be equally applicable to all countries, the need eniyi, "National Development Planning and Plan Admin- istration in Nigeria," Journal of Administration Overseas to strengthen capacity in project vetting and (London), vol. 19. no. 3 (July 1980); and Wouter Tims and in policymaking is a high priority in all Af- others, Nigeria: Options for Long-Term Development (Baltimore, rican countries. Planning and finance minis- Md.: Johns Hopkins University Press, 1974). For later sim- the keyagencie involed, nee rein- liar conclusions, see Tony Killick and ).K. Kinyua, "On tries, the key agencies involved, need rein- Implementing Development Plans: A Case Study," 0OD forcement if they are to play the important Review (London), no. 1 (1980). 32 planning ministries do, and the fact that many to become true priority-setting agencies for economists and much of the available analytic development expenditures, respected con- capacity have been absorbed in this activity tributors to policy discussions, and the train- helps, in no small measure, to account for the ing ministry for government economists (see shortage of policy-relevant analysis. Box D). Observers and practitioners of planning in developing countries saw in the 1960s and Finance Ministries and Policy Analysis. 1970s that the plan-writing activities of plan- Almost everywhere the financial and budget- ning organizations were without much effect. ary instruments of African public sectors are MvIany then concluded that the thing to do overburdened. More effective execution of was to build up capacity at the levels of gov- development programs and better policy for- ernment where the action took place-the mulation require stronger institutions in this executing ministries. Thus, they recom- area. This is a theme that will reappear in the mended that governments emphasize sector discussions of training in Chapter 6 and tech- planning and project work. But this may have nical assistance in Chapter 9. contributed to a weakening of the project se- Finance ministries also need greater capac- lection process since it left unchanged the ity in policy analysis-for tax planning and "vetting" capacity of the center, while policy, assessment of budgetary requests, and strengthening the capacity of spending min- determination of overall fiscal and monetary istries to generate projects. To the extent that policies. Finance ministries, like planning this occurred, it contributed to a lowering of ministries and other bureaus at the center of the quality of investment programs. government, have special roles to play in The appropriate response now is to rein- economic decisionmaking. They represent the force the central planning agencies, and to general interest in the bureaucratic struggle endow them as quickly as possible with the for resources. They therefore should have the investment evaluation capacities they need. capacity to make credible technical contribu- For countries to which this analysis is appli- tions to policy decisions at both the macro- cable, one implication is that planning min- economic and sectoral levels. The istries should consider changes in their work strengthening of budgetary review proce- priorities. Certain "traditional" planning dures and establishment of policy analysis functions might be given lower priority. Thus, units in finance ministries would help them medium-term planning exercises-now the perform their central role. major activity of most planning ministries- Policy analysis units in the technical min- could be done less intensively without much istries and parastatals would also make large loss of impact. The weight given to other ac- contributions to improved decisionmaking. tivities, and their compatibility with the prior- And two other elements are important. The itv-setti ng functions of planning ministries, first is better data; the pool of information should also be reconsidered-project prepa- available to support policy analysis is very ration, project monitoring, and regional plan- limited. Statistical agencies have come under ning, for example. the same budgetary pressures as other arms In general, it would seem appropriate that of government. Turnover among qualified planning agencies concentrate on two tasks statisticans is extremely high and many work other than investment programming: policy abroad. Many African services are therefore analvsis and the training of economic ana- providing very limited supplies of data. lvsts. Their policy analysis role will be dis- The second element is procedural. Even cussed below. As for training, in most where data are at hand, they are frequently countries planning agencies are well placed not brought to bear on policy decisions be- to take on the critical job of training practical cause the process of making decisions is in- economists for the government as a whole. adequately structured-consultation proce- The vocation of planning ministries, then, is dures may not be well defined, the decision 33 Box D: Botswana's Planning and Budgeting System The Economic Planning Department plays a central technical ministries. This resulted in a steady improve- role in Botswana both in determining the investment ment in both the definition of the sectoral development program and in policy analysis. The most crucial fac- strategies and in project preparation. However, all the tors have been, first, a steadfast commitment of the economic staff employed by the Government belonged President and senior ministers to the planning process to a single economist cadre with the Director of Eco- and, second, the full integration of the planning and nomic Affairs responsible for assignments and career budgeting system, which has ensured that Develop- development. ment Plan priorities were respected in both recurrent The Department of Economic Affairs has partici- and capital budget decisions. pated actively in setting overall and departmental bud- Throughout the 1970s, when Botswana achieved a get guidelines and in projecting government revenues. real annual GNP growth rate of around 14 percent, the It has also been responsible for negotiations with ex- Ministry of Finance and Development Planning came ternal aid agencies and for preparing and monitoring directly under the control of the Vice-President. The the capital budget. In addition, the Department has Plan was regularly updated by the central planning initiated policy papers and a broad range of economic organization (the Department of Economic Affairs), and issues including fiscal reform, incomes and employ- a new Plan document, incorporating a rolling three- ment policies, customs union and tariff negotiations, year public investment program, was issued promptly and monetary matters. Planning staff have also been every two or three years. After approval of the Plan nominated to the boards of the key parastatal bodies. by Parliament, the Govemment was bound by legis- Finally, close links have been forged with the Central lation not to finance projects which had been excluded. Statistics Organization, which also comes under the The Plan spelled out in some detail the macroeconomic Minister of Finance and Development Planning. and sectoral objectives and the policies to be adopted In summary, the successful impact of planning in for their achievement; it was essentially a pragmatic, Botswana in helping to rationalize economic manage- action-oriented document firmly rooted to ongoing ment and to darify development objectives may be operations and elaborated with the close involvement attributed to: a strong political commitment to prag- of the technical ministries. matic planning and economic rationality in govern- When economic planning was initiated in Botswana ment decisionmaking; the close integration of planning, in 1966 at the time of independence, only two profes- budgeting, and economic policy formulation under a sional economists were employed throughout the Gov- powerful senior minister; the recruitment of a highly emnment. The central planning unit in the Ministry of competent economist cadre to staff the planning or- Finance was first built up, and available competent ganization; focusing planning efforts first and foremost staff were concentrated there to vet the list of invest- on determining the short- and medium-term public ment projects submitted 'by the technical ministries. investment program which was closely adhered to by As more staff were recruited (both locally and inter- the Cabinet; establishing a strong policy analysis ca- nationally), economic planners familiar with the pro- pability; and the continuous and active involvement cedures were reassigned from the Department of of the planning staff in budgetary and economic man- Economnic Affairs to the small planning units in the agement decisions. meetings may be called too quickly, staff objective; the Economic Commission for Af- memos may not be called for, or may arrive rica is giving systematic attention to the prob- too late for consideration, and so forth. These lems involved and action required. Donors are, of course, general problems in develop- should give sympathetic support to proposals ment administration; their solution requires aimed at institution-building in this area. long-term institution-building efforts. Donors should also recognize their special priority-setting responsibility in the African DONOR CONTRIBUTIONS context. Official development assistance pro- Most of the implications for donors of the vides over a third of total public investment above analysis refer to training and technical in this region, and more than that in the poorer assistance for the building of stronger deci- countries. Donors therefore play a major role sionmaking institutions-planning agencies, in project selection, and in shaping the pro- budget bureaus, policy analysis units ject selection process. In their own opera- throughout the public sector. The building of tions, donors could give closer attention to solid statistical infrastructure is a long-term the development of orderly local decision- 34 making procedures and institutions-a posi- possible opportunities for economizing on the tive force for long-term development. scarcity of experienced managers required for Finallv, donors could do more than they running large organizations. In particular, have in the past to associate local analysts organizational formats for the public and pri- with their own technicians in their sectoral vate sectors that are smaller and more easily and macroeconomic work, and in project ap- managed, and are, therefore, more efficient praisals. The World Bank, in particular, should need to be encouraged. The further devel- regard its economic and sector work as major opment of cooperatives is one such possibil- vehicles for training local people, as well as ity. The development of parastatals which are for gathering information useful to the Bank. smaller and, perhaps, answerable to local governments, is another. More extensive re- (.'0?Zn 'i: YfAac7N'11lC1t2 liance on the private sector is a third option. Economic progress in any society requires that PRIVATE SECTOR resources be used efficiently by organiza- This third option is in danger of being insuf- tional units in both public and private spheres. ficiently examnined and developed by African This applies not onlv to the production of governments. The reasons for this are many goods in agriculture, industry, and mining and varied. One of these reasons is a fear that but also to the provision of services, such as a greater reliance on the private sector will marketing, transport, and health care. This run contrary to other objectives of some Af- requires that organizational structures and the rican governments-in particular, a concern important role of management be continu- over the equitable distribution of income and ously examined (see Box E). improvement of the plight of those in abso- Agriculture is the most vital sector in Sub- lute poverty. However, achievement of these Saharan Africa. Chapter 5 will emphasize the objectives is heavily dependent on higher rates central role that will have to be played by of growth of output in the private sector, par- smallholders if agricultural growth is to be ticularly by small farmers, who are the largest stimulated. All the evidence points to the fact occupational group in tropical Africa. that smallholders are outstanding managers Possibilities for emerging private sector op- of their own resources-their land and capi- tions are discussed in later chapters. The fol- tal, fertilizer, and water. They can be counted lowing are indicative. on to respond to changes in the profitability of different crops and of other farming activ- Agricultural Marketing. As discussed in ities (e.g., dairying). African farmers can Chapter 5, rural areas are frequently served "manage" the use of resources in the agri- by de jure public monopolies that sell agri- cultural sector to achieve the objectives of cultural inputs, market outputs, and often higher food output and export growth, if the even monopolize the sale of consumer goods. price, tax and subsidy structures are provid- One justification for these agencies is that they ing adequate and appropriate incentives, and prevent exploitation of farmers by unscru- input and output markets are efficiently op- pulous private traders. To the extent that this erated and organized. is a problem, it is not solved by replacing The incontrovertible evidence that small- private monopoly with public monopoly but holders are excellent managers has wider im- rather by increasing the competitiveness of plications for the organization of production markets. The best way to increase competi- activities in Sub-Saharan economies. For his- tion is to encourage private trading. In fact, torical reasons, there are too few experienced private traders now handle most trading ac- managers for large organizations. Their ranks tivity almost everywhere, but often in semi- are now being augmented by training and by legality. Legalization and encouragement of the recruitment of foreign managers. How- private trade would reduce costs and uncer- ever, equal attention needs to be given to tainties in the market environment, relieve 35 Box E: Macroeconomic Indicators, Economic Management and Growth Statistical evidence from a number of African coun- ilar to those of the fast growers, they would have tries suggests that the quality of management of the grown fairly rapidly, at more than 4.5 percent an- economy probably had a significant impact on eco- nually. nomic growth in the last decade. What can explain the disparity in IcoRs? Poor climate The table in this box compares macroeconomic in- (drought), political upheaval, or severe changes in the dicators for eleven African countries-five of which income terms of trade often account for high ICORs and grew rapidly in the 1970s (6.6 percent per year), and low growth. But none of these factors operated more six of which grew more slowly (1.0 percent per year). strongly in one group of countries than in the other Both groups of countries exhibit similar tax ratios and during the 1970s*. Better public management of the used nearly the same percentage of GDP for govemment economy in general and of government resources in expenditure. Although the high-growth countries in- particular has probably contributed to the lower ICORS vested half again as much as a proportion of GDP as of the fast growing countries. Better management may did the slow growers, this only partly explrins why also have had some impact on the export performance the former grew six times faster than the latter during of the two groups: the fast growing countries increased the last decade. Of far greater importance is the fact their volume of exports, on average, 2.5 percent an- that, on average, only $3 of investment was made for nually, while the average volume of exports for the each additional dollar of output in the fast growing slow growers actually declined 1.7 percent per year. countries, while in four of the slow growers, an av- erage of $16 was required to produce each additional dollar of output. An in two slow growing countries, *None of these countries experienced serious civil strife nonwithstanding substantial investment, output ac- or, as was the case in Zaire and Zambia, extremely tually declined. These relations are commonly referred adverse terms of trade. Two fast growing countries, to as incremental capital output ratios ([coRs)-the in- Botswana and Swaziland, have been excluded because crease in investment divided by the increase in output. of their atypical integration with the South African Had the slow growers been able to generate ICORS sim- economy. Macroeconomic Indicators for Fast and Slow Growing Countries in the 1970s (ratios as percentages) Average annual Gross Growth of growtiz rate Ratio of Ratio of govern- domestic exports of GDP taxes to ment expenditure investment to (volume) Country 1970-79 GDP 1973-77 to GDP 1973-77 GDP 1970-77 ICOR. 1970-79 High-growth countries 6.6 15.8 21.0 21.0 3.2 2.5 Mauritius 8.2 18.6 23.8 240b 2.9 Ivory Coast 6.7 20.6 24.0 19.0 2.8 5.2 Kenya 6.5 15.3 20.6 21.0 3.1 - 0.5 Malawi 6.3 11.3 21.3 22.0 3.4 4.6 Cameroon 5.4 13.0' 15.3d 19.0 3.7 0.5 Low-growth countries 1.0 15.4 19.9 14.3 16.2 -1.7 Senegal 2.5 17.7' 19.4 15.0 6.0 -0.8 Sierra Leone 1.6 15.0 23.7 13.0 8.2 -6.5 Liberia 1.8 21.2d 24 9d 19.0 10.6 2.3 Ghana -0.1 10.4 18.1 11.0 -7.2 Upper Volta -0.1 12.9 13.7 16.0 3.1 Madagascar 0.3 15.2d 19.7' 12.0 40.0 -1.0 Note: Country-group averages are unweighted. c. 1974-77. a. The incremental capital output ratio (iCOR) is defined d. 1974-78. as the increase in investment divided by the increase in e. 1975-78. output. f. 1972-73. b. 1970-75. the marketing parastatals of their most diffi- Transport. Transporthastraditionallydrawn cult tasks, and in some countries reduce on the resources and energies of small entre- budget drains arising from the deficits of grain- preneurs in Africa. Recently there has devel- marketing agencies. oped a tendency by governments to exclude 36 private buses or vans and to introduce or ex- countries demrand for drugs is satisfied, very tend monopoly rights for urban bus compa- imperfectly, in black market transactions. A nies, almost always parastatals. There is also legalization of private drug sales would cer- a tendencv to discriminate againstprivate truck tainly result in lower prices and steadier, more operations in the allocation of foreign ex- widely dispensed supplies-though it would change, where public and private transport entail some risks. Rural areas in particular structures coexist. The experience in this area would be better served (see Chapter 6 for a is unambiguous. Private trucking is dynamic further discussion of this point). and highly competitive; it tends to offer ef- In market-oriented countries this approach ficient and low-priced services, especially to could be developed further. For example, disadvantaged groups. In urban areas, the training paraprofessionals in health and san- experience in cities such as Abidjan and Nai- itation, and encouraging them to set up on robi offers a revealing contrast. In Abidjan, a their own, would limit the public sector role public sector bus monopoly was imposed in to training, certification, and supervision. Ru- the mid-1970s, leading to a tripling of the bus ral health workers would then circulate, or fleet andI a heavy deficit, despite the fact that establish themselves in the villages, and pro- the system serves only close-in populations. vide services for fees. In contrast, the public- In Nairobi, private vans were allowed to com- sector strategy (village health workers serving pete with the city's bus company, with the a group of villagers) involves great financial result that the entire city is efficiently served and logistic difficulties, thus far unresolved with transport facilities, including a public bus in most of Africa, or indeed elsewhere. This fleet, w;hich has remained small in size and example serves to illustrate a general point: also profitable.6 much of the administrative burden in the public sector arises from the need to organize, Civil Vorks. No better training exists for motivate, and control people; and this burden construction industry entrepreneurs than the could be reduced if the private sector were use of small contractors, and it is highly likely allowed a greater role under government that greater efficiency will come from con- leadership. tracting out such activities as road main- This kind of redefinition of the frontier be- tenance and building repair. There are, of tweenpublicandprivatesectorswill, of course, course, problems-in particular, the devel- be of varying acceptability or attractiveness to opment of simple contract-letting procedures African governments. It should be under- and controls. But these problems are easier scored that the question is not one of radical to solve than are those involved in govern- shifts in the social division of labor, but rather ment maintenance. of marginal changes. By shifting some activ- ities to private hands, significant gains in out- t ru distribution. In mos ican cu- put may be possible with relatively little strices ra dtribuio of dru iegally re- sacrifice of sociopolitical objectives. It is for stricted to the public health care network. But ti esnta nagmn ftesoeo , . , , ~~~~~this reason that enlargement of the scope of for reasons of budget scarcity, logistics, and private activity along those lines has occurred difficulties of supervision and control, in many I in recent years in many places-among them, countries these legal sources are commonly China, Hungary, Mozambique, Zaire, and without drugs for much of the year. In such Guinea-Bissau. THE PARASTATAL SECTOR 6. For an examination of alternative approaches to urban passenger transport, see A. A. Walters, "Costs and Scale The parastatal sector grew rapidly after in- of Bus Services," World Bank Staff Working Paper, no. 325 dependence. Before then, African participa- (Washington, D.C.: World Bank, April 1979). This analysis tion in the modern private sector, especially concludes that minibuses operated bv small firms or owner- drivers are generallv more efficient in meeting urban trans- industry, was very rare. Following indepen- port demands than larger buses and bus companies. dence, change took two main forms: nation- 37 alization of existing enterprises, which were ernment sector, so procedures end up very almost invariably foreign owned, and invest- much like those of general government ment of a very substantial share of govern- administration. In addition, parastatals are ment resources in parastatal enterprises in often undercapitalized and starved of work- transport, public utilities, and manufacturing. ing capital. And, most critical, policymakers The hope was that these public enterprises have found it difficult to accept a "nonpolit- would be self-financing, rapidly expanding ical" role for parastatals. They are pressured institutions. They would not only generate to increase employment, to deliver outputs at surplus for additional investment, but could low prices to key groups, and to shape in- also plav an important role in modernization vestment decisions other than with economic by developing skilled labor and enhancing and financial returns in view. managerial capacity. In the parastatal sector, what is needed is These expectations have not been realized. straightforward, at least on a general level: With the exception of the mineral-exporting * a clear definition of objectives and terms parastatals and some of those trading in ex- of reference; port crops, public enterprises have thus far . an explicit understanding between govern- caused serious fiscal burdens. They do not ment and the parastatal entity defining the pay taxes. Most of their investment costs are annual financial and production plan within covered by transfers (from government bud- the framework of agreed long-term objec- gets, the banks, or marketing organization tives. The "contract plan," as it has evolved surpluses); in some cases their cash surplus in France and some francophone countries, is less than their depreciation; and in a few is an example (see Box F); instances cash flow does not even cover run- * an incentive system conducive to efficient ning costs. A number of the manufacturing performance; parastatals-and mixed public-private enter- * independence in day-to-day management; prises-are moderatelv profitable. But this is * independent personnel management; usually because they enjoy very high levels * proper accounts and records; and of protection from the world market, explic- * acceptance of the principle that under cer- itlv in the form of a heavy duty on competing tain circumstances liquidation of an enter- imports, or implicitly because components are prise may be desirable. imported duty free. In many cases their value Reform of parastatals, particularly in the added at international prices is but a fraction industrial sector, has far-reaching develop- of their value added at domestic prices; in mental significance. Economic development some cases value added may even be nega- inevitably implies that the nontraditional sec- tive. In general, because the parastatals in the tors grow fastest: manufacturing, public util- commercial sectors generate so small a sur- ities, transport, and modem services. In Africa, plus, their grow.th has been limited by the these are predominantly in the parastatal sec- availability of the resources they can com- tor. Unless there is a change in the operating mand from governments. effectiveness of parastatals, particularly in in- There are many reasons for this poor per- dustry, they will not take their proper place formance. Managerial and technical capacities as growth points. They will, instead, continue are not easily established even though para- to be fiscal drains and major contributors to statals theoretically have flexibility in their the slow growth. Reform of the parastatals is, abilitv to operate and, specifically, to hire staff therefore, not just a question of increasing autonomously. Managers and technicians fre- resources available to government in the short quentlv come from the civil service. Budget run, important as that is; it is far more fun- and account systems tend to be taken from damental, involving the whole question of the government administration, and budget long-term growth prospects. analysts, auditors, accountants-few in num- Since improved performance in the para- ber anyway-also tend to come from the gov- statal sector is so important for faster growth 38 Box F: Public Enterprise Contracts in Senegal One promising approach to improving operations of the role of subsidies and tariff increases in company op- parastatal enterprises consists of performance contract- erations; and (3) a reduction in the diversity of com- ing between govemment and individual firms. The pany objectives. Third, the negotiation of company enterprise contract is a negotiated agreement between contracts should be supervised by neutral government government and a parastatal company describing the staff who report to the central political authority. Thus, objectives govemment assigns to the parastatal as well conflicts between companies and government author- as the resources it will provide and the degree of con- ities can be resolved by the President or the Prime trol which it will exercise. The company, in turn, prom- Minister. Moreover, once a company contract is ne- ises certain results; performance is to be judged by gotiated these neutral staff must supervise its imple- indicators mutually accepted over a given period. mentation and expose failures by any of the parties to In Senegal, such contracts were first prepared in the fulfill their obligations under the contract. transport sector (bus, airline, and railroad companies), The enterprise contract allows political choices to be in which quantification of objectives, means, and per- made in full awareness of their costs: excessive em- formance is relatively easy. Negotiations started in 1981 ployment, or selling output below costs will be less on company contracts between government and re- likely to persist if they are discussed in a forum of gional development agencies. Hiere quantification is senior government officials and management and if more complicated, since performance depends in part they are subject to codification in a contractual rela- on exogenous factors such as climate and world market tionship. Government leaders are usually reluctant to prices for agricultural products. give enterprise managers autonomy in crucial deci- Senegal's experience has already produced severat sions such as prices of output, employment, wages, useful lessons. First, effective use of enterprise con- or investments, because they fear substantial diver- tracts requires strong support at the highest levels of gence between the goals of the managers and those government: individual ministers and senior civil ser- of the community. But a contractual relationship, based vants fear losing control over public enterprises, in- on clear goals and criteria for good management per- cluding the power to make fairly routine decisions. fonnance, would permit reduction in the detailed con- Second, company contracts are enthusiastically en- trols of ministries and other central authorities. This dorsed by company managers, who see them as a would conserve their valuable time and energy, while means of obtaining: (l) clear objectives on which their giving the enterprises the autonomy needed to operate performance can be judged; (2) a precise definition of effectively. and increased access to services, ways to to suffer from similar constraints-of short- achieve it wvarrant close attention by govern- ages of managerial skills combined with over- ments and donors. The most common ap- manning, of wide responsibilities combined proach to parastatal reform in Africa as with inadequate administrative structures and elsewhere involves the appointment of one- insufficient resources. time studv commissions. While useful, this Effective administrative reform is noto- approach has a number of limitations, most riously difficult to accomplish: nevertheless, notablv wvith respect to implementation of there are four steps which can now be taken recommendations. A more continuous and to accelerate the process of building public longer-termn approach might be more produc- organizational capacity and to begin correct- tive: for example, the establishment of high- ing some of the anomalies which have arisen level organizational reviewbodieswhichcould in the decades of rapid administrative growth. do ongoing studies and provide continuous Such measures are largely independent of advice to governments. Similar units, strate- particular policy choices about privatization, gicallv placed (e.g., attached to the chief of since the central responsibilities of govern- state's office) have developed useful ap- ment for economic development and human proaches to reform in some countries, as has welfare will remain, even though the exercise been the case in Senegal. of particular functions may change. In many African countries the parastatals First, there is a need to improve the cost- present mereiv the most urgent and visible effectiveness of government manpower, par- problems of public sector performance. The ticularly near the top (where managers need traditional ministries and agencies, too, tend longer tenures and more delegated authority) 39 and near the bottom (where the presence in not only from expansion of government per the field of large numbers of thinly spread se, but also by extension of the state into agencies often actually detracts from the pop- commercial or productive activities-manu- ulation's effective access to productive and facturing, mining, transport, marketing-ac- welfare services). tivities which were largely in private hands Second, delegation and consolidation of before independence. A recent World Bank service output points require better defini- study, limited to seven countries for which tions of who is accountable for performance, data were available, showed that the public and an improved structure of incentives, to sector now employs between 40 and 74 per- reward that performance rather than mainly cent of those recorded in paid employment to recognize hierarchy and length of service. and that public sector employment has grown Third, resource constraints demand novel much faster than that of the private sector approaches to community involvement in (see Table 4.1). The growth of the govern- service provision in cooperation with the ment sector alone has also been very rapid. administration-with the implication also that As indicated in Table 4.2, spending on public bureaucratic agencies will, in the long run, administration and defense, which are (with be answerable for their performance to their education) the largest components of the gov- "clients" in the community as well as to their ernment sector, has grown far more quickly administrative heads. than national output. And as will be shown Finally, while initiatives in some of these in Chapter 6, growth in expenditure on public areas are alreadv in existence in some coun- education has also been extremely rapid. tries, there is a great need for high-quality The fast growth of the public sector since analysis and prescription suitable to each independence is also indicated by the high countrv's conditions-a need which is seldom share of output commanded by governments. satisfactorily met by periodic review. It may, This feature is not reflected in the conven- therefore, be advisable to extend the respon- tional ratio of tax revenue to GDP. For Sub- sibility of the high-level public sector units Saharan African countries these ratios are discussed above to include the problems of similar to those found in other low- and mid- the public administration itself. dle-income countries-a median ratio of 15 percent for low-income countries both in Af- rica and in the rest of the world and 21 per- cent for middle-income countries in Africa uw Si'e .t Uoucr?2UlL?U compared with 18 percent in the rest of the There is a special set of concerns that relates world. However, many African countries also to the growth of those govemment-adminis- generate substantial surpluses from govern- tered activities which make heavy and in- ment-owned mineral complexes or agricul- creasing demands on financial resources. First, tural marketing parastatals. In the late 1970s, the mobilization of resources for these activ- for example, Madagascar, Malawi, and Ivory ities precludes their use elsewhere in the Coast all generated financial surpluses from economy. Second, in order to finance these public agricultural marketing organizations that activities, governments must generate reve- exceeded a quarter of tax revenues. Chapter nues-primarily through -.axation-which wil 5 shows that "taxation" of farmers is rela- have adverse effects on production in the sec- tively high; levies of 40-50 percent are com- tors that are taxed. monly imposed on export crops (in addition to any "taxation" resulting from an overval- ued exchange rate or from an inefficient mar- GROCsTH SINCE INDEPENDENCE keting system). During the past 20 years the public sector has On top of their considerable access to do- greatly extended its economic role in Africa, mestic resources, African governments also as it has elsewhere. This growth has come receive sizeable aid flows-excluding the oil 40 exporters, more than 6 percent of GNP in 1979, quarter and, in many cases, even 40 percent and equal to 38 percent of gross domestic Of GNP. This is a high share compared with investment. Total resources marshalled by other regions. In India, for example, in 1977, governments, therefore, typically exceed a taxes of central and state governments com- Table 4.1. Public Sector Employment and Growth Rates, Selected Countries "Formal" Public "Formal" employment employment employment as a as a growvth rates percentage of percentage of (percent) working age "forrmal" Public Private Total Country Year population employment Ghana 1957 51.449-2223 1972 10.1 73.9 499 -2.2 2.3 Tanzania 1962 27.0 10.7 -4.8 2.3 1974 6.3 66.4 Zambia 1976 14.2 71.5 Ivorv Coast 1970 10.2 Kenya 1963 29.6 6.0 2.4 3.7 1977 12.4 41.7 Malawi 1968 33.49. 0 6 1976 9.6 39.2 8.0 8.6 Uganda 1962 41.8 4.0 4.8 4.5 1970 5.9 42.2 Sou.rce: World Bank data files. Table 4.2. The Growth of Public Administration and Defense relative to GNP Expenditure on puiblic Average annual growth rates. 1970-79 administration and defense as Expenditure on public percentage of GDP administration Expenditure on public coulntries" (annual average, 1970.79)b and defense relative to GDP administration Mauritania 18.0 13.9 1.8 12.1 Kenva 17.1 8.4 6.5 1.9 Sudan 16.1 4.5 4.3 0.2 Tanzania 14.9 10.2 4.9 5.3 Chad' 14.2 6.1 -0.2 6.3 Somalia' 13.7 6.5 3.1 0.2 Congo 12.7 2.3 2.9 - 0.6 Botswana' 12.1 16.3 13.5 0.5 Madagascar 11.5 0.2 0.3 -0.1 Benin 11.0 4.8 3.3 1.5 Rwanda' 10.3 6.5 4.1 2.4 Lesotho 9.2 3.3 7.0 -3.7 Upper Volta 9.1 6.2 -0.1 6.3 Uganda 9.1 5.9 -0.4 6.3 Sierra Leone 8.3 9.6 1.6 8.0 Ivory Coast 8.1 8.9 6.7 2.2 Liberia' 8.0 6.7 1.8 4.9 Cameroon, 7.1 6.2 5.4 0.8 Nigeria 6.4 13.3 7.5 5.8 Mauritius 4.7 12.9 8.2 4.7 Burundi 4.4 5.9 3.0 2.9 Mean (unweighted) 9.9 7.4 3.9 3.3 a. In descending order of proportion of expenditure devoted to public administration and defense. b. Value added. c 1970-78. d. 1972-79. Source: World Bank data files. 41 bined came to 15 percent of GDP, foreign aid much-needed reduction in anti-trade and to 1.1 percent, and expenditures by central anti-agriculture bias in African tax structures. and state governments to 20 percent. African In most countries these sectors are taxed less governments, therefore, have been very suc- heavily than agriculture, minerals, and in- cessful in mobilizing economic surplus for dustry. A shift toward heavier taxation of public uses. nontradeables would also give African tax The question of a large and growing vol- systems more "buoyancy," that is, more au- ume of national resources being allocated to tomatic increases in revenues would accrue government uses would be a matter for ex- as output increases. amination by each government, even if na- Sales taxes, for example, are very buoyant tional output were growing rapidly. However, because their base rests on the rapidly ex- with output rising so slowly, it is critical to panding sectors in the economy. Properly compare the resources being allocated each designed, sales taxes are much simpler to ad- year to basic government services (defense, minister than property and income taxes. A public administration, education, health, and single-stage ad valorem sales tax collected at highways) with the use of resources to sup- the point of local manufacture is especially port, say, agriculture or industry. Further- suited to African circumstances. In some more, it becomes important to reexamine the countries (Nigeria, Gambia, Somalia, Sudan) financial cost of providing government serv- sales taxes have yet to be introduced and in ices and, most especially, the wage and salary many others they generate very modest rev- levels in the government sector. enues, usually because of failure to exploit the tax base fully. In contrast, Kenya, Malawi, RF' ENLE GENERAXTION Niger, and the Congo all generate more than The revenue genieration problems posed by a one fifth of their taxes through the contri- growing government sector also need to be bution of sales taxes alone. examined. Once again, in a rapidly growing Excise taxes also have the advantages of economy, revenues generated by a given tax ease of administration and a base that in many svstem would tvpically be sufficiently buoy- cases grows more rapidly than national out- ant to generate adequate additional revenues. put. They can reinforce a more general sales In the African situation of slow overall eco- tax by selective taxation of luxuries. For in- nomic growth, the danger has arisen that the stance, Sierra Leone imposes excises on to- additional revenues required to finance in- bacco, petroleum products, beer, stout, and creased government expenditure can only be confectionary products, which amount to 15 generated by introducing new tax measures percent of total revenues. In Zambia, the tax which then become a further drag on output on beer alone came to 18 percent of the total growvth. This is particularly worrisome in the taxes in 1976. The excise on motor vehicle case ot agriculture, as discussed in Chapter fuels deserves particular attention. Adminis- 5. The answver to the revenue problem is to tratively it is ideal since it is easily measured find forms of taxation that impose less of a and collected at low cost, either at the refinery burden on the "motor" sectors of the econ- or the point of import. The tax has been very omy as well as to check expenditure by using buoyant, it is progressive, and it is an im- fees and charges to-a greater degree. portant component of energy policy. Many The development of sales and excise taxes African countries have yet to respond fully to on production which does not enter into in- the new economics of very expensive energy; ternational trade (services from hotels, res- compared with taxation in Western Europe, taurants, transport, banking, insurance, for example, motor fuel taxes in Sub-Saharan power,construction),vwhichaccountsformuch Africa are low. In 14 African countries for of-the two thirds of value added generated whichdataareavailablefroml979,taxesadded outside of agriculture, would contribute to a 41 percent on average to the retail price of 42 gasoline, compared with 125 percent in France, * In projects involving primary health care, Italy, and the Federal Republic of Germany. a number of countries are experimenting Simplv replacing the specific taxes on petro- with village revolving funds to finance leum, which are now used in many African medicines and other supplies. countries with ad valorem taxes at a fixed * Some governments continue to encourage rate, would ensure a very buoyant tax in the private sector activities, including those of future. nonprofit organizations, in such areas as education and health. Private schools, and PAYING FOR BASIC SERVICES especially clinics, are active in many rural One major goal of African as well as of other communities. goveomake basic services (sani- * In some countries villagers provide both gatiovenmsafenwatse todern medical care, and classrooms and teachers' houses. In others, tation, safe water, modern medical care, and village pumps and wells are constructed education) available to all, although at pre- only if the village covers a large share of sent these services are enjoyed by few, usu- the costs of installation. ally the minority living in cities. This outcome * Housing projects often include provision w\ as inevitable in that the governments have for payment in installments by the serviced followed the current worldwide practice of sites, and users of irrigation and household concentrating on expanding free services water in some projects are charged accord- through the use of public funds and public ing to the amount used. agencies. The public funds required for the . In Swaziland "study-loans" at the univer- provision of services in this traditional fash- sity level cover all costs to the student- ioni e\ceed bv far what governments are able tuition, board, books-and include a cash to generate now or are likely to generate in allowance. In the first five years after em- the next several decades. ployment begins, the student repays half l'ublic services are highlv valued. African the study-loan. A government bureau is used to recover costs. consumers are willing to give up other ex- penditures to buy such services as bus trans- These examples do not suggest that serv- port, irrigation water, health care, and ices should be provided solely for payment. education. But because many of these serv- In many instances general welfare is also sat- ices are regarded as something only govern- isfied by services which benefit individuals. ment should provide and should pay for, It obviously makes a great deal of sense to alternative modes of supply and of financing subsidize services such as preventive medical have remained unexplored. The outcome is care and primary education. paradoxical. While it is widely believed that The biggest barrier to cost recovery is goodsofsuch fundamentalimportanceshould administration. Just as many taxes are ad- be supplied without payment, governments ministratively too demanding, so too have cannot afford to provide them; therefore, in various cost recovery schemes failed because the face of substantial demand, the services of unsolved administrative problems. The languish accordingly. World Bank's recent urban sites and services It is clear that the only hope of broadly projects in several African countries illustrate based provision of services in a self-reliant both the difficulties and approaches to their Africa is through greater emphasis on charg- resolution. These projects involve direct pay- ing beneficiaries for the services they receive. ment by beneficiaries for services (water, roads, It is also clear that there is an increasing in- street lighting, refuse collection) and goods terest among African countries in examining (house construction or business loans) pro- alternative approaches, all of which would vided under the project. An organization had conserve government resources. The follow- to be developed to bill clients and handle re- ing possibilities should be among those more ceipts. Sanctions-such as site repossession, widely considered: or disconnection of services-had to be im- 43 posed on those of the clients who did not lend themselves to decentralized, self-financ- meet their payments.7 ing approaches. But they need a long time to The development of adequate collection evolve into low-cost replicable programs. Do- organizations and the willingness of benefi- nors could play a substantial role in fostering ciaries to pay depend on the degree of sup- such experimentation, both by providing re- port from political leaders and administrators. sources to support new approaches and by Continuing experimentation is needed to de- assuring support for long enough to see which velop approaches that conserve fiscal re- of these approaches work. sources while expanding the provision of basic Donors can also play a basic role in de- services. Provision of primary rural health care signing projects so as to emphasize user through reliance on paramedicals funded by charges and cost recovery. Until recently, too village revolving funds, village water supply, little attention has been given to this issue, veterinary services, and education-all may which is now an important matter for dis- cussion. Donors and governments together X Similar problems arise in student loan programs, which can bring about a wider awareness that user are an indispensable accompaniment to higher fees at sec- ondarv schools and universities. These programs every- charges are a desirable means of achieving where run into collection difficulties. the governments' own social goals. 44 AC d-v§' TDLUN Agriculture is at the heart of African econo- than the rate of population growth almost mies. Most of the population earns its live- everywhere; lihood from agriculture. Officially its share of * agricultural exports stagnated, and African GDI' in most countries of the Sub-Saharan re- shares in world trade declined for many gion is between 30 and 60 percent. But this commodities; is an underestimate, since agricultural output * food production per capita was at best stag- is valued at the prices governments pay to nant in the 1960s and fell in the 1970s; producers, which are below export or import * commercial imports of food grains grew paritv prices, and since the value of produc- more than three times as fast as popula- tion in the secondary and tertiary sectors is tion, and food aid increased substantially; overstated due to subsidies and protection. and Moreover, the transport, processing, and trade * more of the population shifted its con- sectors depend on the production of agricul- sumption to wheat and rice (as evidenced tural commodities, and incomes earned in this by the soaring imports of these food- sector provide markets for domestically pro- grains), which increased food dependency ductod grood and services,and created in many countries a mismatch between local production possibilities and Thus, agricultural output is the single most consumer demand, since wheat and rice in important determinant of overall economic these countries can only be grown at costs growth and its sluggish record of recent years far above import parities. is the principal factor underlying the poor economic performance of the countries of this OVERALL PERFORMANCE region. For this reason, growth-oriented pol- icies for this sector are crucial for improving overall economic performance. This chapter volume by 2.3 percent a year, or roughly at reviews trendsover the last 20yearsand then the same rate as population growth. In the sets forth the main elements of a production- 1970s, however, production dropped to about focused rural development strategy. 1.3 percent a year,2 while population grew at about 2.7 percent. In the low-income and oil- exporting countries, agricultural growth was -, reu)z,'s :s A' ! rlEz Dtve?o?ment, the slowest-about 1 percent annually-while ' ) !)-,>, ? the middle-income nations achieved a growth rate in line with that of rural population (which The crisis in African agriculture is reflected in five trends which evolved over the past two decades: 1. Crops and livestock. 2. See the Statistical Annex for a discussion of some of * the growth rate of agricultural production the problems involved in estimating average growth rates began to decline and, in the 1970s, was less in agriculture. 45 Table 5.1. Growth of Agricultural Exportsa Annual growth rates (percent) Sub-Saharan Oil-exporting Oil-importing Africa countries cozuntries 1961-63 1969-71 1961-63 1969-71 1961-63 1969-71 to to to to to to Export 1969-71 2977-79 Z969-71 1977-79 1969-71 1977-79 Thirty main agricultural exports Volume 1.9 -1.9 -0.7 - 8.6 2.6 - 0.7 Unit value 223 16.2 3.1 16.8 2.1 16.1 Value 4.3 14.0 2.3 6.8 4.8 15.3 Value of other agricultural exports 4.6 8.9 33.4 -1.6 3.4 9.8 a. Crops and livestock. Sourcas: World Bank data and FAo Tradle Yearbook tapes. is estimated to have increased by a little more marked drop in response to shrinking world than 2 percent a year). All groups of countries demand. Cattle and meat exports grew steeply registered declining per capita production, in the 1960s, but, following the 1972-74 West however, and only eight out of the 39 coun- African drought, failed to regain their pre- tries for which data are available (which ac- vious levels. Timber also enjoyed a brisk in- count for less than 15 percent of the region's crease in the 1960s, but then stagnated in the population) showed rising agricultural pro- next decade. Perhaps the most spectacular duction per capita (see Table SA.25). decline in exports was in oilseeds-especially While production statistics are highly ten- groundnuts-which was due to increased do- tative (particularly for subsistence foods) and mestic demand (in Nigeria, in particular), the must be regarded with caution,3 other evi- disintegration of Zaire's rural economy, a dence substantiates the poor performance switch from groundnuts to cereals in a num- in this sector: the mounting domestic food ber of countries (Mali, Niger, and Nigeria), prices in most countries; the steep rise in cer- and the spread of "rosette" (a plant disease). eals imports; and the export crop figures, As a consequence, Africa's share of world which are more reliable than production sta- trade declined for most of these commodities. tistics and indicate a substantial decline (see While world trade in those commodities ex- Table 5.1). ported by the Sub-Saharan countries grew in volume by 1.8 percent a year, and 3.3 percent EXPORT PRODUCTION in value (constant prices) over the two de- By the end of the 1970s, agricultural exports cades, the growth rates of exports from Africa ~~~~~~~~~~ I were zero and 1.8 percent, respectively . vere no greater than in the early 1960s. In fact, a modest rate of increase of 1.9 percent FOOD PRODUcrioN a year in the 1960s was offset bv an equal decrease in the 1970s. In terms of volume, the The key crops produced for domestic con- onlv crops registering gains were coffee, co- sumption are millet and sorghum in the Su- coa, tea, sugar, and cotton.4 Sisal suffered a dan and Sahelian countries, maize in Eastern and Southern Africa, rice in Madagascar and 3. Limitations of agricultural data are discussed in the Statistical Annex. coffee in the total value of agricultural exports increased 4. In terms of prices, coffee and cocoa recorded substan- from one third to nearly two thirds between 1961 and 1979. tial real gains over the two decades, rising by some 5 percent 5. This applies to the 30 main agricultural exports (crops annualiv relative to import prices, while the relative prices and livestock), accounting for more than 90 percent of the (net barter terms of trade) of the other export crops (when total. For the volume of total exports, the rate of increase taken together) fell. Consequently, the share of cocoa and was 1.6 percent a year. 46 Table 5.2. Growth of Production of Selected Food Crops (in volume) Annutal growvth rate (percent) Sub-Saharan Oil-exporting Oil-importing Africa countries countries 1961-63 1969-71 1961-63 1969-71 1961-63 1969-71 to to to to to to Crop 1969-71 1977-79 1969-71 1977-79 1969-71 1977-79 Cereals Rice (paddy) 4.0 2.9 6.3 10.7 3.8 2.2 Wheat 3.8 -0.2 -1.2 - 0.6 4.0 - 0.2 Maize 5.2 1.3 5.1 0.9 5.2 1.4 %lillet-sorghum 0.9 1.0 0.2 0.7 1.3 1.2 Roots and tubers 2.0 1.8 2.3 1.6 1.8 1.9 Pulses 3.3 . 1 5.2 0.0 2.8 1.4 Groundnuts 0.7 -0.9 -1.2 -14.0 1.7 2.7 Palmoil 1.7 2.2 0.1 2.7 3.7 1.6 ourc c: sAL Proiductiont Yearbook tapes. parts of West Africa, roots and tubers (cas- a year for roots and tubers and close to 1 sava, yams, and sweet potatoes) mainly in percent for cereals), stagnated in the next the forest zone of West and Central Africa, decade. Thoughyield data are especiallyweak, and pulses (cowpeas, in particular) through- they describe general tendencies, and the de- out Africa, but largely outside the forest zone. clining trend in Africa is in marked contrast Important oilseeds are groundnuts and palm- to those in other developing regions, such as oil. the Indian subcontinent (where, however, Food production, as measuredbythegrowth much of the growth has come on irrigated of these major crops,' rose by about 2 percent land). And while land productivity is not the a vear in the 1960s (see Table 5.2), or at ap- decisive factor for a thinly populated conti- proximately the same rate as rural popula- nent such as Africa, when growth of total tion. In the 1970s, however, production food production is compared to that of the increased only bv an average of 1.5 percent rural population, it suggests that labor pro- a year and the drop was sharper among the ductivity stagnated in the 1960s and fell in oil-exporting countries, mainly because of the the 1970s.8 decline of groundnuts in Nigeria.7 In fact, for What is significant is that this decline oc- Africa as a whole in the 1970s, growth of food curred over a period when the various gov- production was not only well below the in- ernments and external sources of finance crease in total population, but also well below focused more strongly on food production that of rural population. projects than ever before. Between 1973 and Where output increasecd, it was due largely 1980, about $5 billion in aid flowed into ag- to an expansion of the areas under cultiva- riculture, $2.4 billion of which was from the tion. On the other hand, productivity was World Bank.9 These projects have so far failed stagnant, both in terms of land and labor. to boost output or have been offset by de- Yields, which grew in the 1960s (by 2 percent clines in other parts of the food economy. 8. The rate of growth of rural population does not take 6. The main items left out are fruits and vegetables (in- into account the changes in rural population structure, how- cluding plantains) and meat and fish, for which data are ever-in particular. the decreasing proportion of able-bod- Very patchy and even more uncertain than for the staples ied males. discussed below. 9. Most of this aid was for food production; of the total ,. Excluding groundnuts, the growth rate of food pro- incremental output expected from the assisted projects, ex- duction in the oil-exporting countries would have doubled, port production accounted for only one eighth-or one- from 11.8 percent a year in the 1960s to 1.6 percent annually third, if crops such as maize and groundnuts (which are in the 1970s. both exported and domestically consumed) are included. 47 Table 5.3. Growth of Imports of Selected Agricultural Commodities, 1961-63 to 1977-79 (in volume) Percent yearly Lowo- and middle- Oil-exporting Szub-Saharan income countries countries Africa 1961-63 1969-71 1961-63 1969-71 1961-63 1969-71 to to to to to to Commodity 1969-71 1977-79 1969-7Z 1977-79 1969-71 1977-79 Cereals' 7.4 6.8 21.5 18.2 9.0 9.5 Wheat 9.3 9.2 26.8 13.3 12.9 10.7 Rice 4.9 7.3 3.7 68.0 4.9 12.1 Maize 8.5 2.6 . 47.3 8.7 5.7 Dairy products' 9.8 5.4 .. 17.0 7.2 7.2 Sugar 2.1 -0.1 6.0 23.4 2.5 5.8 Meat 1.1 .4 2.3 33.1 1.3 13.3 Animal and vegetable oils 11.6 5.4 9.1 70.3 11.5 13.0 a. Including those not itemized below. b. See technical notes for Table SA.29. Source: r-v< Trade Yearbook tapes. AGRICULTURAL IMPORTS extent, therefore, a low rate of growth of food Food imports grew steadily over the past 20 production may be an effect rather than a years, except for sugar (see Table 5.3). Im- cause of rising cereals imports; both factors ports of cereals soared by about 9 percent a are related to overvaluation of exchange rates. vear since the earlv 1960s. Commercial im- ports increased from 1.2 million tons a year in 1961-63 to nearlv 5 million tons in 1977- The figures in Table 5.3 actually understate 79 (see Table SA.29). If exports of cereals are the real hike in imports, since, in a number deducted, net imports averaged about 4.3 of countries, food aid is not reported in for- million tons during the late 1970s. The fastest eign trade statistics; yet it now accounts for growing are wheat and rice, which together more than 20 percent of total net cereals im- rose bv about 11 percent a year in the 1970s, ports (see Table 5.4). The bulk of the aid con- the increase partlv reflecting Madagascar's shift sists of wheat and wheat flour (75 percent of from a position of net exporter to major rice U.S. food aid in 1979-80), while maize and importer. Cereals imports are heavily concen- sorghum, the staple commodities in many of trated in ten countries: some of the most pop- the countries, constitute only minor shares. ulous (Ethiopia, Ghana, Nigeria, Sudan, Food aid is targeted to the areas experi- Tanzania,and Zaire), severalsmalleroneswith encing wars and refugee flows, and to the a marked European lifestyle in the cities and Sahelian countries (see Table SA.24). It rose many expatriates (Ivory Coast and Senegal), from about 800,000 tons in the mid-1970s to and two with a high degree of urbanization more than 1.3 million tons in 1978, because (Congo and Zambia). Wheat and rice now of the poor crop year of 1977-78; and the account for 82 percent of gross cereals imports World Food Program estimates that the vol- and their share is expected to rise further. ume reached 1.5 million tons in 1980, with While this is partly due to lagging growth of pledges for 1981 placed at 1.8 million tons. domestic food production, it is also the result of the rapid rate of urbanization and of eco- SOURCES OF SLOW AGRICULTURAL GROWTH nomic policies. Consumption patterns shifted There is a fairly widespread consensus as to from the traditional staples to wheat and rice, the main factors behind the present rural cri- a practice exacerbated by overvalued ex- sis, although perhaps not on the weight to change rates which often make imported cer- be attached to each. These include (as dis- eals the cheapest source of supply. To some cussed in Chapter 2) the disruptions caused 48 Table 5.4. Food Aid and Commercial Imports of Cereals, 1975-79 Item 1975 1976 1977 1978 1979 Food aid (thousands of metric tons) Low-income countriesa 834.7 648.5 708.2 1,001.0 953.0 NMiddle-income oil-importing countriesb 113.2 98.4 147.7 336.9 234.6 Total Africa, 957.8 752.0 868.6 1,359.2 1,201.5 Commercial net cereals imports (thousands of metric tons) Low-income countries 1,471.8 1,309.7 1,532.9 1,488.7 1,272.0 NMiddle-income oil-importing countries 646.9 779.5 1,324.0 1,389.9 1,439.1 Total Africa 2,818.3 3,050.5 4,212.6 4,747.2 4,282.6 Total net cereals inflow (thousands of metric tons) Low-income countries 2,306.5 1,958.2 2,241.1 2,489.7 2,225.0 Middle-income oil-importing countries 760.1 877.9 1,471.7 1,726.8 1,673.7 Total Africa 3,776.1 3,802.5 5,081.2 6,106.4 5,484.1 Food aid as a percent of total net inflow Lo-w-income countries 36.2 33.1 31.6 40.2 42.8 Middle-income oil-importing countries 14.9 11.2 10.0 19.5 14.0 Total Africa 25.4 19.8 17.1 22.2 21.9 Foodi aid per capita (kilograms) Low-income countries 4.9 3.7 4.0 5.5 5.1 MAiddle-income oil-importing countries 2.2 1.8 2.7 5.9 4.0 Total Africa 3.1 2.4 2.7 4.1 3.5 Total net cereals int7ow per capita (kilograms) Low-income countries 13.6 11.3 12.6 13.6 11.8 .Middle-income oil-importing countries 14.6 16.4 26.6 30.2 28.3 Total Africa 12.4 12.2 15.8 18.5 16.2 a. Includes Comoros and Cape Verde. b. Includes Djibouti, Equatorial Guinea, Sdo Tome and Principe, and Seychelles, but excludes Zimbabwe. c. Includes otl exporters: excludes Zimbabwve. Source: Footi Aid Bulletin, no. 4 (Rome: FAO, October 1980), Table 2, pp. 8-10. by wars and civil strife, drought and poor Table 5.5). The downward trend is particu- rainfall patterns during the 1970s, and rapid larly ominous since it has occurred despite population growth, which pushed cultivation major investment efforts and additional use into less productive areas. Agriculture was of off-farm inputs (see Table SA.28). The shift also neglected for a long time by government in consumption toward wheat and rice is also and donors, as it was by development theo- worrisome because it involves cereals that are rists. Further, there was a misallocation of more costly to produce than millet, sorghum, investment, notably an excessive emphasis on and maize, and-except for traditionally grown large-scale government-operated schemes. rice in some countries-have high costs rel- Also, agricultural and economic policies and ative to imports of the same products. institutional frameworks were not conducive to increasing output: official prices were too Action tor fo-Rural Dceuclopn, eilt low; marketing systems too uncertain, inef- ficient, and uncompetitive; input supplies too A major action program for agriculture, the irregular; and participation of farmers in rural Food and Agriculture Organization (FAo) Re- affairs too limited. The agricultural extension gional Food Plan for Africa,i° was written in effort was weakened by unfavorable policies, 1978 and endorsed by the Organization for deficient research output, and the organiza- African Unity (OAU) in Arusha (1978), Mon- tional deficiencies of the public sector agen- rovia (1979), and at the Extraordinary Eco- cies which were responsible for spearheading nomic Summit in Lagos (1980). It postulates rural development. a growth rate for agricultural production of Low agricultural growth rates typify all but a few African countries in recent years (see 10. Regional Food Plan for Africa (Rome: FAO, 1980). 49 Table 5.5. Growth Rates of Agricultural Production, 1969-71 to 1977-79 (average annual growth rate in volume as a percentage) 4 , 3-4 2-3 2-2 0-1 < 0 Kenya Cameroon Benin Botswana Ethiopia Angola Malawi Ivory Coast Burundi Chad Gabon Congo Swaziland Rwanda Central African Guinea-Bissau Gambia Ghana Republic Lesotho Guinea Mauritania Liberia Madagascar Somalia Mozambique Upper Volta Mali Togo Zambia Mauritius Uganda Zimbabwe Niger Nigeria Senegal Sierra Leone Sudan Tanzania Za'ire Source: FAO Production Yearbook tapes 3.9 percent a year for 1980-85 and 4.2 percent * a focus on smallholder production; for 1985-90. This amounts to a tripling of the * changing incentive structures (by hiking growth rate achieved in the 1970s. To reach producer prices, developing more open and this level, the Plan calls for investments of competitive marketing arrangements, and 565 billion in 1975 prices, or $125 billion in involving farmers in the decisions that af- 1981 prices, over a 15 year period (from 1975- fect them); 90). Of the total, 45 percent is for food crops, * expanding agricultural research; and 30 percent for livestock, and 25 percent for * undertaking quick-yielding activities in ir- support services (storage and transport). These rigated agriculture. figures do not include complementary in- vestTnent in infrastructure (marketing, proc- essing, and feeder roads), other support Focus ON SMALLHOLDERS services (research, extension, and training), There are three reasons why the smallholder or inputs. sector should be the focus of a growth-ori- The Plani is careful to note that its so-called ented rural development strategy. First, al- maximum feasible demand and production though it accounts for the bulk of agricultural hypothesis is highly optimistic; it postulates output in most African countries, its massive rapid economic growth, greatly accelerated potential has yet to be realized: use of off- agricultural production, and a high degree of farm inputs is still quite limited, yields are government and external support. Moreover, very low, and specialization is uncommon. the FAO contrasts this alternative with a much Second, recent studies confirm what most of more dismal projection based on extrapolat- the literature on African economies sug- ing existing trends which it considers the most gests-that poverty on this continent, unlike likely result if major actions are not taken. in Latin America, is still predominantly a ru- Nevertheless, it provides a useful macro- ral phenomenon." Thus, raising the output economic framework and helps indicate the and income of small farmers is the best way outlines of an investment program in agri- culture as well as possible financial con- 11. See Deepak Lal and Paul Collier, "Povertv and Growth straints. In this Report the emphasis is on the in Kenya," World Bank Staff Working Paper, no. 389 (Wash- policy framework, a necessary supplement to ington, D.C.: World Bank, 1980), pp. 12, 31, where it is the indicative investment plan adopted at shown that only 3 percent of the total urban population in Kenya live under poverty conditions. This compares with Lagos. The action agenda described in this 30 percent of the total smallholder population living under chapter includes the following: such conditions in 1974. 50 Box A: Kenyan Smallholder Agriculture Several lessons can be drawn from Kenya's rapid products, and sugar cane-previously believed to be agricultural growth, which strongly attest to the de- bevond their capacity. sirability of promoting smallholder agriculture. Kenya's These impressive increases were in part made pos- experience shows that African small farmers are very sible by new production on redistributed land. But of responsive to opportunities for profitable innovation, greater importance is the increased production per and that small farms are frequently far more productive hectare throughout both existing and new smallholder than large farms. areas. Surveys of settlement areas in the 1960s showed In the mid-1950s, the colonial authorities reversed that small farms tended to have both higher output previous policy and began to encourage smallholders and higher employment per hectare than large farms. by lifting restrictions on coffee and livestock produc- Skeptics attributed this to differences in land quality. tion and by land adjudication to provide freehold title. But a rural survey in 1974 showed that the smaller the After independence, smallholder agriculture benefitted farm the higher per hectare of both output and em- from widespread land redistribution and settlement. ployment. For example, on holdings of less than one- The result was a very rapid growth of smallholder half hectare, output per hectare was 19 times greater output. Total agricultural production increased 4 per- and employment 30 times greater than on holdings cent per year from 1955-72 and a disproportionate over eight hectares. The same general relationship held amount of this growth came from smallholders. Two within areas in which all land was of equal quality. developments were especially noteworthy. The first Nor was the higher output per hectare simply due to was the expansion of hybrid maize production, which higher employment. Combining the value of labor, spread more quickly among Kenyan smallholders be- equipment, land, and other inputs, smallholdings were tween 1964-73 than it had among American farmers still generally more productive than larger farms. during the 1930s. Theincreasedyields per hectare raised This does not mean that smallholdings are always total production of this staple food crop and freed land more productive than larger units. Coffee and tea es- for other CrOpS. Second, under the aegis of the Kenyan tates in Kenya still have higher yields and provide just Tea Development Authority, smallholder tea produc- as much employment per hectare as small farms. In tion in Kenya grew from a negligible amount in the semiarid areas, too, more extensive farning may be early 1960s to one third of the total by the mid-1970s. superior under rainfed conditions. But over a wide Prior to this, tea had been regarded strictly as a plan- range of mixed farming on land of medium to high tation crop. Smallholders also rapidly expanded output potential, smallholdings have proven to be superior in of several other products-coffee, pyrethrum, dairy Kenya. to meet basic needs. Finally, attention to it was considered a reasonable solution to la- smallholders is a more cost-effective way to bor shortages, where these existed. Further, raise output than other alternatives currently it was a response to the drought of the early allow-at least for most crops and areas (see 1970s, which convinced govemments in the Box A). semiarid countries that a substantially larger During the 1960s and 1970s, for example, share of total food requirements should be manv African countries directed a substantial produced under drought-proof conditions (for proportion of their agricultural investment to example, modern irrigation programs were large-scale, government-operated estates begun in Chad, Mali, Mauritania, Senegal, which involved heavy capital outlays for Sudan, and Northern Nigeria). And last, it mechanization (as with the rainfed crops) or wasreasonedthatwhileproductivitywasoften irrigation schemes, or both. 12 Why did they lower on state farms, the share of marketable follow such a course? First, there was the surplus would be much higher; thus an em- notion that only a rapid transition to mech- phasis was placed on such enterprises in anized, high productivity schemes, as prac- Congo, Ethiopia, Somalia, and Tanzania. ticed in the industrialized world, would But most of these ventures did not fulfill overcome the stagnation linked with the tra- expectations, and their contribution to growth ditional low-input, low-output methods. Also, was small when compared to their cost. They were beset with problems of management, 12. Examples can be found in Nigeria (irrigation). Ghana overemployment of staff, underutilization of (state farmsl, Ivorv Coast (sugar scheme), Congo, Ethiopia, expensive machinery, and maintenance of and Zambia (all state farms), and in many other countries. equipment and infrastructure. 51 Priority attention to smallholders must be ments should also consider giving more room selective-targeting those areas where the to agro-industrial enterprises (perhaps through physical resource base and existing human concessions) whose external capital and tech- and phvsical infrastructure provide the pre- nical know-how could be applied to planta- conditions for rapid payoff from additional tion or irrigation crops as well as used in investment (see Box B). In the 1970s, agri- industrial processing. cultural programs were often aimed at mar- Private investment-both domestic and for- ginal regions, since the postindependence eign-need not be associated only with large- governments wanted to respond to the needs scale farming. In fact, some of the smallholder and aspirations of people in remote or dis- cotton growing schemes in francophone Af- favored regions of their countries. They hoped rican countries are organized by agencies with to achieve greater regional balance in devel- mixed private-public ownership and are among opment, since in the previous decade re- the more successful ventures on the continent sources had been concentrated on the export- (see Box C). Sugar, tobacco, rubber, and oil oriented sectors and regions. And while the palm projects provide additional examples of 1970s' policv was applied unevenly, in the these kinds of initiatives. poorer countries it absorbed a significant share In a smallholder-based strategy which places of the new investment in agriculture during production first, larger farmers can be used the decade. to spearhead the introduction of new meth- But the marginal areas are only suited for ods. (This has been done in many cases in food production and, even then, not well the past, but the approach has been frowned suited; manry are located in zones of low and upon in recent years, as it has conflicted with unreliable rainfall, where knowledge about the equity concerns, not so much of govern- technical and social conditions of production ments as of foreign sources of finance.) In is limited. Thus, these investments had low practice, this means a recognition that small- yields in terms of increased output. Even when holders with a few hectares more than the "successful" in terms of meeting the social average are legitimate beneficiaries of devel- objectives, the opportunity cost was high. In opment efforts. manv cases, these programs increased the A more production-focused investment peripheral regions' dependence on outside strategy, which concentrates on regions with subsidies (since they could not return much relatively high potential in each country, does of the invested resources in taxes) and des- not mean neglect of disfavored regions. More tabilized agricultural production since the re- intensive research should be launched which source-poor areas generally have low and aims at finding technologies appropriate to uncertain rainfall. these mainly arid and semiarid regions. Until Priority to smallholders does not mean that this technology is discovered, these areas only thev warrant attention. The agricultural should be provided with basic economic and sector is highly dualistic in some countries, social infrastructure which will eventually en- with larger private farms providing major able the local population to make use of fu- shares of marketed output (as in Kenya, Ma- ture opportunities. Finally, programs should lawi, Swaziland, Zambia, and Zimbabwe, be devised to facilitate the migration of people among others); any growth-oriented strategy from the poorest regions to those which are must include these islands of high productiv- better endowed. ity agriculture. Also, there is surely scope- at least in some countries-for larger-scale Improvedi Incentive Strutctures mixed public and private enterprises. Export crops are particularly suited to this option Smallholder energies and skills can only be because the marketing skills and market con- mobilized voluntarily; the incentive structure nections that are often associated with foreign must be right. "Incentive structure" refers to investment could be well utilized. Govern- all those aspects of the farmer's environment 52 Box B: Northern Nigeria Agricultural Development Projects In 1974-75, Nigeria launched the first three of a se- farmers. Effectiveness was enhanced by the high ries of agricultural development projects assisted by population density in the areas selected. The proj- World Bank loans. They were located in the Gusau, ects also did not ignore larger farms, which had Funtua, and Gombe areas of Northern Nigeria. By the political clout and provided an informal channel end of the 1970s, half a dozen other projects had been to the Government for reporting successes, obsta- started in areas further south. In the early 1980s, the cles, and failures. Eventually, the larger farms pro- Government started to expand the development ap- vided a demonstration (over the fence) to smaller proach to entire states, thereby paving the way for the farmers. This trickle-down theory worked because reintegration of the autonomous project structures into the larger farmers proved to be the greater risk- general administrative structures with an important takers and thus were more innovative. role for local government administrations. * Size. The projects were very large and thereby af- The three projects (costing t58 million, $42 million, fected a "critical mass" of people, both farmers and 537 million in 1974 prices, respectively) had the and nonfarmers. They were thus highly visible and following objectives and components: provision of ex- attracted the attention and interest of authorities tension, input supply, and marketing services; agri- to a higher degree than small pilot projects could cultural credit; seed multiplication; rural road have done. development; water resource development; and an ad- * Incentives. The projects were conceived as services ministrative center, training facilities, and evaluation offered to farmers, with no pressure to adopt cer- unit. The projects were executed through project man- tain techniques or crops. The main inputs (fertil- agement units with managerial autonomy; supervi- izer, seed, seed dressings, and chemicals) were sion, coordination, and review were assured by an made available at many accessible outlets. The fer- executive committee at the state level and by a coor- tilizer was heavily subsidized. The initial subsidy dinating committee at the federal level. Project man- undoubtedlypopularizedfertilizersales,butitcould agement was strengthened, in each case, by about one have been quickly removed with little adverse ef- dozen expatriate technicians and managers, including fect because, except for cotton, farmgate prices of the project director. crops were high. Marketing of all crops, again with Over the project period (1975-80) the three projects the exception of cotton, was in the hands of private constructed nearly 1,700 kilometers of feeder roads and traders and, except in circumstances of unexpected 132 dams to provide water for domestic and livestock surpluses due to the project's impact, crops were use. The roads linked a network of 163 farmn service moved out of the project area quickly. (Not sur- centers which were the hub of an input distribution prisingly, cotton is the only crop whose production system. Nearly 150,000 tons of fertilizer were distrib- dedined in the project period.) uted in five vears, reaching 47,000 tons annually in * Management. The projects contained few novel ele- 1980. Sales of other inputs were low as farmers were Ments. The was conta few tove ea- selective in adopting recommended practices and in- ments. Success was due in a large part to the rea- puts. Evaluation results indicate farmer participation rates in excess of 60 percent and average production matters and to the Governent's willngness to gains of more than 5 percent annually for the main compensate local manpower deficiencies through crops-millet, sorghum, and maize. Only cofton pro- hliberal recourse to expatriate managerial and tech- crops-millet, sorghum, and maize. Only cotton pro- nical skills. duction declined significantly. Compared with tradi- c tional plots, yields on fields cultivated with recom- Not all features of these projects are replicable. First, mended practices and commercial inputs were more many countries lack the critical mass for setting up than twice as high. For sorghum and maize, results rural development projects of this size and must thus were particularly spectacular: improved local varieties forego the economies of scale that go with them. Sec- reached 1,400 kilograms/hectare for sorghum and close ond, many countries lack the agricultural potential, to 2,000 kilograms/hectare for maize. i.e., well-watered areas with good soil and moderate What were the main factors contributing to project population density. Third, few governments can afford success? the 80 percent subsidy on fertilizer which Nigeria pays out of its oil-based wealth. (It is true that the subsidy * Location. The project areas were selected on the was not a decisive element in the project's success-. basis of agricultural potential-areas of good soil in fact, it was partially offset by the scarcity of subsi- and adequate rainfall. Pressure to include zones dized inputs, which in turn drove up the consumer further to the north was resisted on the grounds price-but it certainly helped to attract farmers in the that there were no valid technical packages for beginning.) Finally, Nigeria is aiming at plowing part those regions, and the application of a new ap- of the oil wealth back to the rural areas and, therefore, proach would have constituted an enormous risk. does not attach great importance to cost recovery and * Project Design. The project focused on rural roads, a financially self-sustaining agricultural sector. Other water resources, farm inputs, and improved ex- countries will have to strike a more delicate balance tension. The foregoing, at the time of evaluation, between farmers' incentives and governments' limited reflected the order of priorities as perceived by financial means. 53 Box C: Cotton-based Projects in Mali and Upper Volta Two cases illustrating the development potential of over the last four years, partially due to the residual projects based on cotton as well as the contribution effect of fertilizer from cotton production in the area. private enterprises can make in the context of mixed In Mali, millet and sorghum yields on some 120,000 public-private companies are the Mali-Sud project and hectares of the project area are estimated to be 40 the West Volta project in Mali and Upper Volta, re- percent higher than yields under traditional methods spectively. The two project areas are contiguous, lo- of cultivation. In recent years when the state grain cated in a zone where annual rainfall equals or exceeds marketing agency (oPAM) found it difficult to purchase 900mm. the desired amounts of cereals at the unattractive of- In Mali, cotton and cereals production is promoted ficial producer price, cMDT agreed to act as oPA.Ys buy- by the Compagnie Malienne des Textifles (cMor), in ing agent and was able to persuade farmers to deliver which the Government has a 60 percent majority hold- some 10,000 tons of cereals annually at this price. In ing; the remaining 40 percent are held by the Com- addition, the Mali-Sud project successfully introduced pagnie Francaise pour le Developpement des Fibres hybrid maize to the project area. The response sur- Textiles (CFDT), a company with long experience in cot- passed all expectations: maize production now covers ton growing in a large number of African countries. 24,000 hectares with average yields of about 2,000 coDT covers the whole range of activities from exten- kg/ha. sion and input supply to the marketing and processing Both projects have actively encouraged village par- of cotton. Exports are handled by a separate state tidpation. In the West Volta project, village organi- agency. In Upper Volta, the corresponding agency is zations act as the primary marketing agents of cotton, the Societe Voltalque des Fibres Textiles (somnEx), which marketing about half of the entire crop; they also have is 63 percent government-owned, 36 percent cnn- collective fields, responsibility for credit, and some so- owned, and I percent owned by Voltaic banks. somnx cial service activities, such as construction of dispen- is involved only in marketing, processing, and input saries. In the Mali project, Associations Villageoises, with supply, while extension, credit administration, and the help of cMDT, are extending their activities to health other tasks are handled by provincial development care (such as self-financed maternity centers) and water organizations, the Offices Regionaux de Developpe- supply (through the collective purchase of pumps). ment (oRD). The project trains blacksmiths who operate and main- Both projects are assisted through IDA credits. The tain village grain mills. Functional literacy programs West Volta Agricultural Development Project ($20.2 are also important elements in the community devel- million) seeks to improve cereals and cotton produc- opment program. tion, establish a financial and administrative unit, build There are four main reasons for the success of these a ginnery as well as 75 small village grain stores, sup- projects; two relate to cultural and human conditions, ply inputs to farmers, develop low-cost irrigation and two to the structural and organizational charac- schemes, and advance the educational and economic teristics of the project. First, both projects are located status of women. The Mali-Sud Agricultural Project in a region with relatively high agricultural potential: ($15.5 million) aims at expanding the production of good soils, sufficient and reliable rainfall, and adequate cotton, kenaf, maize, and rice; developing applied re- transport infrastructure. Cimatic hazards, while not search, livestock, and health; and establishing a black- absent, remain within tolerable limits. Second, both smith training program. At the Government's request, projects are based on populations known for their in- project activities were extended to cereals production dustriousness and community spirit; this facilitates col- during the project's implementation. lective action. A third feature of key importance is the Through the use of well-known suitable packages of well-established structure of cotton projects, which is improved cultural practices, bothprojectshaveachieved based on a proven system of extension organization substantial increases in cotton production during their and a confirmed technical message. This structure also first three years (1976/77 to 1979/80). In 1979/80, seed provides farmers with an assured outlet, prompt cash cotton production peaked at 65,000 tons for the West payments at fixed dates, and considerable external Volta project and 142,000 tons for the Mali-Sud proj- economies through the aftereffects of fertilizer on cer- ect-increases of 40 percent and 22 percent, respec- eals production. Finally, the minority partner, cnn-, can tively, over three years. Yields have been remarkable claim credit for a good portion of the success. The for rainfed cotton: about 1,000 kg/ha in Upper Volta company provides technical know-how and manage- and about 1,150 kg/ha in Mali. This is twice the average ment (about half a dozen expatriates hold technical or recorded for Sub-Saharan Africa as a whole. managerial positions); the international structure of the Both projects have achieved substantial progress in company permits it to keep abreast of market devel- cereals production as well, but data are less reliable as opments and research results in other countries and most of the production is not marketed through official to take advantage of cr's expertise in bulk buying of channels. In Upper Volta, production of millet and certain inputs. sorghum is estimated to have increased by 25 percent .54 which affect his willingness to produce and portance of good prices, export crop produc- to sell: the level, structure, and predictability ers have been heavily taxed, and prices of of prices; the efficiency, fairness, and stability food crops have been systematically set at ofmarketingarrangements;theavailabilityand below-market levels for most of the past de- prices of off-farm inputs and of consumer cade. These aspects of price policy are dis- goods; and (especially in societies where non- cussed below. material incentives are stressed) the degree of participation in decisionmaking. While all of these are important, price and marketing pol- Pricing of Export Crops. Export crops are icies are the most general, and these will be heavily taxed; African producers have re- the center of analysis here. ceived only a fraction of the world market prices of major exports. Their tax burden, PROBLEMS OF PRICING POLICY defined as the ratio of farmgate producer price It is now widely agreed that insufficient price to economic value at the farmgate, is on av- incentives for agricultural producers are an erage in the 40 to 45 percent range (see Box important factor behind the disappointing D). Subsidies on inputs and other services growth of African agriculture. The impor- provided by government partly compensate tance of price policy comes out strongly in for taxation of cash crops, though they soften project experience. A recent review of 27 ag- the tax impact very little-by 10 to 15 percent ricultural projects undertaken by the World in most cases. Bank noted "the almost overriding impor- Availableinformation does notpermit broad tance of producer prices in affecting produc- generalization about how farmers' relative tion outcome and production levels, often prices and incomes have changed over time. cutting across the qualitv of technical pack- Data for 12 countries suggest that stagnation ages and extension services. Seven out of nine or decline was the most general tendency in projects implemented under favorable prices the 1970s, both for farmer terms of trade and achieved or surpassed their production ob- the purchasing power of farmer cash incomes jectives; 13 of the 18 under unfavorable prices (see Table SA.31). These data provide some failed to do so."'3 This idea is also borne out confirmation for the view, widely expressed stronglv in micro-level studies, which indi- in Africa and outside and implied in the pro- cate substantial farmer responsiveness to duction and population growth data, that price.s4 farmer real incomes have fallen in many Despite this general appreciation of the im- countries in recent years. Heavy taxation and unfavorable terms of trade do not necessarily have quick and/or 13- Sixtf; Annual Re-view of Project Performance Auidits, Sep- observable effects on output trends. But the tembtr 1980. paragraph 3.71. high level of taxation of export crops through 14. Hossein Askari and John Cummings, Agrictlttural Suip- export taxes, marketing board levies, exces- ;pv Response: A Surueti of the Economzetric Evidence (New York: Praeger, 1976). The literature and common observation in- sve marketing costs, and overvalued ex- dicate that farmers respond strongly to changes in relative change rates have kept export production in prices. The question of aggregate supply response is more many countries below what it could have been, nuanced. In the short term, farmers' possibilities are indeed and hence contributed to the steep fall in sharply constrained, and they respond to changed incentive structures by switching to the more profitable crops [see Africa's share in the world market noted ear- Rai Krishna. "Agriculture Price Policv and Economic De- lier (see box on agricultural exports from Tan- velopment" in H. Southworth and B. F. Johnston (eds.), zania and Ghana in Chapter 4).ls Since in the Agricultural Developmezcnt and Economic Growth (Ithaca, New York: Cornell lniversitv Press. 1967), especially pp. 505 ff]. case of most export crops African countries In the longer run, a more congenial set of marketing con- ditions will motivate them to invest in equipment, to hire labor, to work harder, and to find other ways of breaking 15. See also John Levi, "African Agriculture Misunder- those "constraints" which derive from inadequate motiva- stood: Policy in Sierra Leone," Food Research Instituite Studies, tion rather than from inadequate technology. vol. 13, no. 3 (1974). 55 Box D: "Taxation" of Export Crops Nominal Protection Coefficients of Selected Export Crops A good measure of the degree to which crops are Crop 1971-75 1976-80 taxed is the "Nominal Protection Coefficient" (NPC), defined as the price paid to the producer divided by Cocoa the amount he would have received had he sold his Cameroon 0.37 (2)- 0.45 (2) crops at the world price minus transport, marketing, Ghana 0.47 (5) 0.40 (4) and processing costs. An NPC value of more than one Ivory Coast 0.56 (2) 0.38 (1) means that the crop is being subsidized; and the more Togo 0.50 (5) 0.25 (4) it is above one, the bigger the subsidy. An ,,Tc of less Coffee than one indicates taxation; the lower it is below one, Cameroon (Arabica) 0.72 (2) 0.60 (2) the heavier the tax. The table at right shows NPCS for Cameroon (Robusta) . . 0.36 (1) major crops in 13 countries, representing more than Ivory Coast 0.68 (1) 0.36 (1) half the region's population. Kenya 0.94 (1) The actual level of taxation of export crops is higher Tanzania 0.80 (5) 0.59 (4) than shown in two imnportant respects. In the first Togo 0.42 (5) 0.23 (4) place, the economic farmgate value of these crops Cotton has been derived on the basis of actual marketing Cameroon 0.79 (1) costs. These costs are, in most cases, those of mo- Ivory Coast 0.79 (1) 1.05 (1) nopolistic agencies working without competitive Kenya 1.07 (1) . . pressure, and thus are generally inflated. If the mar- Malawi 0.68 (5) 0.75 (2) keting cost of an efficient marketing system were Mali 0.55 (2) 0.44 (4) used instead, the economic value of crops would be Senegal 0.65 (2) higherandthedegreeofimplicittaxationevengreater. Sudan 0.78 (2) 0.60 (1) The level of taxation is also higher than shown be- Togo 0.62 (5) 0.79 (4) cause the i.'Pcs do not reflect the influence of over- Upper Volta . . 0.79 (1) valued currencies, which reduce the proceeds of Groundnuts exports in terms of domestic currency. Taking into Malawi 0.70 (5) 0.59 (2) account the effect of an overvalued currency, pro- Mah O.7 (2) 0.43 (4) ducers in a number of countries listed in the table Senegal 0.48 (4) 0.66 (4) received less than half the real value of their crops Zambia 0.70 (5) 0.71 (4) in recent years. Maize' Kenya 0.96 (1) 1.33 (1) Malawi 1.68 (5) 1.34 (2) have distinct comparative advantages, this Zambia 0.72 (5) 0.78 (4) implies a loss of growth opportunities for the Sesame economv as a whole. Sudan 0.83 (1) 0.59 (1) Upper Volta . . 0.88 (1) Tea Pricing of Food Crops. In most African Kenya 0.89 (1) Tobacco countries producer and consumer prices for Malaw 0.42 (5) 0.28 (2) basic foodstuffs are legally controlled. Gov- Zambia 1.09 (5) 0.88 (4) ernments have dual policy objectives in set- Wheatb ting and regulating their prices. They want Kenya .. 1.43 (1) to provide adequate incentives for increasing . . not available. food production, and they seek to protect the a. Figure in brackets indicates number of observations (years). interests of consumers at the same time. 16 In b. Maize and wheat have been alternately exported practice, the objective of ensuring a regular and imported in these countries. supply of staples at "affordable" prices for Source: World Bank data. consumers has been the dominant criterion in most countries. This is accomplished in costs; food imports are encouraged when do- various wavs: producer prices are fixed at mestic food price levels rise; and imported below market levels; subsidies are provided foods are given an implicit subsidy because by selling imported foods at below landed of currency overvaluation. In most instances over the past decade, 16. See Lagos Plani of Action, paragraph 33. official food prices in African countries have 56 been set "too low"; prices in parallel markets Box E: Nigerian Food Imports are often two to three times as high. The official prices have thus been only partially Government trade policy in Nigeria has had a par- effective; producers have been able to sell part ticularly strong influence on prices for major staples. of their marketings at the free market prices Wheat and flour imports grew from 400,000 tons in 1975 to 1.3 million tons in 1978, and then declined and most consumers have been forced to buy to an estimated 1.0 million tons in 1980. Rice imports at open market prices. This happens despite have grown even faster: volumes increased eleven- the existence of state marketing organizations fold between 1976 and 1978-from under 50,000 tons that are frequently endowed with legal trad- to over 550,000 tons-then dropped sharply due to thatnar frequentlys endoweda wihin grain trading. import restrictions. In October 1979, licenses were ing monopolies--especilaly igrntadg. issued for only 200,000 tons for 1980. Prices tripled While direct price effects on production are by September of that year. The Government re- thus diluted, the policy of setting low official sponded to this surge in rice prices by lifting import producer prices undoubtedly has negative quotas on rice, cereals, and flour. The same pattern had occurred earlier: in 1979, wheat and flour prices effects on farmer incentives to produce and fell markedly as a result of big increases in imports. to sell basic foods. In short, the sharp increases in imports in recent The effect of government import policies is years help explain the fall in prices in 1979 while later in manv cases much more important than low import restrictions fueled their recovery in 1980. Gov- ernment trade policy, particularly forrice, has changed controlled prices of staples. Imported wheat frequently in recent years as indicated in the table and rice are now becoming steadily cheaper below. than domestic staples because of the over- Summary of Rice Trade Policy valuation of many African currencies. More- over, intent onl keeping urban food prices low, Prior to April 1974 66.6% tariff April 1974-April 1975 20% tariff many governments have in recent years had April 1975-April 1978 10% tariff periodic recourse to massive injections of food April 1978-June 1978 20% tariff imports, thereby causing sharp reductions of June 1978-October 1978 10% tariff October 1978-April 1979 Imports in containers domestic prices (see Box E). These policies, under 50 kilograms which stimulate wheat and rice consumption banned. and discourage producers of substitutable lo- April 1979 Imports in containers 50 cal cereals, have been reinforced by food aid, kilograms and above under restricted i- w hich has been maintained at consistently cense. high levels after a brief reduction in the mid- September 1979 Six-month ban imposed 1970s. Relative price changes favorable to do- on all rice imports. January 1980 Impor-t licenses issued for mestic staples have been checked, and urban 100,000 tons. preferences for wheat and rice reinforced. October 1980 Rice placed under general These pricing policies are widely attributed import license-no quantitative restric- in the scholarly literature as well as in donor tions. circles to an urban bias among policymakers. But a balanced assessment has to recognize Erratic trade policy not only has had a dramatic the constraints at work, and the conflicts of effect on price levels, sending confused price signals to producers, but also has increased the risk to trad- objectives involved. Policymakers in Africa are ers who market domestic supplies. fully aware that the raising of producer prices for export crops would stimulate production and is in general therefore a desirable objec- sides: overvalued exchange rates mean that tive. But thev are also aware that other ob- the foreign currency obtained from exports is jectives may be sacrificed. Thus, the "taxes" converted into a relatively small amount of levied on export crops are a principal source domestic currency, making it difficult for gov- of finance for public sector activities; for non- ernments to pay higher producer prices. And mineral economies there is no other impor- on the domestic marketing side, marketing tant domestic source. Moreover, the scope for margins absorb large shares of total proceeds, higher producer prices is obstructed on two reducing the share available to producers. 57 Finally, a positive price policy for export crops only at high cost. It can be grown efficiently might lead to reduction-at least in the short under rainfed conditions or in small-scale run-in food crop production, and this would schemes in swamps or riverine valleys while have consequent effects on food self-suffi- large-scale irrigation schemes are extremely ciency objectives. costly both in terms of initial investment and Similar conflicts of objectives-real or sup- recurrent costs. Therefore, import substitu- posed-underlie food price policy deficien- tion of these crops, where technically feasible, cies. Abundant worldwide experience in the will preempt a substantial proportion of in- past decade certainly indicates that the polit- vestible resources. Since urban consumers ical risks involved in raising food prices are would be the main outlet for these cereals, hardly negligible. Moreover, there exist gen- ambitious rice and wheat production pro- uine concerns over the impact of higher food grams will reinforce the traditional bias against prices on real incomes and the nutritional sta- the rural populations. In addition, this trend tus of the poor. To see food price policies toward rice and wheat consumption, un- solely in terms of political will or commitment checked by suitable pricing policies, will is too simple. maintain the traditional rainfed cereals (and It is true, nonetheless, that the price poli- roots and tubers) in the position of inferior cies described above have proved self-defeat- goods, reducing the extent and stability of ing. The policy of attempting to control prices their markets. and supplies of foodstuffs has, by and large, succeeded in securing only a limited supplv of low-priced (and often low-quality) food- PROBLEMS O MARETING AND INPUT SUPPLY stuffs for a relatively small group of urban consumers. It has increased farmers', and The central problem in marketing and input traders', risks in producing and marketing food supply is the very general tendency to give surpluses. It has failed to stabilize and indeed too large a set of responsibilities to public has actuallv destabilized supplies over the sector institutions, and too few to other course of the vear. Further, through its effects agents-individual traders, private compa- on farmers' supplv response, it has probably nies, and farmers' cooperatives. It is the ma- resulted in a higher overall level of food prices jor manifestation of the organization and than wvould have pertained without govern- management problem discussed in the pre- ment attempts to control supplies. ceding chapter. Reliance on imports, moreover, is creating a potentially very costly structural depend- ence on wheat and rice. These now account Marketing Agencies. Export crops are al- for 82 percent of net commercial cereals im- most everywhere in Africa marketed by state ports, wheat alone for nearly half the total. trading organizations; often they use "li- Projections of Africa's cereals import require- censed buying agents," private traders, to help ments in 1990 vary from 6 million to 28 million in village-level purchases. Government mon- tons, depending on the assumptions used. opolies also exist in many countries for food Most estimates cluster around 11 million to crop purchases, though these are generally 12 million tons, most of which would be wheat less well organized than export-crop market- and rice. This is a conservative estimate since ing and are in most cases unable to purchase the 1981 imports may already be 6 million to more than a minor share of marketed output. 7 million tons (including nearly 2 million tons The performance of the export crop mar- of food aid). Except in Sudan, Ethiopia, the keting agencies is of major importance in sev- East African Highlands, and parts of South- eral respects. First, their degree of efficiency ern Africa, wheat cannot be grown or can be affects the share of export proceeds that can grown only at prohibitive costs. Rice can be be paid to producers. Because of long dis- grown in a wider range of countries, but often tances and the frequently difficult problem of 58 access, the cost of marketing tends to be high marketed production may be as high as 25 to even under conditions of efficient marketing 50 percent (in some East African countries), operations." Second, the crop marketing or as low as 1 to 2 percent. Price and mar- agencies are the major point of contact be- keting controls are conspicuously absent for tween peasants, the money economy, and the roots and vegetables, no doubt because of the state bureaucracy. Unless the marketing problems and risks associated with the per- transactions are done fairly and efficiently, ishability of these crops. Most govemments there are high risks of peasant disaffection do not put much trust in the private sector's from both the bureaucracy and the market ability to cope with the task of providing sta- economy. ble supplies of food to the urban masses, al- Serious inefficiencies characterize the op- though private traders handle the bulk of the eration of most marketing agencies. Some of trade almost everywhere. In most cases pri- these arise from problems found in almost all vate traders are tolerated openly or tacitly as parastatals-overmanning, inadequate non- indispensable partners but are not allowed to salary budgets, and management scarcities. work in an economic environment that would There are also inefficiencies peculiar to the enable them to realize their full potential. The export crop parastatals due to the lack of com- uncertainties associated with the ambiguous petition. And there are additional problems position of private trade and traders discour- in these agencies when marketed volumes age full-time involvement in food marketing, stagnate or decline: decreasing turnover is investment in transport and storage, and a compensated by higher overhead per unit, systematic approach to developing an ade- the producer price being the residual. The quate supply network. result is an upward spiraling of costs and a Official marketing agencies are responsible parallel downward spiraling of exports. Clas- for collection, transport, sometimes process- sic examples of this are groundnuts in Mali ing (as with rice), and distribution to the and several export crops in Tanzania.18 wholesale and sometimes even retail level. In food-crop marketing, parallel marketing But producer prices and consumer prices are channels exist in many countries of the re- fixed by government with little regard to the gion; the legal and official marketing agency actual cost of collection and distribution. Mar- coexists wsith a semiclandestine private trad- keting agencies are also not always or not ing sector. This is most often the situation fully reimbursed for losses incurred in the with respect to foodgrains. In these markets, process. Several of them have accordingly attempts at controlling marketing and prices accumulated large deficits reflecting opera- are most extensive but they are effective to tional inefficiencies and the cost of govern- varvingdegrees. Incountriesimportingwheat, ment-imposed subsidization of consumers. In price controls are often fairly effective for flour several cases, deficits have reached striking and bread; for rice, the degree of control de- proportions, given the rather modest quan- pends primarilv on the share of paddy grown tities of foodstuffs controlled by these agen- in govemment-controlled schemes. For do- cies. Some agencies are passive, buying mestic cereals, the share of official trade in whatever small quantities are offered to them at official prices in postharvest periods when market prices are low, or during bumper crop 17. In Kenya. for instance, charges for marketing, stor- years, buying all they can pay for. Others age, transport. and administrative overheads averaged 34 exercise varying degrees of compulsion, oc- percent ot the f.o.b. border price for maize, 23 percent for wheat, and 48 percent for rice during the 1972-79 period- casionally bordering on outright requisition. and the agency in question is not regarded in Kenya as a particularly inefficient marketing institution. Input Supply. Input distribution agencies 18. See Frank Ellis, A Preliminary Analysis of the Decline in are another part of the rural marketing system Tanzania Casheuwnut Production 1974-79: Cases, Possible Rem- edies arid Lessons for Rural Development Policy (Economic Re- that has contributed to the poor performance search Bureau. University of Dar es Salaam, December 1979). of agriculture. Unless farm inputs are made 59 Table 5.6. Relative Frequency of Government and Private Sector Control in the Procurement and Distribution of Agricultural Inputs, 39 Countries Percentage of countries Farm Fertilizer Seed Chemicals equipment Item sulpply suipply supply sluppIy Government control' 64 61 47 42 Private sector control' 11 11 17 22 Mixed government and private sector involvement 25 28 36 36 Total 100 100 100 100 a. Procurement and distribution activity is considered "private" if more than 80 percent of it is in the hands of the private sector, and "government" if more than 80 percent of it is in the hands of the public sector. Source: World Bank data files. (See Table SA.32 for country-specific information.) available to farmers on a regular basis and at ficial prices of inputs uniformly for the entire the right time, there is little chance that ag- territory without regard to actual transport ricultural production and productivity will costs-which impedes private trade from ef- move forward.'9 Unfortunately, there are only fectively competing in remote areas. a few countries in Africa where this important There is no a priori reason why government condition is fulfilled. Procurement and dis- agencies should not be able to fulfill the input tribution of inputs is another field monopo- supply functions efficiently, but due to the lized by governments or parastatal agencies. structural problems besetting many public In more than 60 percent of African countries, agencies-scarce management, lack of incen- governments reserve full control of the pro- tives, conflicting objectives, overstaffing, and curement and distribution of fertilizer, seeds, lack of control-they have rarely succeeded and most other services as well (see Table in meeting the rigorous requirements of their 5.6). The motives for entering this field are clients-input delivery at the right time, at similar to those advanced for government in- the right place, and in the right amounts. volvement in food crop marketing: inputs are Government agencies have failed to meet seen as vital commodities that should not be these needs because they have difficulties in left to the care of the private sector, which is adapting bureaucratic, financial, and admin- regarded as exploitative and unreliable. Pol- istrative procedures to commercially oriented icymakers also frequently perceive a need to operations. For example, they fail to buy in- subsidize the service provided, which is a ra- puts on a phased basis because they are geared tionale for monopolizing its distribution.20 to the time of release of funds from the budget, Many officials also believe that only by public and these are not necessarily the optimal times. distribution will inputs be made available to Likewise, pay scales and hiring and promo- the remote areas that private trade is assumed tion procedures tend to be similar to those in to neglect because of low profitability. While government. This leads to reduced individual this may be true in some cases, it is mainly initiative, unwillingness to make quick and the policy of panterritorial pricing-fixing of- independent decisions, and consequent effi- ciency losses. The absence of competition in input supply 19. Input supply involves more than just the provision also leads to a lack of innovation. Inputs are of fertilizer and seeds. It can include the supply of farm ordered in routine fashion without regard to equipment. fencing and building materials, tractor hire location-specific requirements. Arecent study services, and spare parts. 20. Subsidies need not involve monopolization. The mar- in Senegal has revealed considerable scope ket mechanism can be used. But most officials do not believe for savings in fertilizer cost to farmers, no- that markets work well enough to be utilized this way. They tably by tailoring the nutrient content more believe, in this case, that subsidies granted to importer- wholesalers or to private traders would not be passed along closely to their needs, by eliminating ineffec- to farmers. tive elements, by reducing transport cost 60 through higher concentration, and by deter- efficient, pushing beyond these propositions mining optimal dosage and composition on is not easy because the problems are complex economic rather than on technical grounds. and involve broad aspects of development Such adjustments, and the supply of inputs strategy. For example, the appropriate level in variable package sizes convenient for farm- of producer prices, the relationship between ers, would be introduced more readily and prices of export crops and food crops, and on a broader base in a system leaving more between prices of individual crops in each scope for private sector participation in input category are all a function of a government's supply. development goals and social policy objec- The general problems outlined above are tives. Nonetheless policy changes are needed exacerbated by the common practice of gov- and the directions of change are discussed in ernment, subsidization of inputs, in particu- the following paragraphs. lar fertilizer. This has a number of negative consequences. First, in monopolistic input EXPORT CROPS distribution systems the funds budgeted for Trade data for the 1970s and the level and input subsidies limit the total amount of fer- trend of export taxation suggest that in many tilizer made available; under the budgetary countries there is scope for increasing pro- constraint that many countries experience, the ducer prices for export crops. The slow growth actual amount of inputs that can thus be pur- of world demand for many primary com- chased remains far below the quantity desired modities is not a valid argument to the con- by farmers at the subsidized price. Therefore, trary as long as Africa does not even maintain rather than supplying more farmers than can its market share. Higher producer prices in be served under private trade conditions, real terms would stimulate production di- governments end up serving considerably rectly. It would also allow elimination of most fewer. Second, since the quantity delivered of the subsidies on inputs, equipment, credit, remains well below the level of demand, price or water that now hamper the distribution of is driven up and, in spite of the subsidy, some these commodities and services and distort users may pay as much as-or conceivably the allocation of resources. more than-they would under free-market Four objections to a high price export crop conditions. Third, even where the input dis- policy were noted earlier: the need for gov- tribution agency has a source of finance in- ernment revenues; the limited freedom for dependent of the budget (such as bank credit maneuver on producer prices due to over- or sufficient working capital as equity), sub- valued exchange rates; high marketing costs; sidization ties its operation to the budget year, and the conflict that develop with food self- causing delays in procurement and untimely sufficiency objectives. delivery to farmers.21 With respect to government revenue, a number of observations are in order: first, revenue preservation should take second place to the need for maintaining or increasing the ,yirn of prg /WvrMt pace of export production; second, reduced 8,Io5V7U I SDf t) FO P lF o)s taxes should raise export levels so that higher volumes would to some extent compensate While there is not much disagreement with for the reduced rates; finally, and even more the general propositions that higher producer significantly, higher producer prices should prices would stimulate production and sales, still leave scope for taxing some of the "rent or that marketing systems should become more elements" prevailing for some crops-coffee, cocoa, tea, and even cotton. The exchange rate issue has to be con- 21. Manv of the problems and svmptoms associated with g the distnbution of subsidized inputs also applv to rural fronted directly or indirectly, as noted in credit institutions administering heavily subsidized credit. Chapter 4. In many cases, lagging export 61 growth and soaring food imports are attrib- were scarce, governments took control. It is utable to this factor. An adjusted exchange now evident, however, from 20 years of ex- rate, or revised tax and tariff policies with perience in many different settings, that a equivalent effects, would allow better incen- high price is being paid to keep the export tives for export crop production and would, monopolies in place. Since new abilities, both if the resulting increase in import prices is private and organizational, have now devel- passed on to consumers, curtail demand for oped, it would seem time to make export crop imported cereals and put food production marketing more competitive. Cooperative programs aimed at import substitution on a marketing could be more widely encouraged, sounder economic basis. At the same time, a as among coffee growers in Cameroon (see corrected price relationship would boost and Box F). And private traders might be allowed stabilize demand for traditional staples, which fuller entry, perhaps for sale to a state export usually can be grown at lower cost than wheat agency. In this, as in all proposals for struc- and rice. Although there are supply problems tural change, it should be stressed that there for traditional staples as well, these problems are no ideal solutions. Rural markets function are not unrelated to the mismatch between imperfectly in many respects, and there are the structure of cereals demand in African risks that some farmers may suffer from un- countries and the associated structure of do- equal bargaining power. But the present ar- mestic supply.22 rangements have proved so generally More will be said about marketing prob- inadequate and the costs are so high that new lems in later sections. Two points suffice here. departures are justified. First, the export crop monopolies suffer greater There is, finally, the issue of export crop- propensities to inefficiency than the food food crop price policy interactions. If export marketing agencies, because they are exposed crop prices rise, it is feared that food crop to no market pressures pushing them to re- production will fall. This is, however, not duce costs. This means they risk becoming a necessarily so, and even if export crop output steadily growing drain on export proceeds, were to grow at the expense of food crop with producers paying the price. production, it is not necessarily bad. Second, the situation has evolved since the Empirical evidence does not support the first years after independence, when many or hypothesis that expanding export production most of the export crop monopolies took their leads to declines in food production. This may present form. The export crop marketing occur in some cases (the northern groundnut monopolies arose because, except where there basin in Senegal may be one), especially in were public agencies, the export trade was the short run. But the bulk of the evidence almost entirelv in the hands of foreign firms points the other way. Countries that have and immigrantmerchants. Fewsocieties, much been doing well in cash crop production have less those newly independent, would accept also been among the most successful in ex- that control over trade in vital commodity panding food production. This is confirmed exports be so completely dominated by for- by aggregate data as well as by examples at eigners. Since nongovernmental alternatives the level of individual countries.23 This com- 22' "Efforts undertaken for the development of irrigated 23. In both Mali and Upper Volta, for example, resource- agriculture should not be relaxed. Otherwise we may wit- poor countries where a tradeoff might be most likely, cotton ness a major increase in the rice and wheat deficits with a output has grown substantially in the past decade and the CorreSpondzno rincrease In the need for food assistanice." See The areas specializing in cotton production are also major pro- Dte'lot'retr!ti ot Irrigated Agriculture in the Sahel (Ouagadou- ducers of marketed cereals (see Box C). More generally, if gou, Upper Volta: Club du SahebCILsS, April 1980), p. 33. the hypothesis that food gaps are attributable to "excessive" (Emphasis added.) This statement suggests that it is a le- cash crop production were true, there ought to be a negative gitimate aim of food aid to assist not only people in distress correlation between countries ranked according to progress but also to supply urban populations with a preferred food in food and nonfood production. An analysis based on FAO that cannot be grown domestically except at high cost, if data for some 40 countries in Africa shows no positive at all. correlation between strong growth performance in export 62 Box F: Arabica Coffee Marketing in Cameroon's Western Highlands Since 1958, arabica coffee producers in Cameroon's delivered performs additional electronic sorting to im- Western Province have marketed their production prove quality. ucCAo also is the central buying agent through six cooperatives under the leadership of the for the fertilizer and agricultural material required bv Union of Arabica Coffee Cooperatives of the West its members. In this, it arranges for delivery and dis- (UCCAO). In recent years, about 100,000 smallholders tribution to cooperatives and helps defray the distri- have used the cooperatives to market about 18,000 tons bution costs from the 1 percent commission it takes of coffee and to purchase about 20,000 tons of fertilizer on the value of the coffee sold. annually. uccAo also represents the interests of the growers Although chartered by the Government, both the with the Government, which sets coffee prices to pro- cooperatives and uccAo fully control their own affairs ducers and determines the amount to be paid from the including finances and terms of employment of staff. price stabilization funds at the end of the crop year. Members of each cooperative elect a delegate assem- uccAo also maintains vehicles and equipment and keeps blv which in turn elects directors of the cooperative the central accounts. Finally, UCCAO manages the re- who appoint an executive committee and a chairman serves accumulated from the difference between the to manage routine operations. These indude the pur- f.o.b. price for coffee and the payments made to pro- chase of coffee at the collection centers throughout ducers minus the operating costs of the Union. By law, each district, where the coffee is weighed, graded, about 20 percent of these reserves must be retained as sorted, sacked and shipped to central warehouses to long-termprotectionforcoffeeproducers'incomes.Most await sale abroad. The cooperatives also distribute fer- of the remainder can be invested in developmental tilizer and equipment, and administer the seasonal projects with Government approval. credits which are used by members to finance the pur- In 1978, uccAo became the implementation agency chase of production inputs. for an integrated rural development project which was uccAo has a board of directors-the chairmen and supported by 1DA. As part of the project, the cooper- selected executive directors of the cooperatives-and atives of uccAo also made additional investments in is run on a day-to-day basis by a director general and their infrastructure, and are diversifying their partici- a central secretariat. Its primary function is to market pation in the development of rural areas to indude coffee overseas. It also arranges for a line of credit to food crop promotion. finance the purchase of the crop, and when coffee is plementaritv is not surprising. First, export markets. The general point is that the benefits crops are the nucleus around which exten- of a changing, dynamic agriculture are not sion, input supply, and marketing services restricted to a single crop or sets of crops. are built; these also benefit food producers. When change accelerates, the productivity of Second, food production directly benefits from the whole farning system also increases.24 aftereffects of fertilizer expended on the com- Even if it could be demonstrated that export mercial lead crop. Third, the existence of a crop increases have come at the expense of commercial crop facilitates the propagation of food production, the conclusion would not productivity-increasing equipment. Finally, necessarily follow that a strategy of self- where individual farmers undertake cash crops reliance requires a substitution of food pro- to such an extent that they develop a food duction for exports. Most African countries deficit (which they usually do only if there is a reasonably well-developed local or regional food trade), cash crop production creates a 24. A detailed analysis by John Cleave provides con- local market for food crop producers that is vincingevidence based on nearly 50farm managementstudies often more secure and stable than distant urban that in most areas a substantial expansion of export pro- duction was supenmposed on the traditional farming sys- tem in such a way that the level of food production for subsistence was maintained. And for the areas that have production and poor performance in food production. On increased their reliance on purchased food, there is related the contrarv, the relationship between performance in food evidence which suggests that this has not led to a decline and nontood production is positive although the correlation in food consumption. See John H. Cleave, African Farmers: is rather weak (Spearman rank coefficient of 0.42 for the Labor Use in the Development of Smallholder Agriculture (New 1969-71 to 1977-79 period). York: Praeger, 1974), pp. 27-30 and Appendix B. 63 have distinct comparative advantage in ex- traders an increased role in these markets. port crop production. An export-sacrificing Some observers object to proposals for more policy of self-reliance would therefore have competitive marketing arrangements on the costs in terms of income (see Box G). And a grounds that rural African markets function policy aiming at food security at the price of imperfectly, that traders would therefore ex- lessened emphasis on exports has a further ploit farmers, and that indigenous traders are pitfall: most methods of intensification imply still few in number in parts of the continent, increased use of inputs such as fertilizers, in- so that trade in foodstuffs might once again secticides, and fuel for pumping (in irrigation come to be dominated by nonnationals. But schemes), i.e., they rely heavily on imported many recent studies suggest that African food inputs. Thus, agricultural production under markets are reasonably competitive, that trader these known methods of intensified cultiva- profits are rarely "excessive" and that farmers tion becomes more vulnerable to external dis- are usually well protected against "exploita- equilibria. If the pursuit of food self-sufficiency tion" by market information and the availa- diverts resources from export crops to food bility of alternative points of sale.25 But even crops, declining export earnings may lead to if this were not so, governments can more balance-of-payments problems jeopardizing efficiently protect farmers by making markets the self-sufficiency objective itself. Sudan and more competitive through better information, Tanzania are countries that have, in recent roads, and marketing facilities than by acting years, deliberately sacrificed export expan- as substitutes for traders. sion for the sake of increasing food produc- In any case, it is important to recall that in tion (see box on Tanzanian exports in Chapter a large number of African countries, food 4). Their present balance-of-payments crises, markets continue to operate, as they have in as severe as those of some mineral exporters, the past, without much public control. Per- are partly related to that policy. formance is generally impressive; in Nigeria, private trade suppliers two very large cities FoOD CROPS AND INPUT SUPPLY (Lagos and Ibadan) and many more towns of 100,000-500,000 people. In Mali, despite the While many details regarding food price pol- uncertainties of public policy, private trade in icies can only be assessed in the context of the mid-1970s supplied two thirds of the ce- individual country situations, two principles reals consumed in the Sixth Region, the most are central. First, food imports should be sub- remote part of the country. ject to duties, so that the import price reflects Indeed, even the most casual visitor to a at least the true cost of foreign exchange. market town in Africa has to come away im- Otherwise, low-price imports will continue to pressed by the range of goods and services replace domestic production, with negative available for sale, their variety and quality effects on rural income and growth. Second, there should be a gradual freeing of domestic food markets, to encourage greater competi- in 25. See Henry Hayes, Marketing and Storage of Food Grains food markets, to encourage greater competi- in Nigeria, Samaru Miscellaneous Paper no. 50 (Samaru, tion. This would in most cases merely rec- Nigeria: Institute for Agricultural Research, Ahmodu Bello ognize the existing reality, which is that, University, 1979) and IDrET-CEGOS, Pre-dtude de la commerciali- whatever the legal or formal situation regard- sation des produits vivriers ade Camneroun (R6publique Unie du Cameroun: Minist6re de I'Agriculture, avril 1980). See also ing public monopoly, the main part of the Van Roy Southworth, William 0. Jones, and Scott R. Pear- cereals trade goes through private channels son, "Food Crop Marketing in Atebubu District, Ghana" in and, consequentlv, the great majority of con- Food Research Institute Studies, vol. 17, no. 2 (Palo Alto, Cal- ifomia: Stanford University Press, 1979), J.T. Mukui (ed.), sumers already pay "free market" prices. "Price and Marketing Controls in Kenya," Institute of De- With regard to food marketing and input velopment Studies Occasional Paper no. 32 (Nairobi, Kenya: supply, the proposal to allow a fuller degree University of Nairobi, 1979); and Guy Nicolas, "Processus d'approvisionnemsent vivrier d'une ville de savane: Maradi of competition means encouraging coopera- (Niger)," Travaux et documents de gdographie tropicale, no. 7. tive actions by farmers and allowing private decembre 1972. 64 Box G: Measuring Comparative Domestic Resource Costs per net unit of Foreign Advantage Exchange for Export Crops and Food Crops in Selected Countries, A country is said to have comparative advantage Export crops in a given commodity when it can produce that com- Cocoa modity relatively more efficiently than most other Ghana (1972) 0.30 commodities. Comparative advantage can be as- Ivory Coast (1972) 0.36 sessed with the help of a measure called "Domestic Coffee Resource Cost" (DRC); this measures the cost of do- Ivory Coast (1972) 0.51 mestic resources (labor, materials, etc.) used to save Kenya (1975) 0.44 Cotton or eam a net unit of foreign exchange. The lower the Ivory Coast (1972) 1.12 DRC. the more efficient the activity. Mali (1972) 0.21 The table at right compares DRCS for food and export Senegal (1972) 0.42 crops in 11 countries. It shows a pattern of Togo (1977) 0.37 strong comparative advantage in exports. These re- Zambia (1974) 0.53 sults are based on price relationships, input costs, Zambia (1977) 0.34 Croundnuts and technologies of the mid-1970s, and can change Maiu (1972) 0.23 over time. But projections for the 1980s do not predict Senegal (1972) 0.36 changes in price relationships on world markets which Palmoil would significantlv alter these conclusions. Ivory Coast (1972) 0.36 Nigeria (1979) 0.39 Tea Kenya (1975) 0.67 gradations, as well as by the evident dyna- Tobacco (1974) 0.54 mism and liveliness of the bargaining that Zambia (1977) 0.82 characterizes the simplest transaction. What Food Crops Groundnuts is particularlv striking is that one finds in these Nigeria (1979) 1.40 markets almost evervthing except goods which Zambia (1974) 0.50 Zambia (1977) 0.94 are sold by monopoly suppliers of the public Maize sector: fertilizer, seed for main crops (though Nigeria (1979) 1.76 one finds seeds for vegetables), and animal- Zambia (1974) 0.58 Zambia (1977) 1.16 drawn implements. Millet The kevstone of anv marketing reform, then Mali (1972) 0.62 1 1 ~~~~~~~Nigeria (1979) 1.21 must be to capitalize on the indigenous trad- Senegal (1972) 0.62 ing svstem, a proven asset, and let it play a Rice Ivory Coast (1972) 1.50 bigger role in the distrbuton system. The Ivory Coast (1975) 1.80 private sector, with its small-scale, decentral- Mali (1972) 0.67 ized and flexible structure, is particularly well Nigena (1976) 2.55 suited for this task. Devolution of marketing Senegal (1972) 1.02 functions to private enterprise may be more Sorghum Mali (1972) 0.62 difficult in some parts of Africa, where the Nigeria (1979) 1.66 tradition of indigenous entrepreneurship is Senegal (1972) 0.62 weak, but this should affect only the pace of a. At official exchange rates. change, not the objective. Source: World Bank data. In most instances, governments will be re- luctant to allow food marketing to become and farmers through greater emphasis on ru- exclusivelv a private sector activity. A variety ral road development and maintenance, by of agents can, of course, coexist; indeed, it providing better information on crop size and should be encouraged. Cooperatives can take prices, via radio and otherwise, and by grad- on manv activities in this area and the state ually introducing uniform weights and meas- role in food marketing would remain sub- ures, a task that governments have neglected. stantial even after considerable liberalization. State grain agencies would also continue to Governments could improve market func- have other major functions: they could man- tioning, easing market access by both traders age grain imports; they might buy and sell in 65 the open market for special purposes (e.g., of fertilizer. Yet, in many parts of the world, localized production crises); they might op- private corporations' research efforts and erate buffer stocks for seasonal price stabili- demonstrations parallel, or even surpass, those zation; they could do grain storage extension of government institutions. Therefore, some work, especially for new grains (e.g., maize African governments might find it useful and in parts of West Africa); they could constitute acceptable to create incentive schemes de- and operate a reserve stock of cereals as a signed to attract private companies willing to first line defense in case of drought or other provide these services.27 food emergencies;26 and they could provide A field with broad possibilities for private for the needs of collective consuming units, sector participation is production and distri- such as the army. This is obviously a large bution of quality seed; poor seed quality and set of tasks; to carry them out well would irregular and late renewal of seed are impor- strain existing capacities in public sector food tant sources of agricultural stagnation, for in- marketing organizations. But they cannot stance in groundnut cultivation. In many perform those tasks well while they are grap- countries, there has been a marked deterio- pling with the intractable problems of trying ration of seed quality through improper pro- to control trade in food grains. duction and multiplication practices by The role of the private sector in input pro- parastatals (Mali, Niger, Senegal, and Tan- curement and distribution should also be en- zania are examples). The success of the Kenya larged. The private sector should contribute Seed Company may be replicable elsewhere. to the distribution of inputs down to the farm Before governments began to monopolize [evel, and to their importation and wholesale input distribution, rural traders handled this distribution. While the latter is a task for larger function in association with produce market- commercial enterprises, distribution offers ing and retailing of consumer goods. Input scope for small traders as well (see Box H). supply activities can be more attractive for The private sector is a major partner in input private trade, and costs of distribution could supply activities only in a few African coun- go down in the process, if private traders tries, and a subsidiary partner in a few more. could engage in marketing of food and export There are many reasons for this, besides fre- crops. The more functions private trade is quent lack of encouragement by govern- encouraged to fulfill, the more scope there is ments. The shortcomings of research and for spreading transport costs and overheads, extension work have held back farmer de- and this will reduce cost to farmers. A broader mand for farm supplies; in countries experi- variety of goods brought to rural areas is also encing foreign exchange shortage import an effective way of inducing farmers to pro- quotas on raw materials have limited the pos- duce a marketable surplus of produce on a sibility of local fabrication of simpler equip- more sustained basis. ment. An important issue is the "critical mass" But a marketing system based on compe- factor. A private company is not prepared to tition of government and private sector is in- establish, for example, research or demon- consistent with the principle of applying stration plots for sales of a few thousand tons uniform producer prices throughout the ter- ritory (panterritorial pricing), a system cur- 26. While there is also a need for additional storage fa- rently applied almost universally in Africa. cilities at various levels, a large part of apparent storage This system has been adopted mainly to help needs is the outcome of the current system of food mar- poorer regions. It does so imperfectly and at keting and could be much reduced, shifted to the farm level, or left to the private sector, in a regime with more appro- priate price and marketing policies. Major public invest- ments in storage should be considered only after a liberalized 27. An excellent example of useful innovation by private system of food marketing has come into being: the regional firms in Africa is the introduction of the ultra low volume, and seasonal pattern and level of commodity flows may hand-held sprayer by leading chemical companies. These turn out to be quite different from those observed under sprayers have tremendous possibilities for insect and weed present distorted marketing conditions. control. 66 Box H: Privatizing Input Supply Systems: The Bangladesh Experience with Fertilizer Distribution The Government of Bangladesh has recently intro- gram to develop greater private sector involvement in duced reforms in fertilizer distribution arrangements fertilizer distribution. which shift functions from govemment agencies to In 1978, BADC and usAID undertook lengthy and de- private traders and may have some lessons for Africa. tailed field studies to develop specific proposals for a Due to heavy population pressure on land, growth New Marketing System (NMs). The NMs is designed to inagriculturalproductioninBangladeshdependslargely reduce restrictions on private traders and move toward upon the increased use of modem inputs; the achieve- a more open fertilizer distribution system. Under the ment of a 3.5 to 4 percent rate of growth of output NMS, BADC is gradually withdrawing from fertilizer sales. depends on an annual increase of about 15 percent in BADC sells mainly to wholesalers at "primary distribu- fertilizer use. The Bangladesh authorities concluded in tion points" while retaining responsibility for sales to the mid-1970s that without reform of the fertilizer dis- retailers in remote and inaccessible thanas. All private tribution system such increases were not likely. dealers and cooperatives are permitted to buy from all Under the Old Marketing System (oMs), the Ban- BADC warehouses. Private movement of fertilizer is un- gladesh Agricultural Development Corporation (BADC), restricted except in the five-mile border area. The Gov- an autonomous body under the Ministry of Agricul- emment agreed to develop a system whereby private ture, was responsible for the procurement and mar- dealers can obtain sufficient credit from commercial keting of all agricultural inputs, including fertilizers, banks, although credit has not yet proved to be a prob- pesticides, seeds, irrigation pumps, and various types lem for traders. usAI also financed consultants to help of agricultural machinery. sArc employed almost 7,000 BAX set up and monitor the new private sector fertilizer people in its fertilizer marketing operations alone, marketing system, and to devise measures to reduce drawing supplies from three ports and three factories internal transport and storage problems. from which it moved fertilizer to 67 intermediate ware- In 1978 and 1979, BADC took the first steps to liberalize houses ("godowns") or directly to 423 thana-level marketing. It increased official dealers' margins, per- warehouses. (The "thana" is an administrative unit 100 mitted farmers to buy from any traders, whether or to 150 square miles in size with an average population not in the farmer's own thana, and made it easier to of about 22,000 comprising about 10 unions or 150 become a trader. Backed by a usAID Fertilizer Distri- villages.) The final distribution level was private deal- bution Improvement Grant, BAnc introduced the NMS as ers. Of the 32,000 licensed dealers in 1978, 20,000 were a large-scale pilot operation in December 1978 in the active, each selling an average of about 25 tons per Chittagong Division, which accounts for one fourth of year under a price structure which gave them too small the area of the country and a third of total fertilizer a profit to provide a real incentive to promote sales. consumption. Dealers accounted for about 75 percent of total sales, The marketing system introduced in the Chittagong cooperatives for the remaining 25 percent. Dealers were Division enjoyed a reasonably successful start. Sales required to register at the thana warehouse of their increased over the previous year and forty-five thana area, to purchase from that warehouse, and to sell only warehouses became redundant, leaving mainly those within their union. BADc held all the storage space at in remote thanas which did not attract wholesalers; transit points and intermediary warehouses, as well as retail prices dropped in areas around the primary dis- in most thanas. tribution points and were below official prices, except While the major problem under the Old Marketing in remote thanas. The new fertilizer wholesalers dem- System was inadequate supply of fertilizer, the fertil- onstrated their ability to move fertilizer cheaply and izer distribution system was also handicapped by in- effectively from surplus to deficit areas, selling to both adequate planning and coordination and unavailability farmers and retailers. of sufficient transport and storage facilities. Conse- Based on the successful pilot in the Chittagong Di- quently, there were frequent local or national short- vision, the NMs was adopted and extended to the rest ages. In order to develop a system capable of effectively of the country. Major accomplishments of the Nms as distributing greater quantities in the future, BArC and of mid-1980 include the following: the U.S. Agency for Intemational Development (USAID) * BADC'S fertilizer points of sale will be reduced by set up a small pilot distribution scheme in 1976. This 55-60 percent; about one third of the original 130 was followed by a detailed study of fertilizer marketing thana warehouses have been closed. and distribution, financed by the IDA. This study pro- *In the Chittagong Division, farmer access to fer- vided basic information concernig fertilizer use by tilizer points of sale has greatly increased. season and district, storage utlization, seasonal and geographical flows, prices, and so forth, to help BADC * Prices paid by farmers for fertlizer under the NMs plan its ferilizer marketing storage program. This pro- are lower than under OMS. gram was developed into a usAm-financed project which * A new class of private wholesalers developed as included, among other things, construction of 350,000 intermediaries. tons of additional storage capacity, and a phased pro- (continued overleaf) 67 D Despite the change in system and a local drought, dealer sales to farmers have not developed as expected. fertilizer sales in the Chittagong Division, as a per- The BADc has not yet worked out a new staffing pro- centage of national sales, remained unchanged. gram for staff rendered unnecessary by the N-Nis. Bangladesh's experience shows the importance of There have been problems: the tAas has worked poorly careful preparation of a marketing reform and also how in underdeveloped areas where transport and com- long a process it can be. It has taken five years of munications are inadequate and fertilizer sales low; intensive effort to bring only the wholesale function whether or not dealers assume the distribution func- successfully into the private sector. BADC is now trans- tion has often depended on whether transport facilities ferring its seed and pump operations to the private are good. Commercial credit programs for assisting sector through similar long-term programs. heavy cost. First, the pricing policy either ab- motivation to produce to the existence of sta- sorbs scarce public resources (if the transport ble and permanent market outlets-which is cost differential is covered by a subsidy) or it not only a matter of prices and marketing penalizes producers in more favorable loca- institutions but also of physical access to mar- tions (if the extra cost is covered by averaging kets.29 Since the early 1970s, investment in the producer price). With transport costs sky- feeder roads has been intensified, often within rocketing in recent years, the principle of pan- the framework of rural development projects, territorial pricing has become more costly than but the initial impetus has been blunted by ever.25 The same conclusion applies to the the lack of local financial and organizational pricing of inputs such as fertilizer. Second, resources needed to maintain them. An in- uniform producer prices, without regard to creasing share of resources spent on transport transport costs, are an impediment to re- development is now devoted to maintenance gional specialization. Finally, the system of and rehabilitation. Rural road development panterritorial pricing distorts competition be- and maintenance should continue to hold tween private and government trade both in prominent places in rural development. They crop marketing and in input supply: private are vital complements for the liberalization of traders occupy profitable markets and leave marketing and input distribution advocated unprofitable ones to the state agency. The above. A complementary necessity is to help private traders buy in the most productive farmers equip themselves with means of regions and sell where unit marketing costs transport (carts), increasing thereby their ca- are low, while the state agency is constrained pacity to deliver produce to an accessible spot to buy and sell everywhere, and at the uni- without inordinate expense in terms of labor. form national prices. There is no way that the If a larger number of farmers owned, or had state agency could avoid deficits under these access to, animal-drawn transport, this would conditions. expand the zone of effective coverage along It is also important to give special emphasis both sides of a feeder road. This process would to transport policv in making distribution sys- also increase the economic rate of return to tems more competitive. After a decade of ru- road development. ral development projects the crucial role of transport, in particular feeder roads, is gairi- 29. In a number of countries, in particular in East and ing enhanced appreciation as a decisive ele- Central Africa, more feeder roads will not have much impact ment in the chain of conditions linking farmers' without more trunk roads, which makes road development a costlier proposition than in other countries where the basic network exists, or where distances are short. Moreover, feeder road development by itself is no remedv if import 28. In Zambia, for instance, comparison of costs of trans- policies and foreign exchange allocations are not handled p ort in an accessible and in a remote area showed that each in such a way as to give priority and encouragement to the hectare of maize grown on the farm costs the nation K28 importation of trucks and spare parts; in many countries, in crop and input transport in the former location, and K128 feeder roads are presently underutilized because of a per- in the latter. Net revenue to the nation is K36 per hectare vasive lack of spare parts, which has effectively reduced in the first location, but there is a net loss of K188 per the number of operating vehicles. Nor will feeder road con- hectare on maize grown in the remote area. struction do much good unless the roads are maintained. 68 Donors have a major role to play in helping changes, but should be designed so as to African governments move toward a restruc- minimize negative influences on domestic turing of incentives in agriculture. Infrastruc- production. The most effective food security ture-especially rural roads-is a high priority objective in Africa today is, after all, a reversal almost everywhere. In some countries, im- of the declining trend of production. If Africa port credits for rehabilitation of the road net- had maintained a 1 percent annual growth work and the transport system are rate in cereals yields from 1961 to 1979, cereals preconditions of renewed growth. Such cred- production would be 6 million tons higher; its can suitably be provided within the frame- this is more than 1979 commercial imports work of a structural adjustment loan, which and food aid combined. Real food security involves policy discussion between donor and comes from a dynamic agriculture. recipient, as in the World Bank Export Re- habilitation Credit to Tanzania. Changes in food policy everywhere pose Agric0ulturl Research and Extension especially sensitive problems. Donors can help With the exception of a few crops, yields in governments that need to make adjustments by prvdn tehia .dieadbign Africa are markedlv lower than ins other con- by providing technical advice and bridging tiens Thyaeas eerlysann,i fiac-oeaml,omohefrsolg ,fhnents. They are also generally stagnant, If domnanic-frice exaplr tosooth eforts tosal not falling, while regions like the Indian sub- dom i pcontinent have recorded substantial increase with world market prices, in yields. This reflects a failure of research to Donors can also help by responding sym- provide answers to the problems which con- patheticallv to African concerns about food pathetically to African concerns aboutfood front African agriculture. Too little money has securitv. Without greater assurance on this gone for research in the past two decades; score, indeed, governments may be reluctant expenditure in the mid-1970s was estimated to engage in restructuring incentives, to be 1.4 percent of the value of agricultural African governments, in their quest for food output in Sub-Saharan Africa, about half the securitv, tend to emphasize buffer stocks of proportion of the industrialized countries.3' cereals. However, buffer stocks are an expen- Despite very considerable achievements in sive and riskv road to food security.30 They export crops, much of the recent research ef- are, therefore, best limited to a bridging role, fort has been inadequately focused, lacking a first line of defense, until imports arrive. in continuity, frequently academic, and gen- Donors should explore with African govern- erally has suffered from recurrent funding ments more cost-effective alternatives, in- scarcities. What is needed is a larger, more cluding the possible use of futures markets focused thrust, at both national and inter- as insurance devices.31 national levels, and for export as well as food Adequate food security facilitates policy crops. At the same time, agricultural exten- 30. Initial investment is high and annual costs (losses, interest, treatment, overheads) may amount to 15 to 20 the balance of positive and negative margins realized on percent of the investment. Stocks need to be rolled over forward contracts. This cost could be established with a every 2 to 3 years in order to avoid deterioration, which reasonable degree of precision, for instance, by export-sim- could disrupt the domestic grain market. Administration is ulation for the 1975-80 period. The costs would probably demanding, and there are high risks of additional losses be only a fraction of the holding costs of those buffer stocks through inadequate management. that could be spared through such a scheme, and could be 31 African countries could protect themselves against absorbed by external sources of finance. An intermediate steep incredses in world market cereals prices in case of solution would be to hold African buffer stocks in the ce- increased import needs, by making use of the futures mar- reals-exporting countries with temperate zones. The advan- ket in grains, for instance, by rolling over a certain volume tage would be lower losses due to more favorable climatic of cereals orders (adjusted continuously in the light of har- conditions, and advanced storage technology. vest prosoectsl. The costs of such a strategy (which could 32. James K. Bovce and Robert E. Evenson, National and be operated jointlv by a group of countries and contracted International Research and Extension Programs (New York: Ag- out to an International broker) would consist of fees plus ricultural Development Council, 1975), p. 8. 69 sion services should be consolidated and disease control. In countries with more so- adapted, with further expansion dependent phisticated research environments, there has on the availability of better technological been a marked shift from broad to narrow packages. Pest control should continue to be spectrum chemical pesticides, so as to cope a major area of research and program activity. with the increasingly serious resistance prob- And in some cases, better utilization of ex- lem. In Africa, only some agencies have been isting technology-for example, improved able to do this, because such approaches are seeds-may require associated changes in seed dependent on detailed scientific study of the multiplication institutions. natural factors conditioning pest and disease incidence. This research effort is costly and not worthwhile below a certain "critical mass. " Also, the more specific pest control programs Effective technical packages are scarce in Af- are costly to administer and require a very rica. It is true that for decades research con- complex organizational effort regarding the centrated on plantation crops, such as palm- method and timing of application. These con- oil, coconut, tea, coffee, cocoa, rubber, and siderations call for intensified internationally other industrial crops, such as cotton and financed research efforts. These should be groundnuts,producingnewvarietiesthatwere divided suitably between national, regional, quick and high yielding. Despite these and international research agencies. achievements, average yields remain rela- In the field of food crops, the problems are tivelv low: for cocoa, they are less than half even broader. Since the early 1970s, when of those achieved in parts of Asia, and for these were recognized as a national priority palmoil, only about half of those found in the in virtuallv every African country, efforts di- Far East, even when grown under fairly com- rected at increasing food production have as- parable farm management conditions. This sumed three forms. First, increased attention may reflect differences in underlying poten- has been paid to the food crop component of tial (ecological conditions), or differences in the farmers' households within the frame- scale and effectiveness of research efforts, or work of existing export crop-based agricul- both. Moreover, the need for defensive re- tural projects, e.g., integrating the food crop search to support and maintain improved component in cotton or groundnut projects, technologies has been neglected. Without this abandoning the block farming system in cot- capability, so-called improved technical pack- ton,33 and introducing hybrid maize in the ages may be shortlived in their impact. Im- rotation, making systematic use of the resid- proved plant varieties, for example, will break ual effect of fertilizers deployed on the lead down in the face of new strains of disease or crop. Second, new projects have been launched pests. Research capacities in most African in locations previously neglected, which were countries are unable to provide the back-up mostly those with marginal soils and/or low research to maintain higher levels of produc- and erratic rainfall where cash crops were in- tion, particularly when new technical pack- significant. Thus, these innovations were not ages are introduced from beyond Africa. only untested but also intrinsically hazard- A related problem in export crop research ous. Third, increased attention was given to is the decline in the quality of disease control, irrigated food crops, either through diversi- though this is only partly a matter of knowl- fication and shifts of emphasis in the crop- edge; organizational and incentive factors en- ter, too. The rapid spread of tree crop diseases along the west coast of Africa, for instance, 33. That is, organizing all cotton plots of a village in one suggests that plant quarantine measures have consolidated block, separate from food crop fields. While not been verv effectively enforced. Also, many this has obvious advantages for the treatment of cotton, it countries have fallen behind the present state puts additional burdens on farmers (distance between cot- ton and food crop plots, etc.) and tends to discourage cotton of the art with regard to the techniques of growing. 70 ping pattern of existing irrigation schemes and others expanded. National programs (Sudan, Office du Niger in Mali), or through should be geared largely to testing and vali- creation of new schemes (Mauritania, Niger, dating, under individual country conditions, Nigeria, and Senegal). the improved technology developed by re- Food crop promotion in projects based on gional and international research organiza- a commercial lead crop has often been more tions, such as the International Institute for successful than in the case of pure food crop Tropical Agriculture (IITA) at Ibadan, the In- projects. This is partly explained by the fact ternational Centre for Research in the Semi- that many successes are associated with hy- Arid Tropics (ICRISAT) at its substation in Up- brid maize, which can be grown only in areas per Volta, the International Center for Maize with sufficient rainfall on relatively good soils. and Wheat (CYMMET) in Zaire and Tanzania, Where these conditions exist, there is usually and the International Centre for insect Phys- an export crop as well (often cotton) and an iology and Ecology (ICuPE) in Kenya. These agencv taking the lead in assisting the intro- should continue to undertake and expand the duction of the new food crop. more basic, long-term research required on In the less-favored agricultural areas, de- African agricultural problems, providing na- velopment efforts had to fall back on tradi- tional programs with promising new tech- tional or slightly improved varieties of millet nology for subsequent location-specific testing and sorghum, cowpeas, and traditional types and validation. Another possible source of of maize. The accumulated results of research tasic study is the research establishment of are limited here. Also, the more marginal the the industrialized countries. As of yet, the ecological conditions, the more a variety needs potential of directing capacities to specific to be adapted to the very specific conditions African needs has not been sufficiently con- of the zone. Thus, tradeoffs have to be made sidered and the Consultative Group on In- between yield increases and drought resist- ternational Agricultural Research (CGIAR) or a ance, and agricultural research has not yet similar body should explore this idea. succeeded in producing varieties adapted to While this Report does not propose a prior- these special conditions.34 ity list of detailed research needs, it is certain that future agricultural research programs must PRIORITIES focus on social and economic aspects as well Because the development of research insti- as on technical considerations. Programs must tutions will take time, given the general scarc- be designed to solve farmers' production ity of skilled manpower, a systematic build- problems, i.e., they should give due attention up should begin now. Attention must be paid to labor of both men and women, the ration- to strengthening and reorientating national ale of traditional cropping patterns, the im- research systems and establishing closer sup- pact of land tenure systems on production portive linkages with appropriate regional and opportunities, emerging changes in the rela- international programs. It is also crucial that tive value of crops, farmers'perception of risks, the extent to which present research pro- input and output prices, and similar ques- grams are directed toward high priority re- tions. This implies that research must be based gional and national production requirements on key commodities in the context of the farm be evaluated. Some programs could be cut system as a whole. Given the diversity of back, but many more will need to be created social customs, farm systems, soils, and cli- mate, this makes the coordination of research activities no easy task. It emphasizes the need 34. One essential element of a "technical package" for for the development of clear channels of com- the drier areas already exists, namely contour ridging, deep munication between economic planners, pro- plowing, and other techniques which retain moisture in the ducers, extension workers, researchers, and soil. These methods have received little attention from ex- tension services, in part because they are not regarded as farmers to ensure the continued relevance of "modern." research work in hand. 71 Experience over the past two decades high- Box I: Controlling the Desert Locust lights the importance of localized testing and in East Africa fine tuning of technical recommendations, In 1962, Djibouti, Ethiopia, Kenva, Somalia, Su- especially for annual crops. Such testing needs dan, Tanzania, and Uganda established the Desert to take account of local physical and socio- Locust Control Organization for East Africa (DLCOEA) economic conditions to establish relevance and to eliminate the infestations of locusts and grass- acceptability. Thus, short season drought-re- hoppers, which in the past devastated the crops of these nations. Armed with a fleet of 12 spraying, sistant cereals developed at international re- transport, and communications aircraft, and large search centers may need some genetic numbers of vehicle-mounted spraying machines, modification at regional or national centers to DLCOEA has effectively performed this role: no major meet localized taste preferences. General fer- plagues of desert locusts have occurred since the cat- tilizer recommendations may need fortifica- astrophic outbreak of 1949-52; the outbreaks in 1967- 69 and 1977-78 were both stopped before much dam- tion to meet localized micronutrient age was done. deficiencies, and so on. DLCOEA'S success is explained by superior forecast- A good deal of agricultural research is ing and painstaking development of techniques to thgooddealof internautioal, reginal, ad control locusts through large-scale research funded underwav at the internatonal, regional, and by substantial contributions from both member coun- national levels. Some of it may not adequately tries and foreign donors. reflect farmers' needs. What is mainly re- The DLCOEA Council recently decided to widen the quired now is a critical review of the relevance scope of the organization to include control of army of research programs, better coordination and worms and quelea birds. To this end, FAO and the Centre for Overseas Pest Research (COPR) are sup- division of labor, better monitoring and con- porting DLCOEA in expanding the forecasting service; trol, efforts to achieve continuity of staffing, in developing a smaller and more efficient aircraft and expansion of key programs, especially unit; and in improving )LcoEs capacity to rapidly ex- those directed at food crops. Special research ecute large-scale campaigns. Country financial contnbutions to DLcoEA have been attention should be given to crops which have paid fairly regularly, although at times several mem- important nutritional impacts. ber governments had broken diplomatic relations, or Further, the international agricultural re- were even at war! This can be taken as a measure of search centers financed through CGIAR provide the importance of the organization to its members. The DLCOEA is a useful model for other regions or for a powerful international and regional base for countries striving to control migrant pests. broad, basic research programs together with valuable training facilities for national re- searchers. The most recent addition to the Also, research on pest control is essential. CGIAR, the International Service for National The FAO Food Plan for Africa and the Lagos Plani Agricultural Research (ISNAR), has been cre- of Action have put much emphasis on the im- ated to assist governments to develop and portance of, and scope for, reducing food strengthen their research organizations. Such losses. Pending the development of further assistance can take the form of preparing in- improved technologies in rainfed food pro- vestment proposals for funding by interna- duction and the consolidation and rehabili- tional or bilateral agencies, together with tation of existing irrigation projects, more providing top research administrators for na- attention to pest and disease control (and farm- tional programs. The services of ISNAR could level storage improvement) may be one of the well prove invaluable in supporting national most productive investments to be made in research initiatives contemplated or underway agriculture. In recent years, many govern- in Nigeria, Senegal, Sudan, Tanzania, and ments have given only half-hearted support other countries in the region. International to various regional agencies dealing with these support for agricultural research in Africa, problems, which have been rendered largely although already substantial, is one of the ineffective due to a chronic shortage of funds. fields which calls for more technical assist- A notable exception is the Desert Locust Con- ance, particularly at the national level. trol Organization for East Africa (see Box I). 72 With more support, regional agencies could the use of insecticides have been undertaken coordinate efforts and exchange and pool in- in a number of countries. The most extensive formation on plant quarantine and other program has been in Northern Nigeria, where matters.35 large areas had been freed of tsetse in the The crucial role of seed multiplication and second half of the 1950s. Another project is distribution should be noted. In many coun- underway on the Adamawa Plateau in Ca- tries, there has been a marked degeneration meroon. Implementation of such campaigns of useful varieties through improper seed requires highly efficient organization to meet production and multiplication practices. More logistic problems, and continuing vigilance to attention should be given to alternative in- maintain tsetse-free status. In Nigeria, for in- stitutional arrangements in this area-for ex- stance, it has not been possible to prevent the ample, delegating seed production and resurgence of trypanosomiasis. Tsetse eradi- multiplication to cooperatives and contracting cation by insecticides also has undesirable out seed production to individual growers are effects on the environment. Continuing re- allowsing private sector activity in this area, search on this and on alternative methods leaving to government the important function (sterile males, fly traps) is needed to devise of qualitv control and certification. methods that are effective and durable, unob- Research needs in agriculture-related ani- jectionable from an environmental point of mal husbandrv are also substantial. Technol- view, and economic. Although the cost of ogy for raising productivity of forage crops is eradication by insecticides seems low ($20 per inadequate. Existing research on vaccination hectare for eradication, and $5 per hectare per has resulted in useful technology for the four year for maintenance), it is more than the or five major animal diseases, but much re- projected benefits from increased animal pro- mains to be done, particularly tsetse control. duction alone. For agricultural purposes, The presence of tsetse flies and the result- however, the cost would seem to be very ac- ing disease, trvpanosomiasis, virtually pre- ceptable. It is mainly the lack of a technically clude the use of some 10 million square sound method of disease control which stands kilometers of higher rainfall areas for live- in the way of a broadly based attack on the stock or agrcultural production. Methods of tsetse in the near future.36 control of tsetse which have been used or are Given the vast scope for expanding rainfed proposed for use include: clearing of vege- agricultural production in well-watered areas, tation which harbors flies; breeding and re- the payoff of intensified research on eradi- lease of sterile male flies; and use of fly traps. cation and control measures would seem to Large-scale campaigns to free extensive areas be high. For several countries, there are im- of Sudanese savanna country from tsetse by portant issues of possible tradeoffs between high-cost investment in irrigation and ex- 35. The Desert Locust Control Organization for East Af- panding the area usable for rainfed agricul- nca (PLCOUA). an intergovernmental body, was established tureunderfairlygoodrainfallconditions. Once in 1962. The Organisation Commune de Lutte Antiacri- a technically sound and economic method of dienne et Antiaviaire (OCLALAV), established in 1965 (Mali, tsetse eradication has been found, an inter- Mauritania, Niger) has the same status. Other regional agencies include: the International Organization for the national consortium, such as the one pro- Control of the Afncan NMigratory Locust (OICMA), established moting the international river blindness in 1962; the International Red Locust Control Organization for Central and Southern Africa (IRLCO-CSA), established in 19H9. the Coimmission for Controlling the Desert Locust in 36. A breakthrough in control of the disease may be fore- North-wVest Afnca. established in 1971; and the Interafrican shadowed by recent genetic engineeening research which Phvtosanmtary Council (tArSC), established in 1956. At pres- affects the impact of the trypanosome on the blood cells of Cnt the lAis has a membership of 48 African countries. The host animals. Genetic engineering is an example of the FAO has also created a Panel of Experts on Integrated Pest potential value of applications of basic research from the Control, which held its first meeting in Rome in March industrialized countries, wider use of which is recom- 1980. mended above. 73 eradication program in the Sahel, could be an 1970s, for instance, some staff were operating appropriate form of organizing a concerted only the first two months out of a six-month effort in this field, since in most cases parallel budget period due to lack of fuel for their and concerted measures in neighboring coun- vehicles, and the situation in Kenya was bet- tries are of the essence.37 ter than in most other countries. Further, there are very few women extension officers, de- EXTENSION SERVICES spite the fact that women in many cases are There has always been some diversity in or- heads of household and almost always pro- ganizational approaches to rural develop- vide a substantial part of the rural labor. Male ment, but generally in Africa agricultural extension staff, however, have limited access extension services have been a major orga- to women in many cultures and therefore talk nizational instrument. Agricultural ministries to the wrong people.35 or parastatals have spread new technology Perhaps the most important problem is that, via extension workers, who usually provide in many instances, the staff does not have a inputs and credit as well. But a successful validmessagetoextend. Wherelandand water extension effort requires an appropriate in- resources are poor, improvements that can be centive structure, an ability to "deliver" what proposed based on current knowledge are the farmers need, and an attractive "package" often marginal. This is in itself an impediment to offer. These elements rarely coexist-which to successful innovation because a substantial explains why resourcesputintoextension have gain is usually required if a technical package produced limited results. is to be readily accepted by farmers: farming No extension effort will be successful if the systemsoperatingunderharsh conditions have pnce of the target crop is fixed by the gov- lower flexibility and farmers, therefore, need ernment at a level which makes the crop un- a stronger incentive to change. The risk fac- attractive for farmers. Although this appears tor, in particular, assumes considerable sig- self-evident, many extension services have nificance, given the fact that many projects been fighting uphill battles against low pro- involve attempts to persuade farmers to incur ducerprices. Marketinguncertaintiesalsohave debts for annual inputs or equipment. Failure the same counterproductive effects on the to appreciate the risk factor fully and other extension program. constraints in a traditional farm system ac- The rural development agencies which run count for misguided extension efforts to in- extension operations suffer the organizational troduce early sowing and high-density planting weaknesses characteristic of almost all public or to eliminate intercropping practices. Im- sector units. But these are aggravated by spe- proved understanding of these subjects has cial problems. Distance from the urban cen- led to a thorough reappraisal of these themes ters hampers recruitment of capable managers and middle-level technicians, and this has affected the quality of the advice and the 38. The following recent description of one country sit- farmers' acceptance of the services. The agen- uation is thought to be fairly representative: ciesneed manyemployeesdispersedoverlarge "The extension work program is usually of an ad hoc nature, defined by the sector chief. Little guidance is distances, who must be physically mobile and available on its content, presentation, or means of ex- capable of adjusting to diverse environments. ecution. There is no monitoring of program effective- Even before the recent budgetary constraints, ness. The extension agent is a 'general agricultural agent,' many extension services were typically short with tasks that range from promoting new technology, organizing credit schemes, and supplying inputs, to of operating funds which immobilized staff general administrative duties. The wide-ranging respon- and undermined morale. In Kenya in the early sibility normally results in little organized work being done. Since the working conditions, general support facilities, and supervision of extension agents are poor, salaries low and education requirements for recruitment 37. A Commission on African Animal Trypanosomiasis minimal, it is not surprising that, in general, extension Control has been set up under FAO auspices. results are negligible, and morale in the service is low." 74 and a rehabilitation of some traditional prac- extension services. Among other things this tices-although these insights have not yet should include testing the relevance and found general entry into extension advice. sharpening the focus of existing extension Theseconsiderationsdonotapply,ofcourse, messages, giving more weight to pest and to all crops and all regions. Valid technical disease control, training staff more system- packages exist for crops such as tea, tobacco, atically, developing feedback from farmers sugar cane, cotton, rubber, and-the out- to researchers, and experimenting with al- standing food crop example-hybrid maize. temate information delivery systems, for But the technological packages are generally example, through the news media, input sup- weak for traditional food crops and are es- pliers, and unpaid rural community pecially uncertain in the climatically harsher leaders. regions. So, for these crops and areas there Until results of intensified agricultural re- is at present an imbalance between potentials search are forthcoming, rural development of the technical packages and the extension projects should, to the extent feasible, be built apparatus set up to spread these packages. around a commercial lead crop (cotton, for In projects based on commercial crops, such example) which offers a confirmed technical as cotton, extension overheads are absorbed package, an assured outlet, a means of cost by the lead crop. Thus, where projects focus and credit recovery, and favorable aftereffects entirelv or predominantly on food crops, con- on cereals production from the fertilizer ap- tinued extension efforts should be subject to plied to the lead crop. Projects based entirely careful assessment.39 on food crops should, during this interim Within these limitations, there is still con- period, be smaller and of a pilot nature. siderable scope for disseminating existing Over the last decade, it has been recog- technology. In addition to the export crops nized that labor bottlenecks are a key con- mentioned earlier, this is the case for such straint to agricultural progress in Africa, but food crops as rice and maize in Guinea and rural development strategies have not fully savanna-tvpe zones. But for Africa's food sta- reflected this insight. Instead, most of the ples, it is verv unlikely that the next decade methods encouraged still aim to increase pro- will vield dramatic production break- ductivity of land (fertilizer and seed pack- throughs. In fact, renewed and greatly en- ages). More emphasis should now be placed largedresearcheffortsareneedednoz-ifthere on measures that increase labor productivity, are to be better results in the 1990s-since the in particular, use of farm implements, ox- development and validation of new technical drawn cultivation,40 use of cereals processing packages normally take at least a decade. equipment (winnowers, threshers), and Pending the development of new messages, equipment aimed at reducing the labor input donors and African governments together of women's tasks (mills, improved water sup- should review and rationalize the existing 40. A breakthrough in ox-drawn cultivation would ob- 39. It is often argued that, even in the absence of a tech- viously have the most powerful effect on labor productivity. nical breakthrough, extension efforts are justified if they However, progress has been strikingly uneven in different only succeed in spreading the so-called best farmers' prac- countries, and on the whole disappointing. This is another tices to all farmers. This argument has some merit, although field in which too little is known about the conditions gov- the scope for progress mav be smaller than appears from erning acceptance or rejection of this innovation and where comparing yields between "best farms" and average farms, more research is warranted in order that officials can better Best traditional practice is visible to all farmers and, to the target future efforts. Prime factors appear to be: varying extent that they are willing and able to emulate the example, costs of maintaining cattle in the off-season; varying costs they can be expected to do without extension advice. This of destumping fields, which is a function of density of suggests that differences in natural ability and assiduity are vegetation and frequency of rotation in shifting cultivation responsible for part of the observed differences in yields, systems; and familiarity with animals and social taboos at- and differences in land quality, structure of the family labor tached to handling animals. In addition, there are some 10 force, etc., for another part. The part due to information million square kilometers in Africa where ox-drawn culti- gaps is probably relatively small. vation is not feasible due to trypanosomiasis. 75 ply).4' This does not mean that research aimed tion. It is equally natural for people to per- at raising productivity of land should be cur- ceive the irrigation solution in large-scale tailed; the land constraint exists in some areas terms-big dams and full water control over now, and will surely become more pervasive thousands of hectares. in the future. For many African countries, large-scale water control will undoubtedly animate agriculture in the future. But recent experience and eco- nomic analysis suggest that this will only hap- Irrigated agriculture has a small place in Af- pen with proper preparation and with the rican economies, except in Sudan and Mad- right sequence of policies and programs. agascar. Irrigation plays a significant role in some of the Sahelian countries (Mali and Sen- RECENT PERFORMANCE egal), to a lesser extent in the northern zones The development of irrigation has made some of Cameroon, Ivory Coast, and Nigeria, and advances in the last two decades, yet the net in the river valleys of Ethiopia, Mozambique, impact of irrigation on total agricultural pro- Somalia, and Zimbabwe. Estimates of areas duction has remained modest in all cases ex- under irrigated cultivation differ widely, since cept Sudan and Madagascar. The problems there is no generally accepted definition of which irrigated agriculture is experiencing in "irrigated cultivation." In some cases it in- the modern sector have been summarized for cludes only full water control (storage or the Sahelian countries by the Club du Sahel pumping), in others, partial water control. and Comite Inter-Etat de Lutte Contre la Sometimes only areas in formal schemes with Secheresse dans le Sahel (CILSS) in a recent some kind of organized administration and document.42 To a striking degree, the defi- field services are counted, in others, formal ciencies and problems identified apply also to and unassisted schemes are included. Count- other countries with an important irrigation ing only formal schemes with full water con- sector (Madagascar, Mozambique, Nigeria, trol, total irrigated area amounts to some 2.5 Somalia, and Sudan). million hectares, of which 65 percent is in In spite of considerable investment in irri- Sudan and a further 15 percent in Madagas- gation development in the 1970s, total culti- car. In Madagascar, irrigated land occupies 50 vated areas hardly increased in a number of percent of cultivated areas, in Sudan, 75 per- countries. While there appears to have been cent; in all other African countries the figure a net increase in total cultivated areas is below 10 percent and in most cases below throughout the 1960s and early 1970s, sub- 5 percent. Apart from the paramount role of sequent additions to the developed area were cotton in the irrigation sector of Sudan, and offset by others that had to be abandoned and sugar schemes scattered all over the conti- require rehabilitation, and this was in spite nent, irrigation schemes are used predomi- of considerable investment in a number of nantly for rice cultivation. countries. Moreover, not all developed areas There is a natural tendency for people in are farmed, and not all farmed areas are har- poor countries dependent on dryland farming vested, either because water control is not to see in irrigation the brightest hope for ag- complete, or because of deficient water man- ricultural development and in particular an agement. In Sudan, extensive restoration is answer to problems of lagging food produc- now being prepared in the White and Blue Nile pump scheme areas and the improve- ments required in Madagascar's irrigation 41. In some cases (for example, Malawi), it was found that the main income effect of agricultural projects resulted 42. The Development of Irrigated Agriculture in the Sahel from the return to labor in activities outside the sector, made (Ouagadougou, Upper Volta: Club du Sahel/CILSS, April possible by the labor-releasing effect of innovations. 1980). 76 perimeters are no less extensive. These major Apart from poor water management, insuf- rehabilitation needs are the result of omitted ficient leveling of land, soil problems, input or insufficient maintenance. This applies not supply bottlenecks, and use of varieties not onlv to the established operations dating from adapted to local needs, poor economic incen- the 1930s (Office du Niger in Mali, or Gezira tives are a prime factor in this performance. in Sudan), but also to very recent schemes Since irrigated rice growers cannot avoid the such as those in the Senegal delta. Poor main- official marketing channel with the same ease tenance is attributable to inadequate organi- as growers of rainfed crops (at least not in zational capacities and lack of money-the government-assisted schemes), rice farming latter due to low rates of cost recovery and/ is often carried out without enthusiasm. or the failure to set aside collected revenue Farmers concentrate on more remunerative for maintenance purposes. (In Sudan, ma- side activities such as vegetables, livestock, croeconomic problems leading to a lack of and rainfed farming. The lack of economic foreign exchange, and hence of spare parts, incentive also helps explain why, in these fuel, and machinery, have been an additional semiarid areas with recurrent drought, culti- contributing factor.) vation intensities are not higher. While vields have generally been disap- Besides lack of price incentives, weak sup- pointing, in particular those in rice, there are port from the agencies operating the irriga- some exceptions. Sugar schemes have re- fion schemes is an important factor explaining corded varving results but have often achieved poor performance of farmers. All the prob- vield levels comparable to those achieved in lems identified with regard to input supply other parts of the world. In paddy produc- impinge with particular severity on irrigation tion, vields of more than five tons per hectare agriculture. In addition, there are important w ere obtained in the Mwea scheme in Kenya training deficiencies. The CILss report on Sa- and the Semrv scheme in Cameroon,43 with helian irrigation observes of extension work- a substantial proportion of double-cropping. ers: "their theoretical training is sometimes High yields have also been achieved in sev- barely credible to farmers whom they are sup- eral minor schemes along the Niger river in posed to guide." Under African, and espe- Niger and on the upper Senegal river in Sen- cially Sahelian conditions, this shortcoming is egal (although here too problems of main- more acute than in other parts of the world. taining soil fertility emerged after a few good Only Madagascar has a deep-rooted tradition years). In these cases, there is pressure on of irrigation; in most of Sahelian Africa, farm- land, the plots owned by individual farmers ers admitted to formal irrigation schemes are are small, and cultivation intensity is high. usually without prior experience in irrigation The Semry and Mwea projects are noted for and the peculiar discipline it requires, and good management, and the small schemes on most are drawn from a peasantry practicing the Senegal river for a high degree of farmer a very extensive type of farming in areas with participation in management. hazardous rainfall. Thus, expansion of irri- In most cases, however, yields have stag- gation is also limited by the speed with which nated or even fallen over the last decade. Not farmers not accustomed to irrigation can ab- more than 3 tons per hectare of paddy are sorb new techniques and the required culti- achieved per hectare harvested, and not more vation discipline. The gap between developed than 2.0 to 2.5 tons per hectare cultivated. and cultivated areas in many. schemes (for example, in Northern Nigeria) is telling in this respect. Poor water management, partly due 43 In Semrr. Upper Volta, yields and cultivation inten- to deficiencies of the infrastructure but also ,Itv appear to have been sagging in recent years, however, to lack of expertise and management control, and the area actuallv cultivated has remained well below i supl the developed area: this reduces yields per developed area gives rise to irregularities in water supply to less than 3 tons per hectare. which increase farmers' risk and undermine 77 their willingness to cultivate intensively. At assessment of the situation, concludes that the same time, these factors severely limit "due to the delays in implementation, irri- both farmers' willingness and ability to pay gated agriculture will not be in a position to the water charges required for financing contribute in an important way to the Sahel maintenance. food picture before the 1990s, and it will only play a significant role in providing food and REHABILITATION NEEDS security for the region toward the end of the These experiences, combined with present ur- century."44 Thus, rehabilitation of existing gency for quick-yielding programs and the need projects, involving the issues listed above, will to prepare for large-scale irrigation efforts in take the remainder of the 1980s under the best the future, suggest that priority be given to of circumstances. Moreover, increased yield raising production in existing schemes to levels levels, when attained, must be shown to be justifving the very substantial investments sustainable before further large-scale invest- planned in the subsector. Unless close to6tons ment can be justified, and preliminary evi- of cereals are grown per hectare per year, in- dence with some projects suggests that vestments of $10,000-$20,000 per hectare can- maintaining soil fertility may pose problems. not be justified. Required consolidation and Rehabilitation involves many tasks for which rehabilitation measures include: external sources of finance can provide useful assistance. First, restoring infrastructure is • improving the operation of existing pr j expensive and may amount to several thou- ects with respect to water management and sand dollars per hectare in individual cases. agricultural services, and closer study of Second, technical assistance will be required soimprovdingonomis inenr .for water management, soil studies, and • improving economic incentives for farmers training of aricultural and infrastructure and increasing their participation in oper- maining off.g ating and maintaining the irrigation maitenance staff. schemes; - rehabilitating infrastructure (drainage, water RIVER BASIN DEVELOPMENT distribution, land leveling); Large-scale irrigation through the develop- - reducing the size of holdings where they ment of river basins has received high priority are too large to be cultivated intensively, in the economic planning of many countries, and introducing double-cropping where in particular since the early 1970s. Mali, feasible; and Mauritania, and Senegal are preparing to de- * increasing the rate of cost recovery. velop the Senegal valley through dams at the Efforts in these directions have been urged estuary (Diama) and upstream (Manantali). from various sides (FAO, CILSS, for example) The total cost of the two dams (without count- since the early 1970s and were initiated in a ing any irrigation infrastructure) is now es- few cases, but rehabilitation programs have timated at $890 million. The Kandadji dam, fallen seriously behind schedule. The so-called at a less advanced stage of preparation, is to First Generation Projects of CILSS (focusing on increase the irrigation potential of the Niger rehabilitation) have made little progress to date valley in Niger; its feasibility is still under and may continue to move more slowly than review. For the Sahel countries as a whole, anticipated. The problems and delays en- CILSs had set a target of 550,000 hectares to be countered in redressing the infrastructure and developed by the year 2000. Major schemes overall organization of the Office du Niger, planned in Eastern Africa include the Kagera the oldest and largest project in the Western basin (Tanzania) and the Badhera dam (So- Sahel, are indicative of the situation; the or- malia), the latter estimated to cost $600 mil- ganizational and technical problems of So- ciete d'Amenagement et d'Exploitation des 44. "Strategy for Drought Control and Development in Terres du Delta (SAED) in Senegal provide an- the Sahel" (Washington, D.C.: World Bank, September 1980), processed. See also The Development of Irrigated Agriculture other example. CILSS, in a remarkably sober in The Sahel. 78 lion. In Nigeria, the Sokoto Rima and easily surpass debt service for the irrigation Hadejia- Jama'are river systems are being infrastructure. The Organisation pour la Mise developed, and several schemes of smaller en Valeur du Fleuve Senegal (oMvs) program scale are under way in the Middle Belt of that in Senegal, Mali, and Mauritania is a good country; the potential of Lake Chad is being illustration of this. Thus, raising domestic tapped by a further large scheme in north- consumer prices of cereals to the projected eastern Nigeria. cost of locally produced substitutes will be River basin development has been attrac- indispensable if such schemes are not to lead tive to governments and donors alike since it to permanent large subsidies. seems the obvious way to food security and Newriverbasindevelopments, then, should agricultural expansion for resource-poor and be undertaken when the technical and insti- arid countries. But without a solid prepara- tutional foundations have been laid and tested, tion in irrigation agriculture and close assess- and the decision has been made that the do- mentofeconomic, social, andecologicaleffects, mestic price structure will be brought more and without restructuring policy to improve in line with the prospective cost of the planned vields on existing projects, building new dams production programs. Given the extensive will prove to be costly, and, indeed, contrary rehabilitation and consolidation needs dis- to the quick-yielding investments most coun- cussed above, governments should prepare tries require to overcome the present state of future schemes more thoroughly than in the stagnation in agriculture. past, giving due attention to pedological is- The financial aspects of river basin devel- sues, watershed management, ecological im- opment need special mention. Most irrigation pact, other economic activities (livestock, schemes are built with the goal of growing fishing, rainfed farming), and land tenure is- food crops, and thereby increasing the pro- sues, and preparing future irrigation farmers portion of cereals grown under droughtproof for their tasks, including maintenance. Re- conditions. But the question must be raised habilitationofexistingschemesmaybeviewed, whether import substitution is economically for many purposes, as a first step for new justified for these commodities even assum- schemes since it will provide a training ground ing that higher levels of efficiency and yields for both govemment staff and farmers, some can be achieved. In Sudan, due to special of whom can fulfill pilot functions for future natural conditions and the effect of existing larger schemes. large infrastructure, new areas can be devel- The international donor community can play oped at costs in the $5,000-$10,000 per hec- a useful role in financing studies, encourag- tare range. But, generally, costs are higher. ing international cooperation between coun- Recent projects in Niger, Mauritania, and tries concerned, and providing direct technical Northern Nigeria have all cost more than assistance to both national and supranational $10,000 per hectare in 1980 prices, some even development agencies. In the past, most su- more than 520,000 per hectare. Even assum- pranational agencies (Lake Chad Basin Com- ing efficient production, a ton of rice pro- mission, Niger Basin Authority) had a duced in a modern irrigation scheme somewhat uncertain existence, since most developed at $10,000 per hectare is estimated member countries preferred to pursue proj- to cost not less than $600 per ton. But im- ects directly and enlisted donor support with- ported high quality rice cost $400-$450 per out recourse to these agencies.45 The ton in coastal countries of Africa in 1980; the effectiveness of these agencies will depend kind of rice ("brokens") imported in Senegal, mainly on the support and responsibilities Gambia, and Mauritania -the largest rice- African countries are willing to give to them consuming area in Africa outside of Mada- in the future, though external assistance can gascar-is 40 percent cheaper. Thus, unless play a significant facilitating role. consumer prices of wheat and rice are raised bv a vervsubstantialmargin, productionwould hav e to be subsidized. Indeed, subsidies might 45. The OMVS is an exception. 79 works which provide supplementary wet sea- SMALL-SCALE IRRIGATION son irrigation, particularly for rice, to blocking While formal irrigation schemes struggle with run-off by contourridgingorbunding forwater the technical, human, and financial problems conservation which increase residual mois- outlined above, there has been vigorous ture for the benefit of rainfed crops planted growth of informal small-scale irrigation in at the end of the rainy season. All are low- flood plains and swamps. In Nigeria, the area investment improvements which can be un- is estimated to have more than quadrupled dertaken by farmers with appropriate exten- between the late 1950s and 1970s. Similar de- sion advice as an integral part of a service velopments, in proportion to local potential package that accompanies the assistance. An and country size, have occurred in Ivory Coast, aspect deserving special attention is the or- Liberia, Senegal, and Sierra Leone, to name ganization of effective maintenance services only a few. The informal schemes have made with farmers' participation, because this is still a much larger contribution to food produc- the critical element in pump-based small- tion, for instance in Nigeria, than the formal holder irrigation. irrigation schemes. In Madagascar they are Utilizing irrigated bottomlands, rice could thepredominantformofproduction.Themain be grown as a wet season crop. Sufficient types are seasonally inundated depressions, technology is available to reach, with a rea- shallow swamps, and river valley bottom- sonable amount of water control, average lands. Most of these lands have been devel- yields of close to 3 tons per hectare. Suitable oped bv the farming population with little, if dry season crops are vegetables (onions, to- any. help from governments, which demon- matoes, carrots, and so forth), cowpeas, and strates the overall attractiveness of this cul- maize. In several countries, there is also sub- tivation system. stantial scope for small-scale irrigation based The potential for such schemes is far from on groundwater development through exhausted in manv African countries. In Ni- pumping: by hand-operated pumps in sim- geria alone, for instance, the area which could pler cases, by small-motor pumps in other be expanded is estimated at 2 million hec- cases. Systematic efforts to tap these re- tares, and of the existing 800,000 hectares, sources are now being undertaken in Nigeria, many schemes could benefit from structural where the farmers are responding positively, improvements and intensification. But the and government is shifting the emphasis of potential use of riverine bottomlands and its irrigation policy from the large-scale sw.vamps is determined by factors such as land schemes favored in the 1970s to support for titles and tenure, availability of labor, and small-scale ones. allocation of labor between rainfed and irri- Donors can help primarily through tech- gated crops. The question of land tenure, nical assistance for watershed management particularly on acquired and improved land, and land-use planning (including the use of is a matter which determines the degree of satellite pictures, which has proved useful for improvements farrners are willing to provide. this purpose); transfer of technology (tools, More attention ought to be given to these pumps), with special reference to the tech- issues, since the degree of self-help that can nology available in Asia; and training of mid- be mobilized will have a strong impact on dle-level irrigation technicians to help farmers project design, construction method, and, in surveying and laying out irrigation plots. above all, costs. Donors can also help by studying the alter- This tvpe of traditional small-scale irriga- native organizational forms which exist in tion depends on readily available water re- large-scale water control schemes in Latin sources in the form of rainfall, run-off, and America and Asia; smallholder-based orga- natural storage, which is carried over into the nizational arrangements are frequently used drv season. Irrigation systems may range from in these regions in ways that might be adapt- simple water diversion and redistribution able to African circumstances. 80 e~~1 - 'i4li 5OURCES The importance of the human factor in Afri- development limited to economic growth. In- can development has been repeatedly stressed creasing attention is being paid to its effect in this Report. This chapter discusses policy on distribution and social equity. issues in education and training, as they re- late to African economic development, and in THE AFRICAN CONTEXT health strategies. Faster economic growth in Africa requires accelerated development of human resources. This involves more and better formal school- ing and intensified training. There is agree- ment among African and other experts that EDUCATION AS AN ECONOMIC INVESTMENT schooling systems must be expanded and im- To begin, a few points need to be stressed proved, especiallyattheprimarylevel. Inabout regarding the complex role of education in one third of all Sub-Saharan African coun- development. As defined here, education tries, fewer than half of all primary school covers all schooling (both formal and infor- age children are in school (see Table SA.38). mal), and must be considered as an invest- Secondary education should also be substan- ment as well as a consumption good. For tially extended; in only about 40 percent of policymakers,thismeansthateducationalcosts the African countries are more than 15 per- should be balanced against potential private cent of the relevant age group in secondary and social benefits. school. And while university education has The impact of education extends beyond spread rapidly in the last two decades, there the traditional production sector into the are numerous places where the output of household. Educated women, even if they do university graduates is still far short of de- not participate in the labor force, can have a mand. In Nigeria, for example, recent man- significant impact on the country's economy power estimates indicate that in certain through lower fertility rates, health informa- specialtiestherearealmosttwiceasmanyhigh- tion, and more "household production."' level job openings as there are university Moreover, education does not relate only graduates. to the modern wage sector. Farmers and self- Expenditures on schooling already claim a employed people in the urban sector are now large part of CDP-around 4 percent in two thought of as contributing more to the coun- thirds of the countries for which data are try's economy if they have a higher level of available. And, more important, they claim a education. Nor is the impact of education on sizeable share of public expenditure-about 16 percent of the total, on average, more than 1. See T. W. Schultz (ed.), The Economics of the Family any other government function except gen- (National Bureau of Economic Research, 1974). eral administration (see Table SA.41). In a 81 Table 6.1. The Social Returns to Education concern on several aspects of educational in Africa strategy. Is it possible, for example, to reduce Rate of return costs without impairing educational quality? Education level (percent) Has the pace of certain levels of education Primary 29 been too rapid? These and other issues are Secondary 17 discussed below. Higher 12 Souirce: George Psacharopoulos, "Returns to Education: THE INADEQUACY OF RESOURCES an Updated International Comparison," in Timothv King (ed.), "Education and Income," World Bank Staff Working Given Africa's extreme shortage of fiscal re- Paper, no. 402 (Washington, D.C., 1980). sources and the many claims on revenues, all educational strategies must have as a key ob- jective greater efficiency in resource use. Af- significant number of African countries, re- rican education is expensive not only in the current expenditure on education is between sense that it absorbs a significant share of 25 and 35 percent of total recurrent spending. public sector resources; it is expensive also in In the 1970s, when government revenues and terms of average costs per pupil, especially expenditures rose rapidly in most of the con- at the higher level. Affican governments spend tinent, the average African country's incre- as much per university student as countries mental share going to education was 13 with per capita incomes at least three times percent-again larger than any other single and as much as eight times higher. By con- item except general administration (see Table trast, primary education is cheap in compar- SA.42). ison with industrialized countries. Table 6.2 That educational expenditures claim a large shows that primary education costs (per stu- and, in some countries, growing share of re- dent year as a percent of per capita GNP) in sources does not necessarily mean that gov- Africa are about twice as much as in other ernments are giving "too much" to education. developing areas; secondary education costs On the contrary, a recent review of the eco- are 4 to 5 times as much, and higher educa- nomic returns to educational investment in tion costs 5 to 10 times as much. African countries has shown the returns to Costs are high in terms of years of instruc- be substantial (see Table 6.1). If one measures tion per graduate, since dropout and repeater the output effects of more education by rate- rates are high in most of the region. At the of-return analysis, investment at all levels of primarylevel,itcommonlytakesl0pupilyears education still appears to produce a relatively to produce a primary school graduate on a high yield. six-year course; only between one third and As far as allocation within the education one half of students who enter primary school sector itself is concerned, the general order complete the usual six-year cycle. Dropout of priority runs as follows: first, primary; next, and repeater rates are lower at secondary secondary; and last, university-although in- dividual country requirements differ, and Table 6.2. Costs of a Student-year countries particularly short of university- as a Percentage of GNP per Capita trained people would find fault with this or- dering. All in all, there is much evidence that Region Primary Secondary Higher generalized primary education has far-reach- Eastern Africa 20 124 927 ing modernizing effects.2 Western Africa 24 142 1,045 In spite of the results of educational cost- Asia 11 27 205 benefit calculation, there exists widespread North Africa 15 47 306 Latin America 11 22 121 2. World Bank, World Development Report, 1980 (New York: Source: World Bank, Education Sector Policy Paper (Wash- Oxford University Press, 1980), Chapter 5. ington, D.C., 1980). 82 Table 6.3. Relative Teaching Costs and Student-Teacher Ratios, Public Primary Schools, Selected West African Countries Teaching-implied Ratio of teacher Teacihing costs per costs of universal salaries to per student as a percent primary education Country (year) capita GNPa Stutdent-teacher ratio of per ca pita GNP (percent of GNP) Liberia (1977/78) 3.8 51 7 1.4 Togo (1978) 6.3 60 10 2.0 Sierra Leone (1973) 4.1 32 12 2.4 Cameroon (1976/77) 6.7-11.6 52 18 3.6 Ivorn Coast (1975) 6.8-10.8 44 20 4.0 Nigeria (1976) 6.2 30 20 4.0 Upper Volta (1978) 24.0 51 47 9.4 a. Includes apprentice teachers and monitors. Souirce: World Bank data files. school level, though still substantial. At the educated people in the 1970s is not yet fully university level, these rates vary. For exam- mirrored in public sector wage rates.3 ple, they have been high in Liberia, very high Given this situation, African governments in Ivory Coast (10 student years required for would need to reduce unit costs by increasing one three-year graduate), and low in Kenya pupil-teacher ratios to the extent possible, and Nigeria. without reaching the threshold where quality The principal factor behind high unit costs would be adversely affected. Other cost-sav- is high teacher salaries, which typically make ing systems (e.g., double-shift, multigrade up 75 to 90 percent of total recurrent costs in teaching, and so forth) should also be con- education. In some countries, salary costs are sidered. Wider use of textbooks and other high in part because expatriates are still widely reading material deserves special considera- used at the secondary and university levels. tion (see Box A). But even at the primary level, salaries are African educational planners and policy- high in relation to per capita income. It should makers and their outside partners can also be noted that the salary weight is so high avoid costly mismatches between types of because all modern sector salaries are high schooling offered and social demand. One relative to average incomes. In any event, Table example is the tendency to establish technical 6.3 gives some West African cost figures; scat- secondary schools in systems where places in tered evidence suggests that costs are lower general secondary facilities are few relative to in East and Southern Africa. Since about one demand. These technical schools are -much fifth of the population is of primary school more expensive to build and to operate than age, universal primary education at the cost general secondary schools. Since the career ratios of Table 6.3 would imply teaching costs prospects for general secondary school grad- alone ranging from 1.4 percent of GNP in Li- uates are usually better than for graduates of beria to 9.4 percent in Upper Volta. This is technical school, the technical schools often to some extent a reflection of market forces; end up serving disaffected people who wanted educated people remain scarce, and the de- a general education but were denied entry to mand for teachers has grown rapidly in the a general secondary school. In this case, the 1970s. But in many instances it is a reflection technical schools provide more general in- of the rigidity of public sector salary scales. Wage rates in the public sector are tied to levels of educational attainment and do not 3. It is interesting that Liberia had no formal civil service wage structure until very recently. Government salaries there readily adjust to market conditions. Thus, in were not rigidly linked to levels of education. This helps many countries, the increase in the supply of explain its position in Table 6.3. 83 Finally, a more generalized use of fees in BoxiA: Ethoa's the public system would certainly help, not only in financing a more rapid expansion of Ethiopia's campaign to eliminate adult illiteracy has the education system, but in bringing social made rapid progress. In 1979, well over 90 percent and private costs and returns closer together. of Ethiopia's population of about 30 million was il- This is especially relevant at secondary and literate. By early 1981, seven million Ethiopians- over 70 percent of them women-had benefitted from unversity levels, where restriction of schol- the National Literacy Campaign. The majority of these arships would bring substantial economies had attained basic literacy and numeracy. The cam- and make private and social returns more paign gained worldwide acclaim in 1980, when the equal. But such changes, often proposed, have International Reading Association, which is affiliated with Unesco, gave Ethiopia its annual literacy award. proved politically difficult to implement, and The Association was particularly impressed with the nowhere in Africa have student loan pro- follow-up of the initial teaching of literacy and nu- grams replaced scholarships. meracy with courses using simply written books to teach hygiene, better farming techniques, and other skills. IMPROVED QUALITY The campaign has overcome formidable obsta- Better quality education stands high on the des-notably diversity of languages and enormous financial and administrative constraints. About a list of educafional priorities. In more than half quarter million instructors have been mobilized-pri- of the countries of the region, 50 percent or marily teachers, students, civil servants, and army more of the teaching staff lacks formal train- personnel. Many of these are active in the campaign ing of the standard required in the countries only in the summer. The instructors work out of in ofsthe standar r re inathe outdies about 35,000 literacy centers. Peasant associations in question. Textbooks are scarce, outdated, across the country make the required logistic arrange- and often inappropriate. Buildings and equip- ments, induding provision of food and lodging for ment wear out more rapidly for want of main- the instructors. The government provides books and tenance. Clearly, more and better teacher supplies such as chalk and blackboards for use by instructors. By September 1980, the Government had trainng, textbooks, and maitenance of fa- distributed some 15 million copies of literacy material cilities would contribute substantially to im- in Amharic and four other languages. Total govern- proved quality of instruction. Adaptation of ment costs under the program have been extremely the curriculum to local needs and reform of low-15 million birr up to September 1980, or less the system of national examinations are also than 2 percent of annual government expenditure- y according to one calculation. These low costs are partly important factors. explained by the fact that many of the instructors are unpaid, while the salaries of teachers on the public PLANNING, ANALYSIS, payroll and military acting as instructors are not at- tributed to the campaign. The Government plans to continue the campaign Cost reduction and quality improvement- enrolling up to several nillion new peasants annually central themes of all education strategies for in five-month programs until 1986, when the entire Africa-require strong institutions and well- population will have been covered, developed analytic capacities. Development of locally appropriate reforms, their imple- mentation and evaluation, cannot be done struction than they should, at much cost, to people who will rarely work as technicians. had some predictable problems in defining its role, but aside It would be much more economical and more from that, its existence illustrates the inefficiency problem. educationally efficient to create additional The unit salary costs of these Folk Colleges are 2.5 times general secondary capacity.' higher than those in general secondary schools. But Tan- zania has one of the lowest rates of secondary school at- tendance in Africa and the world (4 percent of the age group, the same as Malawi, Somalia, and the Yemen Arab 4. The same point applies to other programs. In Tanza- Republic). It also suffers acute scarcity of middle- and high- nia, a program to create Folk Colleges on the Swedish model level manpower. In terms of efficiency, resources spent on was begun in the mid-1970s. This was an innovative attempt the Folk Colleges would seem to be much lower yielding to provide practical adult education in rural areas. It has than those in general secondary education. 84 without strengthened institutions and ana- arise for expansion of secondary education. lytic capacities. Larger donor efforts in train- The budget impact of some of these pressures ing, technical assistance, and policy-focused can be deflected by allowing private second- research could contribute greatly to this end. ary schools to expand. Recent experience in Kenya and Ivory Coast, however, shows that CHALLENGE FOR THE FUTURE public secondary enrollments are ultimately The principal challenge facing African policy- expanded, and substantially. Since, as shown makers as they look toward the end of the in Table 6.2, secondary school unit costs are century is how they can reconcile what their many times higher than primary school costs, people want and need with what their soci- heavy expenditure obligations can be ex- eties can afford. A good example of this is pected to follow. Given the constraints on the pace at which primary education should financing, gradual expansion of primary ed- expand, at least in the short run. ucation would therefore seem in order in As noted earlier, many recent studies find countries where primary education has al- that people with primary education produce ready achieved 50 to 80 percent coverage. more and are generally more open to inno- Efforts to reduce unit costs should proceed at vation than others, making primary educa- all levels, along with attempts to improve tion's economic rates of return relatively high quality. At the same time, the search for bet- almost everywhere. On the basis of these re- ter adapted, more economical systems of basic sults, some people argue that African gov- education should be intensified. ernments should concentrate resources on the While different countries require different expansion of primary education, to achieve priorities, in all cases the objectives of better universal primary education as quickly as quality and management should receive spe- possible. Other observers, however, are more cial attention from donors, and in the frame- concerned with the cost of universal primary work of policy reforms, the cost reduction education than with its potential benefits, and effort is particularly relevant. In general, do- worry whether Africa can afford it now. Al- nors should consider more lending for edu- though rapid expansion to the entire popu- cational software-recurrent-cost kinds of lation might eventually lower unit costs expenditures-especially to improve plan- because of economies of scale, a study of the ning and management. As a key sector and costs of expanding primary education in 27 a major user of resources, education should countries in the 1950s and 1960s found that be a high priority claimant for nonproject loans on average recurrent costs tripled when en- in the coming decade. rollments doubled, because of escalation in the teacher wage bill.5 Of course, if efforts are made to improve the quality of teachers and Traininig other inputs at the same time, costs will be higher still, though the corresponding edu- Training outside the formnal school system highertstioal, b itshough te bcorrespond edu- covers a vast set of activities, from in-service Another concern is that if numbers of pri- classes for civil servants to nutrition educa- mary school graduates rise, and no changes tion for mothers in well-baby clinics. Some are made in curricula, pressures on secondary training matters are discussed in Chapter 9, areo m ladesi cricula, pressure oince sheoaory particularly those related to externally aided school places will increase. Since the major projects and technical assistance. Here, train- function of present primary curricula in most ing only in certain key skill areas will be con- countries is to prepare pupils for entry into sied:pbcmagmntndcomc. secondarv schools, enormous pressures will sidered: public management and economics. These are chosen because they bear so di- rectly on one of the central themes in this 5. Philip N. Coombs and Jacques Hallak, Managing Ed- Report-the need for increased efficiency of ucatiorial Costs (New York: Oxford University Press, 1972). resource use. 85 Virtually all African countries now have their For example, one of the most crucial and per- own facilities for training in public manage- vasive weaknesses in development adrninis- ment and administration, and there are a tration in Africa is in the area of financial number of regional institutions with this as management, and a sustained effort is re- their principal function-the African Training quired across the board to improve budget- and Research Center in Administration for ing, accounting, and auditing standards, to Development (cAFRAD), the East African Man- augment the supply of qualified accountants, agement Institute, the Pan-African Develop- and to develop effective national systems of ment Institute, and several others. Numerous financial control. courses for African officials are offered by These initiatives to broaden and diversify universities abroad, not only in Europe and the supply of formal management training North America, but also in Asia and the Mid- deserve the support of the donor community. dle East. Aid donors have helped to finance Management problems in Africa, as in all de- a great variety of training schemes for middle- veloping areas, are only due in part to a scar- level African managers and other profession- city of training facilities, but this part is als employed in the public sector. Several amenable to elimination and should be ad- projects of the World Bank have been devoted dressed directly. exclusively to training for public service.6 The training of economic analysts presents Many of the African public administration a somewhat similar picture. In the past 20 schools have been going through a period of years, substantial resources have been in- assessment, reviewing the suitability of their vested in faculties of economics by African methods and curricula to the tasks at hand governments, bilateral donors, and private in African public administration. There are foundations. While this has resulted in some new efforts to more effectively combine teach- strong centers of economics education (Nai- ing, research, and consulting roles and to re- robi, Dakar, and Ibadan, among others) ca- late these more closely to the workaday pacity generally remains weak. Meanwhile, problems of government administration and/ most donors have reduced their support for or public enterprise management. In some various reasons, among them changes in cases, healthy diversification is occurring- priorities in their programs. for example, in the new School of Manage- Strong faculties in economics (buttressed ment Science in Dakar, which may become a by science and technology faculties), are the West African regional training center. This cornerstone of formal training in public institution promises to provide an approach administration, business adrninistration, and to public management training other than that many of the social sciences. Stronger uriver- of the traditional national administration sity training in economics is necessary but schools, and hence offers some diversity in does not directly increase the supply of policy- this field. Moreover, while training alone can oriented economists and policy analysts, peo- never be an answer to management prob- ple who are essential for strengthening de- lems, there are some basic value-neutral skills cisionmaking procedures and institutions. Just and techniques that must be widely dissem- how such people can best be trained is not inated if government is to function efficiently. well known. It is likely that the best training is on the job, and that it should be done mainly within government agencies. In the 6 One of the more ambitious ventures underway in this discussion of planning strategy (Chapter 4), area is a program in Madagascar for the training of ac- the training role of national planning countants and auditors, for which a national center has been agencies established. In West Africa, three groups of consultants was mentioned. These agencies have strong were engaged in 1979 to advise on the training of agricul- advantages in a training role; they could be- tural managers, and the Pan-African Development Institute is cooperating with the World Bank's Economic Develop- come the centers of traimg of economc an- ment Institute (EDIt in launching a series of courses to ad- alysts for government in general. dress this particular need. There is a widespread feeling among man- 86 agers and some educators that effective man- every physician and 3,300 persons for every agement training requires a greater "hands- nurse in Sub-Saharan Africa (see Table SA.37). on" character and that on-the-job training is Physical resources are also scarce and poorly in fact the best instrument. One problem with distributed within countries; the majority of on-the-job approaches in the African context health professionals and a disproportionate is that manpower-scarce and financially hard- share of health facilities are located in urban pressed agencies cannot do a great deal of it. areas, where only about 20 percent of Afri- This is an area where donors might move in cans live. Most Africans simply do not have new directions. It would seem possible, for access to modern medical care. example, for bilateral agencies and interna- Financial resources to expand health care tional institutions to take in as trainees a sig- systems are also scarce. In recent years, pub- nificant number of African civil servants and lic expenditure per capita on health varied parastatal staff and put them in operating between $1 and $4, compared with $100-$500 assignments. Their on-the-job assignments per capita in industrialized countries.7 Public could be supplemented by specially designed spending on health has increased over the formal training at universities or, better, in- last decade, although it has barely kept pace house at such facilities as the World Bank's with population growth. In 1979 it claimed 5 Economic Development Institute (EDI) or other to 7 percent of central government expendi- training institutions. Under similar "cadet," ture in most African countries. Continuation "intern," or "research assistant" approaches of low rates of economic growth, coupled with African trainees could also be placed in man- high population growth, would make signif- agement consulting firms and other institu- icant increases in government spending for tions with interests in Africa-banks and health on a per capita basis unlikely. multinational industrial corporations in both Careful consideration of alternatives sug- industrialized and Third World countries- gests four key areas where action can be most perhaps under some umbrella organization conducive to improving the health situation: set up by appropriate donor agencies. The increasing access to low-cost health care; con- basic idea is simple: to use external institu- solidating and upgrading health systems; re- tions as instruments for managerial and tech- search improvements; and improving access nical training on the job. Of course, much to safe water and adequate sanitation. remains to be spelled out in this proposal, which has some obvious pitfalls but merits A STRATEGY FOR INCREASING ACCESS further exploration. TO HEALTH CARE The disparity between the need for care of the uncovered majority and the meager re- sources to provide it-not just financial but also technical and administrative-has led to The African life expectancy at birth is 47 years, emphasis on expanding primary health care bv far the lowest of any region in the world. systems-that is, community-based ap- Much of this is explained by the fact that so -funding, many children die before the age of five; in- and the use of paraprofessionals. Thls ap- fant mortalitv alone is on the order of 150 per proach pse a speia ocalln To a. 1,00 liv bi,rth. Mst o thee dathsare proach presents a special challenge to Africa. 1,000 live births. Most of these, deaths are Wolwd,teehvbenumrspit . . ~~~~~WoTldwide, there have been numerous pilot caused bv' malnutrition and infectious dis- projects for small groups but there are few eases, diarrhea and dehydration, malaria, measles, and respiratory infections. Im- proved diet, sanitation, and health care would 7. Estimates for private expenditures for 12 African coun- eliminate most of these deaths. tries in the early 1970s suggest that private health expend- itures were, on average, about the same as public African health care systems are very lim- expenditures. Together, they accounted for 2 to 4 percent ited. In 1977 there were 25,000 persons for of GDP. 87 examples of successful organization and fi- and irregular drug supplies, however. In many nancing of such schemes on a nationwide scale. cases, budget allocations for pharmaceuticals This is due not only to the fact that mass are used up in the first few months of the expansion of such demonstration projects is year. Most rural people must therefore either too expensive for Africa, but also that the do without, or buy on the illegal market-if needed administrative and managerial skills they can afford it. Reductions in the legal are so scarce in the region. restrictions on sales of basic pharmaceuticals Because health resources in Africa are al- would increase the public's access to these ready spread thin, rapid expansion of exper- drugs and reduce their cost, making a major imentalstrategiescouldcompletelyoverwhelm contribution to the health care of the majority existing structures and fail to provide the ex- of Africans. tra coverage that is their aim. Thus, expan- At the same time, governments should seek sion of African health systems and the ways to generate revenues from at least some adoption of rural health strategies should pro- beneficiaries of publicly provided health ser- ceed gradually, on a pilot basis. Donors should vices. Methods such as industrial insurance encourage a variety of approaches, aimed at schemes and user fees for public services may finding answers to key questions, such as: be unpopular, but may be the only alternative the appropriate speed of expansion of health to systems which are too poor to provide many posts; suitable relationships between com- services at all. munity health workers (CHWS) and staff in the formal health system; recruitment strategies CONSOLIDATING AND UPGRADING HEALTH for cHws; alternative means of payment; and, SYSTEMS particularly, the feasibility of revolving funds. The credibility of primary health workers far One question should underlie the design of from the cities depends on logistical and tech- all projects in this area: Can an externally nical support, drug supplies and distribution, financed program survive financially and or- and appropriate referral mechanisms. African ganizationally once the donor leaves? health systems, however, are weak with re- It follows from the scarcities of financial spect to all of these functions-particularly in and institutional resources that African gov- rural areas. In a number of Sub-Saharan coun- ernments would do well to use a variety of tries, hospitals, clinics, and equipment have institutions and approaches in the health sec- fallen into disrepair. Facilities are often staffed tor, and notably to mobilize private as well at levels far below what is needed, by persons as public energies. Productive use of individ- who have worked too long at their posts with- ual initiative is as possible in the health sector out benefit of "refresher courses" to improve as in others, although there are also risks and their skills. Essential health service support inconveniences. systems, including supervision, drug supply, For example, African countries could also and manpower development, have been un- consider wider encouragement of voluntary derfinanced. As a consequence, services have agencies-local and foreign-many of which generally deteriorated. have done innovative and successful work in Arresting and reversing this trend has be- the field. In this context, experiences in Latin come a central need in many countries if the America could prove particularly relevant. systems are to provide reasonable services, Some governments might find other attrac- much less support new efforts at outreach. tive possibilities along these lines-for ex- This requires, among other things, rational- ample, freer trade in pharmaceuticals. In much izing staffing patterns and consolidating and of Africa, drugs can be legally bought only in rehabilitating hospitals and equipment, as well government dispensaries or pharmacies. In as the rural clinic systems. Another crucial rural areas, often the only legal source of drugs aspect of consolidation is the simplification is the public sector pharmacy or dispensary. and standardization of procedures. For ex- The public sector frequently offers inadequate ample, simple criteria for diagnosing specific 88 or bv soiled water containers. With respect to sanitation, successful use of the simple low- Box B: Provision of Rural Water Supplies cost technologies most appropriate in Africa in Malawi will require fundamental changes in behavior In Malawi, a long-standing, relatively strong gov- that are very difficult to bring about. emment commitment and effective community par- This situation suggests that an appropriate ticipation have brought about steady progress in the long-term strategy should focus on training provision of rural water supplies. Initially, some 4,000 9 . . . deep wells (more than 45 meters) were drilled with and the development of viable institutions, heavy machinery, and expensive handpumps were adoption of low-cost technologies and real- installed. More recently, light drills, plastic screens, istic goals with respect to levels of service, and simple handpumps have been used to exploit and the development of hygiene education shallow aquifers, replacing the earlier, more complex technology. In other programs government-sup- programs. In the short term, an attack on ported private groups have supplied piped water to water and sanitation problems in urban areas villages through gravity systems fed by mountain is most technically feasible and is capable of streams. As of 1977, community labor had laid some rapidly improving the welfare of a large num- 750 miles of pipes feeding 1,800 communal taps, with ber of people. Experience in smaller towns in a cost to the Govemment of only about $6 per person served. The program is well organized, with com- Botswana, Ivory Coast, and Kenya, for ex- munities fully involved in planning, operation, and ample, suggests that there is potential for maintenance. mobilizing community support in these ef- Mistakes were made at the outset, but the lessons forts. Moreover, technologies appropriate to were well learned. The Malawian experience dearly demonstrates that commitment, a disciplined ap- local financial and administrative capabilities proach, and genuine community participation can are receiving official acceptance; an example sustain a continuous, if modest, program. Service is the transformation of the Dar es Salaam coverage had reached 70 percent of urban and 30 (Tanzania) "Sewerage Masterplan" into a percent of rural areas in 1978, and there is every reason to assume that by the end of the decade all "Sanitation Masterplan." The original plan Malawians, urban and rural alike, will have reason- recommended comprehensive waterborne pipe able access to safe and adequate water supplies at sewerage at very high cost. The revised plan affordable cost. proposes onsite, low-cost sanitation facilities (pit latrines, septic tanks, and so forth) for trine programs are taking hold in Botswana about 75 percent of the city's population, re- and Lesotho. taining piped sewerage only in the central A durable supply of water and sanitation part of the city. A similar revision is in prog- facilities depends on the development of de- ress for the Accra/Tema Masterplan in Ghana. centralized local capacity. Shortage of skilled Experimentation in rural water and sani- manpower for planning, design, operation, and tation projects should continue broadly and maintenance will have to be overcome by in- receive more international support.8 For ex- tensive training and, in the short term, by tech- ample, rural water supply projects are being nical assistance. International training is now started or designed in Malawi and Kenya us- being given to small private contractors in the ing simple, lowv-cost technologies, such as lo- construction of pit latrines in Tanzania (as in callv made plastic well screens instead of India and Bangladesh). In Kenya, a World Bank- imported stainless steel, and handpumps in- financed water project is encouraging local stead of piped networks (see Box B). Pit la- "plumbers" to construct small sections of a water distribution network. In Ghana, a local con- tractor has been "loaned" construction equip- S. Research and development programs coupled with ment for major water works. sozoocultural studies are being carried out by a number of agencies to find the least-cost and most acceptable solutions. These efforts to build decentralized local Bevond research work, the UNor Global Project is promoting capacities should be paralleled by programs pilot and full-scale projects in several countries in low-cost of nutrition and hygiene education, which in sanitation. Project Preparatorv Teams financed by UNDP' will commence work in Africa in early 1982 in an effort to assist the long run can have the most substantial governments to prepare projects for external finance. impact on health conditions. 90 diseases, treatment protocols, and criteria for and tuberculosis-could be substantially im- referral might be established. proved. A better measles vaccine would be It should be noted that this emphasis on particularly beneficial, as this disease is more consolidation and gradual evolution differs severe in Africa than elsewhere in the world; from the traditional approach to health plan- it affects more people than do the other EPI ning, in which planners use international diseases, and delivery problems are especially "norms" to determine the "required" number constraining. In addition, millions of Africans of hospital beds per thousand people, dis- suffer from afflictions for which no vaccine pensaries per health center, and nurses per exists, above all malaria, schistosomiasis, and clinic. They then calculate capital costs and onchocerciasis; development of vaccines for the associated manpower and supplies, and these conditions is of particular interest to end up with requirements for future capital Africa. Although much support now flows to and recurrent costs. tropical disease research-including Bank fi- This approach is helpful in setting out nancing-more resources can quicken the pace the dimensions of "needs," but it has one of discovery. disadvantage: it leads to misperceptions Until vaccines are discovered, research on as to the nature of the problem and encour- control of the major vector-borne diseases ages diffusion of effort and excessively grand should receive more attention-malaria, attempts in particular programs. The alter- schistosomiasis, onchocerciasis, and trypano- native approach sketched above, of consoli- somiasis, in particular. Present technology for dation and cautious expansion, can lead to the control of these diseases is expensive and closer analysis of priorities. It can more real- requires complicated large-scale administra- isticallv assess administrative and financial tive and managerial structures. Research on constraints. methods of treatment, including chemother- In fact, the shortage of funds for spending apy, should also have high priority. on health gives special urgency to improved Diarrheal diseases are the most common planning, policymaking, and management. A cause of death in infants and children and an useful beginning is a "sector analysis"-a important contributory factor to malnutrition. study of the problems, policies, and resources Trials in several developing countries, which in the health sector, its detail and depth de- have documented that oral rehydration with pending on how much is already known. As sugar and salt preparations is effective in in other key sectors, donors should expand markedlyreducing deathand disability, should assistance to strengthen capacity in health be replicated widely in Africa. policy analysis and related data gathering, in planning and budgeting, and in program WATER AND SANITATION analysis and general management. According to World Health Organization (wHo) estimates, half of infant deaths are related to RESEARCH NEEDS contaminated water and poor sanitation. In One key area requiring donor assistance is addition, collection of water from distant places research on immunization technology. Many severely burdens African women. Past efforts existing vaccines are unstable, must be kept to improve water supply and sanitation have chilled during all stages of transport and dis- proven vulnerable to technical, organiza- tribution, and often require that the immu- tional, and social constraints, however. Pump nized person return for one or more boosters. maintenance and repair have become key These problems increase the difficulties of problems, especially in rural areas; in many carrying out immunization campaigns in African countries over 60 percent of hand- tropical rural areas. Technology for the six pumps are inoperative within a few months diseases targeted for the Expanded Pro- after installation for want of maintenance and gramme of Immunization (EPI)-measles, repair. In addition, clean water sources are diphtheria, tetanus, whooping cough, polio, often contaminated by animals at wellheads 89 7. OTHER PRODUTIVTE SECTORS A turnaround in agriculture is the precondi- of the benefits expected of it in Africa during tion of renewed growth for most of the eco- the past decade. Respectable rates of growth nomicallv lagging African countries. But of manufacturing production were achieved vigorous growvth will also require attention to for several years after independence, but large productive sectors other than agriculture be- savings of foreign exchange upon which much cause they can contribute to more rapid growth industrial investment was based have not in the 1980s and because of their long-term materialized. Some of the specific shortcom- significance. Given the range and complexity ings of the pattern of industrialization have of the questions involved, only some of the been analyzed elsewhere in this Report: in- main issues are addressed in four major sec- dustry has tended to be a burden on agri- tors: industry, nonfuel minerals, energy, and culture rather than being supportive; it makes transport and communications. large claims on scarce foreign exchange; it has not generated domestic savings and govern- ment revenue as anticipated; and it has not l I .i~ 1. f> ,-.provided as sound an industrial basis for fu- ture growth of the sector as would have been Industrialization has a crucial role in long- desirable because of high costs associated with terrn development: it is one of the best train- stagnant productivity. ing grounds for skill development; it is an Chapter 4 examined the distorted incentive important source of structural change and structure which has led to many of these diversification; and it can increase the flexi- shortcomings, particularly the trade and ex- bilitv of the economy and reduce dependence change-rate policies, which are systematically on external forces. Industrialization also pro- biased against exports and in favor of indus- vides employment, foreign exchange, and tries producing consumer goods with little domestic savings. Although these develop- local value added-packaging and assembly- mental benefits justify incurring some addi- type industries, for example-some of which tional cost to promote industry, they do not actually lose foreign exchange rather than save justify the promotion of industry at any cost. it. Poor project selection has also led to in- Manufacturing is only a small sector in Africa vestments with very low rates of return, par- and can make only a modest, though grow- ticularly during the commodity boom of the ing, contribution to development during the late 1970s. next decade. Excessive investment in indus- There are reasons for optimism about the trv can starve other sectors of capital, foreign future, however. Some countries have built exchange, and high-level manpower, while up an industrial base, which will permit in- expensive manufactured products can raise creased growth if an appropriate policy costs in other sectors and limit their growth. framework is established. Several efficient Industrialization has failed to provide many processing industries have been established 91 Box A: Industrial Growth in Malawi only encourages export growth but also maintains ex- ternal balance. Malawi is a good example of how a small African These policies help explain why manufacturing has country with little apparent industrial potential can grown even faster than agriculture in Malawi. During enjoy a high rate of manufacturing growth while fol- 1968-77, total industrial value added in real terms grew lowing an agriculture-oriented development strategy. at an annual average rate of 6.5 percent, while agri- Malawi is one of the poorest countries in the world. culture grew at 4.5 percent. Equally important, because It is landlocked, has no significant mineral resources, of the kind of industry that has developed,* and be- and has a verv small domestic market (population 5.8 cause of a wage policv that has held down urban wages, million). Its main natural resource is good land; but manufacturing employment also grew at 6.5 percent even good land has become scarce in relation to the per year during 1968-77. Few African countries have dense and rapidly growing population. had a higher rate of growth of employment in man- The approach to industry has been far from laissez- ufacturing. Given the obstacles to industrial growth in faire-the Government has provided protection for in- Malawi, it is doubtful whether industrial output would fant industries and has actively promoted industrv have grown any faster with higher protection, while through parastatal investment-but strict limits have agricultural output, manufacturing employment, and been set on industrial promotion. There is a moderate total GDP would almost certainly have grown more protective tariff, which ranges from 7.5 to 40 percent. slowly. Parastatals, which account for more than half of in- dustrial output, have been profitable and have gen- *Food processing, textiles and footwear, and tobacco and erallyremainedfreefromgovernmentinterference.Most tea processing account for about two thirds of manufac- important of all, quantitative restrictions have not been turing output. Most other industries (metal manufactur- used to restrict imports nor to protect industrv, and ing, plastics, paint, and cheniicals) are also relatively labor- the exchange rate has been kept at a level that not intensive. (in Cameroon and Zambia, for example) and rican country (Nigeria, with a population of some have expanded manufactured exports over 80 million) has a GDP greater than that to Europe (such as Mauritius). There are also of Hong Kong (population 5.0 million).' The some promising experiments with regional whole of Sub-Saharan Africa, including Ni- industrial cooperation in West Africa. Finally, geria, has a GDP which is only a third greater few African countries have yet started the than that of the Netherlands, with a popu- expensive stage of import substitution of in- lation of 14 million. termediate products, and many small coun- tries have not pushed industry to the detriment Population Density. Many African countries of agricultural development (see Box A). The are very sparsely populated. This raises the central issue is how to build on these prom- cost of infrastructural development and limits ising aspects to promote long-term industrial the market for some industries-a cement development. plant, for example, usually only serves a 200- to 300-mile radius. High transport costs give natural protection to some industries but limit CONSTRAINTS the achievement of economies of scale. At the Five main considerations bear on industrial same time, the difficulty of policing extensive strategy: market size, population density, frontiers leads to smuggling, which can un- wages and productivity, management costs, dermine attempts to establish high-cost im- and capital and infrastructure costs. port-substitution industries. Market Size. The small population and low Wages and Productivity. African wages are per capita incomes of most African countries high compared with those of Asia. An ILO severely constrain their choice of industries survey in 1979 showed that the median wage because most industries require markets larger for textile workers in 10 African countries was than those existing in most African countries. Only eight African countries have popula- tions greater than 10 million. Only one Af- 1. Nigeria's GDP is four times that of Hong Kong. 92 50 percent higher than in Pakistan and more than twice as high as in Bangladesh.2 Higher STRATEGY OPTIONS African wages reflect both government wage These factors obviously have a profound im- policv, which in many countries sets indus- trial wages above the level they wouldother- pact on the industrial strategies open to many wrise bage, andbovettherleve y oportun otiesor- African countries. Not all countries are equally wise be, and better opportunities for affected, of course. Nigeria, by virtue of its agnicultural employment. Afncan labor pro- afetdofcus.Nga,bviuefis . .tal endseto compar uavorably market size and density of population, has d 'ctivity s broader options than most other Sub-Saharan with many other parts of the world. A com- nations. The potential varies considerably over parison of six African and four South Amer- the rest of the continent, however, in coun- ican textile mills financed by the IFC showed t * disparate as Niger and Zimb that the average number of spinners per 1,000 stie of this diversity it is possible to make spindle shifts and the average number of pite some generalizations about desirable forms of weavers per equivalent loom shift was more istr deverlopmns t than twice as high in Africa. This lower pro- ductivitv of labor is primarily a reflection of Import Substitution. Import substitution can Africa's early stage of industrial development be a sound policy, and most industrialization and should improve; meanwhile, it continues has started on that basis. But in many African to slou development. countries it has been badly implemented. There M4anage?ne7t Costs. African industry relies also is always the risk that the protection af- much more heavily than other parts of the forded in the early stages may be maintained developing world on expatriate management for excessive periods. Most African countries and technicians. In manufacturing firms in will still find that the majority of investment Ivorv Coast, for example, expatriate salaries opportunities with an acceptable rate of re- turn will be in production for the local mar- account for one quarter of value added. Eu- ket. The challenge is to establish an incentive ropean managers and technicians usually cost and institutional structure that directs invest- two or three times as much in Africa as in Eurpe.Thesurey f Ic-sppotedtexile ment toward industries that are productive mlroeferre vey note thatiSu and can be competitive in the future. This micanmills used no expatriates whereS means that attention must be paid to costs expatricat wages added 25nto 50patria, peren and to growth in productivity from the be- expatriate wages added 25 to 50 percent to ginin. Beasvfteiprac feoo th Afia parol ginnng. Because of the importance of econo- the Afrcan pavroll. I mies of scale, it also means that many import- Capital and Infrastructure Costs. Industrial substitution industries should be set up with projects in Africa typically require investment a view to becoming exporters. costs that are 25 percent higher than in de- For countries that have nearly completed veloped countries, and for some industries the first stages of import substitution, such the margin mav be as high as 60 percent. This as Kenya, Ivory Coast, and Tanzania, few extra expense is associated with transport costs new import substitution opportunities exist and construction delays. Although there is no based on the internal market. The next step evidence that the extra cost of investment is in import substitution usually involves going svstematically higher in Africa than in other into intermediate goods production in which developing areas, supporting infrastructure, economies of scale are more important. But such as roads, ports, utilities, and financial such import substitution in a small domestic institutions, is less available than in most parts market cannot sustain industrial growth. This of Asia and Latin America. is because even if high-cost intermnediate in- dustries are started, they will curtail the growth of other industries. 2. ILO Bulletini of Labour Statistics (Geneva: ILO, 2nd Quarter, 1980). Regional Integration. Economic integration, 93 which allows production for a larger regional processed products were the fastest-growing market, is one way in which small African category of African exports between 1963 and countries can expand the scope for efficient 1975, two thirds of the increase in value of industrialization. But there are four obstacles processed products came from refined cop- to such integration in the near future. First, per. There is much scope for increasing local transport and other links among African processing. But the determinants of the lo- countries are poorly developed. Second, the cation of processing (capital costs, structure distribution of industries is likely to be un- of the market, tariffs, infrastructure availabil- equal in a union between countries at differ- ity) are so diverse that it hardly makes sense ent levels of development or with different to speak of processing as a general industrial locational advantages. Unless counteracting strategy as opposed to prospects for partic- measures are taken, industry will tend to con- ular processing activities. Careful project centrate in the more industrially advanced analysis is necessary because of variations in countrv, at coastal locations, and in the larger local conditions, such as the quality of raw country. Third, inefficient industries may be materials and other input costs. unable to withstand competition from partner states following integration. Although a Manufacturing for Export. Neither the past pruning of the industrial deadwood is an im- record nor newly uncovered special advan- portant benefit of integration, it naturally pro- tages suggests that concentration on exports vokes resistance from the industry or country of labor-intensive manufactures is a promis- affected. Finally, political disputes, actual or ing strategy for most of Africa. Africa's share potential, hamper prospects for integration, of world manufactured exports is low (0.2 since thev increase the risk of investing in an percent in 1977), and its growth rate of man- industry that depends on an integrated mar- ufactured exports is the lowest of any devel- ket. oping region. Many of Africa's manufactured Two conclusions can be drawn. First, it will exports under the conventional definitions are probably be easier to approach integration in fact slightly processed resources: 30 per- through the least formal channel possible, such cent are diamonds and precious stones. Labor as a regional project, rather than through a costs and productivity plus high management formal customs union. This is the thrust of costs place Africa at a severe disadvantage most of the integration efforts in West Africa compared with Asia. at present. Second, regional integration is far Nonetheless, important potential for man- more likely to succeed if costs of production ufactured exports does exist. African manu- are fairly close to world levels. Governments factured exports (excluding diamonds and resent the foreign exchange and tariff revenue special transactions) to industrialized coun- forgone when purchasing from a high-cost tries grew at an average annual rate of 8.5 plant in a partner state. Full economic inte- percent between 1970 and 1979. This was from gration in the future will be much more dif- a very small base and was partially offset by ficult if governments establish high-cost a decline in exports to other developing coun- industries that will be unprofitable or perhaps tries, but it is an encouraging sign of the po- not even survive in an integrated market. tential for growth. In any case, past Consequently, only those national strategies performance is no guide to future prospects of development which emphasize efficient because of the policy bias against exports and production for the domestic or world markets the low productivity of early-stage industrial- are likely to be compatible with the process ization. Overvalued exchange rates and pro- of economic integration. tective wage policy have contributed to a high wage level. Replacement of high-cost expa- Processing Raw Materials for Export. African triate managers and technicians with person- exports are dominated by primary products nel from other developing countries or local (see Tables SA.8 and SA.11), and although managers could lower these manpower costs. 94 Improved incentives at the macro and plant time, provide demand for basic metal plants. levels should stimulate growth of productiv- But setting up a basic metal industry will re- ity. Moreover, preferential access to the Eu- tard the expansion of metal-using industries, ropean market offers an opportunity to expand for it is these which are the really important some labor-intensive exports; Ivory Coast and agents of development. Mauritius already export some clothing to the An agriculture-oriented development strat- EEC. Finally, excess capacity in many countries egy with industry in a supporting role does could be turned to advantage if incentives not mean that Africa would forgo industrial were restructured to encourage exports. development. Long-term industrial growth This does not imply that any African coun- might, in fact, be higher with this approach. try is yet in a position to follow the path of Although agriculture would be the driving Korea or Taiwan. But there is no reason why force, industry would still grow faster than some of the relatively more advanced coun- agriculture. In fact, higher agricultural in- tries, such as Ivory Coast, Kenya, Mauritius, comeswillstimulatedemandforproductsfrom and Zimbabwe, should not be able to increase a number of industries-textiles, metal man- the volume of manufactured exports by at ufactures, building materials, and light con- least 10 percent per year. This might come sumer goods-that could be produced more from resource-based than labor-inten- relatively efficiently in most African econo- sive industries. It might also come from new mies. Efficient industries generate their own import-substitution industries built with an momentum. And then export possibilities are eye to the export market as well. Breaking opened up and the local market for inter- into export markets is not easy and consid- mediate goods widens. erable marketing and other support will be In most countries, metal engineering and needed, but for many African countries ex- industries producing local materials for con- port production will have to play a more im- struction have substantial growth opportun- portant role in overall economnic growth. ities. Export processing, regional projects, import substitution, or any other kind of in- vestment would be consistent so long as it INCREASING INDUSTRY S CONTRIBUrION had prospects of becoming efficient and did No single industrial strategy will fit the di- not burden agriculture. The historical expe- verse conditions, prospects, and goals of all rience of Denmark, which has essentially fol- African countries. There are, however, a few lowed an agriculture-oriented development generally applicable principles. To begin with, strategy, demonstrates that giving priority to a conscious effort should be made to seek out agriculture does not impede industrial devel- profitable industrial export opportunities. Even opment. And the experience in Malawi shows though the bulk of investment opportunities that the same is true for small African coun- will be in production for the domestic market, tries (see Box A). sooner or later most countries will also have to increase manufactured exports to maintain industrial growth, expand employment op- PROMOTION OF INDIGENOUS ENTREPRENEURSHIP portunities, and diversify exports. Promotion of indigenous industry is a basic Moreover, the pace of industrialization objective. It is important, therefore, to review should not be forced. In many cases the choice how development of African entrepreneur- is not between having or not having an in- ship would be affected by policy reform or dustry, but between having a small-scale, high- reorientation of industrial strategy. cost industry now or an optimum-scale, ef- Many African countries emphasize direct ficient industry a few years from now. Proper controls to promote large-scale import sub- sequencing is vital. Rapid growth of metal stitution, a policy which discriminates heavily engineering, for example, depends on com- against local small enterprises. Small indige- petitive supplies, and this growth may, in nous firms typically lack the administrative 95 resources to deal with import and industrial ration of incentives available for new invest- licensing regulations or to obtain special ment in each sector should be spelled out. concessions, such as duty drawbacks. The Some promotional devices-lengthy exclu- policy of keeping interest rates on loans and sive production rights, bans on competing deposits artificially low also discriminates imports, and tariff waivers on inputs-should against local small business. Banks respond be ruled out. Discriminatory concessions to low interest rates by rationing credit and among firms should be eliminated. Above all, favoring traditional large customers with low the investment decision must be separated risk of default and low administrative costs. from the decisions concerning appropriate in- Any policy reform in this area would clearly vestment incentives. Guidelines for infant in- benefit local small businesses. dustry protection and tax incentives should An agriculture-oriented development strat- be laid down in advance and not tailor-made egy would also benefit many of the subsectors for each new investment. In much of Africa with greatest potential for small-scale devel- at present, investors actively seek govern- opment: footwear, clothing, furniture, food ment participation in joint ventures because processing, manufacture of small implements, they realize that this is the most certain way and production of local construction materials. to obtain protection and concessions and thus The metal engineering sector is another in which ensure financial success of their project, re- opportunities for small-scale business are ex- gardless of its economic desirability. ceptionallv promising. Many of these indus- A third area of reform should be in policies tries can also be decentralized in small towns concerning public industrial enterprises. The or rural areas, thus providing an altemative problems that parastatals typically encounter source of income for the rural population. do not stem from their public ownership, but World Bank research in Kenya suggests that rather come from their not being treated as off-farm income, in turn, can be a major source commercial enterprises. They should not be of finance for agricultural innovation. In this burdened by requirements to hire more peo- way, agricultural and industrial expansion in- ple than they need, to provide services with- teract to boost the growth of income and out- out payment, or to hold down prices of the put in both sectors. goods and services they sell. Governments must still pursue social objectives, of course, but to the extent possible parastatals should CHANCES IN POLICIES AND INSTITUTIONS be maintained as commercial enterprises and The first and major change needed to im- compensated for any social services they are prove industrial performance and lay a sound required to perform. At the same time, para- er base for long-term development is a reform statals should be subject to the same tax re- of the incentive structure for industry. The quirements as other private industrial reforms have been spelled out in Chapter 4 enterprises. Parastatals should pay taxes on and can be briefly summarized here. The es- profits and on imported inputs, and they sential requirements are to increase the in- shouldreceiveonlyreasonableprotectionfrom centives for industrial exports, to reduce outside competition. protection for import substitution, to reduce Fourth, and finally, governments can use the extreme variation in protection among in- tax incentives more effectively to promote in- dustries, and to phase out direct controls. The dustrial investment. Tax credits can be limited method and pace of reform will be different to an amount that is some share of the in- for each country, but the direction of change vestment, say 100 percent, although, in gen- needed is clear enough. eral, they should not be extended solely for Second, procedures for selecting projects equipment and fixed capital investment. Many should be strengthened. The most important countries have sought to avoid such fiscal measure that can be taken here, too, is a change waste and economic distortions by requiring in the incentive system. The amount and du- substantial local value added before granting 96 tax benefits. Mexico, for example, has re- ance, along the lines being provided by the quired that imported parts of assembly-type United Nations Industrial Development Or- industries be less than 40 percent of direct ganization (UNIDO) to the Economic Commu- costs. Some countries-Pakistan and the Phil- nity of West African States (ECOWAS), would ippines, for example-have simply excluded also help overcome obstacles stemming from packaging and assembly-type operations from inexperience and shortage of high-level man- tax benefits. power. The effectiveness of fiscal incentives to for- A third objective of donor support should eign investors should not be overestimated. be assistance for rehabilitation and consoli- Investors are attracted more by political sta- dation of the industrial sector. This is closely bility, a low probability of confiscation, a sta- related to the first objective of helping in- ble, predictable, and reasonable tax regimen, dustry adjust to new incentives, but it is and the ability to repatriate profits. The evi- broader. Much of African industry is plagued dence is overwhelming that where the in- by low productivity, underutilized capacity, vestment climate is favorable, including and poor management. Some of these prob- measures to avoid double taxation with the lems stem from overall incentives, but others home country, most foreign investment will are specific to the subsector or individual firm. occur without a need for tax concessions.3 The pToblems are sufficiently widespread to DONOR SUrPORT suggest concentration on improving the per- formance of existing industry rather than on Donors should give priority to three areas. major programs of industrial expansion. New First, they should support reform of indus- investments would still be made in industries trial incentives and other institutions. This with promising rates of return, but emphasis will require assistance to existing industTies, on consolidation implies continued support including technical assistance to improve pro- for institutions, such as development finance ductivitv and loans to finance further invest- corporations and manpower and infrastruc- ments or increased working capital tural development. requirements made necessary by price or ex- change-rate changes. An obvious corollary of supporting policv change is that donors should help finance newv investments with long-term Nonfitel Mlinrcnls development potential. Second, donors should actively seek out The African continent has always been re- regional industrial projects. Greater effort by garded as one of the great storehouses of outside agencies to find viable regional proj- mineral wealth, and some African countries ects could make an important contribution to have extensively exploited these riches. Af- regional integration. More technical assist- rica is the prime supplier of Europe's minerals and contributes substantially to nonfuel min- eral exports worldwide. The potential of ex- ploration, however, has only begun to be 3. A number of African nations have already signed mul- realized in most countries of the region. Ex- tilateral treaties that standardize and coordinate incentives so as to avoid a revenue-losing competition in offering in- ploitation of existing mineral capacity iS centives. and to promote integrated areawide development. threatened by civil strife in some countries The Union Douanitre et Economique de i'Afrique Centrale and exploration is proceeding very slowly. (,.) treaty. tor example, sets forth a standard incentives Moreover, the pace of mineral investment lw tor all member countries; multinational approval is needed for grannng some of the incentives. The Economic cor- worldwide has slowed in the recent past; for munity of West Afncan States (ECOWAS) treatY, in contrast, this and other reasons, market prospects for merely calls for coordination among member countries but does not set forth a model. These attempts at coordination prevent harmful competition in incentive granting and should the mid-1980s. Minerals production, there- be pursued. fore, can become a great source of growth in 97 Africa in the decade ahead. To make this hap- pends heavily on their location, their acces- pen, it is essential to rehabilitate existing min- sibiity to power and transport networks, and eral facilities, attract capital for new ventures, the presence of existing mining activities that and give increased attention to exploration. have built-in managerial and technical know- how. In addition, the risk of political or eco- REHABILITATION OF EXISTING FACILITIES nomic instability affects the willingness of Top priority for many African producers is private investors to develop new mineral de- the rehabilitation of existing mines and proc- posits. For all of these reasons, most African essing facilities. During the past 10 to 20 years, countries are at a disadvantage compared to investments in existing facilities in Africa have other mineral-rich countries such as Australia declined. Reasons for the decline are varied. or Brazil. They include reluctance of countries with bal- In these circumstances, the World Bank and ance-of-payments difficulties to reinvest in other donors have a valuable catalytic role to nationalized mining companies, declining play. Even their marginal participation can profit margins, civil disturbances, and dis- serve as a guarantee of fair conditions, pro- agreements between foreign partners and viding governments with an independent, govemments on investment strategies. As a expert assessment of investment proposals, result, there is an investment backlog for many and assuring mining companies of an atmos- African mining enterprises. Rehabilitation is phere in which to negotiate reasonable urgently needed in Ghana, Liberia, Zaire, and concession agreements. In exceptional cases, Zambia, and, to a lesser degree, in Uganda. the international financial institutions could Under the Lome II agreement a new mineral also consider financing some of the equity rehabilitation facility (SYSMIN) has been created contribution, if this is necessary, to attract that will provide up to 280 million European commercial investment. In fact, this concept Units of Account (EuA)-or $365 million-over of a catalytic role has been accepted; a min- five years to African, Caribbean, and Pacific erals lending program of $700-800 million was countries. Minerals covered by the scheme approved by the Board of the World Bank for include copper, cobalt, phosphates, man- the fiscal years 1977 to 1981. Because of the ganese, tin, iron ore, and bauxite and alu- depressed world mineral market and the mina. Although SYSMIN will play an important shortage of commercially viable projects, less role in donor support for mining rehabilita- than half of this amount has been committed. tion investment, limitations of funds and of However, with improved market prospects in coverage will certainly require lending for re- the 1980s and with new opportunities arising habilitation from other sources as well. from more exploration, lending to Africa and other developing countries should increase. NEW INVESTMENT The role of external assistance in financing There are large, well-known mineral deposits new mineral investments will necessarily be and energy sources in Africa that await com- modest. In the first place, the capital require- mercial development. These include the sub- ments are enormous. Annual niining invest- stantial iron ore deposits in West Africa and ment requirements in developing countries in the enornnous hydropower potential for baux- six main minerals are estimated at an average ite-alumina-aluminumdevelopmentinGuinea of $4 billion per annum (in 1977 prices) in the and Zaire. Unfortunately, economic and lo- 1980s, of which at least $3 billion will have cational factors, as well as geological condi- to be sought abroad.4 Africa's share of the tions, affect mineral development prospects. Since infrastructure investments and skilled manpower costs are important factors in de- 4. Marian Radetzki and Stephen Zom, Financing Mining cost competitiveness of potential Projects in Developing Countries: A United Nations Study (Lon- termining cost competiveness of potental don: Mining Joumal Books Limited, 1979), p. 31. The six deposits, the economic value of deposits de- minerals are iron, copper, aluminum, zinc, nickel, and lead. 98 total might reach about 25 percent, but the opment. In recent years, exploration expend- proportion of external financing would be rel- iture for new minerals has been concentrated atively greater. This is a very large amount in a few developed or newly industrialized in relation to flows of official assistance, and countries. This focus reflects the known dis- any attempt to finance the bulk of mining tribution of mineral wealth, the availability of investment through foreign aid would require skilled manpower and low-cost infrastruc- substantial cuts in the aid being made avail- ture, the mining companies' perceptions of able for other purposes. political risks, and the instability of conces- Such a tradeoff is not necessary. There is sion agreements as well as slack demand and no evidence of a shortage of funds for com- low prices for minerals. Exploration, basic mercially viable projects. Mining companies mapping, and survey work all deserve higher have ample proven reserves to meet near- priority, both from African governments and term requirements, and the main constraint donor agencies. Based on UN and World Bank on new mining development recently has not studies, an estimated additional $75-$100 been lack of finance but lack of viable projects million per year is needed in Africa to help with adequate rates of return. Few new min- finance minerals exploration. ing projects provide more than a 12 to 14 percent rate of return. Because of high capital costs, shortage of skills, and underdeveloped Energy infrastructure, deposits that are equivalent The energy problem has three elements: the from a geological point of view tend to pro- fue crgy expas new commercial vide lower retunls in developing than in de- fuelwood crisis, expanding new commercial veloped countries. When the mineral market energy sources, and improving the efficiency revives, as it is projected to do during the of energy use. 1980s, financing from intemational mining companies and commercial sources is ex- THEFuELWOO) CFISIS pected to be forthcoming. During the early stages of development, in- What concerns the governments of many creases in energy use generally involve move- developing countries, of course, are the terms ment away from reliance on noncommercial on which private ,apital will become avail- fuels (wood, dung, and agricultural residues) able. After a period of hesitancy, host gov- to use of commercial fuels, particularly those ernments in many countries have come to based on petroleum. Africa is still in an early realize that mining companies are valuable stage; its per capita commercial energy con- sources of technical and marketing expertise, sumption is at present only 2 percent of that that they bear very high risks, and that it is of the developed world. necessary to assure them an adequate rate of As a result of population growth, the need return and early repayment of foreign in- for agricultural land and fuelwood is increas- vestment. The companies, for their part, have ing. This has caused serious shortages of fuel- accepted that host countries have the primary wood, especially in densely populated right to their own natural resources and, in settlements. As a result, people in rural areas particular, to any exceptional profits after the are spending more time on fuel collection, at foreign investment has been repaid. Thus, the cost of productive activities. In parts of the basis for more stable and equitable in- Tanzania, for example, provision of the an- vestment agreements exists. nual fuelwood requirements for a household of five persons now requires 250 to 300 man- days of work. Fuelwood shortages in the vi- EXPLORAnON cinity of urban townships such as Niamey, In one final area, donors can make an even the capital of Niger, have led to the gradual greater contribution to African mining devel- elimination of all savanna woodlands within 99 a 50-kilometer radius from the capital. A sim- commercial energy from this source (see Table ilar phenomenon is beginning to develop SA.6 for data on oil-import dependency).5 In around Lusaka, Zambia, and a number of other the low-income countries, petroleum pro- major African cities. Indirectly, overcutting vides 80 to 90 percent of total commercial the savanna woodlands for use as fuelwood energy consumption; in the middle-income is a contributory factor to the "desertification" countries, the proportion is 44 percent. taking place in parts of the Sahel, since it Policies that encourage production of en- leads to wind erosion of topsoil. It has been ergy supplies are critical for economic growth estimated, for example, that at the present and for reducing the dependence of African rate of cutting, an area of more than 300,000 countries on oil imports. Oil-importing Afri- hectares of savanna woodland around Nia- can countries must take further steps to iden- mey will have been destroyed by the turn of tify and assess their domestic oil, gas, coal, the century, an area that otherwise would and hydropower resources through geologi- have been capable of sustaining something in cal surveys. Noncommercial and smaller-scale the order of 3,000 farming families together renewable energy sources, such as solar en- with their livestock. ergy, also merit attention. Prospects for the The growing scarcity of fuelwood is the en- various fuels are discussed below. ergy problem in much of Africa. Except in a few countries-those located in zones of hu- Oil and Gas. Oil has thus far been found in mid forest, such as Zaire-potential fuelwood only a few African countries. At the begin- crises can be expected in the coming decades. ning of 1981, the estimated proven oil re- According to estimates made by the World serves of the region were 19 billion barrels Bank, fuelwood consumption in the region is (see Table 7.1). But these proven reserves are projected to increase to about 280 million cu- not regarded as a true indicator of the region's bic meters in the year 2000-an average an- oil potential, which has yet to be determined. nual increase of over 3 percent. Taking account For most oil-importing countries, the pe- of substitution possibilities of various com- troleum prospects are judged fair to good by mercial and other forms of noncommercial many geologists, though the uncertainties of energy for fuelwood, residual fuelwood de- these judgements are indicated by the fact mand in the year 2000 would call for 19 mil- that Ivory Coast, now so promising, was be- lion hectares of fuelwood plantations-100 lieved to have modest potential only a few times more than now exists in the region (0.2 years ago. As for gas, a recent study identi- million hectares). fied 16 countries in Sub-Saharan Africa with Because this Report is focused on the 1980s, some potential for natural gas production.6 and fuelwood supply is a long-term concern, Eight of these countries have proven gas further discussion is deferred to Chapter 8. reserves7; about 87 percent of these reserves are located in Nigeria, and most are not as- sociated with crude oil. INCREASING COMIMERCIAL ENERGY SUPPLIES In most of Africa, commercial energy is con- 5. In Zimbabwe and Zambia, the share of oil in 1978 was, sumed by a verv small fraction of the popu- respectively, 22 percent and 40 percent. The main source lation. Per capita commercial ener gy of commercial energy in these countries is hydropower, consumption in the region is thus very low, most of which is generated from the Zambezi River. about one quarter that of the developing 6. These countries include Angola, Benin, Cameroon, Chad, Congo, Gabon, Ghana, Ivory Coast, Madagascar, world as a whole. Petroleum meets about 70 Niger, Nigeria, Rwanda, Senegal, Sudan, Tanzania, and percent of commercial energy needs in the Zaire. See Bureau d'Etudes Industrielles et de Coop&ration oil-importing countries of the region; only de l'lnstitut du Petrole (BEIcIP), "Survey on the Utilization of Gas in the Developing Countries" (Paris: March 1980). Malawi, Mozambique, Zambia, and Zim- 7. Angola, Cameroon, Congo, Gabon, Ghana, Ivory Coast, babwe derive less than 60 percent of their Nigeria, and Zaire. 100 Table 7.1. Proven Reserves of Oil and Oil Production Estimatehd oil prodtictioti Proven reserves (thotisands of lbzarrels a iail) Courntry (billions of barrels)' 1977 1978 1979 1980 Angola 1,200 171 147 144 150 Cameroon 200 - 12 34 57 Congo Republic 660 34 47 53 56 Gabon 450 223 210 196 145 Ghana 6 - - - 2 Ivory Coast 50 - - - 3 Nigeria 16,700 2,079 1,905 2,301 2,100 Zasre 130 23 18 21 22 Total 19,396 2,530 2,339 2,749 2,535 a. As of January 1, 1981. Sources- Oil and Gas Journal and other petroleum industry sources. Exploration efforts have already begun in amounts of engineering and economic analy- a number of countries (Gambia, Ghana, Kenya, sis from donors. Mali, Mauritania, and Sudan, for example). But unless exploration activities can spread Coal. Although geological coal resources are soon to many other countries, there is little considerable-close to 135 billion tons-eco- chance of substantial, broadly based increases nomnically and technically recoverable re- in domestic production during the coming sources are only a fraction of this amount. decade, given the time needed to mount an About 94 percent of this geological potential exploratory campaign and develop a discov- is located in Botswana and Zimbabwe. A large ery to the point of commercial production. To number of other countries have small coal attract risk capital from private sources, some deposits, primarily of steaming coal quality. African govemments may need to revise leg- Production in Sub-Saharan countries reached islative and contractual provisions that deter a total of about 7 million tons in 1979. Coal foreign participation. consumption in the region has remained very low-18 percent of commercial energy con- Hydropower. Hydropower accounts for sumption in 1978-due largely to: the im- nearly 70 percent of installed electricity- mense hydro potential of many African generating capacity in the region, and 30 per- countries; low oil prices in the past; relatively cent of total primary energy consumption. small domestic demand; low coal quality and Hydropower resources in Africa are believed high coal development costs arising from dif- to be vast: present potential is estimated at ficult mining conditions; high transport costs; 223 gigawatts. Only 2 percent of this potential and lack of transport facilities to export coal so far has been exploited. to consuming centers. The large increases in oil prices have given Nevertheless, high oil prices have stimu- new urgency to development of hydro re- lated new interests in the exploration and sources on the African continent. Present oil development of coal for domestic use (in coal- prices justify a capacity cost approximating fired power stations and in industry) and for 1.5 to 3.5 times that of recently built hydro export, primarily from Zimbabwe, Botswana, plants in developing countries. Moreover, not Swaziland, and Mozambique. Private com- only is there a substantial opportunity for panies, such as Anglo-American, Shell Coal, exploiting hydroelectric potential on a re- Union Carbide, and Rio Tinto, have negoti- gional basis in Africa, but also the potential ated exploration concessions in these coun- for small-scale hydroelectric power has only tries. The success of coal development for been touched and could benefit from large export will depend on the results of the pre- 101 sent exploration campaigns, which will re- consumption from lower to higher value uses, quire at least two to three years, followed by reduce the energy cost of output, and pro- four to six years of development work, and mote a switch from more to less costly sources on the development of an extensive coal- of supply. Measures aimed at reducing oil transport network through Namibia, South dependence need to concentrate on improve- Africa, and Mozambique. It will also depend ments in the energy efficiency of the industry on the willingness of potential consumers, and transport sectors, and on determining primarily those located in Europe, to enter suitable pricing policies. into long-term supply contracts with South- The industrialization process usually en- ern African producers. tails a rapid increase in the use of petroleum. Industries such as oil refining, steel making, Renewable Energy. For a small number of paper manufacture, and cement are all energy countries with surplus production of molas- intensive and are logical targets for programs ses or sugar, alcohol has become a possible in energy conservation. Improved manage- substitute for or a blend with gasoline. Among ment and training will make significant energy these countries are Kenya, Malawi, Mauri- savings possible within two to three years. tius, Sudan, Swaziland, Tanzania, Zambia, Transport is the largest consumer of petro- and Zimbabwe. The economic substitution of leum in many developing countries. It is this alcohol for gasoline will depend on a number sector for which alternatives to petroleum are of factors, including the economic price of the most difficult to identify. Many countries gasoline, the cost of the agricultural feed- need sectoral studies to evaluate the energy stock, and competing uses of agricultural land. intensity as well as the economies of current In most of the Sub-Saharan African countries and alternative transport patterns and modes. with alcohol production potential, these fac- Energy efficiency can be achieved through tors need to be studied further. changing the mix of transport methods, shift- Technologies that make direct use of the ing traffic from less to more efficient carriers elements-sun, wind, and water-may be- (such as public passenger transport), increas- come important sources of energy, particu- ing load factors, and adopting traffic control larly in rural areas, but in most cases large- schemes. In all sectors, but especially in trans- scale, cost-effective technologies are some port, changes in energy consumption will also distance away. require substantial public investment. An essential tool for increasing energy ef- Enerra Pla mostcotes in- ficiency is a pricing policy which ensures that, Saharan Africa demonstrate a lack of plan- as far as possible, the price of energy in var- ning for the efficient use of energy, and re- ious uses reflects its real economic cost. In sponsibility for policy and administration of many cases, the achievement of economic energy resources is widely fragmented. Be- pricing of energy products requires either the cause energy concerns cut across usual ad- removal of inappropriate government- ministrative lines, control over energ reoa ofiapopit.ovmet miitrtvelne,cotoloereegy imposed pricing restrictions or adjustments programs often stimulates bureaucratic strug- in government policies. Relative energy prices gles. Most countries could benefit from a cen- in many African countries remain distorted. tral energy planning unit or department to In particular, recent price increases have undertake the task of preparing a coordinated mostly been borne by gasoline, while relative overall energy policy, prices of kerosene and diesel oil in many cases have declined.8 INCREASING THE EFFICIENCY OF ENERGY USE Even though per capita consumption of en- 8. In most West African oil-importing countries, how- ergy is small in most African countries, de- ever, virtually all major products are at or above interna- mand management policies can shift tional ex-refinery prices. 102 Domestic energy prices in some countries energy planning or to implement an energy need to be raised to bring them up to world policy. Economies of scale are significant in prices, although many oil-importing African this kind of training activity. A study of the countries have raised prices substantially. cost and feasibility of a regional training cen- Many countries could also significantly in- ter in energy planning and policy analysis, or crease tax revenues from motor-fuel taxes.9 the development of such training in existing The difficulty of increasing domestic prices to regional facilities would be useful. international levels in countries where there Second, the development of some of Af- is substantial discrepancy between the two rica's abundant energy resources-coal, pe- should not be underestimated: in some coun- troleum,andnaturalgas-mayrequireregional tries, it would take a doubling of domestic cooperation, for example, in the import and prices in real terms for four consecutive years export of electricity, as well as in the siting or more to eliminate the subsidies implicit in of energy-extensive industrial projects. The the existing price structure. Bank and other donor agencies should help There is also scope for fuelwood conser- promote this approach, initially by financing vation measures. The demand for wood can feasibility studies. be reduced by improving the efficiency with Third, the regional approach could reduce which it is used. Many traditional stoves waste costs through better use of oil refinery ca- 90 percent of their heat, and energy require- pacity. There are a number of small refineries ments for open fires are some five times those in the region and most of them operate below of kerosene stoves. Small improvements in capacity."' A number of new refinery projects chimnev and stove design could double the are also being undertaken. Overall, the region useful energy obtained from fuelwood. Char- is relying very heavily on small, inefficient, coal-which is now produced by felling live and expensive refining of crude oil to meet trees and burning them in sand-covered pits- partial product demand. A comprehensive could be made substantially more efficient by regional study might lead to acceptable pro- use of kilns. Conservation is probably the best posals for a more economic approach. medium-term path to reduction of total en- ergy consumption and warrants closer atten- DONOR SUPPORT tion bv both local authorities and aid donors. The important role that external assistance REGIONAL APPROACHES can play in helping developing countries in- crease domestic energy production and adjust There are a number of areas where regional to higher energy prices has been described in cooperation presents special advantages: in other World Bank reports." training for energy policymaking, in explo- ration, and in resource development. Donors rahiouland helprAfricangovern pments exploiese 10. In 1978, the total installed capacity utilization of the should help African governments exploit these region was 65 percent. In Zaire, the 17,000 barrels per day advantages of regional cooperation, which not (b/d) refinery operated at only 24 percent of its rated ca- only present viable options for energy devel- pacity. A small refinery in Mozambique (16,000 b/d) was opment and independence, but which are in operating at 52 percent of capacity. Some larger refineries operating below capacity also exist in the region; in Kenya, line with longer-term African objectives, as the 95,000 b/d refinery was operating at only 55 percent of specified in the Lagos Plan of Action. capacity. First, almost all countries in Africa lack the 11. See Energy in Developing Countries (Washington, D.C.: World Bank, 1980) and Chapter 4 of the World Development specialized manpower needed to undertake Report 1981. The World Bank's involvement in hydrocarbon exploration in Africa is larger than in any other region. q As discussed in Chapter 4, part of the erosion in rev- During rY1980-81, seven projects totalling $69 million were enue from motor-fuel taxes is explained by the fact that financed through IDA; four of them included energy plan- most Atncan countnes use specific taxes on motor fuel whose ning assistance. For rY1982, 12 lending operations totalling real value falls as the general price level rises. Governments $230 million are in preparation, 11 of which are for explo- should move to ad valorem taxes at a fixed rate. ration. 103 Box B: Public Aid as a Catalyst for Private Capital: Petroleum in Mali Recent experience in Mali illustrates how a small l'Institut du Petrole (BEICP), which submitted its recom- amount of public development assistance can stimulate mendations for hydrocarbon exploration in June 1980. private capital inflows from petroleum exploration. Specific exploration targets were identified in the In 1979, the World Bank's Intemational Develop- Southeast and the Northeast. In order to obtain from ment Association (IDA) granted Mali a $1,000,000 ad- oil companies commitment to an intensive and rapid vance under the Project Preparation Facility to prepare program of work, BEICIc recommended special incen- a petroleum exploration promotion project. The prep- tives, such as accelerated depreciation and favorable aration focused on (a) a program of technical assistance financing of pipelines. to the Government to modernize existing petroleum The Mali Govemment endorsed the approach pro- legislation and reinforce Government technical and posed by BEICIP and undertook a promotion campaign administrative capabilities in the sector; and (b) geo- aimed at attracting companies to explore the Taoudeni physical prospecting and exploratory drilling in the Basin. The technical report prepared by BEICIP on the Taoudeni Basin, which had been subject to some stud- Basin was advertised to industry dunng the summer ies, as well as limited exploration by the French com- of 1980. It elicited a good response: nine companies pany, Societe Generale Elf-Aquitaine. The advance bought the report within a few months. In December financed the reevaluation and updating of existing geo- 1980, the Government, with the assistance of BEICIP, logical surveys in the light of more recent information began negotiations with Esso for an exploration agree- and improved techniques. These were complementary ment on the northem part of the Basin. A contract was to the work being done by Elf-Aquitaine and, together signed in January 1981 providing for a seismic survey with it, provided a better assessment of the resources and the drilling of a well at a total cost of over $20 potential of the area. The advance allowed definition million. At the same time, Elf-Aquitaine decided to of a strategy for attracting foreign companies on the drill a well. Conclusion of the Esso contract provided basis of adequate data, improving Mali's negotiating an opportunity for cross-participation between the two position, and at the same time making the country companies. Agreement was reached in February 198 more attractive to foreign oil companies. that Elf-Aquitaine and Esso would finance jointly 106 For the technical assistance component of the prep- percent of the costs of a well at Yarba, which wil. aration, the Govemment hired a French consulting firm, amount to some $20 million. Bureau d'Etudes Industrielles et de Cooperation de Foreign oil companies are sometimes re- quired increases in energy investment with- luctant to do business in developing re- out affecting growth in the rest of their gions, including Africa, fearing changes in economies.12 If investments are to be made the rules once significant discoveries are without concomitant internal dislocations, made. Participation by the World Bank can substantial capital inflows and technical as- reduce this fear. sistance will be required, even with increases * Multilateral lending institutions are in a in domestic savings. There are four priority unique position to help in accelerating en- areas where development assistance will be ergy development. They can assist African needed in the 1980s: countries to evaluate geological risks and develop exploration strategies, help oil * tni and gas exploraton. The oil and gas po- companies and host governments to reach tential of African countries is virtually un- agreements on joint exploration and ex- known, despite exploration efforts begun ploitation, and assure both parties that po- in some countries. Appreciable amounts of litical risks can be minimized, oil and gas will undoubtedly be found, and, litical risks cn be minimized.if the experience in offshore drilling in West * The infrastructure financed by multilateral if the inc tinofsore dillng fnW , . ~~~~~~~Africa iS any indication, more will be found institutions can also encourage private than was earlier expected. companies to expand exploratory work. Multilateral lending institutions can have an important leveraging effect, and their 12. The World Development Report 1981 estimates that, as contribution can be boosted by attracting a proportion of total developing countries' investment, en- additional private capital into the explora- ergy capital requirements in the 1980s may rise from 5 per- tion phase (see Box B). cent in the past five years to 10 percent. In the African case, the required investment would absorb much greater shares Few African countries can achieve the re- of locally available investible resources. 104 * Evaluation of renewable energy resources. Such funds, and the special demands of road and a review can help to identify some of the other maintenance in the African setting have most promising options, in particular, reas- combined to cause widespread deterioration sessment of hydroelectric possibilites in the of existing facilities. light of changing relative costs. Good planning machinery is especially crit- * Improvement of rural energy supplies. Inter- ical in the transport sector because of the large national technical assistance in forestry has volume of resources it absorbs and the com- received little attention, even though the application ~ ofmdr.ildmngmn plex problems of intermodal complementarity application of modern field managyement methods could increase the contribution of and competitiveness which have to be as- forests to development and help discour- sessed-relations between roads and rail- age deforestation. Moreover, doubling or ways, between air and rural roads, between tripling of the current level of fuelwood coastal and ocean shipping, and between tele- planting is attainable in some African coun- communications and road construction. tries over the next five years. Choices must be made about whether to favor * Energy planning. "Energy assessments," such operations that have high fuel costs per traffic as those undertaken by the Bank in Kenya unit, such as low-density aviation, or road and Mauritius, identify key policy issues, transport, which is more fuel-efficient but may suggest responses, and provide a basis for involve much larger construction costs and future energy planning. maintenance burdens. Foreign exchange budgeting is involved in decisions about which . . :- .' - .t tf7Zal!WiLnttX1>part of the vehicle fleet to keep in operation, and how to do it. Transport has already been referred to in sev- The critical role of transport and commu- eral different contexts: the disadvantage of nications at Africa's present stage of devel- long routes and light traffic, and the role of opment was explicitly recognized by African rural roads in agricultural development. Aside Governments in March 1977, when the Coun- from the general growth effects of increased cil of Ministers of the UN Economic Commis- access and reduced transport costs, improve- sion for Africa (ECA) declared 1978-88 a ment of transport and communication facili- Transport and Communications Decade in ties has special importance for Africa. Africa.'3 The proposal has since drawn strong Transport costs weigh very heavily on the one support from many international bodies, most third of African countries which are land- notably by the OAU in the Lagos Plan of Action locked. Better transport and communications and by the Southem African Development systems contribute significantly to the process Coordination Conference at its meeting at of political integration and administrative Maputo in November 1980. The following consolidation-high priority objectives every- paragraphs sketch out the main lines of a where in the region. Development of rural strategy for transport development in the transport infrastructure also offers special op- coming decade. The discussion draws heavily portunities for community action and en- on the Global Strategy prepared by the Eco- hanced self-reliance. nomic Commission for Africa and included in At the same time, many problems prevent the Lagos Plan. the transport sector from making its full con- tribution. The first is resource scarcity. In most INVESTMENT PRIORITIES AND STRATEGIES African development plans the construction The possibilities in the transport sector are of transport infrastructure absorbed 30 to 40 many. They clearly exceed the financial and percent of total investment over the past two decades. During the 1980s, a lower propor- tion of development expenditure will proba- 13. United Nations, UN Transport and Commutnications Decade for Africa 1978-88, Volume 1: Global Strategy and Plan of Action, bly be available. Moreover, the large national First Phase, 1980-83 (Addis Ababa, Ethiopia: May 1979). transport networks, the scarcity of operating Hereafter referred to as the Global Strategy. 105 manpower capacity of most governments in concerned with new construction. Ministries the decade ahead, even with a great deal of of works and their consultants should pay capital and technical assistance from outside. special attention to designing projects-es- It is essential, therefore, to focus on the high- pecially surfaced road projects-in such a way est priority tasks, and to avoid activities that as to minimize total system costs (construc- divert resources from them. Many worth- tion, maintenance, and vehicle operating costs) while projects will have to be postponed, and with realistic, rather than theoretically or le- improvements delayed. gally perfect, maintenance and axle-loading In all modes of transport and communica- requirements. tions, large-scale capital-intensive projects In countries facing the severe budget con- should be examined very critically by donors straints now characteristic of much of Sub- and governments, whether highways, air- Saharan Africa, aid-supplying countries should ports, railways, seaports, or telecommunica- take a liberal attitude toward financing re- tions systems. Tendencies in African countries current costs, local or foreign, of routine to prefer capital-intensive solutions and, in maintenance. All costs of periodic mainte- donor countries, to seek projects offering nance, of course, represent capital costs, and markets for their manufacturers, have some- this is now generally recognized by aid agen- times greatly reduced the real contribution of cies. Such recurrent-cost financing is partic- foreign assistance to development, and even ularly important in cases where the imposed serious long-term burdens. Recipi- participation of local contractors would other- ents and donors alike must be aware of these wise be discouraged by the occurrence of un- dangers and seek more economic altema- predictable or delayed payments. tives, such as measures to improve operations Now that basic infrastructure networks have and maintenance on the existing facilities, re- been widely developed, that educated na- habilitation of existing infrastructure, im- tionals are increasingly available, and that provement of traffic flow by better regulation public policy is beginning to focus on and control, and reliance on less capital-in- strengthening domestic capacity to exploit tensive modes of communication, such as transport opportunities, both public and pri- aviation and telecommunications. vate sectors should be fully utilized (see Box Top priority should be given to mainte- C). The Global Strategy refers to the need for nance, especially of trunk roads, which are special efforts "by Governments and lending now the mainstay of the transport systems of agencies to develop local African civil works most countries. The Global Strategy puts it well: contractors."'-' "As there is no economic substitute for timely The same point applies to transport oper- basic maintenance, and since neglected roads ators. It means enhanced attention to subdi- can only aggravate the transport problems of viding jobs so that they can be managed by the region, no effort should be spared by the local people, providing specialized training road authorities in African countries and in- and appropriate on-the-job supervision, sup- ternational agencies involved in road devel- plying credit to assist with bidbonds and the opment to protect the existing road network purchase or leasing of construction or trans- from deterioration and destruction."''4 port equipment, technical assistance for small Aid suppliers, in addition to financing proj- contractors, reduced economic regulations for ects oriented to maintenance as such-con- transporters, and, above all, maintaining a struction of regional offices and workshops, clear and stable legal and financial framework procurement of equipment, periodic mainte- in which local initiative can flourish. nance, and training-should include com- Many governments have begun to use pri- ponents contributing to the development of vate agents more efficiently: Mozambique and maintenance capacity, even in projects mainly Sierra Leone in trucking, and Sudan and Zaire 14. Ibid., p. 45. 15. Ibid., p. 44. 106 Box C: Building Local Capacities In earlier years aid agencies focused almost exclu- which revealed initiative and absorptive capacitv be- sivelv on the end-product of their project financing yond expectation, the external donors are proceeding and on economic benefits. But now increased attention with further assistance. is being given to the process by which the end-product Programs in Kenya have demonstrated that govern- is achieved, and especially to participation of domestic ment ministries can also innovate. Labor-intensive con- enterprise. struction methods have been used efficiently and Ghana provides an example. Despite extremely un- economically in the execution of a Rural Access Roads favorable macroeconomic conditions, the Ghana Gov- Program (RARP) in that country. External assistance ernment, with external assistance, increased started in 1976; many agencies participated. The RARP participation of private domestic contractors in road originally included the construction and subsequent works from a negligible amount to about 50 percent maintenance of 15,000 kilometers of rural roads by 72 of the total expenditure for highways in a period of construction units over an eight-year period in districts about five years. Deliberate government policy, allo- covering nearly 80 percent of Kenya's population. Con- cation of external financial aid, and technical assistance struction units of 200 to 300 workers were established, developed a domestic road construction industry that with labor to be used to the maximum feasible extent. now includes well over 100 active contractors. Due to significant differences in labor costs between One major aid project in this area emphasized active Government (or local communities) and contractors, participation by domestic contractors. It contained a and organizational requirements (including necessary planning mechanism which assured a continuity of consultations with local communities), the work pro- demand on regravelling and resealing work. And it gram could be carried out only by the Government's had a number of other features: strongly enforced re- Public Works Department during this initial period. quirements on quality of work; close cooperation of After the usual delays in starting such a huge program, the Ghana Highwav Authority (GHA) and the Bank for its scope was reduced in 1977 to 44 construction units; Housing and Construction (BHc) in prequalifying con- its time horizon was extended to 1986. Now 42 units tractors for road works; funds for procurement of areoperationalandemployaboutlO,000laborers. About equipment and spare parts (through BHc for private 3,500 kilometers of rural roads had been constructed contractors) and working capital; equipment for hire by December 1980; the actual construction costs amount from an Equipment Rental Pool, a subsidiary of BHC; to about $6,000 per kilometer, which is reasonably low. managerial advice and training by a uNDP-financed ex- Progress is now largely determined by the availability pert attached to BHc; training of contractor personnel of labor and the wages Govemment is willing to pay. in technical skills by CHA together with its own per- The quality of work is satisfactory and considerable sonnel; and improvements in material supply in par- experience is being gained by maintaining 2,100 kilo- ticular in quarry capacity. meters of rural road by labor-intensive methods. Under this project, 29 contractors benefitted from One lesson to be drawn is that African countries SHC loans. In 1980, they carried out about 70 percent have a significant untapped potential for supplying of their assigned work, while normally the Public Works domestic resources to the construction sector. Another Ministry could only accomplish about 20 percent. En- is that organizational innovations are possible in both couraged by the performance of the private sector, public and private sectors. in river transport. The Kenyan Government cially for rural roads. In Rwanda and Kenya, is encouraging transport services invented in local communities have helped pay for the the inforrmal sector in response to local needs. roads they use. In numerous countries, regulations that have proved inappropriate, such as controls on Road Networks. In support of agricultural transport, tariffs in areas with poor roads, production strategies and wider geographical and on back-haul loading bv own-account spread of development, high priority should trucks, are being relaxed. In a number of be given to improving rural access, mainly by countries, Ih greater attention is being given to roads. Rural access roads include all nonur- the development of domestic contractors for ban roads that link with the trunk network. road construction and maintenance, espe- These roads have been assuming a more im- portant place in many countries' develop- ment plans and will continue to do so. It has 16. Benin, Burundi, Cameroon, Ethiopia, Kenva, Le- been estimated, for example, that Nigeria's sotho, Liberia, NMalawi, and Nigeria. food self-sufficiency plan will require as much 107 as 25,000 kilometers of new rural roads. Such (particularly important for rural links and long- roads need particularly careful maintenance distance networks), and the buildup of do- to stay open year-round. Maintenance and mestic networks. Continuing improvements renewal arrangements are therefore crucial. in technology, such as more efficient tele- Interesting experiments exist in this area. phone instruments, plastic-clad cables, inte- Kenya's svstem of local resident "contrac- grated switching and transmission, solid-state tors," for example, is highly promising for technology, and solar batteriesfor VHFand UHF populated areas. Each contractor is respon- links, are lowering the capital costs of tele- sible for 0.5 to 2.5 kilometers of local road communications expansion and simplifying and paid a comparatively low part-time wage maintenance. at the end of each month if the road is in Sub-Saharan African governments and aid- satisfactory condition. The system is now ap- supplying countries should, therefore, give plied to about 2,000 kilometers at a current higher priority to the rapid development of annual direct cost of only $250 per kilometer. telecommunications services, and, particu- There is considerable scope and need in most larly, to the buildup of sound organizations countries for further experimentation with for running them through training and man- "appropriate technology" solutions to con- agement assistance. It is especially rare in this struction, maintenance, and transport on ru- field that manufacturer-generated projects, ral roads, adjusting standards to the vehicles involving simple supply of equipment and that will actually be employed,'7 and using provision of directly related technical train- more locally available manpower, equipment, ing, will on their own constitute a valid form and supplies. of foreign assistance. A broader approach aimed, above all, at developing local capabil- Telecoinmumications. Development of these ities is essential. Highest priority should go services should receive higher priority than to expanding domestic local and long- in the past: qualitv and quantity are grossly distance telecommunication networks, to be inadequate in most countries and basic costs complemented as soon as possible by the up- of telecommunications are falling relative to grading and the expansion of intercountry costs of other forms of communication. Lack facilities. of speedy and accurate telecommunications svstems is hampering growth in all sectors, Urban Transport. As African cities and towns prodigiously wasting managerial talent, and grow by 6 percent a year, poorer people con- unnecessarily straining transport systems. gregate in peripheral areas with poor access Despite the dispersion of their populations, to transport, cars and congestion choke city which would argue for more reliance on tele- centers, and public transit systems sink deeper communications, Sub-Saharan African coun- into deficit. As noted earlier, the number of tries in the same range of per capita incomes cities with more than 500,000 inhabitants in- as the poor countries of South Asia have lower creased from three in 1960 to 28 in 1980, and telephone densities, and the rate of growth urban population may well exceed 40 percent in the number of telephones over the last few of the greatly expanded total by 2000. The years has been much lower in Africa than in costly solutions to urban transport problems low-income South Asia (4 percent a year com- adopted in most industrialized countries, such pared with 10 percent). The Global Strategy asnumerousgrade-separatedroads,andlarge, places major emphasis on strengthening subsidized public transit systems, are clearly management, staff training, maintenance unsuitable. Rather, approaches must be in terms of traffic regulation schemes, such as Abidjan and Nairobi have initiated, parking 17. For instance, African countries appear to have at most controls and taxes, inexpensive arrangements 10 bicycles per 100 households (and, in many instances, less), compared with twice that number in India, and ten bicycles mopeds, times as many in China. construction of accessways into poorer quar- 108 ters, reserved lanes for fast movement of buses been amply demonstrated by the port of Dou- on main arteries, and encouragement to the ala in Cameroon, where productivity in the informal sector to develop safe paratransit movement of general cargo showed steady services (shared taxis, minibuses, converted and substantial progress during the 1970s. trucks, etc.) adapted to the city's particular A few countries will need berthing capacity needs and resources. extensions and, in one or two cases, even the development of new port sites. Aside from Railways. Railways continue to be of critical the need to improve container-handling fa- importance in about half of the Sub-Saharan cilities in some places, however, the expan- countries and of some significance in two thirds sion requirements of most AfTican ports are of them. Railways have particular importance limited because of substantial investments in Southern Africa, where mineral production during the 1970s in some cases, disappointing in landlocked countries is important. Many past rates of growth in international trade, railws ays have lost a greater share of traffic to and the possibility of improved productivity other modes of transport than would have from existing wharfs. Shipping services for been the case due to inefficient operation; the better regional distribution and collection of operating efficiency of government-owned containers need to be improved-particularly railways has seriously deteriorated over the for the island countries and for coastal trade. last two decades, straining production and They could also provide valuable opportuni- exports alike. ties for the development of domestic enter- The Global Strategy relegates proposals for prise if controls were reduced and credits were construction of link lines among the existing made available. On the other hand, invest- railways to a categorv of "possible study in mentsinintercontinentaloceanshippingneed the second half of the Decade." It urges, in- to be approached very cautiously because of stead, major concentration on strengthening the riskiness stressed in the Global Strategy, of management, on training of all types and the typically very high cost per job so created, grades of staff, ancd on maintenance, includ- and the minor net savings in foreign ex- ing track renewal. Lightlv trafficked lines change involved. Bulk shipping is a good should continue to be closed when their eco- entry-point for countries looking to shipping- noomic role is eliminated bv development of line investments in the future. Shippers' road services or exhaustion of mines. Equally, councils with supporting technical secretari- some line extensions will be required, mainly ats should be set up more widely to strengthen in connection with new mineral develop- African countries' capacity to bargain with the ments, such as the Nimba/Mifergui scheme international transport companies they use. in Liberia and Guinea and possibly the Aja- There have been some moves toward regional okuta iron and steel complex in Nigeria. The cooperation on port and shipping matters, pooling of orders forequipment among groups but these need to be accelerated to secure an of railwavs, so as to increase the railwavs' efficient development of container services, bargaining power on matters of appropriate and to avoid the high freight rates that can design, price, ancl after-sales service, merits result from the conference system and also further investigation. from the application of the proposed UNCTAD "Code of Conduct." Ports andii Shipping. Staff training, manage- rial improvement, and rehabilitation and pro- Intercouneltry Tranisport. The most urgently vision of some complementary equipment to needed steps to improve intercountrv trans- speed up traffic handling are also principal port are the wide range of "facilitation" meas- needsin theportsofSub-SaharanAfrica. These ures stressed in the Global Strategy- are urged in the Global Strategy/, which under- simplification of border-crossing, adoption of scores the prevailing low level of port effi- bilateral reciprocal arrangements on vehicle ciency. The possibilitv of improvement has movements, strict control of unofficial tolls. 109 agreement on common axle-load limits, har- trols is essential. As regards international monization of driving rules and signs, intro- service, the Global Strategy stresses training duction of some appropriate insurance and the urgent need for fuller cooperation arrangements, coordinated operation of inter- among African countries, such as that which country railway services, and standardization the African Airlines Association (AFRAA) has of documentation-to which may be added successfully developed for maintenance of the the measures of simplified customs, currency, B-737. In general, caution is called for in as- immigration, and phytosanitary regulations."' sessing expansion plans in this subsector. The Global Strategy justifiably urges that higher priority be given to roads linking land- INSTITUTION-BUILDING AND TRAINING locked countries with the sea than to the des- Project appraisals to be undertaken by donor ignated Trans-Africa Highways, 19 and that the agencies throughout the transport and com- plan for the latter be "indicative,"20 because munications field should always include a intervening economic development may re- careful assessment of the adequacy of staff quire changes of routing. Economically war- training. Where a project is expected to in- ranted improvements in routes that serve clude a significant expansion of training, landlocked countries should be considered preparation should start well ahead of time. high-priority claimants for support from con- African governments and the aid commu- cessionarv aid funds allocated for regional nity should give very high priority to multi- projects. Otherwise, such improvements will national training in transport management and be slighted, since most costs fall on the transit planning for top-level staff from both public countries, and many of the benefits accrue to and private agencies. The Eastern and South- the landlocked country, including intangible ern African Management Institute at Arusha, but important insurance benefits. Tanzania, which has received the support of 17 countries in the area since 1977, should be Air Transport. Despite relatively high fuel encouraged to develop courses extending be- costs, conventional small aircraft still have yond transport infrastructure planning and potential for opening up isolated areas and project appraisal to the management of trans- fortransportingmanagersandurgentlyneeded port operations. Second, West African gov- spare parts to places with long surface links. ernments, perhaps within the framework of This is also true for the larger countries still ECOWAS, should also begin as early as possible lacking in surface routes, such as Chad, Congo, high-level multimodal public and private sec- Mali, Sudan, Tanzania, and Zaire. In addi- tor courses in transport management in one tion, experiments on use of lighter-than-air or more centers in their region, as has been vehicles (airships) are now beginning in Latin suggested above for Arusha. Existing physi- America; thev warrant attention for their po- cal facilities should be used to avoid the delay tential applicability in Africa. and expense of new buildings. In all cases, Operation and growth of domestic air training should be aimed principally at staff transport has suffered severely in many Af- already working in the sector so as to com- rican countries from two public policy prob- plement past higher education; initiallv, such lems: the failure of the users of national airlines training should not attempt to substitute for to pay their bills, and tight governmental con- higher education in transport alreadv avail- trols on national airline tariffs. Firmer policies able elsewhere. In the case of each institute, on the bills problem and a loosening of con- short seminars should be organized for chief executives of agencies concerned with trans- IS. Ste the twvo-volume report prepared by consultants port and trade, as well as longer courses for to the Economic Commission for Africa, "Transafrican senior management staff. The faculty of the Highway: Study of Legal and Administrative Barriers," Sep- institutes should also be available to act as tember 1974. 19. [bid., p. 44, paragraphs 175-176. "flying squads" to provide technical assist- 20. [bid., pp. 8 and 9. ance in transport management. 110 Finallv, in line with the emphasis given to planning and policy analysis even more im- extended on-the-job training in institutions portant than before-for instance, in the al- both inside and outside Africa, the potential location of limited resources available for of such training in the transport field should maintenance across a network, in concen- be assessed by a suitable unit in the donor trating upgrading efforts on highest-priority community, such as the World Bank's Trans- stretches, in making intermodal comparisons, portation Department, among others. and in projecting overall financial require- ments over several years. Progress has been INFIRASTRUCTURE AND POLICY PLANNING made, but most countries need to improve Capabilities for setting and following rea- systems for collection of relevant data, such soned and well-coordinated priorities among as regular, accurate traffic counts and road possible investments in transport and com- inventories, and many countries must munications infrastructure and for develop- strengthen institutional arrangements for as- ing policv improvements must continue to be sessing and screening projects. In addition to built up. Resource shortages make careful establishing planning units within modal agencies, arrangements are needed for coor- 21. Better arrangements must also be made than now dinating investments and policies among dif- exist for Coordination of activity almong aid suippliers in the trans- ferent modal agencies. port and communications sectors, particularly because of the large number of small investments required in the com- ing vears. the importance of consistent technical and finan- cial as,;istanice and the manv donors that will have to be mobilized It needs are to be met. :111 The preceding chapters have discussed meas- Table 8.1. Population Projections for Eight ures for accelerating economic growth in the African Countries and Sub-Saharan Africa short term. But the urgency of these imme- (in millions) diate problems does not diminish the impor- Cotrntry 1980 2000 2010 2020 tance of certain longer-term development Zimbabwe 7 15 23 31 questions: the twin issues of rapid population Cameroon 8 14 20 25 growth and expanding urbanization; pro- Ivory Coast 9 16 23 31 grams and policies to conserve Africa's soils Ghana 12 23 34 45 Kenya 16 37 58 81 and expand its forests and fuelwood supplies; Tanzania 19 37 54 72 and ways of encouraging more regional eco- Zaire 28 52 73 95 nomic integration. Nigeria 85 172 255 341 Eight-country total 184 344 540 721 Sub-Saharan Africa 353 679 980 1,411 (100) (192) (278) (400) Note: The projection assumes constant fertility and de- The serious consequences of rapid population clining mortality. growth are increasingly recognized by Afri- Source: World Bank data files. can leaders. In 1973, nine Sub-Saharan coun- tries supported family planning for reasons of health and as a human right. By 1978 this projections are based on the assumption of group had increased to twenty. Five of these constant fertility and declining mortality. countries had, in addition, explicit policies to While fertility will probably fall somewhat, reduce population growth. Nevertheless, the numbers illustrate the powerful momen- within many African countries there is sub- tum of high fertility. stantial ambivalence about population growth, and occasionally there is even the suggestion THE NEXT Two DECADES that Africa would be better off with more rapid In the next two decades, Africa's population population growth. A different view is pre- will continue to grow rapidly-by 3 percent sented here. a year according to World Bank estimates. It The consequences of rapid population increased by an average of 2.7 percent an- growth for economic development and wel- nually in 1970-79-the highest rate of growth fare are very negative. The numbers per se in the world-and is likely to grow even faster are dramatic. If fertility does not decline from as improved health and nutrition reduce in- its current average rate of 6.6 live births per fant mortality (now estimated to be about 150 woman, the population will grow as pro- per thousand) and child death rates. Fertility jected in Table 8.1. Forty years from now, may also increase because of improved ma- Nigeria will have a population of 341 million, ternal health. This is already the pattern in Zaire 95 million, and Kenya 81 million. These Kenya, Zimbabwe, and Zambia, where pop- 112 ulation growth exceeds 3 percent annually able at nearly 4 percent annually. Expansion (see Table SA.34). at 3 percent-ordinarily a welcome achieve- What are the consequences? Most of the ment-will mean that a decreasing share of additional people will begin life in rural areas Africa's children will be able to attend pri- because the urban population is relatively small mary school. (21 percent in 1980). Even if cities were to grow at unprecedented rates, they could only POLICY IMPLICATIONS absorb a part of the rural increase. If just 60 The above scenario for the remainder of the percent of the increase remains rural, declin- century envisages pressure on the land, ex- ing to 50 percent by the year 2000, the rural tremely rapid urbanization with declining population would still be 50 percent greater quality of life, and little increase in the share in twvo decades. Thus, rather than decrease, of population provided with basic services. It rural population would grow very rapidly. is true that population growth usually de- The population of African cities grew 6 per- clines as urbanization proceeds and as edu- centa vear in the lastdecade (see Table SA.36), cation and other aspects of modernization in response to both opportunities in the cities become more widely available. Parents tend as well as reduced incomes in traditional ag- to want small families in urban settings where riculture. An expanding rural population will the extended family no longer exists to pro- push more people out of agriculture into cit- vide child care, where children are no longer ies, particularly in the many parts of the re- producers but are "consumers" in school, and gion -where diminishing returns to traditional where mothers are more highly educated and cultivation are already common (see Chapter have opportunities to earn income outside the 2). Urban pressures will grow in the 1980s home. In brief, modernization brings a de- and the 1990s, as city populations continue cline in fertility-the "demographic transi- to expand, probably at close to current rates. tion." But, excluding Mauritius, no country Mlany cities will double their population each in Sub-Saharan Africa is yet on this road. decade and it is almost certain that the ma- Instead, population growth has recently in- joritv of inhabitants will lack basic ameni- creased in several countries because mortality ties-water, sanitation, and electricity. Other from infectious disease has been reduced; the services will be very scarce: health care, waste factors (listed above) that lead to a decrease disposal, street paving, and communications. in family size have thus far operated only to The nonagricultural economy will need to a small degree. generate jobs at an unprececented rate. Thus, it is crucial to take steps now to re- The question of food self-sufficiency will duce fertility. There is widespread recogni- become more pressing. For example, if the tion that efforts at family planning can be population grows at 3 percent and per capita effective, even prior to modernization. Use of incomes grow between I and 2 percent, there new contraceptive techniques has accelerated will be an annual increase in food demand of in many Asian countries, in part as a result about 4 percent. But in the last decade, ag- of information and education campaigns. For riculture grew at less than half this rate (1.8 example, in Indonesia, techniques were de- percent annuallv). Thus, unless there is a rad- veloped for gaining local community support: ical increase in agricultural production, more village leaders endorsed the programs and rather than less imported food will be needed villagers staffed contraception distribution in the coming decade. centers. These results, however, are due to Then there is the issue of basic services. two decades of experimentation. Rapid population growth slows progress In most of Africa, on the other hand, in tovward universal education and health care. both urban and rural areas there have been Merelvmaintainingtheexistingprimaryschool few efforts of this kind. Before 1972, there enrollment ratio in the next two decades will were no African "family planning experi- require expanding the number of places avail- ments"-that is, attempts to provide family 113 planning services to large groups. And while * Family-planning components should be built the Asian experience can be a useful model, into the health care system; it must be adapted to the specific conditions * Activities in support of family-planning in Africa. At present, the region has only be- policies-notably dissemination of infor- gun to build a body of knowledge on how mation-should be encouraged; and best to proceed. * Governments should develop units-per- The widespread and traditional practice of haps in planning ministries-to undertake childspacinginAfricaisaccomplishedthrough policy analysis that would also emphasize prolonged breastfeeding and abstinence. The the effects of future population growth on practice of and desire for child spacing offer basic goals, such as employment, literacy, a vehicle to promote the acceptance of mod- food, and security. ern contraceptive techniques. In fact, there is The basic point is clear. The population already a small but significant use of modern problem is not merely one of size; it is one contraceptives in West Africa for these rea- of urgency. Where growth is rapid, time is of sons. To increase the use of modem tech- the essence. For example, if the time between niques, the programs proposed must involve children could be increased so that the av- methods that can be easily reversed and do erage number of births per woman decreased not interfere with lactation. from the current average of 6.6 to 4.8 by the The traditional attitude toward spacing pro- year 2000, the total population in Sub-Sa- vides a rationale for incorporating family haran Africa would be 640 million rather than planning within matemal and child health 679 million. And by 2010, assuming a further programs. Once people use the new tech- decrease in average fertility to 3.7 children, niques to achieve traditional goals, they can the total would reach 821 million rather than adapt the technology to other goals, such as 980 million. reducing family size, as external conditions change. In this context it is revealing that the Urban Gro.7t?t World Fertility Survey has shown that al- though many African families have achieved Although most 4fricans currently live in the desired family size, most are not using any rural areas, and it is recognized that rural type of modern birth control. development is the key to economic growth, To conclude, emphasis on modern contra- urban areas are rapidly expanding. This cre- ceptives to provide birth spacing appears to ates not only immediate problems but also have great potential in accelerating Africa's urgent long-term concerns. demographic transition. These considerations As increasing numbers of the population suggest that population policy in Africa should have flocked to cities to find employment, be largely concerned with slowing population urban populations have mushroomed overall growth by the following actions: by 6 percent a year, and 8.5 percent annually . Family-planning advice should be recog- for 35 major capitals-a rate at which they nized as the right of every couple, and pro- will double in size every nine years. There viding such advice to all who desire it should are now 28 African cifies of over 500,000 pop- be a basic goal of every govemment; ulation, where, just 20 years ago, there were * Governments should encourage wide- only 3. In fact, the urban population is ex- spread family-planning services and sup- pected to quadruple again in the last quarter plies, including availability of contraceptives of this century. at a very low price, perhaps even free; Most of these urban populations are de- * Female education should be encouraged for prived of basic services. A large proportion many reasons, including the fact that it leads live in slums or squatter settlements, very few to reduced child mortality and eventually households have interior water supplies, and reduced fertility; large numbers have no access to running 114 water. Sanitation services are minimal. In at reaching the largest numbers possible. In Freetown, for example, 95 percent of the pop- the 1960s, policies stressed relatively high- ulation use shared pit latrines, while in Abi- quality construction, often financed by exter- djan, 65 percent use open pits or unlined water nal sources through housing parastatals. In courses. Only 20 percent of Abidjan's resi- former British colonies, the Colonial Devel- dents are served by a sewage system-and opment Corporation financed national hous- this is much higher than is common through- ing corporations and building societies that out the region. built houses on the metropolitan model. In Until now, few governments have had re- francophone countries, many societes im- sources to devise systematic strategies ad- mobilieres were created with the same ob- dressing urban problems; day-to-day demands jectives. Ivory Coast's Societe Ivoirienne pour have absorbed municipal authorities. Unless la Construction et la Gestion Immobilire the new, sprawling urban population is in- (sicoGi) was perhaps the most active, con- tegrated into the economy and its needs ad- structing 3,000 to 4,000 apartment units a year dressed, the situation is likely to be politically in the mid-1960s. Its Senegalese equivalent, disruptive and to stunt economic progress. Societe Immobilire du Cap Vert (sICAP), built Further, the cities, which now produce about extensively in high-income areas of Dakar. half of national output, will become less ef- Public housing construction under this type ficient, as labor productivity and economic of policy never reached the mass of urban growth will decline. poor, or even the middle-income groups. Over An urban strategy should focus on the fol- 90 percent of the urban population in all Af- lowing three areas: planning and administra- rican cities had to seek other solutions. tive structures to cope with the cities' In the last decade, a "sites and services" organizational and financial problem; mech- approach to housing developed and evolved anisms to provide jobs and services, and safe, into "upgrading." These schemes provide ru- sanitary, and affordable housing; and the en- dimentary infrastructure but leave construc- couragement of small- and medium-size sec- tion of housing (or units) to individual ondary centers. occupants. Between 1972 and 1981, a total of Many of the existing municipal institutions $533 million was spent on such projects in 15 are vestiges of the colonial era. They were different countries, the World Bank alone designed to serve only that small portion of contributing about $270 million (see Box A). the population which lived in the city centers, With respect to employment, efforts must and in their present form are not equipped be made to encourage the informal sector and to deal with today's problems. The new gov- labor-intensive activities. This could involve ernments have sought to provide a high level promotingthe construction industryand small- of services, for understandable reasons. But scale cottage industry, where the cost of job- given the current scarcity of resources and creation is low. The cornerstone of such a the enormity of unmet needs, unless "stand- strategy is the encouragement of small entre- ards" are set at modest levels, affordable for preneurs. As noted in Chapter 7, private en- both the municipalities and the consumers, trepreneurs have not only been ignored in the urban masses will still be poorly served. many countries, but indirectly discouraged by In the provision of water supply or sanitation, unnecessary regulations. To encourage their for example, where per capita income is $350 growth, creditcouldbeprovidedthoughquasi- a year, standards should set a minimum level governmental agencies, along with services of service that can be improved over time. like electricity and water. In this area, donor Interim measures would be standpipes and agencies can play a useful role by helping aqua privies, instead of fully developed san- provide the capital for small-scale industry itation or water systems. and also, through policy discussion, a more In housing, policies should also be aimed congenial local policy environment. 115 Box A: Two Approaches to Urban Housing Squatter Housing in Lusaka. In Lusaka, the largest and of paying for these services (among all income groups fastest growing city in Zambia, housing problems are in Zambia) and the weakness of the institutions and acute. Thus, a "sites and services" project was devel- mechanisms required for the collection process. Thus, oped (with $3 million of World Bank funding) to up- procedures are being strengthened and eviction meas- grade and service 17,000 dwellings in fourmajorsquatter ures initiated. settlements, to prepare 12,000 residential plots, and to provide loans for home improvements and construc- Francistown Urban Project. The Francistown Urban tion. It was also designed as a pilot project to dem- Project, approved in 1974, was Botswana's first attempt onstrate the feasibility of low-cost housing and todevelopastrategyfororderlyurbangrowth.Itsought infrastructure programs. to develop simple, low-cost, technically appropriate While the project was under way, the Zambian Gov- standards for urban infrastructure that would benefit emnment experienced an acute financial crisis which the poor and immediately improve the living and pub- led to delays, cost overruns and, eventually, cutbacks lic health conditions of the squatter population. in the original plan. Nevertheless, much was accom- The plan called for upgrading existing squatter areas plished: over 31,000 plots-accommodating approxi- which contained about 1,000 households, providing mately 30 percent of Lusaka's population-were 1,000 plots for traditional housing with minimum serv- provided with basic infrastructure; half the sites and ices, and offering 800 sites and services plots for which services units were allocated to low-income house- loans (for building materials) would be made available holds-the poorest 40 percent of the population; up- and deeds granted. It also involved constructing a small grading reached over 90 percent of the households in number of community centers, servidng areas for in- the targeted area; plots and loans were allocated ac- dustrial and commercial plots and for 25 plots for high- cording to a point system which favored the poorest cost residential settlement, improving electricity dis- residents; and houses built under the project cost ap- tribution, building roads, and installing water pumps proximately one tenth of the amount required under and mains. conventional schemes. The targets were met. Almost 95 percent of the town's The project succeeded for several reasons. First, population had access to clean water, roads, street technical assistance was provided to the Lusaka City lighting, and tenure to the land on which they were Council and to a special office, the Housing Project squatting. A total of 800 sites and services plots were Unit (HPU), which was set up to focus solely on this allocated and 1,000 upgraded plots were provided with project. The budget was large enough to attract highly infrastructure. A full 80 percent of the beneficiaries qualified staff, and cooperation from the community were low-income residents, previously without basic was carefully cultivated. The American Friends Service services. In addition, new jobs were created. Committee and uNIcEF, which had done similar com- The project succeeded primarily because it was de- munity outreach elsewhere in Zambia, worked with veloped on a small scale and for a small town. Also, the project, bringing to it their previous experience. tasks were decentralized. While the central (national) Most importantly, the community was involved in the bureaucracy was responsible for awarding construction planning and did not feel the project was being im- contracts, the municipal government handled "people- posed from above. related" aspects, such as allocating plots, providing Once the project was completed, the HPU was dis- loans for building materials, offering technical assist- banded. But a new office, called the Peri-urban Areas ance in self-help construction techniques, and collect- Unit, was created to repeat the program in other squat- ing loan payments and user fees. ter areas on the outskirts of the city. Cost recovery still remains -a problem and the self- Collections have been a continuing problem. As of help housing agency, which has been vital to com- March 1980, 80 and 93 percent of the households were munity development, still requires technical assistance at least three months behind in service charges and and a more clearly defined relationship with the central loan repayments, respectively. This problem is not due government. to residents' inability to pay but to the lack of tradition Wherever possible, services should be as- set at minimum standards, governments can- sociated with cost-recovery schemes, since not afford to supply them nor can low-income surveys have shown that even in the poorest groups afford to pay for them. User charges areas, much of the population can afford are crucial so that governments can generate minimal charges and are ready to pay for basic the funds needed to provide additional serv- services. However, unless these services are ices to the ever-increasing population. Also, 116 when populations are aware that they will be the demarcation between privately and pub- charged for water, electricity, and sanitation, licly owned land is often vague. Local political they often reduce their expectations to the and traditional interests are, of course, in- level provided. And while it is understood volved in land ownership and use, compli- that cost recovery is no easy task, if reason- cating the development of a coherent national able standards are adopted it will become more policy regarding the demarcation and use of feasible. public lands. National policies and legislation While the discussion above applies partic- dealing with the distribution and use of ground ularly to the larger cities, it is important to and surface water are also lacking, with the note the vital development role played by result that sharp conflicts between the various secondary centers. Small towns serve as a link categories of users have become more and between urban and rural development: they more frequent. are natural distribution centers for agricul- The resource data that are most relevant tural goods produced in the rural areas and for soil and water conservation are: physical manufactured goods from the cities. When properties of soils; water retention capacity of agriculture prospers, these towns become soils and vegetation; quality and quantity of major providers of off-farm employment op- vegetation cover; drainage patterns; density portunities for rural people. of human and livestock population; and pre- sent land use (farming and grazing practices, use of forests and trees). This kind of infor- .''K'>":i.> ;' f-'K.l u1'iu1g mation is rarely available, is patchy in time and space, and is only rarely synthesized in Three issues in the natural resources area are usable form. of special significance: soil conservation, re- Land use planning agencies, if they exist, forestation, and fuelwood supply. The fact is are usually not equipped to use modern that Africa's people are primarily dependent methods of land classification such as satellite on what the land provides for grazing and images or infrared aerial photography. The crops, shelter, and firewood. If we are to ex- application of these techniques should be far pect this fragile resource to continue produc- more development-oriented and geared to ing into the next century, conservation steps users' needs in rainfed agriculture, small-scale must be taken now. irrigation, forestry, and animal husbandry. As regards erosion control and watershed SO1L CONSERVAIION management, it is useful to distinguish be- Rainfed agriculture will necessarily be the main tween micro-solutions, which are applied at source of incremental production in the great the farm or village level, and macro- and long- majority of African countries for some time term solutions on public lands, which aim at to come. But the long-term potential of ag- controlling erosion and runoff caused by in- riculture is being diminished-in some coun- discriminate tree-felling, overgrazing, and tries seriously-because of more intensive slash-and-burn cultivation. The methods to cultivation and grazing. Conservation of ex- be propagated and supported at the village isting resources is therefore the crucial long- or farm level should focus on soil and water term task in agriculture. conservation techniques that are within the A strategy to stop the accelerating degra- technical and economic reach of farmers and dation of soils and vegetation is overdue. The hold sufficient short-term benefits and incen- design and carrying-out of such a strategy, tives. Pilot efforts should be made an integral hovwever, are hampered by political con- part of area-based agricultural development straints, lack of data, and the rudimentary projects with due recognition of the role that capacity of land-use planning institutions. the traditional local authorities or village or- The knowledge of land tenure patterns and ganizations can play in mobilizing local labor 117 for implementation and maintenance. Exper- low the rate needed to solve the long-run iments of this type in Upper Volta, for ex- fuelwood problem in Africa. The present rate ample, have had encouraging results. External of planting of fuelwood species in Africa is assistance can play an important role in the in the order of 70,000 hectares per year. But compilation of resource data and the orga- for African countries to gain self-sufficiency nization of land-use-planning units, in the in wood energy by the year 2000, this rate dissemination of experience from other parts would have to increase at least fifteenfold to of the tropical world, and in developing large- a level of one million hectares per year. Fur- scale methods of soil conservation, which have thermore, massive fuelwood programs are yet to be found. subject to the constraints of limited land avail- ability, the weakness of forestry services and REFORESTATION research programs, the lead time required to Reforestation is, of course, an important ele- develop an institutional capability to handle ment of watershed management and erosion the program, and budgetary constraints at the control, but it is also justified in most coun- national level. Because the inputs into refor- tries to maintain or increase the supply of estation are largely land, labor, and time, ru- fuelwood, which is being seriously depleted. ral people themselves can generate fuel Reforestation and soil conservation pro- supplies at low cost. Woodlots have to be grams have in common that they can be selected carefully, however, as forests take stepped up in line with perceived needs only land away from food crops or grazing. More- to the extent that countries make progress in over, since planting does not solve immediate organizing these works in a cost-effective wood needs, it is often difficult to enlist the manner. In most cases, this can only be ac- cooperation of farmers and landless laborers, complished bI mobilization of the rural pop- especially if they are not assured of their rights compishd bvmoblizaionof te rral op- to the mature trees. ulation; without active and voluntary Nonetheless, a doubling or possibly tri- cooperation, programs cannot be maintained. ofethelent a lng of planti- Contributing labor to works with long-term pling of the current (1980) levels of planting benefits often has high opportunitv costs for should be attainable in some countres. High farmers, however, and the key task is to arouse priority should be given to farm forestry, in- sufficient interest before the ecological situa- cluding planting treesaround homesteadsoand tion becomes desperate and the need for ac- along farm boundaries. Achieving a doubling tion obvious. At that late stage, ecological of the current rate of reforestation over the deterioration may have become either irre- next five years in Africa as a whole would be versible or much more costly to redress, expensive, requiringexternal investmentfunds There is a major role for foreign sources of of $350-$500 million. The main focus of such J p~~~~~~~~rojects would be on creating the institutional assistance to help African institutions shift pand i tuld freatingtheineededtior the emphasis in the training curricula toward larger planting programs in the future. The rural and environmental forestry, and to or- larger higherorams of planting de ganize training courses for African foresters. justification for higher rates of planting de- Funding of more training programs for for- pends, in each country, on the experience estry extension workers in farm forestry and with past and ongoing reforestation pro- support of forestry research are further prior- grams, most of which nevertheless remain in ities. A recently formed research agency, the the pilot stage. International Council for Research in Agro- Forestry (icRAF), located in Nairobi, is just get- ting under way. Regionalism FUELWOOD In the Lagos Plan of Action, the African Chiefs of State approved the target of an economi- Current efforts at reforestation are much be- cally unified, self-reliant continent by the year 118 2000. The donor community should help Af- rect short-run costs and benefits of cooperation rican governments move toward this objec- schemes are often not favorable. Whatever tive, since regional economic cooperation and the activity-a regional research effort, a joint ultimate integration are important for the re- faculty, or a collaborative health campaign- duction of long-term obstacles to develop- high administrative costs are involved. It in- ment, in several respects. variably takes longer to implement such ef- First, broadly speaking, small states have forts,asspeciallegalandinstitutionalquestions limited development alternatives. And most arise. The benefits, moreover, may be uncer- African states are small-in population and tain or intangible: economies of scale may not in market size. Only six have more than 15 be quickly obtainable. In addition, financing million people (Ethiopia, Kenya, Nigeria, Su- for these kinds of regional efforts is often dan, Tanzania, and Zaire). Twenty-four num- sought from external sources; but donors ber less than five million each, and twelve usually find the execution of national projects fewer than one million. challenging enough, and are reluctant to bear It is true that much remains to be done in the risks, delays, and sometimes higher ex- the short and medium run for which larger pense involved in regional programs. Second, size is not a particular requirement: establish- during the past two decades, internal political ment of effective administrative machinery, consolidation has been the primary goal in development of a more productive monetized most states, and the security, continuity, and agriculture, creation of physical and social in- political confidence required to push through frastructure, spread of suitable education, and economic cooperation schemes have not been similar tasks. But after the institutional adequate. Finally, because of well-known groundwork has been laid and the agricul- problems involved in equitable distribution of tural and mineral potentials of small African benefits, both potential gainers and losers have countries are more fully exploited, further become more hesitant to enter into deep in- development, notably of industry, requires volvements with neighboring states if poten- larger markets. This almost surely will entail tial mutual gains are not both obvious and larger inter-African economic units. Thus, sure to those involved. economic integration is needed to widen the Regional economic cooperation then, while range of options available in the future. essential in loosening long-term development Second, long-run solutions to the special constraints facing many African states, will problems of the landlocked, usually ex- not come easily. It will require changes of tremely poor states, require regional ap- great substance, including strengthening of proaches. This is true for the Sahelian states, transport links, reduction of monetary and Swaziland, Lesotho, Rwanda, and Burundi, commercial policies that inhibit and distort among others. The question of economic vi- intraregional trade, promotion of joint proj- ability is highlv relevant for these states and ects in industry, education, and research, and for those who think about their development regional institutions with adequate staff and options. Most of these states can achieve rea- budgets that could become major instruments sonable levels of economic welfare and growth of cooperation and integration. only in association with the better-endowed countries in their regions. Market forces al- DONOR ROLES ready reflect this reality-notably via migra- Some of the ways that economic integration tion flows. The point is that the economies and functional cooperation in Africa can be of some of the landlocked states will develop facilitated by external aid are indicated in ear- more fully in a framework of regional eco- lier chapters: the financing of transport links nomic cooperation or integration. between countries; shared hydroelectric facil- The record on economic cooperation and ities; factories serving integrated markets; integration efforts thus far shows few suc- multinational research and teaching insti- cesses, for three main reasons. First, the di- tutes; and the provision of hydrological, me- 119 teorological, and other technical services, to investmentstudiesofintegrationprojects.This name a few. Foreign aid donors could also is an area in which financially hard-pressed provide support for regional institutions that African governments are often unable to com- have clear and sustained backing from Afri- mit scarce funds. Finally, donors can, in gen- can governments. Donors could play a cata- eral, raise the level of priority they ascribe to lytic role by financing feasibility and pre- integration and cooperation. 120 > f 2CT -,$sT XI A 4T s ' '5¢J '' r- < r -e ST4 -i 7 Only the joint effort of African nations them- than elsewhere in the developing world. Thus, selves and the international community will to the extent that Africa's problems are the lead to progress in solving the problems that result of unsuitable project concepts or per- face Africa today. A commitment from both petuation of inappropriate policies, the donor partners will be necessary to transform stag- community shares in the responsibility. Af- nation into growth, and the prospect for end- rican states and donors have been bound to- less poverty into hope for a better life. African gether in development efforts to a unique governments must lead the way because do- degree, and so they must remain if the pre- mestic policy issues are at the heart of the sent crisis is to be solved. crisis, and no real turnaround is conceivable Aid to Africa must be augmented in the unless these policv matters are dealt with. An 1980s because, first of all, the continent con- equally profound commitment is necessary tains many of the poorest and most vulner- from the international community-a com- able people in the world. Twenty of the thirty mitment to increase aid and to provide as- least-developed countries are in Africa, and sistance in ways more suitable to Africa's needs the remaining African countries are little bet- than in the past and in support of the reform ter off. Oil-importing Sub-Saharan Africa as programs defined by the African govern- a whole has a per capita income of only $316. ments. Policy reforms supported by substan- By almost any other indicator-life expect- tiallv increased aid flows promise substantially ancy, child mortality, literacy, access to safe improved growth prospects for Africa in water, or supply of educated people-Africa the 1980s. is extremely poor. Second, Africa's medium- term prospects for growth are worse than those . ::. .* .. *, t! .' . ,-;l7>CL1 .AiLt of any region. Projections in the World De- velopment Report 1981 indicate per capita in- External assistance to Africa is already at a come growth of only 0.1 percent per year for relatively high level compared with other de- the period 1980-90, on optimistic assump- veloping regions (net ODA per capita in 1980 tions (see Table 9.1). Third, Africa is even less was $13.70 for Africa compared with $9.60 for well situated than other low-income devel- all developing countries), and has grown oping areas to take advantage of the growth quickly in the past decade. Aggregate num- of world trade. The share of nonfuel mer- bers on aid flows alone do not tell the full chandise exports is far higher in Africa, and storv of the contribution of donor agencies- the region is specialized in a group of primary both bilateral and multilateral-to the devel- commodities with slower growth prospects. opment of Africa. The extent to which outside Fourth, Africa remains highly dependent on personnel have played a role in projects, in concessional capital (aid) because of limited policy advice, and in formulating develop- creditworthiness. Finally, increased aid to- ment strategy has been much greater in Africa gether with policy reform holds the promise of 121 Table 9.1. Growth of GNP per Person, 1960-90 GNP per person Annual growth of GNP (1980 current Lo7v case High case Countrn grouvs dollars) 1960-70 1970-80 l980-90 1980-90 Sub-Saharan Africa Low-income oil importers 260 1.7 -0.4 -1.0 0.1 Middle-income oil importers 520 1.7 0.4 0.0 0.3 Oil exporters 730 0.4 2.6 2.0 2.3 All developing countries 850 3.5 2.7 2.2 3.3 Low-income 250 1.8 1.6 1.5 2.6 Middle-income 1,580 3.9 2.8 2.2 3.4 Source: World Bank, World Development Report 1981 (New York: Oxford University Press, 1981), Table 1. 1. a highly fruitful investment which will be able Table 9.2. Projected Performance of to accelerate growth during the 1980s. Oil-importing African Countries, 1980-90 Average annual growth 1980-90 (percent) Without policy With policy reform reform : : : . ,: J>- ,.- Performance Small aid Stubstantial Small aid Substantial indicator increase aid increase increase aid increase Gross domestic As noted above, the World Development Report product 2.4 3.1 4.2 5.0 GDP per capita -0.5 0.2 1.3 2.1 l981 projects a very gloomy economic outlook Agriculture 2.3 2.8 3.5 3.8 for oil-importing Africa in the 1980s-vir- Exports 2.6 3.3 4.1 5.2 tually no growth in per capita income for the Imports 0.7 2.3 2.3 3.9 decade even with optimistic assumptions about Source: World Bank projections. the external environment. While these pro- jections are not simply an extrapolation of Without policy improvement there is insuf- past trends, they assume no fundamental ficient structural adjustment to get the econ- changes in domestic policy. They project a omy back onto a faster growth track. Import modest growth in exports corresponding to requirements remain high because of the im- maintenance of constant shares of world trade port-intensity of industry, a growing share of of Africa's principal primary products and they energy imports, and continued importation assume stabilization of exports in countries of cereals. Export growth continues to be which have experienced declines due to war. sluggish. Domestic savings stagnate. All that We have used the World Bank's global model higher aid can do is help to sustain the level to project development trends in oil-import- of imports and also that of investment. ing Africa assumning extensive domestic policy With appropriate policy reforms, the pros- reform and different levels of aid.' pects brighten. The reforms are those rec- Table 9.2 summarizes the results of these ommended earlier-changes in agricultural simulations. With continuation of most pres- policies, improvements in the efficiency of ent policies and only a small increase in aid- resource use in the public sector and more ODA to oil-importing Africa rising only slightly open trade and exchange-rate policies. These over the decade (see Table 9.3)-per capita changes in policy orientation are expressed in GDP is projected to fall throughout the 1980s. the model as follows: . Increased incentives for agricultural pro- duction permit an expansion of production 1. The "small aid increase" assumption is that annual with minimal investment and at low for- ODA to African oil importers (excluding technical assistance) rises in real terms from S4.7 billion in 1980 to $5.8 billion eign exchange cost; in 1990. Under the 'substantial aid increase" assumption, * Reallocation of spending for essential ODA rises to S9.0 billion (1980 prices) in 1990 (see Table 9.3). maintenance and recurrent costs allows 122 existing capacity to be used more effec- policies and approaches often involve untried tively and increases growth with minimal instruments whose feasibility and efficacy must investments; be tested. There is always resistance to changes * Slower growth of government expenditure from bureaucratic inertia or vested interests. and higher private (including parastatal) In short, outside help is necessary to lubricate savings result in more domestic savings and the process of change and of policy adjust- a higher investment/GDP ratio; and ment, to finance new departures, and to help * Increased incentives for exports enable ex- African governments soften the effects of port growth to increase and imports to grow change on groups that fear losses of income, faster, for a given amount of aid. such as urban consumers. Policv reform without substantially in- Table 9.3 shows the overall magnitude of creased aid, however, does not provide a sat- ODA disbursements to Africa, and Table 9.4 isfactory solution. Investment is required to indicates the level of ODA disbursements that take advantage of improved incentives for will be required of multilateral and bilateral export and agricultural production and to donors during the decade ahead. The "sub- produce and conserve energy. Policy reform stantial aid increase" figure in Table 9.3, which can boost growth, but without greatly in- helps bring about the faster growth shown creased aid there will be insufficient foreign on Table 9.2 (Column 4), represents an almost exchange and investment funds available to fourfold increase in net disbursements of ODA allow full structural adjustment. Higher aid to Africa between 1980 and 1990-from S4.9 permits African governments to increase billion to $17.8 billion (a near doubling in real maintenance and recurrent expenditure dur- terms). This growth of aid combined with ing 1980-85, wvhile simultaneously sustaining policy reform could increase per capita in- investment levels that have a payoff in the comes in oil-importing Africa by nearly one latter part of the decade. This will further quarter during the coming decade compared increase growth during the period and lay the with virtual stagnation without it. groundwork for more sustained growth dur- While these aid flows will require a major ing the last decade of the century. Moreover, effort by all donors, the rate of growth is less mans African countries could not undertake than was achieved in the previous decade. It reform without additional assistance. New should be feasible to achieve these increases, Table 9.3. Net Disbursements of Official Development Assistance (ODA) to Africa 198O-90O (millions of dollars) Average annual growth rate Smnall Substantial Small aid increase Suibstantial aid increase aid increase aid increase .NCi ,l:s&rtct'writPls l9SO 1985 1990 2985 1990 1970-80 1980-90 1980 -90 Al cizirrent przces Africa 4,883 7,740 11,889 10,219 17,839 18.7 9.3 13.8 Oil importers 4,706 7,482 11,482 10,101 17,627 19.9 9.3 14.2 Oil exporters 177 258 407 118 212 4.1 8.7 1.8 .-l! cos.}P'Y.:it 19S" ,rict-s Atnrcj 4,883 5,269 6,044 6,956 9,069 5.0 2.2 6.4 Oil inrnortcr- 4,706 5,093 5,837 6,876 8,961 6.1 2.2 6.7 Oil e,portcrs 177 176 207 80 108 -8.0 1.6 -4.8 Atnca a, pcrcen!age 0f oi)A to all developing couLn- trivs 22.5 21.9 22.2 25.0 27.1 ad Excludes technical assistance. Suirce iVorld Bank projections. 123 but they will require early action to increase Table 9.4. Net Disbursements of ODA to commitments. The World Bank will continue Africa, 1979 (Actual), and 1985 and 1990 to give priority to Africa in the allocation of (Projected)' r 1 T ' 1 1 ~~~~~~~~~(millions of current dollars) funds from the International Development mon ent Association (IDA). Approximately 30 percent Percent of IDA funds are planned for Africa in FY82- Acttal share 1979 Of 86. However, the volume of resources that net total Projected ODA net this share of IDA will represent depends on disburse- ODA disbursettzentsb the successful implementation of the agreed Source ments 1979 1985 1990 sixth replenishment of IDA and the yet to be DAC bilateral 3,570 60 8,840 15,480 OPEc bilateral 270 5 680 1,190 discussed level of the seventh replenishment. Total multilateral Bank lending to Africa will continue to receive of which: 1,830 31 4,540 7,940 priority, although the volume will be limited Arab financed agencies (370) (6) (910) (1,590) by creditworthiess considerations. Support All other sources 270 5 670 1,160 from the International Finance Corporation Grand total 5,940' 100 14,730 25,770 (IFC) for private sector development in Africa a. Projections are calculated using the rates of annual is expected to grow. increase of the substantial aid increase projections for ODA disbursements in Table 9.3. Technical assistance is included here; it is not included in the ODA figures of Table 9.3. b. Total net disbursements of ODA in 1979 were 68.5 per- cent of commitments. Assuming that this relationship re- mains constant, total commitments of S21.5 billion will be AssISTANCE IN FORIMULATING AND SUPPORTING required in 1985, and $37.6 billion in 1990. PROGRAMS OF AC-ON c. Technical assistance is estimated to be between 20 and 25 percent of ODA. Policy reform needs to be based on detailed Source: OECD, Geographical Distribution of Financial Flows to macro and sector analysis. The level and Developing Countries, various issues. structure of agricultural prices, structure of protection, wage and salary policy, marketing and storage programs, and other aspects of In addition, donors can provide assistance development policy require examination in the in examining specific policy issues when re- context of a particular country if precise policy quested by governments. For its part the World measures are to be formulated and effective Bank undertakes a continuous dialogue on action is to be taken. In addition, attention economic and sector policy issues based on needs to be given to the formulation of re- its regular economic work. This macro and alistic public sector investment programs. sector analytical work is the basic documen- These should reflect limitations on resources, tation for consultative group meetings for those sectoral priorities, and appropriate project se- countries for which such a group exists. The lection criteria. importance of generating a policy framework, The capacitv to undertake this policy-fo- which is more growth and development ori- cused analysis should be a priority objective ented, implies that more policy-focused eco- of governments. Donors can assist in building nomic work will need to be undertaken. The up this capacity and in so doing can help World Bank is expanding its economic work governments, both directly and indirectly, in Africa, and is prepared to form consultative formulate macro and sectoral action pro- groups for countries that request this pro- grams. The World Bank is prepared to re- vided that effective donor participation is spond to all requests of this nature. This probable. assistance will probably require an expansion of field staff of donor agencies. The World DONOR SUPPORT FOR PROGRAMS OF Bank, in the past few years, has considerably POLICY REFORM increased its own field resident staff in Africa The level and pattern of donor assistance to and is willing to consider further expansion. a country must be determined in the frame- 1 24 work of programs of action prepared by in- assistance the priorities attributed to infra- dividual governments, which address the structure (transport, telecommnunications, and critical development policy issues outlined in power), to urban development, to the pro- this Report. In this way, donor financial as- vision of safe drinking water, and to mining, sistance will effectively support the attain- can be transformed into a supply of projects ment of development objectives, and avoid that can, and will, attract increasing amounts financing projects that do not reflect a gov- of financing from a variety of sources. ernment's priorities or even run counter to The formulation of sound projects, whether these priorities. for domestic implementation or donor fi- Donor assistance can take many forms- nancing, should increasingly be within an project lending and nonproject lending (sec- agreed policy framework which assures that tor lending, program lending, structural ad- the project contributes to increasing the rate justment lending); foreign exchange cost of development in the economy. This policy financing; local-cost financing; and recurrent- framework should encompass the issues dis- cost financing. All of these have a role to play cussed earlier, and should be in the context in different countries and in different circum- of a public investment program consistent with stances. Donors should be prepared to be the resources available to the economy. Do- flexibLe and use the tool most likely to in- nors as well as African govemments must crease the contribution of their aid to the have clear project selection criteria to ensure process of development. that projects support current national priori- ties and to avoid investments which have low PRZOJECT LE-NDING, rates of return. Conventional project lending will be an im- Because project financing is the dominant portant vehicle of resource transfer from the means of assistance, it must be sufficiently developed world to African countries. Not flexible so that projects can be designed to only is this frequentlv the preferred form of respond to African realities. Three aspects are lending for many donors; it is also particularly of particular importance. relevant to Africa's development needs in view of the limited planning, project identification, Administrative Complexity. Programs and and implementation capacity. The need for policies that will reduce the burden on the increasing levels of project lending is most overextended administrative capacity of Af- obviously seen in the case of agriculture, which rican countries have been a major theme of has to be the centerpiece of development pro- previous chapters. Donor financial assistance grams in Africa. For instance, the FAO Regional should be structured to support the devel- Food Plan for Africa called for major invest- opment of smaller and organizationally more ments (S125 billion in 1981 prices) in food manageable operational units in both the public crops, livestock, and agricultural support and private sectors. In particular, donors services for the 1975-90 period (see Chapter should support governments in devising ar- 5). This emphasis on agriculture has been re- rangements through which external assist- flected in World Bank lending, where the share ance can be channeled to small firms in of agriculture in total lending to Africa rose industry, agriculture, and the ancillary serv- from 20 percent in 1969-73 to 33 percent in ice industries, such as transport. Assistance 1974-78. It is expected to stay at about this of this kind can partly be "projectized" through level in the future. financial intermediaries-development fi- In other sectors, too, the identification, nance companies, agricultural credit organi- preparation, and implementation of projects zations, and commercial banks. Commercial ws ill continue to have high priority as a means banks, in particular, have a network of of developing the institutional, human, tech- branches, and are generally well staffed and nical, and phvsical capacity central to the de- managed; they constitute an institutional ca- velopment process. With appropriate pacity that could be much more widely used 125 for development financing purposes. Donors increasing proportion of total project costs and could assist in this process. In addition to they are increasingly abandoning their pre- investment funds, many small firmns and farms vious objections to local-cost financing. For simply need to be able to purchase supplies the same reasons, the approach to local coun- of imported spare parts, tools, and fertilizer. terpart contribution is also changing. While a Availability of such inputs requires that for- local contribution is often still required-usu- eign exchange be made available to a country ally 10 to 15 percent in the case of the World in a form that does not tie its use to a specific Bank-in-kind contributions are now more project but which, at the same time, can be widely accepted by donors. Today, with heavy channeled to the intended beneficiaries. pressures on budgets and large investment Project designs, developed and supported expenditures ahead, these more flexible pol- by donors, need also to recognize more fully icies are appropriate. the overextension of administrative capacity which has occurred in many African coun- Recurrent-cost Financing. In Africa, it is now tries. Whenever feasible, existing organiza- widely acknowledged that the shortage of tions should be used and their administrative budgetary resources limits the operating funds and managerial capacity strengthened as part available to utilize investments already made of project activity. and causes inadequate maintenance of past There is also a need for pilot studies. More investments. Vehicles and equipment fre- good projects will be ready for financing only quently lie idle for lack of spare parts, repairs, when there has been a large expansion of gasoline, or other necessities. Schools lack project-focused research. In agriculture this operating funds for salaries and teaching ma- might mean a thorough review of the agro- terials, and agricultural research stations have nomic evidence, farm systems research, com- difficulty keeping up field trials. Roads, pub- parative and pilot studies, a deeper lic buildings, and processing facilities suffer involvement of both borrower and lender in from lack of maintenance. the preparation process, and use of high-level As with the local-cost issue, donors now consultants on short-term assignments. The recognize that adequate external financing of need for more experimentation with new recurrent costs may be necessary to ensure technical packages and new forms of social the successful completion, maintenance, and organization has been underlined above. While operation of development projects, and that the number of pilot projects financed by ex- expenditures along these lines frequently have ternal sources has increased in recent years, higher yields than traditional capital projects. there is scope for more experimentation. There is widespread acceptance of the need for donors to finance recurrent costs during Local-cost Financing. The volume of local-cost the construction phase of a capital project. financing for projects has been growing in The difficult cases involve projects or sectors recent years; a survey of DAC members shows which are not directly productive, where no that local-cost financing averaged over 13 per- end to the need for recurrent-cost support can cent of gross disbursements in 1977.2 Since be foreseen. Rural development projects projects in Africa typically have a high foreign sometimes have this characteristic, but it exchange content, this implies that between mainly concerns such sectors as health and one fourth and one third of local costs are education. Here donors should remain sym- being financed by aid. Because of the resource pathetic to recurrent-cost financing over rel- limitations of African governments, donors atively long periods, in recognition of the long- recognize that it is desirable to finance an term returns possible from these expendi- tures. Prudence is required in adding new 2. OECD, Development Cooperation: Efforts and Policies of the staff and starting new activities that must be Members of the Development Assistance Committee, 1979 Review absorbed by the government's budget, and (Paris, 1979), p. 102. whenever feasible priority should be on more 126 productive uses of existing govemment staff uation confronting African countries, foreign and facilities. Budgetary implications should exchange for these purposes is typically be explicitly addressed, since, ultimately, re- squeezed out by the combination of the prior- sponsibility for financing current operations ity demands for debt servicing, food, and fuel must be transferred to the regular budget. on the one hand and project-tied foreign ex- change on the other. Nonproject assistance Nonproject Financing. While project financ- which is linked to policy changes intended to ing will continue to be the major channel of stimulate the small producers would help to donor financial assistance to Africa, further relieve this need. development of nonproject lending will be There is a comparable need for nonproject desirable. This can take many forms, varying assistance to meet the operation and main- from subsector lending (which might in some tenance requirements of ongoing projects. In cases approximate project lending) and sector part, these requirements can be met through lending, to lending for a range of imports. regular project lending. For example, high- The growing importance of nonproject way maintenance programs can be supported lending is related to the importance of major by flexible project lending. However, there changes in policies by African governments. will remain a range of needs for foreign ex- Nonproject lending can generally be used more change to meet operation and maintenance effectively than project lending for agreeing requirements which are too diverse and frag- on a program of action that addresses major mented to be "projectized" in this way. These policy issues. The formulation of such pro- can best be financed by donor lending which grams can be made the sole focus of negoti- is linked to this broad range of sectoral or ation, and the disbursement of donor funds economic needs. can then be made quickly in support of the The World Bank has recently introduced agreed policy action. Formal arrangements forms of nonproject assistance to address these between donors and governments will, of needs. Structural adjustment lending (SAL) was course, vary and some donors might be will- introduced in 1980 to assist countries that had ing to make their own nonproject assistance formulated comprehensive programs of ad- available in support of an agreement reached justment to meet the deteriorated balance-of- by another donor. payments prospects they would face during The second important role of nonproject the 1980s. SAL has the distinguishing features assistance is that it makes available foreign of providing: (a) foreign exchange to finance exchange that can be used for imports which imports not linked in advance to specific in- are of high priority for development but which vestment programs; and (b) finance over a cannot easily be financed through project number of years in direct support of specific lending. Reference has already been made to policy reforms. Comprehensive programs of this in the context of the need for donors to structural adjustment are required not only give support to changes in government pol- to enable countries to return to the growth icies that would stimulate the growth of small path they had before the deterioration in their production units in the economy-farmers, external economnic circumstances occurred but, artisans, small firms, road builders, and co- in addition, to improve on a previous unsat- operatives. Many of these small producers isfactory rate and pattern of growth. Struc- have either no access to financial intermedi- tural adjustment programs comprise those aries or no need for such access. Certainly changes in policy over a range of sectors (ag- many of them would be excluded from con- riculture, industry, and energy in particular) ventional donor project concepts. They need needed to generate the required changes in to be able to purchase critical inputs for their the structure of output and growth in the operations through the regular distribution medium term; policy changes must lead to a network of shops and traveling salesmen. In higher rate of growth of expoTt earnings and the very constrained balance-of-payments sit- a lower rate of import growth than would 127 otherwise occur. Such loans are being made have a long-term commitment to the country, to Kenya, Malawi, Mauritius, and Senegal in this risk should be minimal. the early 1980s. Of course, development assistance is not In cases in which a comprehensive adjust- the only source of nonproject finance. The IMF ment program has not been formulated and has played, and will continue to play, a major adopted by a government, the Bank has pro- role in providing balance-of-payments assist- vided assistance in support of programs of ance through its regular operations, the Ex- policy change that are more limited than in tended Fund Facility, and the Compensatory the case of SAL. These operations have had a Finance Facility. Under the Lome Conven- sector or subsector focus and finance expend- tion, nonproject assistance in the form of iture which is likely to yield quick returns and compensatory financing is available. quick effects on production. In the case of Another important source of freely avail- Sudan, support has been given through lend- able foreign exchange is the private capital ing for the rehabilitation of the agricultural market. Although relatively few African sector. This assistance seeks to achieve a bet- countries have adequate creditworthiness, one ter upkeep of capital assets through adequate objective must be to improve debt manage- allocations for maintenance, rehabilitation, and ment so that maximum use of this market can rationalization of existing projects, to be made. Initially, borrowing in the private strengthen the planning and policymaking capital market may be limited to project- capacity of government at the national and related finance, but as creditworthiness im- sectoral levels, and to address the problems proves it can also be used in support of gen- of policy framework and institutional envi- eral sector programs and ultimately for general ronment that so frequently determine the suc- balance-of-payments financing. cess or failure of projects. A similar operation Finally, the experience of the World Bank in Tanzania focused on the rehabilitation of and the IFC demonstrates that there is a grow- the export sector with an emphasis on the ing opportunity for private investment in required changes in incentives and in the al- Africa. Lack of knowledge and uncertainty on location of budgetary funds to agriculture. the part of investors restrict the flow of such An intermediate form of financing can also capital and the associated managerial talent be used more broadly for longer-term pro- and technological know-how. The growing grams, whether in education, irrigation, or involvement of the IFC in Africa indicates that maintenance of infrastructure. It is possible with appropriate assistance in these areas, for donors to review the longer-term program private investment opportunities can be ex- and, if it is mutually acceptable, to finance panded and donor govemments can do much onlv a fixed number of years of its imple- to assist by expanding information flows to mentation. Thus, donors might finance three their business community and assuring that years of a six-year highway maintenance pro- adequate insurance and credit facilities are gram, as is sometimes done already. The ad- available. The work of the World Bank, par- vantages are that a smaller amount of resources ticularly in the areas of oil and gas and other has to be committed at the outset and that minerals, suggests that the provision of geo- the support can be provided for the total pro- logical information, analysis and restructur- gram rather than to select specific compo- ing of tax laws and other incentives, and nents for project financing and thus reduce relatively small financial participation to re- the flexibility of the operating agencies in the duce political risk can attract foreign private use of funds. The risk involved is that the investors. In turn, they will contribute not commitment cannot be made for the full pe- only their own equity but can obtain Bank riod of implementation and thus it is possible capital that otherwise might not be available. that a portion of the program may be without The scope for action in this area is still very adequate financial support in the later years. large and donors can assist in providing sim- However, if the program is sound and donors ilar services to potential investors. 128 DEBT MANAGEMENT Table 9.5. Actual and Projected Sub-Saharan The World Development Report 1981 indicates African Debt-Service Ratios that overall debt-service ratios are likely to rise p moderately for oil-importing Africa through the Category 2977 1978 1980 1985 1990 1980s (see Table 9.5). (The debt-service ratio is Oil importers 8.4 10.6 15.8 17.6 19.8 the ratio of interest and amortization payments Low income 8.4 10.4 19.2 9.5 19.9 to export earnings.) While the ratios for the Middle income 8.1 10.1 13.6 16.2 19.3 P ~~~~~~~~~~~~~Oil exporters 1.8 3.3 3.5 4.6 4.1 low-income countries are estimated to remain Sorterl p . unchanged at around 19 percent, those for the middle-income countries are expected to rise from 13.6 percent in 1980 to about 19 percent African governments themselves can take in- in 1990. For oil-exporting countries the debt- clude the following: service ratio is projected to rise through 1985 * strengthening their reporting systems to and then decline. provide accurate information on the extent, Projections of debt-service ratios are highly composition, and trends of external bor- dependent on assumptions about trade pros- rowing to serve as a guide for policy; pects, import requirements, and capital flows. * introducing or strengthening legislation to Further, aggregate projections often mask wide streamline and centralize the authority to country variations. Some countries that ex- contract debt; and perienced severe debt-servicing difficulties * actively reviewing debt-service implica- during the 1970s (Liberia, Sierra Leone, Su- tions of government financing plans, dan, Zaire, and Zambia) are likely to continue medium-term development programs, and to do so in the 1980s due in part to the impact the cost of borrowing for purposes (such of the earlier debt reschedulings and to am- as balance of payments or current budget ortization requirements for outstanding obli- support) that do not contribute to future gations. Other large borrowers in the 1970s foreign exchange earnings. (such as Cameroon, Ivory Coast, Senegal, and Donors, and particularly the World Bank, Tanzania) could face difficult debt-servicing which has at its disposal extensive debt data problems if export performance falls. and experience with rescheduling exercises, Debt-service obligations are likely to be- should significantly increase their training and come a more important element in the bal- technical assistance efforts in these debt- ance of payments of many African countries3 related areas. and may compel more of them to use the Much closer collaboration than has existed Paris Club4 and other arrangements for debt up to now would be desirable between pro- relief. It will also require that African govern- viders of ODA and other agencies of DAC and ments strengthen their capacities in debt re- OPEC member governments who have respon- porting and debt management, so that they sibility for managing export or suppliers' credit. can identify timely corrective measures to head Given the economic prospects in most coun- off the emergence of debt problems. tries, and the fact that even with substantial Preventive actions that keep debt problems changes in policy growth will accelerate sig- from arising are to be preferred to curative nificantly only in the second half of the dec- actions that must be taken once the debt and ade, debt problems should be considered in balance-of-payments situation becomes un- the broader framework of the general prob- manageable. Actions of this nature that the lems of development, and longer-term solu- tions for debt crises should be sought. The 3 The iollowing African countries had recourse to mul- present practice of separating aid and debt tilateral debt renegotiations during the period 1975-80: Zaire decisions may be counterproductive. It is in (1976. 1977, 1979); Sierra Leone (1977, 1980); Gabon (1978); the interest of creditors and donors that eco- Togo (1979); Sudan (1979); Liberia (1980). nom 4. The Paris Club refers to ad hoc meetings of represen- nomic and financial health be restored to the tatives of the governments of western creditor countries. economies of the borrowing countries, and 129 that the separate treatment of aid and debt mously complicate the tasks of national not jeopardize a promising attempt on the administrations; and under cofinancing ar- part of a debtor government to put through rangements, which are more and more com- a viable policy-reform package. mon, individual donors often leave to the host agency the demanding task of arranging total HARMONIZING AID PROCEDURES project financing. Aid inflows have grown rapidly in the past It is easier to identify these problems than decade-nearly 20 percent per year in current to find practical solutions for them. National, prices-and new donors, such as the OPEC organizational, and administrative styles are countries, and institutions have become im- involved, as are a thicket of legal require- portant contributors. The rapid build-up of ments imposed on aid transactions by donor aid volumes and the accompanying diversi- governments. There are, nonetheless, certain fication of aid sources have created significant ways to simplify things, and the adoption of problems. streamlined procedures would help all par- First, the stepped-up aid activity, injected ties.5 Certainly one major contribution donors into a situation of manpower scarcity and weak can make to Africa is to subordinate, to the administrative structures, has compounded extent legally and politically practicable, dif- administrative problems and contributed to ferences in style, documentation, criteria, and distortions in project execution. To bypass local analyses, to the overall need for administra- administrative inefficiencies, donors tend to tive simplicity. set up autonomous project authorities in one Coordination of project selection criteria and form or another. They also compete for scarce aid allocations is a bigger problem. There are skills by bidding up salaries and fringe ben- hesitations about greater donor coordination efits. From each project authority comes a call on both sides. Each donor institution sees for higher wages, not always tempered by the things differently, and each has different con- realization that higher wages in one sector straints and objectives. Many find the idea of cannot resolve the problem of generalized skill orchestration objectionable. Some African scarcity. These attempts to bypass the exist- governments are also unenthusiastic. They fear ing structures do not reduce problems of wage "ganging up," as well as a loss of "maneu- disparities and incentives; often they exacer- verability"; individual spending agencies bate them. A general result of the competition would almost certainly see their scope of ac- between donors for people and for projects tion limited by effective donor coordination. is to raise the economic or opportunity cost Nonetheless, some coordination does take of new projects, as scarce factors are made place and more would be desirable. The Arab- more costly, coordination is rendered more OPEC agencies have been particularly success- burdensome, and accumulating commit- ful in harmonizing their activities through the ments tie up fiscal resources with little con- concept of the lead-donor for each project. sideration of the costs in terms of sacrificed Local coordination arrangements that allow a financing for existing activities. more complete exchange of information and Second, the multiplicity of donors, each dovetailing of activities should therefore be operating independently, puts an especially encouraged. heavy burden on small countries with limited administrative capacities. These burdens are TRAINING AND TECHNICAL AsSISTANCE well known: competition for projects among Accelerated growth in Africa depends more donors often undermines decisionmaking than anything else on training people and procedures and increases the difficulty of holding to sectoral and national priorities; different donor requirements and procedures 5. See Chapter 7 of oEcD, Development Cooperation: Efforts and Policies of the Members of the Development Assistance Com- regarding project identification, appraisal, mittee, 1979 Review for a fuller discussion of progress being procurement policies, and so forth, enor- made in the simplification of donor procedures. 130 building institutions. Although technical as- ect-often the lowest priority. Training usually sistance can help overcome particular scarci- begins late in the cycle of the project. Little ties in the short run, its major purpose is, in-service training occurs; most of it consists and should be, training and the building up of sending local trainees abroad. Most donors of institutions. and African governments are aware of the Training and technical assistance represent problems and are trying to make improve- major areas of donor activity. In the late 1970s, ments. Some of what should be done is fairly about 25 percent of ODA was in the form of clear: training needs should be identified very technical assistance grants. The identifiable early in the project cycle, and trainees chosen project-related training in World Bank proj- quickly; the training component should re- ects alone amounted to $55 million in 1979, ceive substantially more attention in projects; or about 8 percent of total Bank/IDA lending and project supervision and evaluation should to Sub-Saharan Africa in that year.' The Af- give greater weight to training performance rican regions of the Bank spend three to four rather than focusing overwhelmingly on the times as much as other regions on project- physical progress of projects. Donors and lo- related training. cal govemments should, either by sector or The contribution of technical assistance and at the center, develop local training capacity, donor-financed training programs has been including training of African trainers, employ substantial. Many, probably most, of the specialists to prepare training components, present cadre of experienced African techni- and employ training officers in large projects. cians and managers have benefited directly Donors, including the Bank, must strengthen from these training and technical assistance their own ability to provide adequate and re- arrangements. Since trained people remain sponsive services in this area. scarce in most of the continent, large tasks As noted earlier, the expansion of project remain for technical assistance and large aid to Africa requires enlargement of the al- requirements for training. A production- ready substantial flow of project-ielated tech- oriented strategy will require greater reliance nical assistance. Much of this would go to on technical assistance in the decade ahead. strengthening project management. Empha- Substantial increases in project lending will sis on the need for management assistance is require increases in technical assistance in most especially strong among donor technicians, of Africa, as will the formulation of sectoral who tend to see in the presence or absence investment programs and the evolution of the of good management the chief determinant policy framework. of project success.' Some modifications in the general ap- It is worth noting that this stress on the proach do seem appropriate, however: management factor may be exaggerated. There strengthening project-related training; em- is at any moment in any system a given stock phasizing on-the-job training, including train- of management skills. No project or other ing outside Africa; and shifting technical activity should expect more than the "aver- assistance more decisively in the direction of age" level of management ability present in training. Special emphasis is given in Chapter the available supply. If project success de- 6 to training managers and economists (or pends on finding a manager who possesses policy analysts), since there are particularly abilities far superior to the average, the proj- critical needs in these fields. ect itself may be badly designed. In any case The problems with project-related training there are always scarcities of "superior" man- are wel[ understood: in many cases the train- agers in any society, since they have special ing component has low priority in the proj- talents. Furthermore, there may be plenty of 6. While most sectors were involved, over half of training 7. In many World Bank ex-post project evaluations "good expenditures ($29 million) were in the agricultural sector management" is the most frequently cited source of project and most of the rest ($20 million) were in transport. success. 131 management ability present in a given coun- between expatriates and that of senior local try, but the environment (social, political, and officials can be as high as 10:1 or even higher). economic) reduces the effectiveness of these All of these factors, combined with a grow- abilities. It is commonly the case that an ex- ing sense of confidence among local techni- patriate manager makes a project or an en- cians, are lowering the threshold of terprise run better not so much because he is acceptability of expatriate technical assist- a better manager than his local counterpart, ance, and they suggest that some shift in policy but because he is removed from many of the direction is in order. First, short-term tech- constraints of the surrounding system. nical assistance should be relied on more The implications of these considerations are, heavily to complement a greater reliance on first, that project design should not proceed local staff, systems, and institutions. Recruit- as though there were no local management ment and logistic problems are much reduced constraints on the implicit assumption that in this way, and the injection of outside ad- managers will be imported. Project designers vice and help can, in some instances, be more should generally make precisely the other effective. Second, local people should be em- assumption: that the project will be locally ployed more frequently as consultants and managed. Its scale and complexity should then staff for externally financed projects. In the be shaped accordingly. There may be tech- case of the World Bank, existing regulations nical or economic considerations that rule this permit the use of consultants. This has been out, particularly for capital-intensive projects tried in some projects and should be encour- and those necessarily embodying modern aged. Third, given the difficulties of combin- technologies (hvdroelectric projects, large ir- ing the functions of management or technical rigation projects, airport operation, modern expert and trainer, resident technical assist- industries), but these should be explicitly as- ance people as well as those on short-term sessed. Second, donors should avoid the cre- assignments should be explicitly named ation of autonomous project management "trainers," except where they have another units, staffed with expatriates and operating task of explicitly higher priority; in general, more or less independently of the local or- training should be their primary and, in some ganizational and political environment. Even cases, their exclusive responsibility. The though there may be some price in terms of training components of programs should be lowered material output of the particular proj- reviewed to see how the training aspect can ect, the long-term process of management im- be strengthened. As noted above, future proj- provement and institution-building is furthered ect-related training components should have only by integration of these projects into ex- heavier weight, and alternations should be isting administrative systems. made in the personnel mix of such projects. Technical assistance needs will continue to be substantial through the 1980s and beyond. If this assistance is to be effective, certain Conclusion problems must be recognized. The main dif- ficulties are the maintenance of the supply of Despite their enormous advances since in- high-quality personnel willing to undertake dependence, particularly in developing insti- technical assistance assignments, the partic- tutions, human resources, and even nations, ular need for people who can develop the Sub-Saharan African countries are in a crisis "counterpart" relationship thereby emphasiz- that can only be surmounted by the joint ef- ing the training responsibility and avoiding forts of African governments and the donor becoming an executive, and the high cost of community. The increased aid and related technical assistance which, in some parts of technical assistance recommended in this Re- Africa, can reach $150,000 per year for a per- port can only be mobilized if they support son hired through a consulting firm and which deliberate and well formulated programs to can be a source of stress (salary differentials reverse the downward trend of development 132 in Africa. The policy reforms required in Af- to be successful. But the rewards of taking rica will be technically difficult and politically these pains will be great. Policy action and thorny. African governments and the donor foreign assistance that are mutually reinforc- community will have to work out a relation- ing will surely work together to build a con- ship that recognizes these realities if the ac- tinent that shows real gains in both tion program recommended in this Report is development and income in the near future. 133 Accelerated Development in Sub-Saharan Africa: An Agenda for Action STATISTICAL ANNEX ConieM+ Key 141 Introduction to the Annex 142 Basic Indicators Table 1. Basic Indicators (by country) 143 Population El Area El GNP per capita o Growth of GNP Cl Inflation o Adult literacy o Life expectancy o Food production per capita Production Table 2. Growth of Production (by country) 144 GDP Li Agriculture o Industry 0 Services Table 3. Structure of Production (by country) 145 GDP c Agriculture 0 Industry o Services Table 4. Growth of Consumption and Investment (by country) 146 Public consumption 0 Private consumption 0 Gross domestic investment Table 5. Structure of Demand (by country) 147 Public consumption 0 Private consumption ol Gross domestic investment oL Gross domestic saving C0 Exports of goods and nonfactor services Ol Resource balance Table 6. Commercial Energy (by country) 148 Growth of energy production and consumption 0 Energy consumption per capita oL Energy imports as a percentage of merchandise exports O Petroleum imports as a percentage of total commercial energy consumption Trade Table 7. Growth of Merchandise TTade (by country) 149 Export values L Import values C Growth of exports o Growth of imports Table 8. Structure of Merchandise Exports (by country) 150 Fuels o Minerals and metals 0 Food and beverages o Other primary products O Manufactures Table 9. Structure of Merchandise Imports (by country) 151 Food o Fuel o Other primary commodities oL Machinery and transport equipment o Other manufactures - Table 10. Principal Merchandise Exports of Sub-Saharan Africa 152 Value of principal exports o Percentage share of Sub-Saharan African exports 0 Principal exporters Table 11. Merchandise Exports by Degree of Processing 153 Sub-Saharan Africa 0 Other developing countries 137 Table 12. Destination of Merchandise Exports (by country) 154 Industrialized market economies al Sub-Saharan African countries 3l Other developing countries Cl Centrally planned economies o Capital-surplus oil exporters Table 13. Terms of Trade (by country) 155 Net barter terms of trade Cl Income terms of trade Cl Growth in terms of trade Table 14. Exports: Commodity Concentration and Fluctuation in Values (by country) 156 Percentage share of three principal exports E Index of export fluctuations Table 15. Commodity Trade: Volume and Prices 157 Growth of volume for Sub-Saharan Africa and the world C Change and growth of prices Table 16. Developed Country Tariff Rates on Selected Commodities 158 European Economic Community El Japan ra United States Aid, Debt, and Capital Flows Table 17. Balance of Payments, Debt Service, and International Reserves (by country) 159 Current account balance before interest payments on external public debt al Interest payments o Debt service as a percentage of GNP El As a percentage of exports L Gross international reserves E In months of import coverage Table 18. Debt and Debt Service (by country) 160 Public and publicly guaranteed private debt outstanding and disbursed El Official sources El Private sources El Total debt o Debt service Table 19. Outstanding External Debt of Sub-Saharan Africa 161 Outstanding and disbursed, value E Total concessional bilateral o Total official export credits E Total multilateral loans E Total private publicly guaranteed loans El Total public and publicly guaranteed loans El Total private nonguaranteed loans El Total public and private debt Table 20. Gross Disbursements of External Loans to Sub-Saharan Africa 162 Annual value u Total concessional bilateral El Total official export credits Cl Total multilateral loans cl Total private loans E Total public and publicly guaranteed loans El Nonguaranteed private sector credit El Total public and private disbursements Table 21. Average Terms of Borrowing for Sub-Saharan Africa 163 Average terms for new commitments of public and publicly guaranteed debt El Total public debt El Total official debt EL Bilateral debt 0 Multilateral debt El Total private debt Table 22. Indicators of Aid and Total Resource Flow, 1979 (by country) 164 Population El Total recorded net flow of resources per capita ol Net official development assistance (ODA), disbursements per capita E Net bilateral as a percentage of total El As a percentage of GNP nl As a percentage of gross domestic investment cl From OPEC as a percentage of total net bilateral ODA 138 Table 23. Disbursements of Official Development Assistance (by country) 165 Net disbursements value D Grants as a percentage of net ODA 0 Technical assistance as a percentage of net ODA Table 24. Food Aid Imports (by country) 166 Thousands of metric tons, grain equivalent 0 Kilograms per capita Agriculture Table 25. Growth of Agriculture (by country) 167 Growth of volume of production, food and nonfood n Growth of production per capita, food and nonfood Table 26. Production of Major Crops 168 Production O Growth rates Table 27. Yields of Major Crops 169 Average yields o Index of yields o Index of relative yields Table 28. Fertilizer Consumption (by country) 170 Annual consumption O Growth rates Table 29. Agricultural Imports 171 Volume C Growth of volume 0 Value Table 30. Agricultural Exports 172-73 Volume O Growth of volume a Share of total of developing countries a Of the world Table 31. Domestic Terms of Trade of Export Crops for Selected Countries 174 Index of barter terms of trade o Index of income terms of trade Table 32. Procurement and Distribution of Agricultural Inputs 175 Government and private sector activity in supply of: Fertilizer 0 Seed O Chemicals o Farm equipment Social Indicators Table 33. Population Growth, Past and Projected (by country) 176 Average annual growth rate 0 Projected population o Hypothetical size of stationary population o Assumed year of reaching net reproduction rate of 1 a Year of reaching stationary population Table 34. Demographic and Fertility-related Indicators (by country) 177 Crude birth rate a Crude death rate C Percentage change in crude birth and death rates o Total fertility rate 0 Life expectancy at birth 0 Child death rate Table 35. Labor Force (by country) 178 Percentage of population of working age o Percentage of labor force in agriculture o In industry El In services O Average annual growth of labor force 139 Table 36. Urbanization (by country) 179 Urban population as a percentage of total c Average annual growth rate ° Percentage in largest city a In cities of over 500,000 persons a Number of cities of over 500,000 persons Table 37. Health-related Indicators (by country) 180 Population per physician C Per nurse r Percentage of population with access to safe water o Daily calorie supply per capita, total C1 As a percentage of total daily requirement Table 38. Education (by country) 181 Number enrolled in primary school as a percentage of age group C Number enrolled in secondary school as a percentage of age group o Number enrolled in higher education as a percentage of population aged 20-24 o Adult literacy rate Fiscal Data Table 39. Central Government Budgetary Operations (by country) 182 Share in GDP of: Capital expenditure c Current expenditure c Net lending c Tax revenue C Nontax current revenue o Capital o Grants C Financing, foreign and domestic Table 40. Central Government Taxes and Expenditure (by country) 183 Ratio of taxes to GDP C Ratio of taxes to expenditure Table 41. Central Government Functional Expenditure (by country) 184 General public services C Defense o Social services o Economic services C Other services Table 42. Functional Distribution of Increase in Central Government Expenditure, 1972-78 (by country) 185 General public services o Defense a Social services o Economic services o Other services Table 43. Military Expenditure (by country) 186 Armed forces per 1,000 population a Military expenditure as a percentage of GNP Technical Notes 187 Bibliography of Data Sources 197 140 In the country tables, countries are listed in following conventions apply: . . not avail- ascending order of income per capita within able; (.) less than half the unit shown; n.a. each country group. The reference numbers not applicable. indicating that order are shown in the alpha- Data for countries followed by a double betical list of countries below. asterisk were unavailable from the standard Figures in colored bands are summary sources used elsewhere in the tables; in these measures for country groups. cases, information from World Bank files has The letters w, m, and t indicate weighted been used. mean, median, and total, respectively. The Angola 36 Mali 3 Benin 13 Mauritania 7 Botswana 33 Mauritius 34 Burundi 10 Mozambique 14 Cameroon 31 Niger 6 Central African Republic 19 Nigeria 38 Chad 1 Rwanda 12 Congo 37 Senegal 27 Ethiopia 8 Sierra Leone 15 Gabon 39 Somalia 2 Gambia 5 Sudan 24 Ghana 26 Swaziland 32 Guinea 18 Tanzania 16. Guinea-Bissau 9 Togo 23 Ivory Coast 35 Uganda 21 Kenya 25 Upper Volta 4 Lesotho 22 Zaire 17 Liberia 29 Zambia 30 Madagascar 20 Zimbabwe 28 Malawi 11 141 introduction to tlhe Annex The Statistical Annex provides information on Sub-Saharan African countries (Cape Verde, the main variables affecting social and eco- Comoros, Djibouti, Equatorial Guinea, Sao nomic development in Sub-Saharan Africa Tome and Principe, and Seychelles) are not (SSA), including: Basic Indicators; Produc- included because of lack of data. The alpha- tion; Trade; Aid, Debt, and Capital Flows; betical list of countries in the Key shows the AgTiculture; Social Development; and Fiscal reference number of each country. Data. Within low-income countries, defined Summary measures (means, medians, and as those with a per capita income of less than totals) are calculated for the country groups or equal to $370, a distinction is made be- only if there are enough countries with suf- tween (1) low-income semiarid countries and ficient data to make the measures meaning- (2) all other low-income countries. For mid- ful. Readers should exercise caution in com- dle-income countries, defined as those with paring summary measures across country a per capita income exceeding $370, a dis- groups as well as indicators across countries. tinction is made betwveen (1) oil importers and The quality of statistics is weaker in Africa (2) oil exporters. (There are no low-income oil than in most other parts of the world, and exporters.) Worldwide averages or medians many of the statistics presented here are very for low-income, middle-income, and indus- approximate. National practices also vary, trialized countries' are usually included for and country data are not always comparable. comparison. The Technical Notes that follow the Tables The countrv tables include 39 countries of outline concepts, definitions, and specific Sub-Saharan Africa with population exceed- data problems and are followed by a Bibli- ing one half million in 1979. The six remaining ography of Data Sources. Tables 1 through 9, 12, 17, and 33 through 38, and their tech- 1. Industrialized countries exclude all centrally planned nical notes, are derived from the World De- economies. The countries in these three groups are as listed in the Annex to the World Development Report 1981 (New velopment Indicators of the World Develop- York: Oxford University Press. 1981), pp. 129-92. ment Report 1981. 142 Table 1. Basic Indicators Average GNP per capita ~~~~~~~~~index GNP per capita ~~~~~ ~~~~~L ife of food Average A vrganulexpec- production Area annual Avrage ofifannual Adult tancy per capita Populatin (thouands groth rate (percent) literacy at birth (1969-72 (millions) of square Dollars (percent) (percent) (years) = 200) mid-1 979 kilometers) 1979 1960-79 1960-70 1970-79 1976, 1979 1977-79 Low-income countries 187.1 t 15,718 t 239 w 0.9 w 2.8 m 10.2 m 25 w 46 w 91 w- Low-i'ncome semi arid 28.0 t 5,745 t 187 w 0.0 w 3.3 m 10.0 m 17 w 43 w 88 w I .Chad 4.4 1,284 110 - 1.4 4.6 7.9 15- 41 91 2. Somalia 3.8 638 .. - 0.5 4.5 11.3 60 44 85 3. lali 6.8 1,240 140 1.1 5.0 9.7 10 43 88 4. Upper Volta 5.6 274 180 0.3 1.3 9.8 5'* 43 93 5. Gambia 0.6 11 250 2.6* . . 10** 42 77 6. Niger 5.2 1,267 270 - 1.3 2 .1 10.8 8 43 89 7 Mauritania 1.6 1,031 320 1.9 1.6 10.1 17* 43 75 Low-income other 159.1 t 9,973 t 247 w 1.0 w 2.8 m 10.7 m 27w 47 w 91 w S. Ethiopia 30.9 1,222 130 1.3 2.1 4.3 15* 40 84 9. Guinea-Bissau 0.8 36 170 ....7** 42 94 10. Burundi 4.0 28 180 2.1 2.8 11.2 25 42 105 11. Malawi 5.8 118 200 2.9 2.4 9.1 25* 47 100 12. Rwanda 4.9 26 200 1.5 13.1 14.6 .. 47 107 13. Benin 3.4 113 250 0.6 1.9 9.2 7-~ 47 97 14. Mozambique 10.2 783 250 0.1 2.8 11.0 . 47 75 15S. Sierra Leone 3.4 72 250 0.4 2.9 11.3 .. 47 87 lb. Tanz.ania 18.0 945 260 2.3 1.8 13.0 66* 52 94 17. Zaire 27.5 2,345 260 0.7 29.9 31.4 15 47 90 IS. Guinea 5.3 246 280 0.3 1.5 4.4 20* 44 86 19. Central Afr-ican Rep. 2.0 623 290 0.7 4.1 9.1 .. 44 102 20. Madagascar 8.5 587 290 - 0.4 3.2 10.1 50* 47 94 21. Uganda 12.8 236 290 - 0.2 3.0 28.3 .. 54 90 22. Lesotho 1.3 30 340 6.0 2.5 11.6 52* 51 100 2.3. Togo 2.4 57 350 3.6 1.1 10.3 18 47 81 24. Sudan 17.9 2,506 370 0.6 3.7 6.8 20r 47 105 Middle-income oil importers 65.2 t 3,690 t 532 w 1.5 w 2.4 m 9.9 m 34 w 50 w 95 w 25. Kenva 15.3 583 380 2.7 1.5 11.1 45* 55 92 26. Ghana 11.3 239 400 -0.8 7.6 32.4 . 49 82 27. Senegal 5.5 197 430 - 0.2 1.7 7.6 10, 43 88 28. Zimbabwe 7.1 391 470 0.8 1.3 8.4 .. 55 100 29. Liber-ia 1.8 1ll 500 1.6 1.9 9.4 30 54 101 30. Zambia 5.6 753 500 0.8 7.6 6.8 39* 49 99 31. Cameroon 8.2 475 560 2.5 4.2 10.3 . 47 110 32. Swaziland 0.5 17 650 7.2** .. 65** 47 109 3.3. Ekotswana 0.8 600 720 9. 1.* . 35** 49 89 34~ Mauritius 0.9 2 1,030 2.3** . 80** 65 100 35. Ivory, Coast 8.2 322 1,040 2.4 2.8 13.5 20 47 102 Middle-income oil exporters 91.6 t 2,781 t 669 w 3.2 w 3.3 m 19.0Om . 48 w 86 w 36. Angola 6.9 1,247 440 -2.1 3.3 21.6 .. 42 85 37. Congo 1.5 342 630 0.9 5.4 10.9 .. 47 81 38. Nigeria 82.6 924 670 3.7 2.6 19.0 .. 49 87 39. Gabon 0.6 268 3,280 6.1** .. 12- 45 94 Sub-Saharan Africa 343.9 t 22,189 t 411 w 1.6 w 2.8 m 10.3 m 27 w 47 w 91 w All low-income countries 2,260.2 t 33,778 t 230 w 1.6 w 3.0 m 10.8 m 51 w 57 w 105 w All middle-income countries 985.0 t 38,705 t 1,420 w 3.8 w 3.0 m 13.3 m 72 w 61 w 107 w Industrialized countries 671.2 t 30,430 t 9,440 w 4.0 w 4.3 m 9.4 m 99 w 74 w 110 w a. Figures marked with an * are for years other than 1976. See technical notes. 143 Table 2. Growth of Production Average annual growth rate (percent) GDP Agriculture Industry Services 1960-~70a 1970-79b, 1960-70a 2970-79b 1960-70, 1970-79", 1960-70, 1970-79b Low-income countries 3.7 w 1.7 w 1.5 m 1.5 in 4.2 mn Low-income semiarid 2.3 w 2.2 w 0.7 m 1.0 m 4.6 m I. Chad 0.5 - 0.2 .. 0.7 .. 0.2 .. -2.6 2. Somalia 1.0 3.1" -1.5 2.7* 3.3 - 2.6" 2.5 6.8* 3. Mali 3.3 5.0 .. 4.2 .. 4.2 .. 6.1 4. Upper Volta 3.0 -0.1 . -3.3 .. 1.0 .. 2.9 5. Gambia** 5.4 2.8 4.6 7.0 2.1 3.3 6.6 -0.5 6. Niger 2.9 3.7 3.3 -1.5 13.9 10.2 (. 4.6 7. M,auritania .. 1.8 .. -1.4 .. 0.1 .. 7.2 Low-income other 3.9 w 1.7 zv 1.8 m 1.9 m 4.0Om 8. Ethiopia 4.4" 1.9 2.2" 0.4 7.4" 0.4 7.8" 4.6 9. Guinea-Bissau. .. .. .. .. 10. Burundi 4.4 3.0 .. 1.8 .. 7.7 .. 4.0 1I. M~,alawi 4.9 6.3 .. 4.1 .. 7.0 . 9.1 12. Rwanda 2.7 4.1 .. . .. ... 13. Benin 2.6 3.3.. . ..... 14. Mozambique 4.6 - 2.9 2 .1 -1.8 9.5 - 5.6 64 -3.0 15. Sierra Leone 4.3 1.6 .. 2.3 .. - 3.8 .. 4.4 16. Tanzania 6.0 4.9 .. 4.9 .. 1.9 .. 5.9 17. Zaire 3.6 - 0.7 .. 1.2 .. - 1.1. () IS. Guinea 3.5 3.6 .. . .. ... 19. Central African Republic 1.9 3.3 0.8 2.4 5.4 5.1 1.8 3.3 20. Madagascar 2.7 0.3 .. 0.1 .. 1.0 . 0.1 21. Uganida 5.9 - 0.4 0.8 .. -7.9 .. 0.1 22. Lesotho 4.6 7.0 .. 1.8 .. 7.0 . 13.9 23. Togo 8.5 3.6 . 0.3 .. 7.8 . 4.0 24. Sudan 1.3 4.3 . 2.7 .. 3.3 . 6.9 Middle-income oil importers 4.5 w 3.1 zv 3.5 m 3.5 in 5.8 in 25. Kenva 6.0 6.5 .. 5.4 .. 10.2 . 5.8 26. Ghana 2.1 -0.1 .. - 0.2 .. -1.5 .. 1.0 27. Senegal 2.5 2.5 2.9 3.6 4.4 3.5 1.7 1.6 28. Z-imbabwe 4.3 1.6 .. - 0.5 .. 1.8 . 2.1 29. Liber-ia 5.1 1.8 . 5.0 .. - 0.6 .. 1.9 30. Zambia 5.0 1.5 .. 2.3 .. 1.5 . 1.2 31. Cameroon 3.7 5.4 . 3.5 .. 6.5 .. 6.3 32. Swaziland"" 8.6 4.6 7.5 3.7 13.4 3.2 6.0 6.4 33. Botswana** 5.7 13.5" 1.6 8.5" 12.6 16.6" 7.6 14.9" 34. Mauritius"" 1.6 8.2 .. -4.1 .. 11.8 .. 23.7 35. Ivorv Coast 8.0 6.7 4.2 3.4 11.5 10.5 9.7 7.0 Middle-income oil exporters 3.4 w 4.6 iv -0.3 mn 10.6 mn -0.1 mn 36. Angola 4.8 -9.2 4.0 -10.2 11.0 -3.9 4.2 -10.9 37. Congo 2.7 2.9 1.0 0.1 7.0 10.6 2.1 -0.1 38. Nigeria 3.1 7.5 -0.4 -0.3 12.0 11.2 4.9 11.0 39. Gabon" 4.1 7.7 .. . . . Sub-Saharan Africa 3.9 w 2.9 iv .. 1.8 m . 3.3 mn . 4.2 m Sub-Saharan Africa (excluding Nigeria) 4.1 w 1.6 wv - 1.8 m .. 3.2 mn . 4.0 mn All low-income counitries 4.5 w 4.7 zv 2.5 in 2.0 in 6.6 mn 4.2 mn 3.8 mn 4.5 mn All middle-income countries 6.1 w 5.5 wv 3.6 mn 3.0 m 7.4 m 6.5 m 5.5 m 6.0 mn Industrialized countries 5.1 w 3.2 zv 1.3 m 0.9 mn 6.2 mn 3.2 mn 4.8 in 3.4 mn a Figure-s marked with an *are for 1961-70. b Figures marked with an are for 1970~-78. 144 Table 3. Structure of Production GDP millions of Distribution of gross domestic product (percent) current dollars Agriculture Industry Services 1960' 1979b 1960' 1979b 1960' 1979b 1960' 1979b Low-income countries 56 w 44 w 12 w 16 w 31 w 40 w Low-income semiarid 61 w 47w 11 w 20 w 28 w 33 w 1. Chad 180 570 52 70 12 11 36 19 2. Somalia 160 1,030' 67 60* 13 11* 20 29* 3. Mali 270 1,220 55 42 10 11 35 47 4. Upper Volta 200 860 62 38 14 20 24 42 5. Gambia** 20* 132 43* 46 18* 9 40* 46 6. Niger 250 1,710 69 44 9 32 22 24 7. Mauritania 70 470 .. 27 .. 33 .. 40 Low-income other 56 w 43 w 12 w 15 w 32 w 41 w 8. Ethiopia 900 3,530 65 46 12 15 23 39 9. Guinea-Bissau- .. 137 .. 54 .. 9 .. 34 10. Burundi 190 730 .. 55 .. 15 .. 30 il. Malawi 170 1,220 58 43 11 20 31 37 12. Rwanda 120 860 81 42 7 21 12 37 13. Benin 160 850 55 43 8 12 37 45 14. Mozambique 830 2,360 55 44 9 16 36 40 15. Sierra Leone . . 790 . . 36 . . 23 . 41 16. Tanzania 550 4,130 57 54 11 13 32 33 17. Zaire 130 6,020 30 33 27 24 43 43 18. Guinea 370 1,540 .. 41 .. 26 .. 33 19. Central African Republic 110 640 51 37 10 18 39 45 20. Madagascar 540 2,810 37 34 10 20 53 46 21. Uganda 540 8,410 52 55 13 7 35 38 22. Lesotho 30 240 73 36 .. 15 .. 49 23. Togo 120 1,000 55 25 16 23 29 52 24. Sudan 1,470 7,640 58 38 15 13 27 49 Middle-income oil importers 30 w 36 w 30 w 24 w 44 w 39 w 25. Kenya 730 5,280 38 34 18 21 44 45 26. Ghana 1,220 10,160 41 66 .. 21 13 27. Senegal 610 2,480 24 29 17 24 59 47 28. Zimbabwe 780 3,640 18 12 35 39 47 49 29. Liberia 220 940 .. 35 .. 26 .. 39 30. Zambia 680 3,240 11 15 63 41 26 44 31. Cameroon 550 5,330 .. 32 .. 16 . 52 32. Swaziland' 346 243* 31' .. 23* .. 46* 33. Botswana- 38* 410* 54* 21 11* 30 34* 49 34. Mauritius- 143 916 22 25 27 28 51 47 35. Ivory Coast 570 9,130 43 26 14 23 43 51 Middle-income oil exporters 58 w 23 w 12 w 44 w 30 w 33 w 36. Angola 690 2,490 50 48 8 23 42 29 37. Congo 130 1,120 23 13 17 36 60 51 38. Nigeria 3,150 75,170 63 22 11 45 26 33 39. Gabon- 167 2,988 33 6 34 65 34 29 Sub-Saharan Africa 49 w 32 w 16 w 31 w 34 w 37 w AlI low-income countries 51 w 34 w 17 w 36 w 32 w 30 w All middle-income countries 22 w 14 w 30 w 38 w 47 w 48 w Industrialized countries 6 w 4 w 40 w 37 w 54 w 59 w a Figures marked with an are for 1961. b. Figures marked with an ' are for 1978. 145 Table 4. Growth of Consumption and Investment Average annual growth rate (percent) Gross domestic Public consumption Private consumption investment 1960-70 1970-79a 1960-70 1970-79 1960-70 1970-79 Low-income countries 4.7 m 4.5 m 4.1 m 3.2 m 5.4 m 3.1 m Low-income semiarid 4.4 m 6.3 m 2.8 m 3.1 m 4.3 m 6.8 m 1. Chad 4.4 -1.7 -0.7 0.3 2.3 -0.5 2. Somalia 3.7 11.7* -0.5 2.7* 4.3 8.5* 3. Mali 6.2 7.7 2.8 5.5 4.9 3.2 4. Upper Volta . . 3.8 . . 1.1 . . 1.2 5. Gambia** 4.9 6.3 4.9 3.1 6.1 25.5 6. Niger 2.0 3.8 3.9 3.2 3.0 6.8 7. Mauritania .. 18.9 .. 5.0 .. 6.9 Low-income other 4.7 m 2.9 m 4.3 m 3.8 m 7.0 m 2.3 m 8. Ethiopia 4.7 4.5 4.7 4.0 5.7 -1.8 9. Guinea-Bissau**.-.. .. . 10. Burundi 19.2 6.0 3.2 3.1 4.3 16.5 11. Malawi 4.6 6.1 4.1 5.7 15.4 2.3 12. Rwanda 1.1 14.0 4.2 1.6 3.5 18.9 13. Benin 1.7 1.0 4.9 3.8 4.2 8.3 14. Mozambique 6.8 -4.0 4.4 -2.3 8.3 -8.4 15. Sierra Leone . . 4.5 . . 1.5 . . -1.3 16. Tanzania b b 5.2 6.0 9.8 3.0 17. Zaire 8.5 -2.2 3.9 - 1.8 9.6 -5.0 18. Guinea .. .. .. .. . 19. Central African 2.2 1.1 3.0 4.4 1.3 0.3 Republic 20. Madagascar 2.7 0.2 2.0 - 0.6 5.4 -1.8 21. Uganda 5.9 1.3 5.6 1.1 9.8 -13.1 22. Lesotho 0.3 12.0 6.0 10.9 18.5 24.4 23. Togo 6.7 10.7 7.6 5.7 11.1 14.5 24. Sudan 12.1 -3.2 -1.2 7.3 -1.3 8.0 Middle-income oil importers 6.7 m 9.4 m 3.9 m 4.3 m 8.2 m 5.2 m 25. Kenya 10.0 9.0 4.6 6.9 7.0 1.2 26. Ghana 6.1 -0.2 2.0 0.3 -3.2 -7.9 27. Senegal -0.2 b 3.2 3.3 1.1 1.8 28. Zimbabwe . . 9.7* . . 0.4* . . -2.1* 29. Liberia 5.6 2.3 1.7 4.3 -4.2 5.2 30. Zambia 11.0 1.8 6.8 -2.2 10.6 -5.6 31. Cameroon 6.1 5.4 2.7 5.3 9.3 7.9 32. Swaziland"* 7.2 12.0 13.6 -0.7 10.6 13.3 33. Botswana* 10.8 16.9 6.9 10.6 25.3 5.6 34. Mauritius- 2.1 13.5 2.8 9.8 -6.7 16.1 35. Ivory Coast 11.8 10.0 8.0 7.3 12.7 13.8 Middle-income oil exporters 8.8 "I 6.8 m 2.6 m 4.6 m 5.2 m 7.9 m 36. Angola 9.1 3.0 4.0 -7.9 9.7 - 9.0 37. Congo 5.4 5.8 -0.3 2.8 2.9 0.2 38. Nigeria 10.0 12.4 1.1 6.3 7.4 17.8 39. Gabon-* 8.5 7.8 8.8 9.7 -2.1 15.5 Sub-Saharan Africa 6.0 m 5.8 m 4.0 m 3.3 m 5.7 m 3.2 m All low-income countries 4.4 m 4.5 m 3.7 m 3.7 m 5.2 m 6.4 m All middle-income countries 6.3 m 7.4 m 5.1 m 5.2 m 7.4 m 7.0 m Industrialized countries 4.8 m 3.7 m 4.3 m 3.6 m 5.6 m 1.4 m a. Figures marked with an * are for 1970-78. b. Public consumption is included in private consumption. 146 Table 5. Structure of Demand Distribution of gross domestic product (percent) Exports of goods Public Private Gross domestic Gross domestic and nonfactor consumption consumption investment saving services Resource balance 2960 1979, 1960 1979, 1960 1979, 1960 1979, 1960 1979, 1960 1979- Low-income countries 10 w 15 w 81 w 80 w 11 w 15 w 9 w 6 w 16 w 16 w -2w -9w Low-income semiarid 11 w 18 w 84 w 78 w 12 w 23 w 7 w 4 w 13 w 22 w -6w -19 w 1. Chad 13 18 82 96 11 13 5 -14 23 33 -6 -27 2. Somalia 8 19* 89 79* 10 16* 3 2* 11 12* -7 - 14* 3. Mali 12 23 79 82 14 15 9 -5 12 16 -5 -20 4. Upper Volta 10 14 94 89 10 24 -4 -3 9 15 -14 -27 5. Gambia* 20 26 72 83 13 22 8 -9 59 65 -5 -31 6. Niger 9 9 79 72 13 28 12 19 9 25 - 1 -9 7. Mauritania .. 39 .. 47 51 . . 14 38 . -37 Low-income other 10 w 15 w B0 w 80 w 11 w 13 w 10 w 6 w 17 w 15 w -1 w -8w 8. Ethiopia 8 17 81 87 12 10 11 -4 9 10 -1 -14 9. Guinea-Bissau` . ,, 102 . 32 . -2 10 . -34 10. Burundi 3 16 92 80 6 12 5 4 13 13 -1 -8 11. Malawi 16 17 88 70 10 29 -4 13 21 21 -14 -16 12. Rwanda 10 16 82 72 6 19 8 12 12 25 2 -7 13. Benin 16 12 75 87 15 21 9 1 12 27 -6 -20 14. Mozambique 11 15 81 85 10 10 8 (.) 14 13 -2 -10 15. Sierra Leone . 18 78 . 15 . 4 . 24 . . -11 16. Tanzania 9 16 72 76 14 21 19 8 31 14 5 -13 17. Zaire 18 b 61 88 12 9 21 12 55 30 9 3 18. Guinea .. 16 .. 70 15 14 24 -1 19. Central African Republic 19 20 72 72 20 20 9 8 23 18 -11 -12 20. Madagascar 20 17 75 73 11 22 5 10 12 17 -6 -12 21. Uganda 9 b 75 96 11 4 16 4 26 4 5 (.) 22. Lesotho 17 16 108 143 2 29 -25 -59 12 21 -27 -88 23. Togo 8 15 88 74 11 39 4 11 19 32 - 7 - 28 24. Sudan 6 11 85 84 9 14 9 5 12 9 (.) -9 Middle-income oil importers 11 w 15 w 67 w 68 w 22 w 19 w 21 w 16 w 37 w 27 w 0 w -4w 25. Kenva 11 20 72 65 20 22 17 15 31 26 -3 -7 26. Ghana 10 9 73 .86 24 5 17 5 28 12 -7 (.) 27. Senegal 17 b 68 98 16 21 15 2 40 34 -1 -19 28. Zimbabwe 11 13* 67 63* 23 15* 22 24* . -1 P 29. Liberia 7 15 58 62 28 27 35 23 39 53 7 - 4 30. Zambia 11 27 48 45 25 21 41 28 56 45 16 7 31. Cameroon .. 10 .. 80 .. 25 .. 10 .. 25 -15 32. Swaziland- 18 22 54 64 13 28 29 14 47 77 16 -13 33. Botswana** 15 25 88 63 8 43 -3 12 23 47 -12 -31 34. Mauritius" 15 14 79 61 30 38 6 25 32 50 -24 -13 35. ivory Coast 10 17 73 56 15 31 17 27 37 35 2 -4 Middle-income oil exporters 7 w 11 w 84 w 57 w 15 w 30 w 10 w 31 w 17 w 26 w -6w 1 w 36. Angola 9 26 77 56 12 9 14 18 20 43 2 9 37. Congo 23 30 98 58 45 22 -21 12 21 -66 -10 38. Nigeria 6 10 87 58 13 31 7 32 15 25 -6 1 39. Gabon- 10 12 40 36 50 30 50 53 32 .. 1 Sub-Saharan Africa 10 w 13 w 77 w 65 w 15 w 23 w 13 w 20 w 22 w 23 w -3w -3w All low-income countries 9 w 11 w 78 w 66 w 18 w 26 w 16 w 23 w 7 w 11 w -2w -3w All middle-income countries 11 w 13 w 70 w 62 w 21 w 26 w 19 w 25 w 16 w 20 w -2w -1 w Industrialized countries 15 w 17 w 63 w 61 w 21 w 23 w 22 w 22 w 12 w 19 w 1 w - 1 w a. Figures marked with an * are for 1978. b. Public consumption is included in private consumption. 147 Table 6., Commercial Energy Petroleum imports as a percentage of Energy imports as total Averageannualgrowth ate (prcent) Energy consumption a percentage of commercial Averae anual rowthrate(peren per ca pita (kilograms merchandise energy Energy production Energy consumptioni of coal equivalent) exports consumption 1960-74, 7974-79 1960-74 1974-79 1960 1979 1960b 7978c 1978 Low-income countries 4.5 w 21.2 w 8.8 w 0.1 w 46 w 70 w 7 w . w 83 w Low-incomne semiarid ..w ..w 12.2 w 8.5 w 10 w 48 w 12 w ..w 90 w 1.Chad . ..7.5 4.6 8 24 23 ..62 2. Somalia . .. 8.7 13.0 17 78 4 ..96 3. Mali .. 8.3 5.6 5.3 15 30 13 ..92 4. Upper Volta . .7.7 10.2 5 29 38 ..94 5. Gam bia.. ........ .... 6. Niger .. .. 14.8 12.8 6 48 6 . 92 7. Mauritania .. .. 21.2 5.5 18 185 39 ..92 Low-income other 4.5 w 21.2 w 8.6 w -0.5 w 53 w 74 w 6 w .. w 82 w 8. Ethiopia 14.1 2.3 14.0 -5.3 9 20 11 20 91 10. Burundi .. 22.0 .. 6.9 .. 17 .. . 11. Malawi .. 6.9 .. 5.7 .. 70 . 259 1 2. Rwanda .. 3.5 .. 10.4 . 30 .. . 13. Benin . .. 9.5 -0.6 40 68 16 ..91 14. Mozambique 3.2 60.0 5.2 1.1 113 139 1 1 ..46 15. Sier-ra Leone . . 9.0 - 1.1 31 89 11 . 96 16. Tanzania 10.6 10.4 9.4 -2.9 43 53 .. . 90 17. Zaire 3.0 18.1 3.8 0.4 98 103 3 18. Guinea 16.0* (. 3.2 1.6 67 87 7 ..66 19. Central African Republic 14.1 4.1 7.6 8.5 38 55 12 1 87 20. Madagascar 6.7 4.1 9.0 3.9 40 94 9 16 89 21. Uganda 5.2 - 4.4 9.1 -8.2 43 39 . 22". Lesotho . ... . 2.3. Togo .. 22.3 12.7 11.8 23 117 10 13* 91 24. Sudan .. 13.7 13.1 - 0.9 54 141 8 24' 94 Middle-income oil i.mporters 3.4 w 2.4 w 6.9 w 3.0 w 249 w 342 w 8 w .w 56 w 215. Kenva 9.6 17.6 3.3 3.5 150 180 18 30 95 26. Ghana . 2.6 12.2 2.3 105 265 7 19* 66 27. Senegal . 4.7 12.4 110 266 8 . 96 28. Zimbabwe 2.5 - 3.1 2.4 -0.3 1,346 791 .. ..2 29. Liberia 31.8 -1.3 18.9 -0.9 88 448 3 17 92 30. Zambia . 5.1 .. 5.2 .. 858 .. Ii' 40 31. Cameroon 1.1 45.3 6.2 7.8 87 148 7 9 76 32. Swaziland . 33. Botswana . ... ... .... . 34. Mauritius.. . . .. .... . 35. Ivorv Coast 9.7 -12.2 14.3 5.5 75 234 5 10 95 Middle-income oil exporters 36.1 w 0.8 w 9.5 w 1.5 w 36 w 95 w 7 w ..w ..W- 36. Angola 35.5 - 2.4 10.3 1.1 90 208 6 37. Congo 15.8 5.1 5.3 7.0 125 213 25 1 38. Nigeria 36.6 1.0 9.4 1.4 29 83 7 2 39. Gabon ... .. . ... Sub-Saharan Africa 34.7 w 1.2 W 8.2 w 1.7 w 76 w 128 w 7 w ..w 65 w All low-incomne countries 5.2 w 8.4 w 4.4 w 8.1 W 356 w 463 w 8 W .W W All middle-income countries 12.7w -0.5w 8.4 w 6.3 w 509 w 1,225 w lO W 20 w W Industrialized countries 4.1 w 2.3 w 5.3 w 2.5 w 4,486 w 7,892 w 11 W 20 w ..W a Figures marked with an are for 1961-74. b Figures marked with an are for 1961. c. Figures marked with an are for 1977. 148 Table 7. Growth of Merchandise Trade Merchandise trade' (millions of dollars) Average annual growth rate in volume (percent) Exports Imports Exports Imports 1979 1979 1960-70 1970-79 1960-70 1970-79 Low-income countries 5,749 t 7,613 t 5.3 m -1.9 m 6.0 m 4.2 m Low-income semiarid 516 t 980 t 5.9 m 4.4 m 4.8 m 5.5 m 1 Chad .. .. 5.9 -34 5.0 -0.1 2. Somalia 11l 287 2.3 5.6 2.6 7.7 3. Mali 177 180 3.0 6.7 - 0.4 5.5 4. Upper Volta 81 254 15.9 3.1 8.5 5.2 5. Gambia .. .. .. .. . 6. Niger .. .. 6.0 11.7 11.9 6.5 7. Mauritania 147 259 50.7 - 1.1 4.5 5.5 Low-income other 5,233 t 6,633 t 5.0 m -2.7 m 6.2 m -0.5 m 8 Ethiopia 418 567 3.6 -2.7 6.2 0.4 9. Guinea-Bissau' 14 61 .. .. . 10. Burundi 105 152 .. .. . ILI Malawi 233 399 11.6 4.6 7.6 4.3 12. Rwanda 115 190 15.8 1.6 8.1 10.5 13. Benin 190 357 5.0 -11.4 7.4 6.3 14. Mozambique . . . . 6.0 -16.6 7.9 -14.4 15. Sierra Leone 205 297 0.3 -6.5 1.9 -3.0 l6. Tanzania 523 1,084 3.4 -6.6 6.0 -0.5 17. Zaire 1,324 597 - 1.8 - 1.1 5.4 -11.9 18 Guinea 373 347 .. .. . 19 Central African Republic 80 70 8.1 -0.5 4.5 -5.0 20. Madagascar 394 641 5.3 - 1.0 4.1 -1.7 21. Uganda 427 230 5.0 -7.0 6.2 -10.5 2 Lesotho .. .. .. .. . 23. Togo 251 441 10.5 - 2.5 8.6 9.8 24. Sudan 581 1,200 0.1 -4.4 1.2 4.5 Middle-income oil importers 9,341 t 9,351 t 7.1 m -0.5 m 6.6 m 2.3 m 25. Kenva 1,104 1,658 7.2 -0.5 6.6 -1.0 2b. Ghana 1,096* 993* 0.2 - 7.2 -1.5 0.1 27,. Senegal 421* 756* 1.2 -0.8 2.3 4.5 2S Zimbabwe 1,194 940 . . .. . 29 Libnria 506 487 18.4 2.3 2.9 2.3 30. Zambia 1,377 755 2.2 -0.7 9.7 -8.1 31. Cameroon 1,129 1,271 7.1 0.5 9.2 7.0 32. Swaziland .. .. .. .. . 33. Botswana .. .. .. .. . 34. Mauritius .. .. .. .. . 35. Ivorv Coast 2,515 2,491 8.8 5.2 9.7 10.1 Middle-income oil exporters 18,192 t 12,641 t 6.6 m -0.1 m 1.6 m 3.3 m 36. Angola .. .. 9.0 -7.9 11.5 -4.2 37. Congo 119' 242* 5.1 8.2 -1.0 3.3 38 Nigeria 18,073 12,399 6.6 -0.3 1.6 20.6 39. Gabon .. .. .. .. . Sub-Saharan Africa 33,282 t 29,605 t 5.9 m -0.8 m 6.0 m 3.3 m Low-income countries 47,199 t 49,699 t 5.2 m -1.0 m 5.0 m 3.3 m Middle-income countries 272,496 t 304,708 t 5.4 m 4.3 m 6.6 m 5.0 m Industrialized countries 1,028,279 t 1,106,543 t 8.4 m 5.9 m 9.3 m 4.5 m a Figures marked with an ' are for 1978. 149 Table 8. Structure of Merchandise Exports Percentage share of merchandise exports Minerals and Other primary Fuels - metals Food and beverages products Manufactures 1962 1978 1962 1978 1962 1978 1962 1978 1962 1978 Low-income countries (.) w 2 w 10 w 21 w 55 w 57 w 27 w 16 w 9 w 6 w Low-income semiarid I1w w (.)w 34 w 75 w 40 w 21 w 23 w 3 w 3 w 1. Chad 4 0 1 11 17 15 75 81 3 4 2. Somalia 0 0 0 0 87 96 11 3 2 1 3. Mali 0 0 1 0 81 47 16 52 2 1 4. Upper Volta 0 0 4 () 85 49 4 46 7 5 5. Gambia- 0 . 100 0 .0 6. Niger 1 0 0 40 92 21 3 4 4 35 7. Mauritania 6 0 0 87 32 9 6 0 55 4 Low-income other (.)w 2 w lO w 19 w 53 w 58 w 27 w 15 w lO w 6 w 8. Ethiopia 0 4 1 0 79 85 18 10 2 1 9. Guinea-Bissau" 0 0 0 0 96 96 4 3 .. 1 10. Burundi .. 0 .. 8 .. 89 .. 2 I 11.Malawi .. 0 .. 0 . 95 . 1 .. 4 12. Rwanda . 0 .. 10 .. 87 .. 3.. C) 13. Benin 0 2 0 4 94 40 3 45 4 9 14. Mozambique .. 10 .. 2 .. 64 .. 21 .. 3 15. Sierra Leone .. 0 .. 8 . 47 I. 1. 44 16. Tanzania 2 2 1 2 34 65 49 25 14 6 17. Zaire 0 1 16 71 49 18 26 2 10 8 18. Guinea 0** 0 70` 98 29" 2** 0** 0* . 0 19. Central Afr-icant Rep. 2 0 0 0 36 32 41 30 21 38 20. Madagascar 0 2 4 6 82 83 9 2 5 7 21. Uganda .. 0 .. 1 .. 92 .. 7.. () 22. Lesotho . 6 .. 26 .. 23 . 9 .. 36 23. Togo 0 0 12 49 75 43 6 2 6 6 24. Sudan .. 5 .. () .. 44 .. 51. () Middle-income oil importers . 5 w . 19 w . 59 w . 11 w . 7 w 25. Kenya .. 19 . .0 60 .. 9 .. 13 26. Ghana 0 2 13 14 71 73 12 6 4 5 27. Senegal () 1 6 12 90 72 1 8 4 7 29. Liberia . 0 .. 63 .. 6 . 29 .. 2 30. Zambia . () .. 94 I. 1 . 0 .. 4 31. Cameroon .) 3 21 3 59 74 16 17 4 4 32. Swaziland .. .. . 33. Botswania .. .. . . .. 34. M3Urit1US- . 0 .. 0 .. 100 .. 0 0 35. Ivory Coast 0 4 2 0 76 74 22 15 1 7 Middle-iincome oil exporters 11 w 90 w 2 w 1 w 59 w 6 w 21 w 3 w 9 w I1w 36. Angola .. 62 .. 2 .. 23 .. 6 .. 7 37. Congo 7 54 0 6 44 10 40 15 9 15 38. Nigeria 10 91 0 () 65 6 17 2 8 1 39. Gabon" 18 81 18 9 3 1 48 7 14 2 Sub-Saharan Africa 3 w 49 w 7 w lO w 62 w 31 w 21 w 9 w 7 w 4 w a. Includes fuels. b. Indudes food and beverages. 150 Table 9. Structure of Merchandise Imports Percentage share of merchandise imports Machinery and Other primary transport Food Fuels commodities equipment Other manufactures 1960a 1978, 1960* 1978, 1960* 1978, 1960* 1978, 1960* 1978k Low-income countries 17 w 15 w 8 w 12 w 4 w 3 w 19 w 33 w 52 w 37 w Low-income semiarid 23 w 22 w 7 w 10 w 2 w 3 w 19 w 32 w 49 w 35 w 1. Chad 19 15.. 12 I4` 4 2-~ 19 28`~ 46 40** 2. Somalia 27* 25** 4* 7`* 0* 6** 18* 28** 51* 34`* 3. Mali 20* 19** 5* 14** 4* 2** 18* 30** 53* 35** 4. Upper Volta 21* 19 4* 9 1* 0 24* 43 50* 29 5. Gambia- 24* 9 3 14 50 6. Niger 24* 10- 5* 12** 4* 2** W3 33-* 49* 43** 7. Mauriitania 5* 3* 3* 39* 50* Low-income other 16 w 14 w 8 w 13 w 4 w 2 w 19 w 33 w 53 w 37 w S.Ethiopia 6 12 4 35 43 9. Guinea-Bissau" 43 5 1 29 22 10. Burundi 23* 11* 8. 27* 31* It.M1alawi 5 12 2 .. 37 44 12. Rivanda. 13. Benin 17 15 10 15 1 2 18 2254 46 14. Mozarnbique. 15. Sier-ra Leone 23 21 12 12 5 1 15 24 45 42 16. Tanzania" 11 19 4 33 33 17 Zaire .. 17 .. 18 . 38 27 18. Guinea 19 Central African Rep. 15 17 9 2 2 2 26 38 48 41 20 Madagascar 17 17 6 14 3 3 23 31 51 35 21. LUganda 6- 3*- 8 30-~ 8* 2- 25' 27*' 53* 34** 22". Lesotho 23. Togo 16 8 6 14 3 4 32 37 43 37 24. Sudan 17 19 8 1 3 2 14 36 58 42 Middle-income oil importers 19w 11 w 6w 15 w 4 w 5 w 26 w 37 w 44 w 32 w 25. Kernva 12* 7 11* 18 8* 3 27* 41 42* 31 26. Ghana 19 9* 5 16* 4 5* 26 26* 46 44' 27. Senegal 30 23 5 12 2 21 19 18 44 26 2,1. Zimbabwe .. 2 30 .. 5 34 29 29.Liberia 16 17 4 18 7 1 34 32 39 32 30. Zambia 6* 16* 3* 71* 4* 31.Cameroon 20 10 8 7 3 2 17 39 52 42 32. Swaziland 33. Botswana 34. Mauritius- 24* 9* 3* 24* 40* 3 5. Ivorv Coast 18 13 6 10 2 2 27 39 47 36 Middle-income oil exporters 14 w 14 w 5 w 2 w 5 w 2 w 25 w 43 w 50 w 39 w 36. Angola 37. Congo 18 21' 6 1* 1 1* 31 32* 44 45' 38. Nigeria 14 14 5 2 6 2 24 44 51 38 39. Gabon- 12 ()2 39 46 Sub-Saharan Afiica 17 w 13 w 6 w 8 w 5 w 3 w 24 w 39 w 48 w 36 w AllIlow-income countries 22 w 17 w 6 w 11 w 16 w 20 w 25 w 24 w 31 w 28 w All middle-income countries 15 w 12 w 9 w 17 w 13 w 8 w 28 w 32 w 35 w 31 w Industrialized countries 22 w 13 w 11 w 19 w 24 w lO w 16 w 25 w 27 w 33 w a Figure-s marked -ith an are for years other than specified. :151 Table 10. Principal Merchandise Exports of Sub-Saharan Africa 1976-78 average Principal exporter Percentage share Percentage share of Value (millions of Sub-Saharan exports from Sub- of dollars) African exports Country Saharan Africa Fuels Petroleum 11,502 43.5 Nigeria 95.5 Mineral and metals Copper 1,589 6.0 Zambia 54.0 Iron ore 432 1.6 Liberia 67.9 Bauxite 188 0.7 Guinea 95.1 Phosphate rock 140 0.5 Togo 58.6 Manganese ore 120 0.5 Gabon 85.7 Zinc 79 0.3 Zaire 57.7 Tin 59 0.2 Nigeria 43.8 Lead 20 0.1 Namibia 63.9 Food and beverages Coffee 2,838 10.7 Ivory Coast 22.5 Cocoa 1,882 7.1 Ghana 34.5 Sugar 432 1.6 Mauritius 48.5 Tea 245 0.9 Kenya 57.7 Groundnuts 194 0.7 Sudan 52.1 Groundnut oil 177 0.7 Senegal 80.8 Beef 78 0.3 Botswana 43.4 Palmoil 58 0.8 Ivory Coast 71.4 Bananas 45 0.2 Ivory Coast 32.8 Maize 30 Nonfood Timber 680 2.6 Ivory Coast 47.4 Cotton 651 2.5 Sudan 46.0 Tobacco 290 1.1 Zimbabwe 47.8 Rubber 128 0.5 Liberia 47.5 Sisal 50 0.2 Tanzania 52.8 All other exports 4,553 17.2 Total exports of Sub-Saharan Africa 26,458 100.0 Nigeria 41.5 152 Table 11. Merchandise Exports by Degree of Processing Percentage share of exports by degree of processing 1977, Other developing Sub-Saharan Africa countries Food products: Coffee Green, roasted 100* 95* Essence, extracts 5* Cocoa Beans, raw, roasted 84 62 Powder and paste 15 29 Chocolate and products 1 9 Tobacco Unmanufactured 94 85 Manufactured 6 15 Groundnuts Green 24 53 Oil 76* 47 Nonfood agricultural products: Leather Hides and skins 77* 18* Leather 22* 76* Leather manufactures 1* 6* Wood Rough logs 77 47 Shaped wood 15 25 Veneers, plywood 7 22 Manufactures 1 6 Cotton Raw 85 53 Grey yam in bulk 1 18 Woven fabrics 14 29 Nonfuel minerals: Copper Ores, excluding matte 3* 21* Unrefined 11* 35* Refined 85* 42* Bars, wires ()* 2* Tubes, pipes ()( Iron and Steel Ore, concentrate 98 56 Pig iron (.) 17 Ingots, primary form 0 6 Worked in various forms 2 21 Aluminum Bauxite 35* 32* Oxide, hydroxide 13 48 Unwrought 50 11* Bars, wire 0 7 Plate, sheet, strip 3 2 Phosphate Natural 98 88 Chemical fertilizer 2 12 a. Figures followed by an * are for years other than 1977. 153 Table 12. Destination of Merchandise Exports Percenfage share of merchandise exports Sub-Saharan Centrally Industralized African Other developing planned Capital-surplus market economies countries countries economies oil exporters 1960 1979 1960 1979 1960 1979 1960 1979 1960 1979 Low-income countries 76 w 64 w 6 w 8 w 16 w 22 w 1 w 3 w 1 w 4 w Low-income semiarid 76 w 71 w 18 w 6 w 6 w 11 w 0 w 0 w O w 13 w 1. Chad 73 30 27 13 0 52 0 .. 0 5 2. Somalia 85 18 0 1 15 2 0 (.) (.) 80 3. Mali 93 68 7 15 0 17 0 (.) (.) (.) 4. Upper Volta 4 75 96 9 0 16 0 0 0 5. Gambia 97 93 3 1 0 6 0 (.) 0 0 6. Niger 74 97 26 1 0 0 0 .. 0 2 7. Mauritania 89 88 11 2 0 9 0 .. 0 1 Low-income other 76 w 63 w 5 w 8 w 17 w 23 w 0 w 3 w 1 w 3 w 8. Ethiopia 69 72 4 (.) 20 11 1 7 6 10 9. Guinea-Bissau- .. 29 32 22 .. 38 .. 1 .. 0 10. Burundi - 89 . . 1 . . 9 . . 1 . . 0 11. Malawi .. 84 .. 12 .. 4 .. 1 12. Rwanda .. 80 . 4 .. 16 . 1 . 13. Benin 90 89 8 2 0 8 2 1 0 (.) 14. Mozambique 29 43 5 4 66 45 (.) 1 (.) 7 15 Sierra Leone 99 98 1 1 0 1 0 .. 0 (.) 16. Tanzania 74 57 4 4 21 36 1 2 0 1 17. Zaire 89 64 5 26 6 10 (.) (.) (.) (.) 18. Guinea 63 69 10 3 9 26 18 .. (.) 2 19. Central African Rep. 83 78 9 2 8 20 0 (.) 0 (.) 20. Madagascar 79 67 18 4 2 29 1 (.) (.) (.) 21. Uganda 62 67 7 3 31 27 0 1 0 2 22. Lesotho .. .. .. .. .. .. .. . . 23. Togo 74 67 26 8 0 17 0 8 0 24. Sudan 59 36 2 (-? 27 45 8 9 4 10 Middle-income oil importers 88w 75w 3w 9w 7w 13w 3w 3w Ow Ow 25. Kenya 77 63 7 21 15 15 0 (.) (.) 1 26. Ghana 88 70 2 2 3 15 7 13 (.) (.) 27. Senegal 89 59 4 27 7 14 0 (.) 0 (.) 28. Zimbabwe .. .. .. .. .. .. . .. . 29. Liberia 100 86 0 (.) (.) 14 0 (.) 0 (.) 30. Zambia .. 82 . 2 .. 16 .. (.) .. (.) 31. Cameroon 93 84 3 6 3 8 1 2 (.) (.) 32. Swaziland . .. .. . .. .. . . 33. Botswana .. . .. .. .. .. .. .. . 34. Mauritius" 97 95 2 4 1 1 0 0 0 0 35. Ivory Coast 84 78 3 6 13 11 0 5 0 (.) Middle-income oil exporters 89 w 84 w 2 w 3 w 8 w 13 w 1 w O w O w (.)w 36. Angola 64 33 7 (.) 27 66 2 0 0 1 37. Congo 93 72 (.) 1 7 27 0 (.) 0 (.) 38. Nigeria 95 87 1 2 3 11 1 (.) 0 (.) 39. Gabon- 87 60 6 8 7 32 0 0 0 (.) Sub-Saharan Africa 82 w 78 w 4 w 5 w 12 w 15 w 2 w 2 w (.) w 1 w All low-income countries 51 w 61 w 29 w, 29 w, 19 w 5 w 1 w 5 w All middle-income countries 68 w 67 w 24 w' 26 w 8 w 4 w (.)w 3 w Industrialized countries 67 w 69 w 30 w' 24 w, 3 w 3 w (.) w 4 w a. Includes exports destined for Sub-Saharan Africa. 154 Table 13. Terms of Trade Average annual growth rate Net barter terms of trade Income terms of trade Net barter terms Income terms (1975 = 100) (1975 = 100) of trade of trade 1960 1970 1979 1960 1970 1979 1961-70 1970-79 1962-70 1970-79 Low-income countries 110 m 117 m 97 m 62 m 116 m 105 m 0.9 m 0.8 m 5.1 m -0.8 m Low-income semiarid 104 m 117 m 94 m 38 m 78 m 82 m -0.1 m -1.3 m 4.8 m 3.0 m 1. Chad 98 93 100 62 116 78 - 1.0 1.6 2.9 - 1.7 2. Somalia 145 135 97 59 71 82 -1.6 -2.7 0.5 2.2 3. Mali 107 117 95 55 122 148 7.0 -0.6 4.5 5.7 4. Upper Volta 88 117 94 20 78 105 3.5 -1.3 15.6 6.2 5. Gambia 104 109 93 38 73 81 -0.1 -1.3 5.4 3.0 6. Niger 98 109 90 32 68 150 0.5 -2.2 4.8 9.0 7. Mauritania 149 133 78 3 102 77 -2.1 -5.2 38.5 -6.1 Low-income other 114 m 118 m 104 m 89 m 140 m 108 m 0.8 m 0.7 m 5.5 m -1.3 m 8. Ethiopia 143 151 142 75 112 113 0.5 2.4 3.3 - 0.7 9. Guinea-Bissau' . . .. 111 .. .. 119 .. .. . 10. Burundi .. .. .. .. .. .. -0.2 -0.9 4.7 -19.0 11. Malawi 115 99 84 40 83 112 -3.0 -0.5 7.9 3.5 12. Rwanda 111 125 145 21 115 140 0.8 6.3 15.1 7.2 13. Benin 114 129 97 127 208 84 1.2 -2.8 8.7 -14.5 14. Mozambique 90 88 75 89 167 32 .. .. . 15. Sierra Leone 121 136 108 146 158 75 1.5 -1.6 2.4 -6.6 16. Tanzania 98 103 102 118 152 104 0.2 0.7 3.8 -4.4 17. Zaire 122 200 91 137 192 90 7.7 -7.8 7.2 -8.7 18. Guinea.. . .. . .. ... .... 19. Central African Rep. 109 118 108 64 124 113 0.9 1.3 8.0 1.1 20. N5adagascar 136 115 105 64 111 89 -0.3 -0.9 4.1 -1.3 21. Uganda 123 130 136 119 211 158 1.3 3.1 5.5 -0.8 22. Lesotho .. .. .. .. .. .. .. .. . 23. Togo 56 59 82 27 89 110 2.2 9.0 11.3 6.0 24. Sudan 83 83 78 96 140 105 0.8 1.4 2.1 -2.8 Middle-income oil importers 112 m 120 m 105 m 60 m 119 m 104 m 1.6 m 2.6 m 6.4 m 0.6 m 25. Kenva 133 119 110 67 120 113 -1.6 2.2 4.8 0.9 26. Ghana 111 121 144 90 125 144 2.3 6.9 1.5 -0.8 27. Senegal 71 79 76 52 71 46 1.3 1.4 1.3 0.2 28 Zimbabwe .. .. .. .. .. .. .. . . 29. Liberia 255 131 88 52 118 94 -4.7 -4.1 12.6 -2.2 30. Zambia 115 227 100 99 238 91 10.0 -9.0 12.3 -9.7 31. Cameroon 106 117 144 90 125 144 1.9 6.1 8.0 6.7 32. Swaziland .. .. .. .. .. .. .. .. . 33. Botswana .. .. .. .. .. .. .. .. . 34. Mauritius 57 47 49 30 48 89 -3.3 3.7 -1.6 8.6 35. Ivorv Coast 113 127 129 31 85 138 2.0 3.0 9.2 8.2 Middle-income oil exporters 54 m 55 m 109 m 18 m 32 m 113 m 1.0 m 12.9 m 7.0 m 12.8 m 36 Angola 60 68 113 30 93 102 1.7 8.5 10.2 0.3 37. Congo 87 89 91 23 34 76 0.1 1.0 2.5 8.6 38. Nigeria 32 32 119 13 30 143 1.0 17.2 7.0 16.9 39. Gabon 47 42 105 11 25 124 1.0 17.2 7.0 17.0 Sub-Saharan Africa 108 m 117 m 100 m 57 m 114 m 105 m 0.9 m 1.2 m 5.5 m 0.6 m 155 Table 14. Exports: Commodity Concentration and Fluctuation in Values Percentage share in total exports of three principal exports 1961 1976-78 Index of export fluctuations- (average) 1955-65 1965-79 Low-income countries 60.55 m 79.6 m 12.5 m 12.85 m Low-income semiarid 60.5 m 79.7 m 17.6 m 19.3 m 1. Chad 90.5 82.1 15.4 12.6 2. Somalia 60.3 90.6* 10.7 13.1 3. Mali 60.5 55.6b 13.9 14.6 4. Upper Voltac 14.3 43.6 27.3 19.7 5. Gambia 95.7 79.7 17.6 19.3 6. Niger 77.2 79.5d 73.2 59.3 7. Mauritania 28.5 86.1 73.2 59.3 Low-income other 60.6 m 79.1 m 8.6 m 9.3 m 8. Ethiopia 63.8 81.5 7.9 8.2 9. Guinea-Bissaw 78.9 76.6 17.3 15.4 10. Burundi 99.1 95.2 6.6 8.2 11. Malawi 93.8 83.1 7.4 8.2 12. Rwanda 91.9 86.5 16.0 18.0 13. Benin 36.4 32.0 10.1 20.6 14. Mozambique 46.9 25.6 5.1 6.8 15. Sierra Leone 19.6 79.1k 14.2 13.5 16. Tanzania 49.2 55.4 7.2 7.6 17. Zaire 52.0 91.1 8.6 18.1 18. Guineas 27.6 72.9 17.1 15.9 19. Central African Republic 74.8 54.0 16.9 0.2 20. Madagascar 43.3 48.1 8.1 8.3 21. Uganda 69.8 96.0 8.2 7.1 22. Lesotho 72.6 100.0 6.5 9.9 23. Togo 60.6 81.4 18.3 10.1 24. Sudan 60.3 70.6 11.1 9.3 Middle-income oil importers 79.9 m 62.8 m 10.5 m 9.4 m 25. Kenya 34.2 52.5 7.2 11.5 26. Ghana 84.5 62.8 5.5 6.3 27. Senegal 79.9 49.7 8.9 7.4 28. Zimbabwe 58.8 21.9 14.9 20.6 29. Liberia 84.3 82.1 13.3 10.4 30. Zambia 93.9 96.2 14.6 12.5 31. Cameroon 59.5 62.8 10.8 9.2 32. Swaziland 22.3 58.9 6.3 8.3 33. Botswana 20.0 99.1 6.3 9.9 34. Mauritius 91.6 71.9 12.0 9.4 35. Ivory Coast 81.2 68.1 10.5 8.1 Middle-income oil exporters 59.9 m 85.1 m 76.7 m 83.05 m 36. Angola 50.0 60.9 55.1 54.5 37. Congo 76.4 82.4 93.0 81.8 38. Nigeria 45.6 97.1 77.5 84.3 39. Gabon 69.8 87.8 75.9 87.7 Sub-Saharan Africa 60.6 m 79.1 m 12.0 m 11.5 m a. See technical notes. b. World Bank estimate = 86.4 percent. c. World Bank estimates = 76.4 percent for 1961 and 91.3 percent for 1976-78. d. Includes uranium = 86.8 percent (1976). See technical notes. e. World Bank estimates = 84.0 percent for 1961 and 91.3 percent for 1976-78. The latter estimate includes fishing products. f. Indudes diamonds = 69.7 percent. See technical notes. g. World Bank estimates = 76.4 percent for 1961 and 96.7 percent for 1976-78. Both estimates include alumina. 156 Table 15. Commodity Trade: Volume and Prices Volume Price Annual average growth rate (percent) Annual Annual average grozwth Sub-Saharan Africa World average rate (percent) change 1960-70 1970-79 1960-70 1970-79 1955-78 1960-70 1970-80 Fuels Petroleum 42.1 7.1 11.1 3.5 14.1 -2.2 18.2 Minerals and metals Copper 2.3 -0.3 2.9 2.5 18.6 8.8 -18.7 Iron ore 25.6 -3.9 7.7 1.7 8.5 -3.9 -13.0 Bauxite 12.5 25.0 5.8 3.0 8.3 5.6 2.2 Phosphate rock 20.2 6.7 7.7 2.7 17.6 -2.0 1.5 Manganese ore 14.1 0.2 5.8 0.0 10.3 -5.7 -1.1 Zinc -0.3 0.2 5.5 3.2 17.9 1.5 -3.0 Tin 5.0 -11.4 0.8 0.9 10.2 4.0 4.6 Lead - 0.3 - 6.4 3.2 0.6 15.8 3.0 0.3 Food and beverages Coffee 4.4 0.4 2.3 0.3 18.1 0.8 3.9 Cocoa 0.8 - 1.1 0.5 -2.2 26.5 3.5 7.5 Sugar 4.4 -2.7 1.2 2.8 37.4 -3.7 -1.3 Tea 9.3 5.0 2.3 2.2 9.7 -4.0 -2.8 Groundnuts -5.5 -8.4 -1.6 -1.9 10.3 0.1 -3.5 Groundnut oil 4.4 -3.5 3.2 1.9 12.6 -0.1 -3.1 Beef 5.2 4.5 5.4 5.4 12.3 4.8 - 1.8 Palmoil -7.7 0.2 4.1 10.2 14.0 -1.7 -2.4 Bananas -1.2 - 1.5 4.9 1.3 7.3 2.3 -3.1 Maize 6.5 8.3 7.0 11.0 9.9 0.9 -0.5 Nonfood Timber 4.4 - 0.4 5.9 3.0 13.7 1.0 0.4 Cotton 6.6 -5.0 0.8 0.3 10.4 0.1 -2.2 Tobacco - 2.7 7.2 1.7 3.5 10.6 0.7 - 1.0 Rubber 3.6 -0.5 2.9 1.9 16.5 -6.0 1.3 Hides and skins 0.1 -2.5 3.7 3.5 . . 0.3 -0.4 Sisal 0.3 -11.1 -1.7 -11.6 .. -7.1 1.5 157 Table 14. Exports: Commodity Concentration and Fluctuation in Values Percentage share in total exports of three principal exports I 1961 1976-78 Index of export fluctuations' (average) 195545 1965-79 Low-income countries 60.55 m 79.6 m 12.5 m 12.85 m Low-income semiarid 60.5 m 79.7 m 17.6 m 19.3 m 1. Chad 90.5 82.1 15.4 12.6 2. Somalia 60.3 90.6'* 10.7 13.1 3. Mali 60.5 55.6b 13.9 14.6 4. Upper Volta, 14.3 43.6 27.3 19.7 5. Gambia 95.7 79.7 17.6 19.3 6. Niger 77.2 79.5d 73.2 59.3 7. Mauritania 28.5 86.1 73.2 59.3 Low-income other 60.6 m 79.1 m 8.6 m 9.3 m 8. Ethiopia 63.8 81.5 7.9 8.2 9. Guinea-Bissau, 78.9 76.6 17.3 15.4 10. Burundi 99.1 95.2 6.6 8.2 11. Malawi 93.8 83.1 7.4 8.2 12. Rwanda 91.9 86.5 16.0 18.0 13. Benin 36.4 32.0 10.1 20.6 14. Mozambique 46.9 25.6 5.1 6.8 15. Sierra Leone 19.6 79.1f 14.2 13.5 16. Tanzania 49.2 55.4 7.2 7.6 17. Zaire 52.0 91.1 8.6 18.1 18. Guineas 27.6 72.9 17.1 15.9 19. Central African Republic 74.8 54.0 16.9 0.2 20. Madagascar 43.3 48.1 8.1 8.3 21. Uganda 69.8 96.0 8.2 7.1 22. Lesotho 72.6 100.0 6.5 9.9 23. Togo 60.6 81.4 18.3 10.1 24. Sudan 60.3 70.6 11.1 9.3 Middle-income oil importers 79.9 m 62.8 m 10.5 m 9.4 m 25. Kenya 34.2 52.5 7.2 11.5 26. Ghana 84.5 62.8 5.5 6.3 27. Senegal 79.9 49.7 8.9 7.4 28. Zimbabwe 58.8 21.9 14.9 20.6 29. Liberia 84.3 82.1 13.3 10.4 30. Zambia 93.9 96.2 14.6 12.5 31. Cameroon 59.5 62.8 10.8 9.2 32. Swaziland 22.3 58.9 6.3 8.3 33. Botswana 20.0 99.1 6.3 9.9 34. Mauritius 91.6 71.9 12.0 9.4 35. Ivory Coast 81.2 68.1 10.5 8.1 Middle-income oil exporters 59.9 m 85.1 m 76.7 m 83.05 m 36. Angola 50.0 60.9 55.1 54.5 37. Congo 76.4 82.4 93.0 81.8 38. Nigeria 45.6 97.1 77.5 84.3 39. Gabon 69.8 87.8 75.9 87.7 Sub-Saharan Africa 60.6 m 79.1 m 12.0 m 11.5 m a. See technical notes. b. World Bank estimate = 86.4 percent. c. World Bank estimates = 76.4 percent for 1961 and 91.3 percent for 1976-78. d. Includes uranium = 86.8 percent (1976). See technical notes. e. World Bank estimates = 84.0 percent for 1961 and 91.3 percent for 1976-78. The latter estimate includes fishing products. f. Includes diamonds = 69.7 percent. See technical notes. g. World Bank estimates = 76.4 percent for 1961 and 96.7 percent for 1976-78. Both estimates indude alumina. 156 Table 17. Balance of Payments, Debt Service, and International Reserves Current account balance before Interest interest payments payments on Debt service as percentage Gross international reserves on external public external public In months debt (millions of debt (millions of Of exports of goods Millions of of import dollars) dollars) Of GNP and services dollars - coverage 1970 1979a 1970 1979 1970 1979 1970 1979' 1970 2979 1979' Low-income countries 1.1 w 2.7 w 6.0 w 12.7 w 1.7 w Low-income semiarid 0.6 w 1.8 w 2.5 w 16.0 w 1.9 w I. Chad 2 - 72' . 4 1.0 3.3 3.9 14.4 2 17 0.5' 2. Somalia - 5 -205 (. 1 0.3 0.2 2.1 1.1 21 54 1.4 3.- Mali - 2 -64 (.) 3 0.2 0.7 1.2 8.5 1 17 0.5' 4 . Upper Volta 9 -68 () 4 0.6 0.8 4.0 3.8 36 67 2.0 b. Niger 1 - 96' 1 7 0.6 0.8 3.8 3.6 19 137 7. Maur-itania -5 - 70 () 16 2.0 13.6 3.2 32.4 3 118 3.6 Low-income other 1.3 w 2.8 w 6.3 w 12.4 w 1.6 w 8. Ethiopia - 26 - 79 6 13 1.2 0.7 11.4 4.9 72 321 5.4 9. Guinea-B3issau- . -68 .. 2b . . 6.0 . 59.0 . . 22 10. Burundi 2 -38 .) 1 0.3 0.4 2.3 3.1 15 95 5.7 11. Malawi - 32 -185 3 16 1.8 2.1 7.0 9.4 29 75 1.7 12. Rwanda 6 44 .) 1 0.2 0.1 1.4 0.6 8 153 5.8 13. Benin -I - 87* () 3 0.7 1.4 2.2 5.1 16 20 14. M,.ozambique.. .. .. .. .. ..... .. . 1-5. Sierra Leone - 14 - 109' 2 12 2.9 8.6 10.1 22.2 39 47 1.3 16. Tanzania - 29 - 457' 6 23 1.2 0.9 8.2 7.4 65 69 0. 9' 1 7. Zaire - 55 -463 9 95 2.0 2.3 4.4 9.1 189 335 1.4 18. Guinea - 30- - 54" 4 24 2.4 5.7 26.7 22.2 13 35 1.0 19. Central Afr-ican Republic - 11 -9 () .) 1.1 (. 3.3 0.1 1 49 2.7 20. Madagascar 12 -425 2 8 0.8 0.7 3.5 3.9 37 5 0.1 21. Uganda 24 32 4 5 0.6 0.3 3.4 7.4 57 221. Lesotho . . -2 (. 1 0.5 0.3 . 0.6 . . 23 -Togo 4 -219 1 16 0.9 6.9 3.0 24.4 35 71 2.0' 24. Sudan - 29 -151 13 86 1.3 4.5 10.7 33.0 22 67 0.7 Middle-income oil importers 2.3 w 4.1 w 5.9 w 12.2 w 2.5 w 25. Kenva - 38 - 419 11 60 1.7 1.8 7.9 7.5 220 669 3.7 26. Ghana - 56 282 12 26 1.1 0.5 5.2 4.2 58 404 4.8 r27.Senegal - 14 -394 2 43 0.8 5.0 2.7 13.7 22 35 28. Zimbabwe - 13 - 61 .. .. . .. .. .. .. . 29. Liber-ia .- -91 6 22 5.5 8.1 .- 13.8 .. 55 30. Zambia 131 264 23 93 3.2 9.7 5.8 19.7 515 193 1.8 31. Cameroon - 26 - 290 4 65 0.8 2.5 3.2 9.5 81 141 0.5' 32. Swaziland . .. .. .. .. ..... .. 33. Botswana.. . . . . ... .. .. . 34. Mauntius . . ..... .. . 35. Ivory Coast - 26 - 560' 11 225 2.8 6.0 6.8 15.2 119 168 1.6' Middle-income oil exporters 0.7 w 0.5 w 1.6 w 4.4 w 36. Angola . . .. .. .. . 37. Congo . 144' 3 38 3.4 10.1 . 7.3' 9 47 0.2' 38. Nigeria - 348 1,429 20 205 0.7 0.4 4.2 1.5 223 5,870 4.5 39. Gabon. .. . .. . . .. .. . Sub-Saharan Africa 1.4 w 2.0 w 5.0 w 6.9 w 3.3 w All low-incomne countries 1.1 w 1.8 w 12.6 w 10.8 w 4.2 w All middle-income countries 1.5 w 3.2 w 9.0 w 14.2 w 5.2 w Industrialized countries 5.0 w a Figures marked with an *are for 1978. 6 Figure shown is for 1980. 159 Table 18. Debt and Debt Service Public and publicly guaranteed private debt outstanding and disbursed (millions of dollars) Official sources Private sources Total Debt service 1970 1979a 1970 1979a 1970 1979' 1970 1979b Low-income countries 1,733.8 t 9,683.9 t 566.7 t 4,619.0 t 2,300.6 t 14,304.0 t 188.3 t 943.9 t Low-income semiarid 406.9 t 2,168.6 t 24.3 t 240.5 t 431.2 t 2,409.2 t 11.9 t 118.8 t 1. Chad 24.6 142.4 7.3 29.3 31.9 171.7 2.7 18.4 2. Somalia 74.9 544.6 2.2 1.1 77.1 545.7 0.9 2.1* 3. Mali 231.5 529.6 6.1 22.0 237.6 551.6 0.7 8.8* 4. Upper Volta 20.2 254.4 0.3 16.7 20.5 271.1 1.9 9.4* 5. Gambia 5.1 44.8 .. .. 5.1 44.8 0.1 0.3 6. Niger 31.2 195.8 0.5 38.2 31.7 234.0 2.3 13.6 7. Mauritania 19.4 457.0 7.9 133.2 27.3 590.2 3.3 66.2 Low-income other 1,326.9 t 7,515.3 t 542.4 t 4,378.5 t 1,869.4 t 11,894.8 t 176.4 t 825.1 t 8. Ethiopia 140.2 603.2 28.7 16.8 168.9 620.0 21.1 26.3* 9. Guinea-Bissau .. 64.1 .. .. .. 64.1 . 10. Burundi 5.8 90.8 1.5 12.2 7.3 103.0 0.6 3.5* 11. Malawi 87.8 311.8 33.3 111.5 121.1 423.3 5.8 27.0 12. Rwanda 1.5 134.0 0.4 0.8 1.9 134.7 0.3 1.6* 13. Benin 29.2 172.6 11.3 13.2 40.5 185.8 1.7 13.3 14. Mozambique .. .. .. .. .. .. 15. Sierra Leone' 32.4 168.1 27.0 130.2 59.4 299.3 12.0 49.2* 16. Tanzania 152.7 1,098.7 95.8 114.6 248.5 1,213.4 15.7 59.8* 17. Zaire 95.3 1,607.7 215.8 2,471.9 311.1 4,079.6 36.8 169.2 18. Guinea 277.5 805.1 36.7 206.8 314.2 1,011.9 28.6 83.0* 19. Central African Republic 17.7 81.4 1.7 68.6 19.4 150.0 2.0 0.2 20. Madagascar 84.8 280.5 8.1 67.2 92.9 347.7 6.9 18.0 21. Uganda 106.3 243.3 21.5 1.7 127.8 245.0 7.4 27.4 22. Lesotho 7.6 44.4 0.5 7.4 8.1 51.8 0.5 1.4 23. Togo 31.9 384.4 7.8 466.7 39.8 851.1 2.3 68.2* 24. Sudan 256.2 1,425.2 52.3 688.9 308.5 2,114.1 34.7 277.0 Middle-income oil importers 1,134.5 t 5,922.9 t 990.0 t 5,948.4 t 2,124.4 t 11,871.0 t 183.8 t 1,751.8 t 25. Kenya 228.7 970.4 84.2 915.3 312.8 1,885.7 27.0 235.6 26. Ghana 264.2 744.1 225.1 242.9 489.3 986.9 23.7 54.6 27. Senegal 78.0 477.0 19.6 261.5 97.7 738.5 6.7 130.0 28. Zimbabwe .. .. .. .. .. .. . 29. Liberia 124.0 308.4 33.9 146.0 157.9 454.3 17.6 76.5 30. Zambia 119.4 1,075.6 476.6 633.0 596.0 1,708.6 54.9 351.6 31. Cameroon 119.6 995.8 11.6 637.9 131.2 1,633.6 8.6 126.3 32. Swaziland 20.9 117.7 16.1 28.5 37.0 146.2 3.3 5.7 33. Botswana 14.2 132.0 0.6 3.6 14.7 135.6 0.6 10.1 34. Mauritius 21.4 134.7 10.3 104.2 31.7 238.9 2.9 24.3* 35. Ivorv Coast 144.1 967.2 112.0 2,975.5 256.1 3,942.7 38.5 737.1 Middle-income oil exporters 569.9 t 1,884.0 t 141.4 t 4,097.2 t 711.4 t 5,981.2 t 75.8 t 793.1 t 36. Angola .. .. .. .. .. 37. Congo 121.7 565.5 20.8 233.1 142.5 798.6 8.8 106.4 38. Nigeria 381.8 1,016.9 96.2 2,952.2 478.1 3,969.1 55.7 341.0 39. Gabon 66.4 301.6 24.4 911.9 90.8 1,213.5 11.3 345.7 Sub-Saharan Africa 3,438.2 t 17,490.8 t 1,698.1 t 14,664.6 t 5,136.4 t 32,156.2 t 447.9 t 3,488.8 t (Sub-Saharan Africa as a percentage of all developing countries) (9.7) (12.7) (9.6) (6.0) (10.0) (8.4) (6.8) (4.9) a. Figures for 1979 are preliminary actual payments. b. Figures followed by an * are actual payments. All other figures are scheduled payments. 160 Table 19. Outstanding External Debt of Sub-Saharan Africa Outstanding and disbursed in billions of dollars 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 Total concessional bilateral 2.3 2.7 3.0 3.7 4.6 5.2 6.1 7.0 7.7 8.5 DAC governments 1.6 1.9 2.0 2.3 2.8 3.1 3.5 4.0 4.4 4.6 OPEC governments 0.1 0.1 0.1 0.1 0.2 0.3 0.7 0.9 1.0 1.4 CPE governments 0.6 0.7 0.9 1.1 1.4 1.6 1.8 2.0 2.1 2.3 Other bilateral 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.3 Total official export credits 0.3 0.4 0.4 0.5 0.7 0.9 1.2 1.5 1.9 2.2 DAC governments 0.3 0.3 0.3 0.4 0.5 0.6 0.9 1.1 1.4 1.7 OPEC governments 0.0 0.0 0.0 0.0 0.1 0.2 0.2 0.3 0.3 0.3 CPE governments 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.1 Other bilateral 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.1 0.1 Total multilateral loans 0.8 1.0 1.3 1.6 2.0 2.5 3.2 4.1 5.4 6.6 IBRD 0.5 0.6 0.8 0.9 1.0 1.2 1.5 1.7 2.0 2.3 IDA 0.2 0.3 0.4 0.5 0.7 0.9 1.2 1.5 1.8 2.2 Regional banks concessional 0.0 0.0 0.0 0.1 0.1 0.1 0.1 0.2 0.4 0.5 Regional banks nonconcessional 0.0 0.0 0.0 0.1 0.1 0.1 0.1 0.2 0.3 0.4 Other multilateral concessional 0.0 0.0 0.0 0.0 0.1 0.2 0.2 0.4 0.9 1.2 Other multilateral nonconcessional .. .. .. 0.0 0.0 0.0 0.0 0.0 0.1 0.1 Total private publicly guaranteed loans 1.7 1.8 2.1 3.2 4.0 4.8 5.6 7.3 10.4 13.0 Suppliers credits 0.7 0.9 1.0 1.3 1.5 1.9 2.2 2.6 3.0 2.9 Financial institutions 0.3 0.4 0.7 1.3 1.8 2.5 3.3 4.5 7.3 10.0 Bonds 0.2 0.3 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 Other 0.4 0.3 0.3 0.5 0.5 0.3 0.1 Total public and publicly guaranteed loans 5.1 5.9 6.8 9.0 11.3 13.4 16.1 19.8 25.4 30.3 Total private nonguaranteed 0.5 0.6 0.7 0.9 1.1 1.3 1.3 1.3 1.4 1.5 Total public and private debt 5.7 6.5 7.6 9.9 12.4 14.7 17.4 21.2 26.8 31.8 of which: Total bilateral 2.6 3.1 3.4 4.2 5.3 6.1 7.3 8.5 9.6 10.7 Total official 3.4 4.1 4.7 5.8 7.2 8.6 10.5 12.6 15.0 17.3 Total private source 2.2 2.5 2.9 4.1 5.1 6.1 6.9 8.6 11.8 14.5 Total concessional 2.5 3.1 3.5 4.3 5.4 6.3 7.6 9.1 10.7 12.4 Total nonconcessional 3.1 3.5 4.1 5.6 7.0 8.4 9.8 12.0 16.1 19.4 161 Table 20. Gross Disbursements of External Loans to Sub-Saharan Africa Millions of dollars 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 Total concessional bilateral 389 402 518 569 851 989 1,054 970 1,028 1,472 DAC govemments 261 283 299 328 450 599 499 507 651 726 OPEC governments 3 6 35 38 118 112 330 200 152 413 CPE govemments 117 105 180 193 252 248 221 234 155 270 Other bilateral 8 9 4 9 31 30 3 29 70 62 Total official export credits 42 46 72 141 237 328 259 277 283 468 DAC govemments 19 36 59 119 143 170 166 230 183 333 OPEC govemments 12 75 72 62 42 70 48 CPE govemments 17 10 12 7 15 17 24 1 (.) 42 Other bilateral 7 3 5 69 7 5 30 46 Total multilateral loans 150 206 316 291 414 616 728 955 1,390 1,939 IBRD 75 124 185 131 154 274 284 291 363 348 IDA 61 63 84 108 133 204 281 339 323 367 Regional banks concessional 12 8 20 22 17 19 29 88 116 158 Regional banks nonconcessional 3 12 27 20 32 37 35 79 107 100 Other multilateral concessional ..) 2 74 81 77 151 447 330 Other multilateral nonconcessional 8 4 21 10 35 25 Total private loans 490 414 576 1,128 1,131 1,531 1,687 2,001 3,308 3,768 Suppliers credits 144 229 250 328 448 626 587 630 440 365 Financial institutions (37) (177) (319) (787) (682) (905) (1,098) (1,366) (2,868) (3,403) Bonds (9) (3) (1) (9) (.) ( ) (2) (.) (.) (G) Other 300 6 6 4 1 5 Total public and publicly guaranteed loans 1,070 1,069 1,482 2,129 2,632 3,463 3,728 4,202 6,008 7,035 Nonguaranteed private sector credit 140 140 220 230 363 361 272 248 390 391 Total public and private disbursements 1,210 1,209 1,702 2,359 2,994 3,825 4,000 4,451 6,399 7,426 of which: Total bilateral 430 448 590 710 1,087 1,317 1,313 1,247 1,311 1,939 Total official 581 655 906 1,001 1,501 1,933 2,041 2,202 2,701 3,267 Total pnvate source 630 554 796 1,358 1,493 1,892 1,959 2,249 3,698 4,159 Total concessional 461 473 621 701 1,075 1,294 1,441 1,545 1,913 2,326 Total nonconcessional 749 736 1,080 1,658 1,920 2,531 2,559 2,905 4,485 5,099 162 Table 21. Average Terms of Borrowing for Sub-Saharan Africa Average terms for new commitments of public and publicly guaranteed debt 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 Total public debt Interest rate (percent) 3.7 4.2 4.5 5.3 5.2 5.7 5.3 5.5 6.3 7.9 Maturity (years) 24.4 22.3 20.9 20.4 20.0 20.3 19.2 19.1 17.6 16.7 Grace period (years) 6.7 6.7 5.8 5.7 5.9 5.2 5.2 4.7 4.9 4.7 Grant element (percent) 46.3 41.5 37.1 33.1 33.8 29.9 31.2 30.1 25.4 16.7 Total official debt Interest rate (percent) 2.0 3.1 3.2 2.8 3.3 4.1 3.5 4.0 3.6 4.1 Maturity (years) 31.9 27.6 28.5 29.3 25.8 27.2 27.4 26.0 26.0 24.9 Grace period (years) 9.3 8.4 7.8 8.0 7.6 6.7 6.9 6.3 6.8 6.3 Grant element (percent) 64.1 53.3 52.1 55.7 50.0 44.0 49.1 44.2 47.0 42.3 Bilateral Interest rate (percent) 1.3 2.1 2.7 1.8 3.1 3.1 3.4 3.9 3.9 4.8 Maturity (years) 31.5 26.2 23.7 25.9 22.4 24.3 22.7 22.3 24.2 20.7 Grace period (years) 9.7 9.3 7.8 8.1 7.6 6.6 6.5 5.7 6.6 5.7 Grant element (percent) 69.8 59.8 53.0 61.0 49.4 49.0 46.6 42.5 45.0 35.3 Multilateral Interest rate (percent) 4.3 5.0 3.9 4.1 3.7 5.2 3.6 4.1 3.3 3.4 Maturity (years) 33.2 30.4 35.7 34.3 32.6 30.2 33.4 30.3 28.4 28.8 Grace period (years) 7.9 6.5 7.8 7.9 7.7 6.9 7.4 6.9 7.0 6.8 Grant element (percent) 46.3 40.0 50.8 48.1 51.2 39.1 52.4 46.3 49.7 48.9 Total private debt Interest rate (percent) 6.8 7.2 6.7 8.4 8.8 8.4 7.8 7.7 9.3 11.4 Maturity (years) 10.0 8.5 8.1 9.9 9.4 8.5 8.6 8.7 8.6 9.0 Grace period (years) 1.6 2.4 2.5 3.0 2.9 2.5 2.9 2.4 2.9 3.2 Grant element (percent) 12.0 10.5 11.7 6.5 4.3 6.0 7.9 8.8 2.0 -7.0 163 Table 22. Indicators of Aid and Total Resource Flow, 1979 Net official development assistancea (disbursements) Total recorded As a net flow of Net bilaterala percentage From OPEC resources per as a As a of gross as a percentage Population capita I Per capita percentage of percentage domestic of total bilateral (millions) (dollars) (dollars) total of GNP investment net ODA Low-income countries 187.1 t 27.2 w 21.1 w 58.6 m 7.8 w 50.0 w 4.0 m Low-income semiarid 28.0 t 45.9 w 36.8 w 49.5 m 18.4 w 76.7 w 8.8 m 1. Chad 4.4 19.6 19.8 56.8 18.0 117.3 0.0 2. Somalia 3.8 67.6 47.6 27.6 .. 109.6 60.4 3. Mali 6.8 29.9 27.9 49.5 19.9 103.7 8.8 4. Upper Volta 5.6 38.5 35.6 66.2 19.8 96.6 0.0 5. Gambia 0.6 67.0 59.5 37.0 23.8 162.3 23.3 6. Niger 5.2 51.5 33.1 67.8 12.3 35.9 (.) 7. Mauritania 1.6 133.6 103.4 21.4 32.3 69.0 61.1 Low-income other 159.1 t 23.9 w 18.4 w 60.0 m 6.6 w 47.0 3.9 m 8. Ethiopia 30.9 7.2 5.7 40.4 4.4 49.5 0.0 9. Guinea-Bissau 0.8 71.4 66.4 63.8 39.1 .. 3.9 10. Burundi 4.0 22.8 23.0 48.0 12.8 104.9 3.7 11. Malawi 5.8 31.5 24.0 66.0 12.0 39.4 0.0 12. Rwanda 4.9 29.9 29.9 60.4 15.0 89.5 0.7 13. Benin 3.4 30.4 23.9 60.0 9.6 45.5 4.1 14. Mozambique 10.2 15.5 15.1 74.2 6.0 65.3 8.0 15. Sierra Leone 3.4 15.7 15.3 54.1 6.1 44.0 12.4 16. Tanzania 18.0 40.4 32.2 78.8 12.4 66.9 0.8 17. Zaire 27.5 26.4 14.7 71.5 5.7 74.5 0.0 18. Guinea 5.3 11.8 10.8 24.8 3.9 24.8 15.5 19. Central African Republic 2.0 45.9 43.8 58.4 15.1 68.4 11.4 20. Madagascar 8.5 25.8 14.7 58.7 5.1 20.3 15.8 21. Uganda 12.8 2.7 2.9 44.0 1.0 10.9 4.2 22. Lesotho 1.3 47.7 48.8 68.8 14.4 91.2 0.7 23. Togo 2.4 91.0 46.5 61.7 13.3 28.6 0.0 24. Sudan 17.9 36.0 31.4 26.5 8.5 52.6 66.1 Middle-income oil importers 65.2 t 45.3 w 26.8 w 75.5 m 4.6 w 24.3 w 0.8 m 25. Kenya 15.3 29.4 22.7 81.8 6.0 29.9 0.0 26. Ghana 11.3 17.6 15.2 51.8 3.8 33.7 15.8 27. Senegal 5.5 63.3 56.2 48.2 13.1 59.3 0.9 28. Zimbabwe 5.6 1.5 2.2 99.2 0.5 2.3 0.8 29. Liberia 1.8 185.4 45.7 37.0 9.1 32.4 32.7 30. Zambia 7.1 45.3 38.1 78.4 7.6 39.7 4.3 31. Cameroon 8.2 68.2 33.1 67.6 5.9 20.4 6.2 32. Swaziland 0.5 147.2 98.2 64.8 15.1 72.2 0.0 33. Botswana 0.8 164.8 116.0 79.3 16.1 52.6 0.0 34. Mauritius 0.9 64.0 35.8 75.5 3.5 9.3 0.0 35. Ivory Coast 8.2 65.6 19.8 85.4 1.9 5.7 0.0 Middle-income oil exporters 91.6 t 4.4 w 2.1 w 64.9 m 0.2 w 1.2 w 0.0 m 36. Angola 6.9 16.1 6.2 69.0 1.4 19.0 0.0 37. Congo 1.5 40.2 54.9 60.8 8.7 33.4 7.4 38. Nigeria 82.6 3.5 0.3 41.2 (.) 0.1 0.0 39. Gabon 0.6 -90.8 60.7 73.6 1.9 4.1 0.0 Sub-Saharan Africa 343.9 t 24.6 w 17.1 w 60.8 m 3.5 w 20.7 w 0.9 m a. See technical notes for definitions. 164 Table 23. Disbursements of Official Development Assistance Grants as a percentage of net ODA (1978179 average) Net disbursements in millions of dollars Technical 1970 1971 1972 1973 1974 2975 2976 1977 1978 1979 Total assistance Low-income countries Low-income semiarid 1. Chad 22 31 31 42 79 65 62 83 119 87 76 25 2. Somalia 28 31 24 36 82 153 105 256 163 181 56 16 3. Mali 21 30 39 70 118 145 89 113 162 190 86 24 4. Upper Volta 22 25 35 57 97 89 84 110 159 199 91 31 5. Gambia 1 4 5 7 10 8 12 22 35 36 57 23 6. Niger 36 38 43 70 137 141 129 97 157 172 95 26 7. Mauritania 8 12 9 25 88 62 180 165 217 166 92 11 Low-income other 8. Ethiopia 40 47 47 67 121 135 141 114 140 175 64 18 9. Guinea-Bissau 7 7 5 4 19 22 38 50 53 . 10. Burundi 18 22 25 27 38 48 45 48 75 92 73 40 11. Malawi 37 32 36 30 42 64 64 80 99 139 97 23 12. Rwanda 22 25 30 39 47 91 79 96 125 146 87 34 13. Benin 15 29 20 27 38 54 55 49 61 81 78 26 14. Mozambique 19 58 51 (.) 1 22 72 80 105 154 46 23 15. Sierra Leone 7 10 10 14 11 18 15 26 40 52 57 30 16, Tanzania 51 61 59 100 163 296 267 340 424 580 94 24 17. Zaire 89 109 121 140 181 205 194 261 317 404 66 42 18. Guinea 10 10 27 11 30 15 12 22 60 57 49 18 19. Central Afnican Republic 14 16 26 26 37 57 38 42 57 88 89 44 20. Madagascar 48 47 56 53 63 85 63 61 91 125 63 29 21. Uganda 33 32 29 15 30 39 25 22 17 37 128 53 22. Lesotho 10 17 14 14 21 30 30 39 50 64 59 22 23. Togo 17 19 23 26 39 42 43 64 103 112 41 22 24. Sudan 6 10 37 43 134 299 369 224 316 563 64 15 Middle-income oil importers 25. Kenya 58 67 72 96 118 131 160 166 245 347 66 31 26. Ghana 59 57 58 41 37 126 64 91 114 171 47 26 27. Senegal 43 53 49 79 139 133 127 123 226 309 61 34 28. Zimbabwe 1 1 1 1 2 4 6 7 9 13 48 27 29. Liberia 13 13 13 11 16 21 27 34 48 82 48 27 30. Zambia 13 22 22 46 58 87 62 109 185 270 100 94 31. Cameroon 59 48 64 61 62 125 134 176 178 272 38 26 32. Swaziland 6 2 9 11 16 19 16 28 48 49 44 24 33. Botswana 14 18 32 37 37 51 48 47 69 93 116 37 34. Mauritius 6 9 11 14 25 29 17 22 44 32 62 13 35. Ivory Coast 53 51 49 64 76 101 108 106 131 162 21 49 Middle-income oil exporters 36. Angola - 2 13 70 (.) (.) 5 38 48 47 43 100 30 37. Congo 16 17 23 27 38 56 73 49 81 82 70 38 38. Nigeria 102 107 83 77 73 83 53 42 37 26 139 124 39. Gabon 24 24 27 34 25 63 34 28 44 36 75 65 Sub-Saharan Africa 1,046 t 1,224 t 1,385 t 1,538 t 2,333 t 3,216 t 3,162 t 3,528 t 4,648 t 5,940 t 65 w 26 w 165 Table 24. Food Aid Importsa Thousands of metric tons, grain equivalent Kilograms per capita 1975 1976 1977 1978 1979 1975 1976 1977 1978 1979 Low-income countries 834.7 th 648.5 th 708.2 t 1,001.0 th 953.0 t 4.9 w 3.7 w 4.0 w 5.5 w 5.1 w Low-income semiarid 396.3 th 240.4 tb 207.9 t 331.9 th 249.7 t 15.6 w 9.2 w 7.8 w 12.2 w 8.9 w 1. Chad 13.0 3.6 35.4 49.2 25.9 3.2 0.9 8.4 11.5 5.9 2. Somalia 109.8 61.5 75.0 79.0 86.2 31.4 17.2 20.5 21.1 22.5 3. Mali 113.6 39.1 4.2 32.8 24.8 18.8 6.3 0.7 5.0 3.7 4. Upper Volta 0.0 17.2 5.6 51.8 49.1 0.0 3.2 1.0 9.3 8.7 5. Gambia 9.0 2.9 5.7 19.6 9.4 17.2 5.4 10.3 34.3 16.0 6. Niger 74.8 90.5 53.1 21.5 23.2 16.3 19.1 10.9 4.3 4.5 7. Mauritania 47.9 25.1 28.9 51.0 31.1 33.7 17.2 19.3 33.0 19.6 Low-income other 438.4 t 408.1 t 500.3 t 669.1 t 703.3 t 3.0 w 2.8 w 3.3 w 4.3 w 4.4 w 8. Ethiopia 58.7 60.1 65.3 76.0 157.7 2.0 2.1 2.2 2.5 5.1 9. Guinea-Bissau 10.9 7.2 20.9 25.2 23.0 15.0 9.7 27.8 33.0 29.5 10. Burundi 5.5 1.7 5.9 4.2 16.0 1.5 0.4 1.5 1.1 4.0 11. M1alawi 0.2 0.8 3.4 3.2 2.2 (.) 0.1 0.6 0.6 0.4 12. Rwanda 19.3 9.8 11.5 14.6 10.3 4.4 2.2 2.5 3.0 2.1 ComOros 0.8 3.6 5.1 6.4 7.4 2.4 10.5 13.8 16.6 18.6 13. Benin 8.8 6.4 6.3 11.5 5.0 2.9 2.0 2.0 3.5 1.5 14. Mozambique 34.0 62.2 94.7 133.1 125.9 3.7 6.6 9.8 13.4 12.3 15. Sierra Leone 9.5 6.0 7.7 7.0 5.6 3.1 1.9 2.4 2.1 1.7 16. Tanzania 147.8 124.1 134.5 104.9 53.2 9.4 7.6 8.0 6.0 3.0 17. Zaire 0.1 17.9 14.5 31.4 79.8 (.) - 0.7 0.6 1.2 2.9 18. Guinea 48.9 34.9 14.3 36.9 33.0 10.4 7.2 2.9 7.2 6.3 19. Central African Rep. 0.6 1.4 0.5 2.5 1.2 0.3 0.8 0.3 1.3 0.6 20. Madagascar 7.1 1.2 15.8 6.2 8.6 0.9 0.2 2.0 0.8 1.0 21. Uganda 15.8 0.0 0.0 0.0 0.0 1.4 0.0 0.0 0.0 0.0 22. Lesotho 13.9 18.5 11.5 23.6 37.8 11.7 15.2 9.2 18.5 28.9 23. Togo 0.0 5.9 6.5 20.8 9.1 0.0 2.6 2.8 8.8 3.8 24. Sudan 49.6 27.0 54.2 111.8 90.9 3.1 1.6 3.2 6.4 5.1 Cape Verde 6.9 19.4 27.7 49.8 36.6 24.2 67.1 93.9 166.0 119.6 Middle-income oil importers 113.2 t 98.4 t 147.7 t 336.9 t 234.6 t 2.2 w 1.8 w 2.7 w 5.9 w 4.0 w 25. Kenva 2.2 8.8 8.3 10.8 9.4 0.2 0.6 0.6 0.7 0.6 Equatorial Guinea 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 26. Ghana 43.1 29.8 57.5 88.7 73.3 4.3 2.9 5.4 8.1 6.5 Djibouti 0.0 0.0 0.0 2.9 7.5 0.0 0.0 0.0 9.7 23.1 27. Senegal 28.3 23.5 31.4 168.0 65.1 5.7 4.6 6.0 31.2 11.8 Sdo Tome and Principe 0.0 2.0 1.0 4.0 2.1 0.0 25.2 12.5 49.3 25.6 28. Zimbabwe .. .. .. .. .. .. .. .. 29. Liberia 3.4 1.9 0.6 1.3 1.1 2.1 1.1 0.3 0.7 0.6 30. Zambia 1.0 5.3 28.6 31.2 49.9 0.2 1.0 5.5 5.8 8.9 31. Cameroon 3.5 1.1 5.8 5.1 7.8 0.5 0.1 0.7 0.6 0.9 32. Swaziland 0.7 0.5 0.0 0.4 0.7 1.4 1.0 0.0 0.8 1.3 33. Botswana 5.4 3.4 3.4 8.4 8.8 7.8 4.8 4.7 11.2 11.4 34. Maunrtus 21.6 21.8 9.2 13.5 6.0 24.5 24.4 10.2 14.7 6.4 35. Ivory Coast 3.6 0.1 0.1 0.6 0.3 0.5 (.) (.) 0.1 (-) SeVchelles 0.4 0.2 1.8 2.0 2.6 6.8 3.3 29.1 31.6 40.3 Middle-income oil exporters 9.9 t 5.1 t 12.7 t 21.3 t 13.9 t 0.1 w (.) w 0.1 w 0.2 w 0.2 w 36. Angola 0.0 0.1 9.6 17.7 7.1 0.0 (.) 1.5 2.6 1.0 37. Congo 2.1 2.6 3.1 3.2 4.2 1.6 1.9 2.2 2.2 2.8 38. Nigeria 7.3 2.2 0.0 0.0 0.0 0.1 (.) 0.0 0.0 0.0 39. Gabon 0.5 0.2 0.0 0.4 2.6 0.8 0.3 0.0 0.6 4.0 Sub-Saharan Africa 957.8 th 752.0 th 868.6 t 1,359.2 th 1,201.5 t 3.1 w 2.4 w 2.7 w 4.1 w 3.5 w a Includes the six countries whose population is less than one half million. b Includes food aid to unspecified Sahel countries. 166 Table 25. Growth of Agriculture Average annual growth rate Average annual growth rate of total of volume of production production per capita 1969-71 to 1977-79 1969-71 to 1977-79 Food Nonfood Total Food Nonfood Total Low-income countries 0.9 w -1.7 w Low-income semiarid 1.1 w -1.2 w 1. Chad 1.0 2.0 1.1 -1.0 0.0 -0.9 2. Somalia 0.6 -0.8 0.6 -1.7 -3.1 -1.7 3. Mali 1.0 9.8 1.4 -1.6 7.2 -1.2 4. Upper Volta 2.0 7.2 2.1 0.4 5.6 0.5 5. Gambia 0.1 . . 0.1 - 2.9 . . - 2.9 6. Niger 1.3 - 7.8 1.3 -1.5 -10.6 -1.5 7. Mauritania -1.3 . . -1.3 -4.0 . . -4.0 Low-income other 0.8 w -1.6 w 8. Ethiopia 0.4 1.3 0.4 -1.7 -0.8 -1.7 9. Guinea-Bissau 1.4 0.0 1.4 -0.2 -1.6 -0.2 10. Burundi 2.7 1.8 2.6 0.7 -0.2 0.6 11. Malawi 3.1 8.6 4.0 0.3 5.8 1.2 12. Rwanda 3.9 4.7 3.9 1.1 1.9 1.1 13. Benin 2.5 -5.2 2.3 -0.4 -8.1 -0.6 14. Mozambique -0.6 -4.7 -1.0 -3.1 -7.2 -3.5 15. Sierra Leone 1.4 4.9 1.7 -1.1 2.4 -0.8 16. Tanzania 1.9 -0.5 1.4 -1.5 -3.9 -2.0 17. Zaire 1.3 -0.6 1.2 -1.4 -3.3 -1.5 18. Guinea 0.5 -11.7 0.2 -2.4 -14.6 -2.7 19. Central African Republic 2.4 1.5 2.2 0.2 -0.7 0.0 20. Madagascar 1.8 0.7 1.7 -0.7 -1.8 -0.8 21. Uganda 1.7 -8.3 -0.5 -1.3 -11.3 -3.5 22. Lesotho 2.4 -7.0 1.4 0.1 -9.3 -0.9 23. Togo -0.2 -4.2 -0.4 -2.6 -6.6 -2.8 24. Sudan 3.1 -3.9 1.8 0.5 -6.5 -0.8 Middle-income oil importers 2.2 w -1.1 w 25. Kenya 2.9 7.5 4.0 - 0.5 4.1 0.6 26. Ghana -0.1 -4.5 -0.1 -3.1 -7.5 -3.1 27. Senegal 1.0 11.3 1.1 -1.6 8.7 -1.5 28. Zimbabwe 2.6 3.8 2.9 -0.7 0.5 -0.4 29. Liberia 3.5 0.2 2.7 0.2 -3.1 -0.6 30. Zambia 3.0 -0.9 2.8 0.0 -3.9 -0.2 31. Cameroon 3.3 1.8 3.1 1.1 -0.4 0.9 32. Swaziland 3.7 14.6 4.6 1.2 12.1 2.1 33. Botswana 1.1 2.0 1.1 -1.1 -0.2 -1.1 34. Mauritius 1.9 3.9 1.9 0.6 2.6 0.6 35. Ivory Coast 4.6 1.8 3.8 -0.9 -3.7 -1.7 Middle-income oil exporters 1.1 w -1.4 w 36. Angola 0.2 -13.3 -3.3 -2.1 -15.6 -5.6 37. Congo -0.1 1.9 -0.1 -2.6 -0.6 -2.6 38. Nigeria 1.7 -1.3 1.7 -0.8 -3.8 -0.8 39. Gabon 0.1 -13.3 0.1 -1.1 -14.5 -1.1 Sub-Saharan Africa 1.3 w -1.4 w 167 Table 26. Production of Major Crops Average annual volume of production Rate of growth (percent) 1961-63 1969-71 Thousands of metric tons to to Crop, 1961-63 1969-71 1977-79 2969-71 1977-79 Cereals: Maize Sub-Saharan Africa 8,105 12,132 13,438 5.2 1.3 Oil exporters 1,136 1,691 1,814 5.1 0.9 Other countries 6,969 10,441 11,624 5.2 1.4 Millet Sub-Saharan Africa 8,083 8,875 9,178 1.2 0.4 Oil exporters 2,689 2,870 3,083 0.8 0.9 Other countries 5,394 6,005 6,095 1.4 0.2 Rice (paddy) Sub-Saharan Africa 3,473 4,735 5,936 4.0 2.9 Oil exporters 232 380 856 6.4 10.7 Other countries 3,241 4,355 5,080 3.8 1.9 Sorghum Sub-Saharan Africa 8,203 8,591 9,768 0.6 1.6 Oil exporters 3,700 3,632 3,768 -0.2 0.5 Other countries 4,503 4,959 6,000 1.2 2.4 W'heat Sub-Saharan Africa 923 1,243 1,220 3.8 -0.2 Oil exporters 36 33 31 -1.1 -0.8 Other countries 887 1,210 1,189 4.0 -0.2 Total Cerealsb Sub-Saharan Africa 31,518 37,701 41,669 2.3 1.3 Oil exporters 7,828 8,648 9,609 1.3 1.3 Other countries 23,690 29,053 32,060 2.6 1.2 Oils and Oilseeds: Coconuts Sub-Saharan Africa 1,350 1,451 1,563 0.9 0.9 Oil exporters 75 86 90 1.7 0.6 Other countries 1,275 1,365 1,473 0.9 1.0 Groundnuts (in shell) Sub-Saharan Africa 4,922 5,194 4,826 0.7 -0.9 Oil exporters 1,864 1,699 503 -1.2 -14.1 Other countries 3,058 3,495 4,323 1.7 2.7 Palm kernels Sub-Saharan Africa 806 /11 664 -1.6 -0.9 Oil exporters 439 306 310 -4.4 0.2 Other countries 367 405 354 1.2 -1.7 Palmoil Sub-Saharan Africa 970 1,112 1,321 1.7 2.2 Oil exporters 573 579 718 0.1 2.7 Other countries 397 533 603 3.8 1.6 Other: Pulses Sub-Saharan Africa 2,973 3,861 4,207 3.3 1.1 Oil exporters 616 925 923 5.2 0.0 Other countries 2,357 2,936 3,284 2.8 1.4 Roots and Tubers Sub-Saharan Africa 57,042 66,694 77,026 2.0 1.8 Oil exporters 23,162 27,674 31,488 2.2 1.6 Other countries 33,880 39,020 45,538 1.8 1.9 Seed Cotton Sub-Saharan Africa 1,315 2,279 1,867 7.1 -2.5 Oil exporters 136 268 195 8.8 -3.9 Other countries 1,179 2,011 1,672 6.9 -2.3 Sugar Sub-Saharan Africa 1,415 2,303 2,806 6.3 2.5 Oil exporters 88 179 109 9.3 -6.0 Other countries 1,327 2,124 2,697 6.1 3.0 a. Major crops that are totally or nearly totally exported (such as coffee, tea, cocoa, rubber) are shown in Table 30, which covers exports of agncultural commodities. b Includes cereals not elsewhere stated. 168 Table 27. Yields of Major Crops Average annual kilograms Index of yields Index of relative yields per hectare (1961-63 = 100) (world = 100) Crop 1961-63 1969-71 1977-79 1969-71 1977-79 1961-63 1969-71 1977-79 Beverages: Cocoa World 273 330 329 121 121 100 100 100 Developing countries 273 330 328 121 120 100 100 100 Sub-Saharan Africa 257 319 278 124 108 94 97 84 Coffee World 442 476 497 108 112 100 100 100 Developing countries 440 475 496 108 113 100 100 100 Sub-Saharan Africa 367 410 355 112 97 83 86 71 Tea World 863 827 881 96 102 100 100 100 Developing countries 868 1,006 1,236 116 142 101 122 140 Sub-Saharan Africa 781 1,153 1,245 148 159 90 139 141 Cereals: M,aize World 2,140 2,558 3,090 120 144 100 100 100 Developing countries 1,197 1,376 1,509 115 126 56 54 49 Sub-Saharan Africa 893 1,020 977 114 109 42 40 32 Millet World 561 617 636 110 113 100 100 100 Developing countries 527 575 568 109 108 94 93 89 Sub-Saharan Africa 600 586 561 98 94 107 95 88 Rice World 2,026 2,320 2,612 115 129 100 100 100 Developing countries 1,628 1,846 2,101 113 129 80 80 80 Sub-Saharan Africa 1,249 1,349 1,419 108 114 62 58 54 Sorghum World 912 1,103 1,320 121 145 100 100 100 Developing countries 638 734 965 115 151 70 67 73 Sub-Saharan Africa 751 693 701 92 93 82 63 53 Wheat World 1,179 1,524 1,784 129 151 100 100 100 Developing countries 998 1,193 1,443 120 145 85 78 81 Sub-Saharan Africa 785 1,255 1,084 160 138 67 82 61 Oils and Oilseeds: Groundnuts (in shell) World 874 926 979 106 112 100 100 100 Developing countries 822 840 882 102 107 94 91 90 Sub-Saharan Africa 832 776 797 93 96 95 84 81 Other: Pulses World 615 694 704 113 114 100 100 100 Developing countries 520 531 524 102 101 85 77 74 Sub-Saharan AfTica 439 381 393 87 90 71 55 56 Roots and Tubers, World 10 11 11 110 110 100 100 100 Developing countries 7 9 9 129 129 70 82 82 Sub-Saharan Africa 6 7 7 117 117 60 64 64 Seed Cotton World 923 1,067 1,214 116 132 100 100 100 Developing countries 665 790 842 119 127 72 74 69 Sub-Saharan Africa 425 565 528 133 124 46 53 43 Sugar Cane, World 49 54 56 110 114 100 100 100 Developing countries 47 51 54 109 115 96 94 96 Sub-Saharan Africa 49 58 59 118 120 100 107 105 Tobacco World 1,117 1,138 1,256 102 112 100 100 100 DevelopLng countries 817 835 984 102 120 73 73 78 Sub-Saharan Africa 767 732 828 95 108 69 64 66 a Metnc tons per hectare. 169 Table 28. Fertilizer Consumption Average annual consumption Average annual growth ratea (thousands of 1962-66b l69-71b metric tons) to to 1977-78b 1969-71 1977-78 Low-income countries 219.42 t 12.0 w 6.8 w Low-income semiarid 31.18 t 31.0 w 22.7 w 1. Chad 6.57 42.7 20.0 2. Somalia . . 22.7 3. Mali 13.00 47.6 19.7 4. Upper Volta 6.65 31.3 45.3 5. Gambia 2.95 22.5 37.8 6. Niger 0.81 18.4 30.7 7. Mauritania 1.20 .. 24.8 Low-income other 188.24 t 11.4 w 5.9 w 8. Ethiopia 27.99 23.6 26.0 9. Guinea-Bissau .. . 10. Burundi 0.65 .. 10.9 11. Malawi 22.42 11.0 14.1 12. Rwanda 0.31 .. 8.4 13. Benin 1.50 40.6 -13.5 14. Mozambique 12.85 4.8 9.3 15. Sierra Leone 2.04 42.0 0.5 16. Tanzania 34.59 20.5 15.7 17. Zaire 10.74 25.5 14.0 18. Guinea 1.45 0.5 -6.2 19. Central African Republic 2.20 27.8 1.9 20. Madagascar 9.60 22.5 -2.2 21. Uganda 2.13 8.7 -11.8 22. Lesotho 1.40 -0.8 19.1 23. Togo 2.37 .. 31.9 24. Sudan 56.00 8.8 5.3 Middle-income oil importers 403.25 t 27.0 w 10.7 w 25. Kenya 52.71 16.6 3.3 26. Ghana 25.05 5.2 40.7 27. Senegal 45.05 -0.8 23.7 28. Zimbabwe 113.50 9.4 2.5 29. Liberia 4.69 52.5 14.0 30. Zambia 67.20 15.9 14.9 31. Cameroon 22.22 21.2 3.5 32. Swaziland 5.20 4.3 -0.5 33. Botswana 2.05 1.9 3.7 34. Mauritius 23.03 -1.1 0.7 35. Ivorv Coast 42.55 9.5 14.4 Middle-income oil exporters 95.80 t 29.9 w 12.3 w 36. Angola 16.05 34.8 7.0 37. Congo 2.38 30.1 -11.3 38. Nigeria 76.50 26.2 33.0 39. Gabon O.B7 . Sub-Saharan Africa 718.47 t 23.1 w 9.7 w Developing countries 16,381.45 t 14.6 w' 10.2 wC World 97,469.10 t 9.0 w' 5.7 w' a Gro-th rates for groups of countries were calculated on the basis of those countries for which data on all three sub-periods were available. b See techmcal notes c Nonweighted mean. 170 Table 29. Agricultural Imports Volume Average annual growth rate 1962-63 1969-71 Value Thousands of metric tons to to (millions of dollars) Commodity 1961-63 1969-71 1977-79 1969-71 1977-79 1961-63 1969-71 1977-79 Rice Sub-Saharan Africa 464 680 1,696 4.9 12.1 63 91 619 Oil exporters 6 8 511 3.7 68.0 1 1 269 Other countries 458 672 1,185 4.9 7.3 62 90 350 Wheat Sub-Saharan Africa 394 1,043 2,352 12.9 10.7 30 81 473 Oil exporters 52 347 940 26.8 13.3 7 28 215 Other countries 342 696 1,412 9.3 9.2 23 53 258 Maize Sub-Saharan Africa 197 385 599 8.7 5.7 12 32 102 Oil exporters (.) 6 133 .-. 47.3 (.) 1 26 Other countries 197 379 466 8.5 2.6 12 31 76 Cereals not elsewhere stated Sub-Saharan Africa 123 239 215 8.7 - 1.3 9 21 40 Oil exporters 30 58 15 8.6 -15.6 2 6 3 Other countries 93 181 200 8.7 1.3 7 15 37 Cereals, total Sub-Saharan Africa 1,177 2,346 4,862 9.0 9.5 114 225 1,234 Oil exporters 88 419 1,599 21.5 18.2 10 36 513 Other countries 1,089 1,927 3,263 7.4 6.8 104 189 721 Dairy products Sub-Saharan Africa .. .. .. 7.2 7.0J 44 109 457 Oil exporters .. .. .. .. 17.0a 13 27 201 Other countries . . . . . . 9.8' 5.4' 31 82 256 Sugar Sub-Saharan Africa 670 816 1,281 2.5 5.8 84 109 429 Oil exporters 68 108 581 6.0 23.4 9 17 216 Other countries 602 708 700 2.1 -0.1 75 92 213 Meat Sub-Saharan Africa 38 42 114 1.3 13.3 20 24 176 Oil exporters 5 6 59 2.3 33.1 3 3 96 Other countries 33 36 55 1.1 5.4 17 21 80 Animal and vegetable oils Sub-Saharan Africa 51 122 325 11.5 13.0 12 32 204 Oil exporters 1 2 142 9.1 70.3 (.) 1 81 Other countries 50 120 183 11.6 5.4 12 31 123 Total agricultural imports" Sub-Saharan Africa .. .. .. (4.3)' (3.5)' 749 1,137 4,227 Oil exporters . . . . . (1.5)' (16.0)' 151 207 1,663 Other countries . . . . (5.1)' (2.8)C 598 930 2,564 a. See technical notes. b. Indudes products not listed above. c. Estimated from an average of price increases for the imports shown in the table and applied to the value of total agricultural imports. 171 Table 30. Agricultural Exports Average annual volume of exports Rate of growth Thousands (percent) SubSaharan exports as a of metric 1962-63 1969-71 percentage of total exports tons to to Of developing countries Of the world Commodity 1977-79 1969-71 1977-79 1961-63 1969-71 1977-79 1961-63 1969-71 1977-79 Beverages: Cocoa Sub-Saharan Africa 687 0.2 -2.6 81.2 76.7 72.2 79.9 75.9 69.2 Oil exporters 167 1.7 -3.4 Other countries 520 - 0.3 - 2.3 Coffee Sub-Saharan Africa 929 3.4 -0.6 26.0 30.0 29.2 25.6 29.3 27.5 Oil exporters 72 3.5 - 11.3 Other countries 857 3.4 1.0 Tea Sub-Saharan Africa 166 9.0 5.8 10.0 17.4 23.8 8.7 14.4 18.5 Oil exporters 0 23.0 Other countries 166 9.0 5.8 Cereals: Maize Sub-Saharan Africa 396 -1.9 -0.1 7.9 3.9 3.5 2.5 1.4 0.6 Oil exporters 0 2.5 0.0 Other countries 396 -3.9 5.9 Wheat Sub-Saharan Africa 2 1.2 -33.3 1.4 1.4 0.0 0.1 0.1 0.0 Oil exporters 0 0.0 0.0 Other countries 2 1.2 -33.3 Rice Sub-Saharan Africa 21 0.7 -13.4 1.4 1.8 0.4 0.9 0.8 0.2 Oil exporters 0 -3.5 Other countries 21 0.8 -13.1 Cereals not elsewhere stated Sub-Saharan Africa 123 -6.0 4.5 16.7 3.6 2.5 4.1 1.4 1.0 Oil exporters 0 -2.9 Other countries 123 -6.1 4.7 Oils and Oilseeds: Groundnut oil Sub-Saharan Africa 202 2.2 -1.3 61.2 70.2 50.1 53.8 57.6 40.6 Oil exporters (.) 3.2 -51.7 Other countries 202 1.7 4.1 Groundnuts (shelled) Sub-Saharan Africa 216 -6.1 -14.0 88.0 77.3 54.7 85.5 69.1 27.0 Oil exporters (.) -6.8 -58.1 Other countries 216 -5.5 -7.6 Oilseed cake and meal Sub-Saharan Africa 729 5.3 -2.8 15.9 17.1 6.3 9.5 8.3 3.5 Oil exporters 66 8.8 -12.0 Other countries 663 4.6 -1.2 Palm kernel oil Sub-Saharan Africa 79 8.9 -1.5 91.7 81.8 31.9 55.2 54.8 28.1 Oil exporters 35 31.8 0.9 Other countries 44 3.9 -3.1 Palm kernels Sub-Saharan Africa 175 -6.2 -9.3 90.8 82.4 84.2 90.4 82.2 83.7 Oil exporters 124 -8.0 -6.7 Other countries 51 -3.1 -13.7 Palmoil Sub-Saharan Africa 94 -8.6 -6.4 57.0 17.1 4.5 55.0 16.4 4.3 Oil exporters 1 - 21.2 -30.5 Other countries 93 -2.9 -4.7 Sesame Seed Sub-Saharan Africa 90 3.8 -6.9 72.8 78.1 42.5 68.6 75.3 41.8 Oil exporters 3 -5.6 -17.1 Other countries 87 5.3 -6.3 (table continues on the following page) 172 Table 30. (continued) Average annual volume of exports Rate of growth Sub-Saharan exports as a Thousands (percent) percentage o total exports of metric 1961-63 1969-71 tons to to Of developing countries Of the world Commodity 1977-79 1969-71 1977-79 1961-63 1969-72 1977-79 1961-63 1969-71 1977-79 Other: Bananas Sub-Saharan Africa 327 - 1.7 - 2-2 11.3 7.1 4.9 10.9 6.5 4.7 Oil exporters 50 40.0 4.3 Other countries 277 - 2.8 - 3.0 Cotton Sub-Saharan Africa 463 5.6 -3.4 18.4 23.1 23.2 10.8 15.5 11.0 Oil exporters 23 1.5 -8.9 Other countries 440 6.0 -3.0 Rubber Sub-Saharan Africa 156 3.0 -2.9 7.6 7.0 4.8 6.8 6.8 4.6 Oil exporters 30 - 1.0 - 7.5 Other countries 126 5.1 -1.4 Sisal Sub-Saharan Africa 139 -2.1 -10.1 65.4 61.1 57.1 60.7 59.7 56.2 Oil exporters 13 -0.8 -17.1 Other countries 126 -2.4 -9.0 Sugar Sub-Saharan Africa 1,189 3.2 0.3 11.4 12.8 9.7 4.7 5.6 4.3 Oil exporters 11 8.1 -23.8 Other countries 1,178 2.9 1.3 Tobacco Sub-Saharan Africa 137 -3.1 6.5 22.2 14.6 18.0 12.1 8.2 10.0 OI expOrters 2 6.9 0.3 Other countries 135 -3.3 6.6 173 Table 31. Domestic Terms of Trade of Export Crops for Selected Countries 1971 1972 1973 1974 1975 1976 1977 1978 1979 (1970 = 100 unless otheroise specified) Cameroon Barter terms of trade 98.2 90.5 85.2 81.2 74.5 73.4 77.5 88.2 Income terms of trade 89.8 96.9 84.0 81.8 81.1 66.4 67.2 91.7 (cocoa, coffee, cotton) Ghana Barter terms of trade 91.2 82.8 88.0 89.4 86.0 58.8 34.0 35.8 46.3 Income terms of trade 85.9 92.4 88.4 75.2 78.0 56.1 26.0 22.6 27.7 (cocoa) Ivory Coast Barter terms of trade 111.2 109.3 97.8 99.7 126.5 111.8 97.3 119.3 101.6 Income terms of trade 113.4 134.1 111.6 134.3 171.5 178.0 144.3 170.6 131.2 (cocoa, coffee, cotton, palmoil) Kenya Barter terms of trade 98.3 93.1 79.3 83.9 123.4 93.6 55.0 49.5 58.8 Income terms of trade 129.2 157.0 177.9 198.5 170.1 280.0 449.6 263.5 218.0 (coffee, tea, pyrethrum, cotton, maize, wheat, sisal) Malawi Barter terms of trade 105.9 129.1 123.0 100.1 94.3 115.4 119.0 116.7 Income terms of trade 108.3 122.2 139.4 104.5 104.6 98.6 120.5 124.6 (tobacco, groundnuts, cotton, maize) Mali Barter terms of trade 81.2 80.6 69.0 61.2 83.1 79.4 65.3 60.0 50.0 Income terms of trade 99.2 98.8 76.8 55.7 94.1 135.8 123.5 90.1 76.4 (cotton, groundnuts) Nigeria Barter terms of trade .. .. 96.5 119.2 125.2 95.9 119.1 93.0 109.0 Income terms of trade .. .. 75.0 102.5 104.6 79.4 80.0 43.6 58.1 (cocoa, cotton, palm kernels) Senegal Barter terms of trade 97.3 111.9 104.2 114.1 120.6 115.1 103.4 101.5 91.3 Income terms of trade 77.3 146.1 81.4 84.5 141.1 204.2 148.7 72.2 104.7 (groundnuts, cotton) Tanzania Barter terms of trade 96.9 95.3 88.5 76.0 63.5 90.7 110.0 79.5 67.4 Income terms of trade 101.8 102.8 96.3 78.8 68.8 84.5 102.5 73.7 62.7 (coffee, tobacco, cashews, cotton) Togo Barter terms of trade 98.5 92.8 88.5 80.0 79.2 76.2 68.4 80.6 90.1 Income terms of trade 110.5 92.8 64.1 57.2 57.0 62.0 49.4 48.8 57.8 (cocoa, coffee, cotton) Upper Volta Barter terms of trade 99.0 101.7 102.8 108.0 91.9 101.7 105.5 101.7 92.6 Income terms of trade 106.9 130.7 110.0 140.2 181.6 214.0 156.0 229.5 245.8 (cotton, sesame) Zambia (1971 = I00) Barter terms of trade 100.0 82.1 113.6 104.8 91.8 84.7 104.6 98.1 127.2 Income terms of trade 100.0 142.6 97.9 126.1 125.6 169.6 133.1 92.9 90.4 (maize, groundnuts, tobacco) 174 Table 32. Procurement and Distribution of Agricultural Inputs Fertilizer supply Seed supply Chemical supply Farm equipment supply Govern- Govern- Govern- Govern- Private ment Mixed Private ment Mixed Private ment Mixed Private ment Mixed Low-income countries Low-income semiarid 1 Chad X X X X 2. Somalia X X X X 3. Mali X X X X 4. Upper Volta X X X X 5. Gambia X X X X 6. Niger X X X X 7. Mauritania X X X X Low-income other 8. Ethiopia X X X X 9. Guinea-Bissau X X X X 10. Burundi X X X X 11. Malawi X X X X 12. Rwanda X X X X 13. Benin X X X X 14. Mozambique . 15. Sierra Leone X X X X 16. Tanzania X X X X 17. Zaire X X X X 18. Guinea X X X X 19. Central African Rep. X X X X 20. Madagascar X X X X 21. Uganda X X X X 22. Lesotho X X X X 23. Togo X X X X 24. Sudan X X X X Middle-income oil importers 25. Kenya X X X X 26. Ghana X X X X 27. Senegal X X X X 28. Zimbabwe X X X X 29. Liberia X X X X 30. Zambia X X X X 31. Cameroon X X X X 32. Swaziland X X X X 33. Botswana X X X X 34. Mauritius X X X X 35. Ivory Coast X X X X Middle-income oil exporters 36. Angola . . 37. Congo X X X X 38. Nigeria X X X X 39. Gabon . . Percent of Sub-Saharan countries 11 64 25 11 61 28 17 47 36 22 42 36 175 Table 33. Population Growth, Past and Projected Hypothetical Assumed size of year of Year of Average annual growth stationary reaching net reaching of population (percent) Projected population (millions) population reproduction stationary 1960-70 1970-79 1980 2000 (millions) rate of 1 population Low-income countries 2.4 w 2.6 w 189 t 334 t Low-income semiarid 2.3 w 2.3 w 29 t 49 t 1. Chad 1.8 2.0 4 7 19 2045 2140 2. Somalia 2.4 2.3 4 6 17 2040 2130 3. Mali 2.4 2.6 7 12 35 2040 2130 4. Upper Volta 1.6 1.6 6 10 28 2040 2130 5. Gambia- 3.2 3.0 0.6 1 3 2045 2135 6. Niger 3.3 2.8 5 10 29 2040 2130 7. Mauritania 2.5 2.7 2 3 9 2045 2135 Low-income other 2.4 w 2.6 w 161 t 285 t 8. Ethiopia 2.4 2.1 31 53 162 2045 2140 9. Guinea-Bissau- 2.6 1.6 0.8 1 3 2040 2135 10. Burundi 1.6 2.0 4 7 17 2040 2135 11. Malawi 2.8 2.8 6 11 36 2040 2110 12. Rwanda 2.8 . 2.8 5 9 29 2040 2110 13. Benin 2.6 2.9 4 6 19 2040 2110 14. Mozambique 2.2 2.5 10 20 51 2040 2130 15. Sierra Leone 2.2 2.5 3 6 17 2040 2130 16. Tanzania 2.7 3.4 19 35 97 2035 2100 17. Zaire 2.0 2.7 28 49 139 2040 2130 18. Guinea 2.8 2.9 5 9 23 2040 2130 19. Central African Republic 2.2 2.2 2 3 9 2040 2130 20. Madagascar 2.1 2.5 9 15 45 2040 2110 21. Uganda 3.7 3.0 13 24 67 2035 2100 22. Lesotho 2.0 2.3 1 2 5 2035 2105 23. Togo 2.7 2.4 2 4 13 2040 2110 24. Sudan 2.2 2.6 18 31 86 2040 2105 Middle-income oil importers 2.8 w 3.2 w 69 t 128 t 25. Kenya 3.2 3.4 16 34 109 2035 2095 26. Ghana 2.4 3.0 12 21 52 2035 2100 27. Senegal 2.4 2.6 6 10 30 2045 2135 28. Zimbabwe 3.9 3.3 7 15 42 2035 2095 29. Liberia 3.1 3.3 2 4 11 2035 2095 30. Zambia 2.8 3.0 6 11 31 2035 2125 31. Cameroon 1.8 2.2 8 14 37 2040 2130 32. Swaziland- 2.2 2.6 0.6 1 3 2010 2075 33. Botswana** 1.9 2.2 0.9 2 6 2040 2130 34. Mauritius- 2.2 1.4 1 1 2 2010 2075 35. Ivory Coast 3.7 5.5 9 15 45 2040 2110 Middle-income oil exporters 2.4 w 2.5 w 95 t 177 t 36. Angola 1.5 2.3 7 12 35 2045 2135 37. Congo 2.1 2.5 2 3 7 2040 2130 38. Nigeria 2.5 2.5 85 161 459 2035 2105 39. Gabon 0.4 1.2 0.7 1 2 2040 2155 Sub-Saharan Africa 2.5 w 2.7 w 353 t 639 t Low-income countries 2.2 w 2.1 w 2,300 t 3,275 t- Middle-income countries 2.5 w 2.4 w 1,008 r 1,569 t- Industrialized countries 1.0 w 0.7 w 675 tP 744 t a. Does not include pTojections foT countries of less than one nillion population in 1979. 176 Table 34. Demographic and Fertility-related Indicators Crude death Crude birth rate rate per Percentage change in Total per thousand thousand Crude birth Crude death fertility Life expectancy Child death rate population population rate rate rate at birth (aged 1-4) 1960 1979 1960 1979 1960-79 1960-79 1979 1960 1979 1960 1979 Low-income countries 48 w 47w 26 w 19 w -2.0 w -24.6 w 6.4 w 38 w 46 w 40 w 27 w Low-income semiarid 49 w 48 w 28 w 22 w -1.8 w -20.2 w 6.5 w 37 w 43 w 42 w 31 w 1. Chad 45 44 29 24 -2.4 -18.4 5.9 35 41 45 35 2. Somalia 49 46 29 20 -5.9 -30.0 6.1 36 44 43 30 3. Mali 50 49 27 22 -0.8 -18.9 6.7 37 43 41 31 4. Upper Volta 49 48 27 21 -1.4 -19.2 6.5 37 43 41 31 5. Gambia** 48 48 26 23 - 0.6 -14.8 6.4 37 42 41 34 6. Niger 52 52 27 22 -0.6 -18.1 7.1 37 43 41 31 7. Mauritania 51 50 27 22 - 0.8 -19.4 6.9 37 43 41 29 Low-income other 48 w 47 w 25 w 19 w -2.0 w -25.4 w 6.4 w 39 w 47 w 40 w 26 w 8. Ethiopia 51 50 28 24 -1.8 - 13.2 6.7 36 40 43 36 9. Guinea-Bissau- 40 41 31 23 0.7 -26.7 5.5 32 42 10. Burundi 47 45 27 22 -3.2 -15.8 5.9 37 42 41 33 11. Malawi 53 51 27 19 -3.6 -31.0 7.0 37 47 41 25 12. Rwanda 51 50 27 19 -2.9 -30.5 6.9 37 47 41 25 13. Benin 51 49 27 19 -3.6 -30.2 6.7 37 47 41 25 14. Mozambique 46 45 26 18 -2.4 -29.5 6.1 37 47 41 25 15. Sierra Leone 47 46 27 19 -2.8 -30.5 6.1 37 47 41 25 16. Tanzania 47 46 22 15 -0.6 -31.5 6.5 42 52 32 18 17. Zaire 48 46 24 18 -4.6 -25.2 6.1 40 47 36 25 18. Guinea 47 46 30 20 -1.9 -32.8 6.2 35 44 45 28 19. Central African Republic 43 44 28 21 3.0 - 23.8 5.9 36 44 43 30 20. Madagascar 47 46 27 18 - 1.9 - 31.6 6.5 37 47 41 25 21. Uganda 45 45 20 14 -0.4 -32.0 6.1 44 54 29 16 22. Lesotho 40 40 23 16 -1.7 -30.7 5.4 42 51 33 20 23. Togo 51 48 27 18 -5.3 -30.9 6.5 37 47 41 25 24. Sudan 45 46 25 18 0.9 -26.2 6.6 39 47 47 29 Middle-income oil importers 48 w 48 w 24 w 16 w -2.3 w -32.5 w 6.7 w 40 w 50 w 36 w 21 w 25. Kenya 52 51 24 13 -1.3 -42.7 7.8 41 55 34 15 26. Ghana 49 48 24 17 -1.4 -30.0 6.7 40 49 36 22 27. Senegal 48 48 26 21 0.2 -18.0 6.5 37 43 41 31 28. Zimbabwe 47 47 19 13 0.6 -30.9 6.6 45 55 28 15 29. Liberia 50 48 21 14 -4.2 -33.0 6.9 44 54 29 16 30. Zambia 51 49 24 17 -2.8 -31.4 6.9 40 49 36 22 31. Cameroon 43 42 27 19 -1.2 -30.2 5.7 37 47 41 25 32. Swaziland- 49 48 27 18 -2.9 -31.5 6.4 38 47 41 27 33. Botswana- 50 51 24 17 2.0 -30.7 6.7 40 49 36 23 34. Mauritius- 41 25 10 6 -38.6 -37.0 3.0 60 65 11 4 35. ivory Coast 50 47 26 18 -6.4 -32.0 6.7 37 47 41 25 Middle-income oil exporters 52 w 50 w 26 w 17 w -4.1 w -31.2 w 6.8 w 39 w 48 w 37 w 23 w 36. Angola 50 48 31 22 -4.0 -27.5 6.4 33 42 49 33 37. Congo 46 45 27 18 -2.2 -29.8 6.0 37 47 41 27 38. Nigeria 52 50 25 17 -4.2 -31.6 6.9 39 49 36 Z2 39. Gabon- 32 33 28 21 0.9 -26.0 4.3 36 45 43 30 Sub-Saharan Africa 49 w 48 w 25 w 18 w -2.6 w -27.8 w 6.6 w 39 w 47 w 38 w 25 w All low-income countries 40 w 29 w 18 w 11 w -27.5 w -38.2 w 4.5 w 42 w 57 w 23 w 11 w All middle-income countries 41 w 34 w 15 w 10 w -15.7 w -32.7 w 4.9 w 53 w 61 w 19 w 10 w Industrialized countries 20 w 15 w 10 w 10w -27.5 w -2.2 w 1.9 w 70 w 74 w 1 w 1 w 177 Table 35. Labor Force Percentage of population of working age Percentage of labor force Average annual growth of labor (15-64 years) In agriculture In industry In services force (percent) 1960 1979 1960 1979 1960 1979 1960 1979 1960-70 1970-80 1980-2000 Low-income countries 54 w 53 w 87 w 79 w 5 w 9 w 7 w 11 w 2.0 w 2.1 w 2.8 w Low-income semiarid 54 w 53 w 93 w 86 w 5 w 7 w 4 w 7 w 1.9 w 2.0 w 2.7 w I.Chad 57 54 95 85 2 7 3 8 1.5 1.8 2.4 2. Somalia 54 54 88 84 4 8 8 8 1.7 2.2 2.0 3. Mali 54 52 94 88 3 5 3 7 2.0 2.2 2.9 4. Upper Volta 54 53 92 83 5 12 3 5 1.2 1.2 2.7 5. Gambia** 54 53 85 79 7 10 8 11 .. . 6. Niger 53 51 95 91 1 3 4 6 3.0 2.6 3.4 7. Mauritania 53 52 91 85 3 5 6 10 2.2 2.4 2.8 Low-income other 54 w 53 w 86 w 78w 6 w 9 w 8 w 12 w 2.0 w 2.1 w 2.8 w 8. Ethiopia 54 53 88 80 5 7 7 13 2.0 1.7 2.6 9. Guinea-Bissau- 61 61 . . 93 .. 1 .. 6 .. . 10. Burundi 55 55 90 84 3 5 7 11 1.2 1.5 2.2 11. Malawi 52 49 92 86 3 5 5 9 2.3 2.2 3.3 12. Rwanda 53 51 95 91 1 2 4 7 2.4 2.4 3.2 13. Benin 53 51 54 46 9 16 37 38 2.1 2.3 2.6 14. Mozambique 56 53 81 67 8 17 11 16 1.9 1.7 2.4 15. Sierra Leone 55 53 78 66 12 19 10 15 1.5 1.8 2.7 16. Tanzania 54 51 89 83 4 6 7 11 2.1 2.7 3.1 17. Zaire 53 53 83 75 9 13 8 12 1.4 2.1 2.7 18. Guinea 55 53 88 82 6 11 6 7 2.5 2.2 2.3 19. Central African Rep. 58 55 94 88 2 4 4 8 1.7 1.7 2.3 20. NMadagascar 55 53 93 87 2 4 5 9 1.7 2.0 2.8 21. Uganda 54 52 89 83 4 6 7 11 3.3 2.5 3.3 22. Lesotho 57 55 93 87 2 4 5 9 1.6 1.9 2.4 23. Togo 53 51 80 68 8 15 12 17 2.2 1.7 2.9 24. Sudan 53 53 86 78 6 10 8 12 2.2 2.4 2.7 Middle-income oil importers 53 w 51 w 79 w 71 w 7 w 11 w 14 w 18 w 2.3 w 2.6 w 3.1 w 25. Kenya 50 48 86 78 5 10 9 12 2.7 2.8 3.9 26. Ghana 53 51 64 54 14 20 22 26 1.6 2.4 3.2 27. Senegal 54 53 84 76 5 10 11 14 1.9 1.9 2.5 28. Zimbabwe 52 50 69 60 11 15 20 25 3.2 2.6 3.5 29. Liberia 52 50 80 71 10 14 10 15 2.4 2.6 3.5 30. Zambia 53 50 79 68 7 11 14 21 2.3 2.4 3.0 31. Cameroon 57 54 87 83 5 7 8 10 1.3 1.3 1.8 32. Swaziland- 54 52 54 52 4 9 42 39 .. 33. Botswana- 51 48 92 83 3 5 5 12 .. . 34. Mauritius- 51 61 40 30 26 24 35 46 .. . 35. Ivory Coast 54 54 89 79 2 4 9 17 3.6 5.0 2.8 Middle-income oil exporters 52 w 50 w 71 w 55 w 10 w 18 w 19 w 27 w 1.7 w 1.7 w 3.2 w 36. Angola 55 53 69 60 12 16 19 24 1.0 1.9 2.7 37. Congo 56 53 52 35 17 26 31 39 1.5 2.0 2.9 38. Nigeria 52 50 71 55 10 18 19 27 1.8 1.7 3.3 39. Gabon- 62 61 85 79 7 10 8 11 Sub-Saharan Africa 53 w 52 w 81 w 71 w 7 w 12 w 12 w 17 w 2.0 w 2.1 w 3.0 w All low-income countries 56 w 58 w 76 w 71 w 10 w 14 w 14 w 15 w 1.6 w 1.9 w 1.6 w All middle-income countries 55 w 55 w 59 w 43 w 17 w 23 w 25 w 34 w 1.9 w 2.3 w 2.6 w Industrialized countries 63 w 66 w 16 w 6 w 39 w 38 w 45 w 56 w 1.2 w 1.2 w 0.6 w 178 Table 36. Urbanization Urban population Percentage of urban population Number of cities As percentage of Average annual In cities of over of over 500,000 total population growth rate (percent) In largest city 500,000 persons, persons 1960 1980 1960-70 1970-SO 2960 1980 1960 1980 2960 1980 Low-income countries 9 w 18 w 5.8 w 6.5 w 24 w 40 w . . .. t l t Low-income semiarid 9 w 17 w 5.9 w 5.8 w 10 w 36 w .. . 1. Chad 7 18 6.7 6.5 .. 39 2. Somalia 17 30 5.3 5.0 .. 34 3. Mali 11 20 5.4 5.5 32 34 4. Upper Volta 5 9 5.3 3.8 .. 41 5. Gambia- 12 .. 5.1 5.0 . 6. Niger 6 13 7.0 6.8 .. 31 7. Mauritania 3 23 15.8 8.6 .. 39 Low-income other 9 w 18 w 5.8 w 6.7 w 26 w 41 w .. 37w i t l t 8. Ethiopia 6 15 6.1 6.6 30 37 37 1 9. Guinea-Bissau- 14 .. 2.2 4.3 . 10. Bur-undi 2 2 1.6 2.5 . 11. Malawi 4 20 6.6 6.8 . 19 12. Rwanda 2 4 5.6 5.9 13. Benin 10 14 5.3 3.9 .. 63 63 1 14. Mozambique 4 9 6.6 6.8 75 83 83 1 15. Sierra Leone 13 25 5.5 5.6 37 47 16. Tanzania 5 12 6.3 8.7 34 50 50 1 17. Zaire 16 34 5.2 7.2 14 28 14 38 1 2 18. Guinea 10 18 6.2 5.5 37 80 80 1 19. Central African Republic 23 41 5.3 5.0 40 36 20~ Madagascar 11 18 5.0 5.2 44 36 36 1 21. Uganda 5 12 7.8 7.0 38 52 52 1 22. Lesotho 2 5 7.5 7.7 . 23. Togo 10 20 5.6 6.6 .. 60 24. Sudan 20 25 6.9 6.8 30 32 31 1 Middle-income oil importers 16 w 28w 5.3 w 6.2 w 28 w 37w . 40 w . 8 t 25. Kenva 7 14 6.4 6.8 40 57 57 1 216. Ghana 23 36 4.6 5.2 25 35 48 2 27. Senegal 23 25 2.9 3.3 53 65 65 1 28. Zimbabwe 13 23 6.8 6.4 40 50 50 1 29. Liberia 21 33 5.6 5.6 . 30. Zambia 23 38 5.4 5.5 .. 35 35 1 31. Cameroon 14 35 5.6 7.5 26 21 21 1 32. Swaziland- 4 .. 8.8 4.2 . 33. Botswana" 2 .. 17.6 15.0 . 34. Mauritius- 33 .. 4.6 3.6 35. Ivory Coast 19 38 7.3 8.5 27 34 34 1 Middle-income oil exporters 13 w 20 w 4.7 w 4.8 w 18 w 22 w .. 56 w 2 t lO t 36. Angola 10 21 5.1 5.7 44 64 64 1 37. Congo 30 45 4.7 4.1 77 56 38. Niger-ia 13 20 4.7 4.7 13 17 22 58 2 9 39. Gabon" 17 .. 4.3 4.5 . Sub-Saharan Africa 11 w 21 w 5.3 w 5.9 w 23 w 34 w 8 w 41 w 3 t 28 t All low-income countries 15 w 17 w 3.7 w 3.8 w 11 w 13 w 31 w 42 w 58 t 144 t All middle-income countries 37 w 50 w 4.1 w 3.8 w 28 w 29 w 35 w 48 w 56 t 125 t Industrialized countries 68 w 77 w JF53w1.8 w 1.3 w 18 w 18 w 48 w 55 w 9t 146 t a. Spaces are left bla.nk to indicate an absence of cities of more than 500,000 persona. 179 Table 37. Health-related Indicators Daily calorie supply per capita Percentage of As Population, population percentage of zwith access total Per physician Per nurse to safe water Total requirement 1960 1977 1960 1977 1975 1977 1977 Low-income countries 50,788 w 32,241 w 7,558 w 3,670 w 24 w 2,072 w 90 w Low-income semiarid 67,302 w 36,781 w 6,157 w 4,498 w 23 w 1,995 w 85 w 1 Chad 72,190 41,940 8,040* 4,810' 26 1,762 74 2 Somalia 36,570 6,220 .. 33 2,033 88 3. Mali 67,050 25,150 4,980* 3,230' 9 2,117 90 4. Upper Volta 81,650 49,810* 4,090 4,510' 25 1,875 79 5. Gambia" 21,800 13,171* 3,930 97 6. Niger 82,170' 42,720 8,450* 6,270' 27 2,139 91 7. Mauritania 40,400 15,160 7,320* 3,430 .. 1,976 86 Low-income other 47,756 w 31,539 w 7,818 w 3,533 w 24 w 2,085 w 91 w 8. Ethiopia 100,470 75,320 14,920' 5,400' 6 1,754 75 9. Guinea-Bissau- . 10,094 1,258' 8' 2,009b 86b 10. Burundi 96,570* 45,020* 6,770* 6,180' . . 2,254 97 11. Malawi 35,250 40,680* 12,920* 2,790' 33 2,066 90 12. Rwanda 138,100* 38,920 11,200 10,490 35 2,264 98 13. Benin 23,030 26,880 .. 3,040' 21* 2,249 98 14. Mozambique 20,390 33,980 4,720* .. .. 1,906 81 15; Sierra Leone 20,420 . 5,900* .. .. 2,150 93 16. Tanzania 18,220 17,550 10,440 3,080 39 2,063 89 17. Zaire 37,620* 15,530' 3,510* 1,940' 16 2,271 104 18. Guinea 48,000* 16,630 3,260* 2,490 10 1,943 84 19. Central African Republic 41,580 17,610 2,760* 1,560 16 2,242 99 20. Madagascar 8,900 10,240 3,110* 3,470' 26 2,486 115 21. Uganda 14,060 27,600 9,420' 4,300 35 2,110 91 22. Lesotho 23,510 18,640 .. 4,340 17 2,245 99 23. Togo 35,760* 17,980 5,340* 2,000 16 2,069 90 24. Sudan 33,500 8,690 3,040* 1,280 46 2,184 93 Middle-income oil importers 20,971 w 11,877 w 4,321 w 1,551 w 28 w 2,181 w 94 w 25. Kenva 10,690 11,630* 2,230* 1,090* 17 2,032 88 26. Ghana 21,600 9,920 5,430' 860* 35 1,983 86 27. Senegal 24,540 15,710 4,110 1,660* 37 2,261 95 28. Zimbabwe 4,790 7,030* 1,010 1,380* .. 2,576 108 29. Liberia 12,600* 9,260* 5,810 2,900* 20 2,404 104 30. Zambia 9,540 10,190* 9,920* 1,930* 42 2,002 87 31. Cameroon 48,110* 16,500 6,150* 2,230 26 2,069 89 32. Swaziland" 10,134 9,185* 3,684 936 37 . 99 33. Botswana" 26,200 9,597 2,380 1,270 45 . 94 34. Mauritius- 4,662 2,410 2,102 641* 60 .. 114 35. ivorv Coast 29,190* 15,220* 2,920 2,370* 19 2,517 105 Middle-income oil exporters 67,250 w 15,494 w 5,889 w 3,973 w 1,970 w 84 w 36. Angola 14,910 .. .. .. .. 2,133 91 37. Congo 16,430 7,290* 1,510* 800* 17* 2,284 103 38. Nigeria 73,710 15,740* 6,020' 4,030* .. 1,951 83 39. Gabon- 9,772 3,029 1,410 .. .. .. 88 Sub-Saharan Africa 50,096 w 23,904 w 6,533 w 3,315 w 25 w 2,065 w 89 w All low-income countries 11,680 w 6,150 w 5,700 w 6,200 w 29 w 2,231 w 98 w All middle-income countries 10,430 w 4,380 w 3,390 w 1,820 w 58 w 2,581 w 109 w Industralized countries 830 w 620 w 450 w 220 w .. 3,377 w 131 w a. Figures marked with an * are for years other than specified. b. Average 1976-79. 180 Table 38. Education Number Number enrolled in enrolled in higher Nunier er-oled i pritiar scioofsecondary education as Numberenrolged on piag ryS/OO school as percentage of as percentageof age grouppercentage of population aged Adult literacy Total Male Female age group 20-24 rate (percent) 1960 1978, 1960 1978, 1960 1978, 1960 1978- 2960 1977- 1960, 2976, Low-incomiecourttries 30 w 56 w 42 w 70 w 19 w SO w 2 w l w .w lw 15w 26 w Low-income semiarid lO w 28 w 15 w 37 w 5 w 19 w (.w 5 w .w (.w 3 w 20 w 1. Chad 17 35* 29 52* 4 19* () 3* . 6 15* 2. Somalia 9 44' 13 57' 5 32* 1 4' () 1* 2* 64) 3. Mali 10 28 14 36 6 20 1 9' . 1 3' 10 4. Upper Volta 8 17 12 21 5 12 (.) 2 () 2* 5* 5. Gambia** 14 37 . .. 3 12 .. .. 6 10 6. Niger 5 23 7 29 3 17 .) 3' . . 1 8 7. Mauritania 8 26* 13 34* 3 17* ( 5* ( 5 17* Low-income other 34 w 61 w 47 w 78 w 22 w 57 w 2 w 11 w . w 1w 19 w 27 w 8. Ethiopia 7 38* 11 .. 3 () 9* (* * 15* 9. Guinea-Bissau- 25 112 . .. . .. 3 10 . 7* 10. Burundi 1s 21 27 26 9 17 1 3 () .) 14* 25 11. Malawi 59* 73* . 51* 1 4' . ) 25' 12. Rwanda 49 64' 68 68* 30) 59' 2 2* .16.) 16 13. Benin 26 60* 38 78* 15 42* 2 12* 1' 8' 7* 14. Mozambique 48 . 60 . 36 . 2 ..()' 1 15. Sierra Leone 23 37* 30 45' 15 30* 2 12* . 1 7 16. Tanzania 25 70)* 33 80' 18 61' 2 4* . 10' 66' 17. Zaire 60 90' 88 103* 32 77* 3 19* . 31* 15 18. Guinea 30 34* 44 46* 16 22' 2 16* 7*b 7 20* 19. Central African Republic 32 78' 53 102* 12 55* 1 9* 1 7* 20. Madagascar 52 94' 58 100' 45 87* 4 12' () 2 . . 54Y 21. Uganda 49 54)* 65 58* 32 41' 3 5* G) 1 35' 22. Lesotho 83 101' 63 82* 102 122' 3 17* . 52* 2.3. Togo 44 102* 63 129' 24 75' 2 25' .. 1 10 18 24. Sudan 25 50 35 58 14 42 3 16 () 2 13' 20* Nliddle-income oil importezs 51 w 85 w 66 w 84 w 35 w 75 w 4 w 19 w O w 1 w 20 w 34 w 25. Kenva 47 99 6.4 105 30 94 2 18 (.) 1* 20* 45' 26. Ghana 38 71* 52 80' 25 61' 5 32' (.) 27 27. Senegal 27 41* 36 54)* 17 32* 3 10' 1 2 6* 10* 28. Zimbabwe 96 97* 107 105* 86 90* 6 9' (.) .. 39* 29. Liberia 31 64 45 80 18 48 2 20 (. 2' 9' 30 30. Zambia 42 98* 51 106* 34 89* 2 16* .. 2 39* 31. Cameroon 65 101' 87 42' 43 91' 2 16' 1 19* 32. Swaziland- 58 92 . .. 5 32 . 3 65 33. Botswana- 42 89 . . 1 20 .35 34. Ma uri tiu s- 98 104 . . . 24 51 2 . so 35. Ivory Coast 46 71' 68 88* 24 69* 2' 14 (. 2' 5' 20 Middle-incomneoil exporters 36 w 64 w 46 w .w 28 w . w 4 w 14 w (.)w 1 w 14 w w 36. Angola 21 28 13 .. 2 ()5* 37. Congo 78 156 103 163 53 148 4 69 1 3' 16* 38. Nigeria 36 62* 46 . 27 .. 4 13* . 1 15' 39. Gabon" 85 202 . . 5 34 3 12 12' Sub-Saharan Africa 36 w 63 w 47w 75 w 24 w 58 w 3 w 13 w .w I1w 16 w 28 w Aflllow-incomnecountries 76 w 83 w 71 w 92 w 37 w 63 w 14 w 36w 2w 3 w 28 w 51 w Aillmiddle-income countries 79 w 95 w 85 w 103 w 72 w 94 w 16 w 41w 4w 11 w 53 w 72 w [ndusftriized countTies 114 w 100 w 109 w 102 w 108 w 102 w 68 w 89 w 17 w 37 w . w 99 w a. Figures marked with an are for years other than specified. b. Pencentage of population aged 19-22. 181 Table 39. Central Government Budgetary Operations Percentage of GDP 1977 Expenditure Revenues Net Nontax Financing Capita! Current lending Tax current, Capital Grants Foreign Domestic Low-income countries 5.6 in 18.5 m 0.2 mn 14.3 mn 2.5 mn 0.0 mn 1.2 m 1.8 m 1.9 m Low-income semiarid 3.2 m 16.1 m 0.6 mn 16.6 mn 3.4 mn 0.0 mn 1.3 mn 1.2 in 1.9 m 1. Chadb 3.5 16.1 0.0 10.9 1.9 0.0 4.4 1.2 2.2 2. Somalia 7.8 28.9 0.0 23.9 6.6 0,3 1.3. 3. Mali 1.6 18.4' 0.3 16.6 2.1 0.0 1.5 -0.3 0.4 4. Upper Volta 2.8 12.7 1.3 14.6 2.5 0.0 0.5 5. Gambia 12.2 21.1 2.0 20.7 4.3 0.0 0.0 2.3 8.0 6. Niger 2.7 12.0 0.6 13.5 3.4 0.0 0.2 0.0 1.9 7. Mauritania .. 51.2' 3.3 20.9 7.5 -0.4 14.1 4.5 7.9 Low-income other 6.5 m 18.5 m 0.1 mn 13.3 mn 2.0 in 0.0 mn 0.9 in 2.5 mn 3.3 mn 8. Ethiopia 3.4 16.0 0.1 12.8 2.0 0.0 1.2 1.5 1.9 9. Guinea-Bissau" 11.1 39.2 . 11.6 6.7 .. . 33.1. 10.7 10. Burundi . . 21.1' 0.0 13.3 0.8 0.1 5.3 1.8 -0.2 11. Malawi 8.1 12.7 1.5 11.6 2.6 0.0 2.1 5.2 0.7 12. Rwanda 4.6 8.9 0.2 11.6 0.6 0.0 0.0 3.1 -1.7 13. Benin" 12.4 0.0 12.2 2.5 0.0 0.0. 14. Mozambique .. .. .. .. . 15. Sierra Leone 4.6 18.6 0.0 14.8 1.5 0.0 0.0. 16. Tanzania 7.8 18.5- 0.2 15.8 2.9 0.0 2.2 17. Zaire 6.5 23.7 0.0 16.1 0.3 0.0 3.3 3 .7 6.8 18. Guinea.. . . . ... 19. Central Afr-ican Republic . .. .... .. . ... 20. Miadagascar.. . ... .. .. ... 21. Uganda .. .. . 2.Lesotho .. . .. 23. Togo 5 .6 27.3 0.3 26.3 2.0 0.0 0.0 0.2 4.7 24. Sudan 10.5 15.2 0.1 14.0 2.6 0.0 0.6 0.8 7.8 Middle-income oil importers 9.3 in 15.8 mn 2.2 m 20.0 mn 2.1 m 0.0 mn 0.5 mn 2.7 mn 1.0 mn 25. Kenva 4.5 15.2 1.2 14.4 2.3 0.0 0.7 1.2 2.4 26. Gha na .. 13.0 0.5 6.3 0.6 0.0 0.2 0.0 6.4 27. Senegal .. 16.4 0.9 17.6 1.0 0.0 0.0 28. Z-imbabwe . . .. .. . 219. Liberia 7.3 23.8 2.7 18.4 0.7 0.0 1.8 4.4 -2.6 30. Zambia 6.2 30.5 4.5 23.3 2.6 0.0 1.7 12.7 1.0 31. Cameroon 13.4 6.5 0.3 18.0 1.0 0.0 0.7 2.9 -2.5 32. Swaziland 21.2 20.4 6.1 31.3 2.7 0,0 0.2 2.4 1.2 33. Botswana 13.0 21.1 2.2 20.5 8.8 0.0 5.5 1.9 -0.5 34. Mauritius 6.9 23.1 2.1 21.2 1.8 0.0 0.0 1.4 7.6 35. Ivorv Coast- 13.0 15.0 9.0 21.0 16.0 0.0 0.0 10.0 -10.0 M(iddle-income oil exporters . . . . .. . 36. Angola .. .. .. . . . 37. Congo .. .. .... . 38. Nigeria 11.6 21.5 6.7 23.9 7.7 0.0 0.0 1.2 7.1 39. Gabon',. 59.2' 1.1 24.6 8.1 0,0 0.0 2.3.5 4.0 Sub-Saharan Africa 7.1 in 16.4 mn 0.6 in 16.6 in 2.5 m 0.0 in 0.7 mn 2.1 in 1.9 mn a. Includes unallocated revenues and adjustments to a cash basis. b. Data for Chad and Gabon refer to 1976 and for Benin to 1975. c. Includes capital expeniditures or unallocated expenditures. 182 Table 40. Central Government Taxes and Expenditure Ratio of taxes to GDP (Average) Ratio of taxes to expenditure 1966-68 1973, Z977, 1978 7973a 1977, 1978 Low-income countries 13.2 m 12.6 m 14.8 in . 72.1 mn 63.8 in Low-income semiarid 13.9 mn 14.6 mn 16.6 in . 81.4 mn 82.6 in 1. Chad 13.9 12.6 10.9* 605 55.3* 2. Somalia 13.6 23.8 23.9 .. 81.0 82.6 3. Mali 15.0 .. 16.6 .. .. 82.8 4. Upper Volta 12.9 11.2 14.6 .. 92.0 94.6 5. Gambia . 16.7 20.7 2.5.9 81.8 62.3 42.9 6. Niger .. .. 13.5 .. .. 94.9 7. Mauritania 15.2 .. 20.9 .. .. 40.8 Low-income other .. 10.9 mn 14.4 mn . 67.9 mn 63.5 mn 8. Ethtiopia 8.6 9.9 12.8 .. 72.8 66.1 9. Guinea-Bissau. .. .. . . . 10. Burundi 10.1 10.2 13.3 .. 54.2 63.2 11. Malawi .. 10.8 11.6 14.3 54.4 56.3 56.7 12. Rwanda 8.5 8.4 11.6 10.9 72.1 84.9 80.7 13., Benin.. . .. .... . 14. Mozambique . 15. Sierra Leone .. 17.1* 14.8 20.0 73.4* 63.8 82.7 16. Tanzania 15.2 14.7 15.8 16.4 65.9 60.4 59.6 17. Zaire 24.5 25.4 16.1 . 65.6 53.2 18. Guinea**. 10.5 16.9 20.2 51.2 74.0 90.9 19. Central African Rep. .. . .. . 20. Madagascar .. 14.8 . . 75.8 . 21. Uganda . . .. .. . 22. Lesotho .. 24.7 .. . 98.0 . 23. Togo 10.3 .. 26.3 .... 79.8 24. Sudan 13.2 10.9 14.0 15.1 67.9 54.5 Middle-income oil importers .. 20.2 mn 19.5 in 20.8 72.1 mn 71.7 mn 67.0 mn 25. Kenya 14.9 13.9 14.4 19.4 67.6 73.1 26. Ghana 14.6 10.0 6.3 6.3 63.6 48.7 39.5 27. Senegal 19.6 15.9 17.6 21.8 87.0 86.1 102.1 28. Zimbabwe . . .. .. .. . 29. Liberia . 20.2* 18.4 18.2 97.4* 86.9 71.4 30. Zambia 30.0Ob 20.9 23.3 21.4 71.9 63.4 73.2 31. Cameroon . .. 18.0 .... 90.7 96.1 32. Swaziland . 2.8 31.3 .. 69.2 99.0 33. Botswana .. 20.3 20.5 25.9 72.2 60.3 56.5 34. Mauritius .. 16.9 21.2 20.1 87.9 70.3 62.2 35. Ivory Coast"* 21.1 22. 1 21.3 21.8 .. 61.8 62.5 Middle-income oil exporters .. 18.4 mn 23.9 mn . 36. Angola. .. .. .... 37. Congo .. 18.2 22.4* .. .. . 38. Nigeria .. 18.4 23.9 .. 139.0 87.6 39. Gabon .. 19.8 24.6* ' 58.3 42.1* Sub-Saharan Africa .. 16.7 mn 17.1 mn . 72.1 mn 66.1 mn AUl low-income countries .. 11.8 in 14.7 mn . 67.4 mn 65.0 in All middle-income countries .. 13.1 mn 17.8 mn . 82.2 mn 77.0 mn 81.3 mn Industrialized countries 32.9 mn 33.9 mn 37.9 mn . 97.4 mn 85.6 mn 84.9 mn a. Figures marked with an * are for years other than specified. b- Excludes social security contributions. 183 Table 41. Central Government Functional Expenditure Percentage shares, 11978 Social services Other Social commu- General security nity and Economic services public Educa- and social Agricul- Other services Defense tion Health welfare Housing services ture Roads Other services Low-income countries, 22.9 mn 11.2 in 15.2 mn 5.8 m 2.6 mn 0.4 mn 1.6 mn 10.3 m 5.3 m 12.9 in 5.3 m Low-income semiarid 24.4 m 19.4 mn 14.8 mn 6.1 mn 2.0 m 0.5 mn 0.8 mn 13.6 mn 3.0 in 10.3 mn 3.9 in I .Chad 22.4 25.8 13.5 4.2 1.9 0.4 0.7 20.9 2.6 3.0 0.2 2. Somalia* 26.8 20.1 14.0 6.1 1.9 5.6 0.0 13.6 1.7 10.3 0.0 3. Mali' 23.5 18.6 21.6 6.2 3.5 0.0 1.2 .. .. 12.7b 5.1 4. Upper Volta' 22.3 21.5 15.6 5.5 0.0 0.6 1.9 1.7 3.0 24.7 3.0 5. Gambia 25.3 0.0 6.5 6.3 2.0 2.1 1.0 22.0 5.3 24.7 4.8 6. Niger' 33.1 6.1 23.3 6.0 2.9 0.1 0.8 6.6 5.4 7.5 12.2 7. Mauritania.. .. . . .. ... .. . Low-income other 21.8 mn 10.8 in 15.2 m 5.7 in 2.6 mn 0.4 m 1.9 in 9.7 mn 5.5 mn 13.2 in 7.6 m 8. Ethiopia* 43.7b . . 11.5 5.0 4.4 0.4 1.6 10.4 10.6 4.9 7.6 9. Guinea-Bissau- 23.4 24.0 15.6 13.6 . .. .. 5.1 7.2 7.9 10. Burundi' 16.2 11.2 20.6 4.7 3.4 0.0 2.8 12.0 5.6 13.3 8.3 11. Malawi 21.3 11.3 11.8 5.3 1.5 D.8 2.5 11.5 11.1 115.0 7.6 12. Rwanda 20.4 12.4 18.8 6.3 1.7 0.0 1.2 10.3 5.6 13.8 2.0 14. Mozambique .. .. .. .. .. . 15. Sierra Leone 26.3 7.8 16.0 7.6 2.6 0.9 2.9 5.4 6.3 13.1 11.4 16. Tanzania' 18.0 10.5 12.5 5.9 0.9 1.2 1.8 8.9 4.6 32.1 3.6 17. Zaire' 18.4 10.8 15.1 4.0 2.0 0.0 2.9 0.2 1.2 16.5 28.8 19. Central African Rep. . . . .. .. .. .... .. . 20. Madagascar 23.6 4.1 15.3 8.1 10.3 0.3 0.8 14.4 5.3 9.3 11.2 21. Uganda.. .. . .. . . ... . . '22. Lesotho 34.7 0.0 21.3 5.7 2.6 4.5 0.5 18.5 5.4 4.3 2.5 23. Togo' 22.3 9.6 13.7 5.8 5.4 0.1 5.4 6.4 4.3 5.2 5.3 24. Sudan 7.4 13.6 5.2 1.7 2.6 0.8 1.9 9.0 0.5 29.8 27.6 Middle-income oil importers, 20.5 mn 6.5 mn 17.8 mn 6.9 mn 1.0 mn 1.9 in 2.0 mn 8.5 in 9.2 mn 8.4 in 11.2 mn 25. Kenya 18.4 16.0 18.7 7.5 0.2 0.5 2.0 8.5 6.9 12.6 8.8 26. Ghana 19.9 5.3 15.6 7.3 9.7 0.0 3.8 12.2 5.1 6.9 16.1 27. Senegal 24.1 10.7 19.0 6.0 4.7 1.8 2.8 5.2 0.8 7.8 17.2 28. Zimbabwe . .. . . . .. .. .. . . 29. Liberia* 2.7 11.3 6.1 1.0 2.9 0.8 2.7 13.0 13.9 30. Zambia 32.7 .. 16.8 7.7 0.2 1.9 1.9 .. 3.1 .. 14.4 31. Cameroon* 27.4 8.3 16.8 4.8 6.5 0.0 2.4 7.1 11.3 11.1 11.2 32. Swaziland' 27.0 6.5 21.4 6.5 0.3 4.6 0.0 13.1 11.3 7.9 1.4 33. Botswana 20.5 0.0 20.5 6.0 0.8 6.5 1.5 10.5 11.1 8.4 5.3 34. Mauritius' 20.5 0.6 14.2 8.0 12.5 2.3 1.1 9.7 13.1 1.3 16.8 35. Ivory Coast" 17.8 7.2 35.7 7.8 0.9 1.5 2.3 2.9 7.3 21.5 -4.8d Middle-income oil exporters .. . . . .. . ... 36. Angola .. . . . . .. . . . 37. Congo .. .. .. . . . 38. Nigeria 13.5 17.9 9.6 2.2 1.1 3.2 1.9 2.6 13.9 29.3 4.8 39. Gabon.. . . . . . . .. . Sub-Saharan Africa 22.4 mn 10.5 m 15.6 in 6.0 in 2.0 in 0.8 m 1.9 mn 9.0 mn 5.5 in 11.9 mn 7.6 mn All low-income countries 18.4 mn 11.3 in 14.0 mn 5.4 in 2.2 mn 0.8 mn 1.8 in 10.3 mn 5.0 mn 13.0 mn 6.5 in All middle-income countries 14.2 mn 11.5 in 14.1 mn 5.6 in 6.6 mn 2.0 mn 1.8 in 4.0 in 5.6 m 11.6 mn 10.2 mn Industriaized countries 8.2 mn 6.9 mn 10.9 mn 10.2 mn 40.8 mn 2.2 mn 0.9 mn 2.6 mn 2.6 in 7.9 mn 8.2 mn a. Countries marked with an * show figures for years other than specified. See technical notes. b. Includes all economic services. c. Includes social security and welfare. d. Includes the unallocated residual. 184 Table 42. Functional Distribution of Increase in Central Government Expenditure, 1972-78 Percentage share of increase Social services Other commu- General Social nity and Economic services public Educa- security social Agricul- Other services Defense tion Health and welfare Housing services ture Roads Other services Low-income countries, 19.1 i 13.5 mn 11.3 mn 5.1 in 1.9 mn 1.0 in 1.7 m 10.6 mn 2.7 mn 18.3 mn 5.6 mn Low-income semiarid 18.6 mn 23.1 mn 11.0 in 4.6 mn 1.4 mn 1.8 in 0.5 mn 22.9 mn 2.1 m 17.0 m 2.8 mn 1. Chad' - 4.3 28.2 10.7 3.9 3.0 0.6 -0.8 30.4 1.6 4.2 22.3 2. Somalia' 17.8 17.9 19.7 5.3 1.9 9.3 0.0 20.3 1.4 6.5 0.0 4. Upper Volta 19.4 30.4 11.3 3.3 - 5.4 0.9 2.5 -3.1 2.6 41.5 - 3.3 5. GambLa 20.6 0.0 5.2 5.5 0.9 2.6 0.9 25.4 5.9 27.5 5.6 7. Miaurntania .. .. .. .. .. .... .. . Low-income other 19.5im 11.lim 11.5im 5.l m 3.2im 1.O in 1.8im 9.8 m 3.8 m 18.3in 7.1im 8. Ethiopia' 52.8b 8.4 4.2 4.2 0.3 1.6 15.3 11.6 1.8 -0.1 9. Guinea-Bissau .. .. .. .. ....... 10. Bur-undi' 17.1 12.0 18.3 3.6 4.0 0.0 4.6 9.8 8.0 10.5 11.9 II. Malawi 20.1 15.1 9.8 5.1 0.2 1.0 1.7 9.1 12.6 18.3 7.1 12. Rwanda' 21.3 6.9 17.3 6.5 1.3 -0.1 1.7 12.8 2.7 17.5 12.2 14. Mlozambique .. . .... .... .. . 15. Sier-ra Leone 44.4 14.9 17.1 11.4 3.5 1.5 4.0 5.2 0.6 18.6 -21.2 16. Tanzania' 18.8 10.2 11.5 5.7 1.1 1.0 1.8 8.3 3.8 34.8 3.1 19. Central African 20. Madagascar . . . . . ..... .. . 21. Uganda . . . .... .. .. . 22. Lesotho . . . . . .. 23. Togoo. . . . . . . 24. Suidan 5.3 9.5 3.7 0.2 3.2 1.1 1.9 10.6 0.6 38.4 25.4 Middle-income oil importerw 20.1 mn 7.5 mn 17.9 mn 7.4 mn 0.6 mn 1.7 in 1.3 mn 11.9 mn 5.5 mn 7.8 in 9.9 in 25. Kenva 17.7 20.5 17.3 7.2 -0.9 0.1 1.2 8.0 3.4 15.5 9.9 26. Ghana 20.8 4.7 14.7 7.5 10.8 0.0 4.0 13.4 5.5 7.3 11.4 27. Senegal .. .. .. .. . 28. Zimbabwe .. . . .. . . . 29. Liberia .. 1.6 9.7 4.6 0.7 3.4 -0.4 0.3 14.5 16.3 30. Zambia 27.4 .. 12.8 8.3 -1.7 5.3 1.5 27.0 0.4 - 7.0 26.1 31. Cameroon. . . . ........ .. . 32. Swaziland' 20.6 9.3 21.8 5.6 0.5 0.8 0.0 13.9 .. .. 1.0 3.3. Botswana 19.1 10.8 23.0 6.0 0.9 1.0 1.4 10.5 12.4 7.8 5.6 34. Mauritius* 19.8 0.4 18.5 7.6 12.3 4.5 0.5 13.2 0.6 7.5 15.1 35. Ivory Coast" 20.1 7.5 37.9 7.7 -1.8 2.3 2.1 2.4 8.3 26.8 -17.3 Middle-income oil exporters,.. . . . .. 36. Angola . . . .. . . ... 37. Congo . .. ....... .. . 38. Nigeria' 12.7 14.6 10.3 2.0 1.2 3.7 1.8 2.3 14.8 32.6 4.0 39. Gabon . . . .. . .. . Sub-Sa.haran Africa 19.8 i 10.5mi 13.8 i 5.6 i 1.2 in 1.0 in 1.7 mn 10.6 mn 3.8 mn 16.3 in 5.6 in a. Countrnes marked with an * show figures for years other than specified. See technical notes. b. Includes defense. 185 Table 43. Military Expenditure Armed Military expenditure forces per as a percentage share 1,000 of GNP population I978, 1968 1978, Low-income countries 3.4 w 1.9 w 2.4 w Low-income semiarid 3.7 w 2.3 w 4.3 w 1. Chad 2.1* 4.5 3.4 2. Somalia 15.9 4.9 13.8 3. Mali 1.3 2.0 3.5 4. Upper Volta 0.9 1.2 3.4 5. Gambia 1.3 0.0 0.0 6. Niger 0.8 1.0 0.8 7. Mauritania 8.6 1.4 7.1 Low-income other 3.3 w 2.1 w 2.7 w 8. Ethiopia 7.5 2.6 2.8 9. Guinea-Bissau 10.0 .. 7.1 10. Burundi 2.0 1.7 2.2 11. Malawi 0.9 0.5 2.1 12. Rwanda 0.8 2.2 1.6 13. Benin 1.2 1.6 2.0 14. Mozambique 1.3 .. 2.4 15. Sierra Leone 0.3 0.7 1.0 16. Tanzania 3.8 1.1 4.2 17. Zaire 1.9 3.2 0.7 18. Guinea 3.5 3.3 19. Central African Rep. 1.8 2.3 2.2 20. Madagascar 2.5 1.6 2.8 21. Uganda 0.5 1.9 3.1 22. Lesotho 0.8 .. 0.0 23. Togo 2.0 1.2 2.8 24. Sudan 4.0 2.3 4.0 Middle-income oil importers 1.9 w 1.9 w 2.4 w 25. Kenya 0.9 1.3 1.6 26. Ghana 1.7 3.0 0.4 27. Senegal 2.4 1.9 2.2 28. Zimbabwe 3.4 1.9 6.8 29. Liberia 4.1 1.3 1.1 30. Zambia 3.6 1.6 3.1 31. Cameroon 1.4 2.2 1.6 32. Swaziland 4.0 .. 1.4 33. Botswana 4.3 .. 3.8* 34. Mauritius 0.4 0.2 0.1 35. Ivory Coast 1.1 1.3 2.2 Middle-income oil exporters 3.2 w 4.4 w 3.4 w 36. Angola 7.2 37. Congo 7.3 2.8 5.2 38. Nigeria 2.8 5.9 4.2 39. Gabon 6.7 1.2 0.5 Sub-Saharan Africa 3.0 w 2.9 w 2.9 w All low-income countries 3.5 w 3.5 w 3.2 w All middle-income countries 7.0 w 2.9 w 2.9 w Industrialized countries 7.3 w 3.4 w 2.8 w a. Figures marked with an are for 1977. 186 :7 :-il :, ; D ,tyL ....Sectoral statistics are also inconsistent. For example, the Tanzanian government's official The data that underlie the national account growth rate of agriculture for the 1973-79 pe- aggregates, particularly for agricultural pro- riod is recorded as 6.1 percent, involving a duction, are very weak in most African coun- 2.8 percent growth in monetary agriculture tries. This is partly due to the large share of and an 8.6 percent increase in subsistence the subsistence sector in most of these econo- agriculture. Yet most of the export crops that mies. The table below, which presents esti- constitute the bulk of Tanzania's monetary mates of GDP growth of Sahelian countries for sector (sisal, cashews, cotton, coffee, and 1960-70 according to seven different sources, cloves) actually experienced steep declines in provides a good illustration. The smallest volume terms over this period, while the range between the highest and the lowest Food and Agriculture Organization (FAo) data estimate is 23 percent, for Mauritania. For show a rate of increase for food production Chad, the highest estimate is eleven times the of just under 3 percent annually for the ag- lowest, for Mati thirteen times, and for Upper gregate of the 10 most important foodstuffs. Volta six times. Estimates of food production in African The problem is a continuing one. For ex- countries also contain many uncertainties. ample, the following average increases in For example, the Niger Republic has consist- nominal GDP were estimated for Sudan for the ently shown much higher levels of food pro- period 1976-78: duction than any of its Sahelian neighbors; -IMF (April 1979) 27 percent in the early 1970s, Niger's estimate of total -IBRD (October 1979) 22 percent food production was 395 kilograms per capita -IMF Tax Survey (1980) 15 percent while Mali and Upper Volta statistics show -Bank of Sudan 13 percent food production of 230 kilograms per capita. Estimates of Real GDP Growth, 1960-70 (Average annual rate of growth) Country planning Country SOEC SIEC UN OECD IBRD UNCTAD ministry Chad 2.1 2.2 1.2 1.5 1.4 0.5 5.5 Mali 3.0 2.5 0.5 2.8 6.6 5.2 Mauritania 7.4 8.0 7.7 7.3 6.5 6.9 Niger 2.4 2.0 4.7 2.0 0.9 2.4 Senegal 1.6 1.6 1.3 2.0 2.1 1.0 Upper Volta 3.9 3.3 3.0 2.0 1.5 0.7 SOEC: Secteur des Etudes Socio-Economiques de Synthnse (Bureau des Programmes, Direction de l'Aide au Developpernent, Ministere de la Coopration, Pans). SIEC Secteur d'lnformation Lconomique et Conjuncture (Bureau des Programmes, Direction de I'Aide au Developpement, Ministere de la Coop&ration, Pans). UN United Nations OECD OrganiLation for Economic Cooperation and Development (Development Assistance Committee). IBRD Intemnational Bank for Reconstruction and Development (The World Bank). UNCTAD. United Nations Conference on Trade and Development. Source: Republique Fran,caise, Ministere de la Cooperation, Direction de l'Aide au Developpement, Bureau des Programmes-Secteur Synthese, Economie, Emploz et Formation: Erolution et Perspectives pour 14 Etats Africains et Malgache, 1'Evolution du P.l.B. 1950-70, Perspectives 1970-90, Septembre, 1974. 187 The fact that cereals production figures for international prices of agricultural exports. Mali and Upper Volta are very similar (206 Agricultural exports grew more slowly than kilograms per capita in Mali, 182 kilograms food crops in the last decade. Hence, a cal- per capita in Upper Volta) while Niger's cer- culation of the annual growth rate for the 30 eals production is estimated at 318 kilograms main agricultural products results in an an- per capita, suggests that Niger's agricultural nual increase of 0.8 percent in the 1970s if production statistics are overstated. intemational prices are used as weights, com- pared with 1.3 percent if domestic price weights are used. In the 1960s, export crops grew slightly faster than food crops, and the Group means in the country tables are based annual growth rate of overall production on growth rates for periods such as a decade weighted with international prices is 2.6 per- using appropriate weights. This procedure cent, compared with 2.3 percent with do- can lead to very different results, depending mestic price weights. Thus, the fall in overall on whether weights from the beginning or agricultural growth between the 1960s and the end of the period are used. This problem 1970s is about twice as large using intema- is critical in Africa because the Nigerian econ- tional prices. omy is so large relative to the rest of the The tables show weighted means of coun- region, and has grown very rapidly (7.5 per- try groups for the most significant growth cent from 1970 to 1979). Thus, the annual rates. In tables using population weights this growth rate for Sub-Saharan Africa is far presents no problem, since population shares lower when weighted by 1970 country GDPS change very slowly. Caution should be used than when weighted by 1979 country GDPS in interpreting group means based on GDP (see table below). The principal reason for components, however, since their weights this difference is that Nigeria has a weight of change rapidly over time. The notes that fol- 23 percent for 1970, but, because of its fast low explain the weights used in the various growth, a weight of 46 percent for 1979. Con- tables of the Statistical Annex. sequently, Table 2 in the Statistical Annex, which uses 1970 GDP weights in calculating N otc group means, shows Sub-Saharan Africa with and without Nigeria. A similar problem arises in choosing weights TABLE 1. BASIC INDICATORS for the calculation of growth rates in agricul- Gross national product (GNP) measures the ture. The production growth rate derived in total domestic and foreign output claimed by Table 25 uses domestic prices as weights, in residents of a country. It comprises gross do- accordance with established FAO methodol- mestic product (see technical notes for Tables ogy. Domestic prices of agricultural exports 2 and 3) and factor incomes (such as invest- are lower relative to food crop prices than ment receipts and workers' remittances) ac- cruing to residents from abroad, less the in- Average Annual Growth of GDP 1970.79 come earned in the domestic economy accruing to persons abroad. It is calculated without Using 1970 GDP Using 1979 GDP making deductions for depreciation. Item weights weights The GNP per capita figures were calculated Sub-Saharan Afca 2.9 47 according to the World Bank Atlas method: GNP SSA less Nigeria 1.6 2.3 in national currency units was expressed first 1970 1979 in weighted-average prices for the base pe- Nigeria weight riod 1977-79, converted into dollars at the (percent of GDP accounted for by GNP-Weighted average exchange rate for this Nigeria) 23 46 period, and adjusted for U.S. inflation. The Source: Derived from data of the World Development Report 1981. resulting estimate of GNP was then divided 188 by the population in mid-1979. This method territory by residents and nonresidents, re- reduces the effect of temporary undervalua- gardless of its allocation to domestic and for- tions or overvaluations of a particular currency eign claims. It is calculated without making and generally assures greater comparability deductions for depreciation. For most coun- of the estimates of GNP per capita among tries, GDP by industrial origin is measured at countries. factor cost, but for some countries without The average annual rate of inflation was cal- complete national accounts series at factor culated from the "implicit gross domestic cost, market price series were used. GDP at product (GDP) deflator," which is calculated factor cost is equal to GDP at market prices, by dividing, for each year of the period, the less indirect taxes net of subsidies. value of GOP in current market prices by the The agricultural sector comprises agriculture, value of GOP in constant market prices, both forestry, hunting, and fishing. The industrial in national currency. This measure of infla- sector comprises mining, manufacturing, con- tion has limitations, especially for the oil-pro- struction, and electricity, water, and gas. All ducing countries, in light of the sharp in- other branches of economic activity are cat- crease in oil prices in late 1973. egorized as services. The adult literacy rate is the percentage of National accounts series in national cur- persons aged 15 and over who can read and rency units were used to compute the indi- write. For some countries the estimates are cators in these tables. The growth rates in for years other than, but generally not more Table 2 were calculated from constant price than two years distant from, those specified. series, the shares of GDP in Table 3 from cur- Life expectancy at birth indicates the number rent price series. of years newborn children would live if sub- The average growth rates for the country ject to the mortality risks prevailing for the groups in Table 2 are weighted by 1970 coun- cross-section of population at the time of their try GDP in dollars. The average sectoral shares birth. in Table 3 are weighted by country GDP in The index of food production per capita shows current dollars for the years indicated. the average annual quantity of food produced per capita in 1977-79 in relation to that in TABLEs 4 AND 5. GROWTH OF CONSUMPTION 1969-71. The estimates were calculated by AND INVESTMENT; STRUcTuRE oF DEMAND dividing indexes of the quantity of food pro- duction by indices of total population. Food GDP is defined above in the technical notes is considered to comprise cereals, starchy for Table 2. roots, sugar cane, sugar beet, pulses, edible Public consumption (or general government oils, nuts, fruits, vegetables, livestock and consumption) includes all current expendi- livestock products. Quantities of food pro- ture for purchases of goods and services by duction are measured net of animal feed, seed all levels of government. Capital expenditure for use in agriculture, and food lost in proc- on national defense and security is regarded essing and distribution. as consumption expenditure. The country-group averages in this table Private consumption is the market value of are weighted by country population. Data are all goods and services purchased or received not more than two years distant from those as income in kind by households and non- specified. profit institutions. It includes imputed rent for owner-occupied dwellings. TABLES 2 AND 3. GROWVTH AND STRUCTURE Gross domestic investment consists of the out- OF PRODUCTRON lays for additions to the fixed assets of the economy, plus the net value of inventory Gross domestic product (GDP) measures the total changes. final output of goods and services produced Gross domestic saving shows the amount of bv an economy-that is, within a country's gross domestic investment financed from do- 189 mestic output. Comprising public and private for fuel and those for use in the petrochem- saving, it is the difference between gross do- icals industry, these percentages may exag- mestic investment and the deficit on the cur- gerate the dependence on imported energy. rent account of goods and nonfactor services, excluding net current transfers. TABLE 7. GROWTH OF MERCHANDISE TRADE Exports of goods and nonfactor services repre- Merchandise exports and imports cover, with sent the value of all goods and nonfactor serv- soexcpin,alrodditrainl ices sold to the rest of the world; they include some exceptions, all recorded international merchandise, freight, insurance, travel, and changes in ownership of merchandise pass- oterc noifa, trervics. The valeo, aco ing across the customs borders of the report- other nonfactor services. The value of factor . services, such as investrnent receipts and mg countries. Exports are valued f.o.b. (free workers' remittances from abroad, is ex- on board), and imports, c.i.f. (cost, insurance, cluded. and freight). These values are in current dol- The resource balance is the difference be- lars. tween exports and imports of goods and non- The growth rates of merchandise exports and factor services. imports are in real terms and calculated from The indicators in these tables are computed quantum (volume) indexes of exports and im- using national accounts series in national cur- ports. rency units. The growth rates in Table 4 are TABLES 8, 9, 10 AND 11. COMMODITY calculated from constant price series, the STRUCTURE OF MERCHANDISE TRADE shares of GDP in Table 5 from current price series. The country-group averages in Table Merchandise exports and imports are defined in 5 are weighted by country GDP in current dol- the technical notes to Table 7. Fuels are de- lars. fined by the Standard International Trade Classification (srrc) Revised Section 3. Min- erals and Metals are the commodities in SITC All data on energy refer to commercial forms Divisions 27 and 28, and include the nonfer- of primary energy: coal and lignite, petro- rous metals of Division 68. Food commodities leum, natural gas and natural gas liquids, and comprise SITC Sections 0, 1, 4, and Division hydroelectricity and nuclear power-all con- 22 (food and live animals, beverages and to- verted into coal equivalents. The use of fire- bacco, oils and fats, and oilseeds and nuts). wood and other traditional fuels, though sub- Other primary products comprise srrc Section stantial in some developing countries, is not 2 (crude materials), less Divisions 22, 27, and taken into account because reliable and com- 28 (oilseeds and nuts, minerals, crude fertil- prehensive data are not available. izers, and metallic ferrous ores). Manufactures The country-group averages of growth rates represent SITC Sections 5 to 9, less Division of energy production are weighted by volumes 68. of country production in 1974, those of growth Other primary commodities in Table 9 com- rates of energy consumption by volumes of prise sITc Section 2 (crude materials), less country consumption in 1974, and those of Division 22 (oilseeds and nuts), plus Division energy consumption per capita by country pop- 68 (nonferrous metals). Machinery and trans- ulation. port equipment are the commodities in sITc Sec- Energy imports refer to the dollar value of tion 1. Other manufactures, calculated as the energy imports and are expressed as a residual from the total value of manufactured percentage of earnings from merchandise imports, represent SITC Sections 5 to 9, less exports. The country-group averages are Section 7 and Division 68. weighted by country merchandise exports in The country-group averages in Table 8 are current dollars. weighted by country merchandise exports in Because data on energy imports do not per- current dollars, those in Table 9 by country mit a distinction between petroleum imports merchandise imports in current dollars. The 190 beef exports in Table 10 do not include live TABLE 15. COMMODITY TRADE: VOLUME AND animals. Data are not more than two years PRICE distant from those specified. The price series are derived from the ratio of TABLE 12. DESTINATION OF MERCHANDISE international prices to the index of prices of EXPORTS manufactured exports from industrialized countries. Both series are expressed in dol- Merclhandise exports are defined in the technical lars; inflationary trends common to both sets notes for Table 7. All trade shares in this table of prices are consequently eliminated. The are based on statistics that show the value of average annual change in the price series ig- trade in current dollars. Unallocated exports nores negative signs. are distributed among the country groups in proportion to their relative shares of allocable TABLE 16. DEVELOPED COUNTRY TARIFF trade. The country-group averages are RATES ON SELECTED COMMODITIES weighted bv countrv merchandise exports in current dollars. All tariff rates are most favored nation rates, applicable after the Tokyo Round of tariff ne- TABLE 13. TEkMS OF TRADE gotiations. The net barter terms of trade are calculated as TABLE 17. BALANCE OF PAYMENTS, DEBT the ratio of a country's index of export unit SERVICE, AND INTERNATIONAL RESERVES values to that of import unit values. Income ternms of trade are an index of the value of The current account balance is the difference commoditv exports divided by import unit between (i) exports of goods and services plus value. Growth rates are calculated from data inflows of unrequited official and private on merchandise exports and imports as de- transfers and (ii) imports of goods and serv- fined in Table 8. ices plus unrequited transfers to the rest of the world. Excluded from this figure are all TABLE 14. EXPORTS: COMMODITY interest payments on external public and pub- CONCENTRATION AND FLUCTUATION IN licly guaranteed debt, which are shown sep- VALUES arately. These interest payments represent those on the disbursed portion of outstanding Commodity concentration expresses the current public and publicly guaranteed debt plus value of the three principal commodities in commitment charges on undisbursed debt. the exports of a country as a percentage of Debt service is the sum of interest payments the total current value of merchandise ex- and repayments of principal on external pub- ports. With the exception of data for Niger lic and private publicly guaranteed debt. The and Sierra Leone, which include uranium and ratio of debt service to exports of goods and diamonds (usually classified as manufac- services is one of several rules of thumb com- tures), respectively, all principal export com- monly used to assess the ability to service modities are primary products. debt. The debt-service ratios in the table do Export fluctuations are defined as the annual not cover nonguaranteed private debt, which percentage deviation from trend, calculated for some countries is substantial; the debt according to the following formula: contracted for purchases of military equip- F oo n 1 Xi - Xii ment is also excluded because it usually is not F = 22 Xj , reported. The average ratios of debt service to exports of goods and services are weighted where x, = value of export earnings at time i; by country exports of goods and services in = corresponding trend value (expo- current dollars. Debt service averages for the nential trend); and industrialized countries are not computed be- n = number of years covered. cause the World Bank Debt Reporting System 191 collects data on external debt solely for de- cial terms by official agencies of the members veloping countries. of the Development Assistance Committee Gross international reserves comprise the sum (DAC) of the Organization for Economic Co- of a country's holdings of gold, special draw- operation and Development (OECD) and mem- ing rights (SDRS), the reserve position of IMF bers of the Organization of Petroleum Ex- members in the Fund, and holdings of foreign porting Countries (OPEC) with the objective exchange under the control of monetary au- of promoting economic development and thorities. The gold component of these re- welfare. Net disbursements equal gross dis- serves is valued throughout at year-end Lon- bursements less payments to donors for am- don prices: that is, $37.37 an ounce in 1970 ortization. and $512 an ounce in 1979. The reserve levels Total recorded net flow of resources includes for 1970 and 1979 refer to the end of the year ODA grants from private agencies (private aid) indicated in current dollars. The reserve hold- and transactions at commercial terms: export ings at the end of 1979 are also expressed in credits, bilateral portfolio investment (includ- the number of months of irmports of goods ing bank lending) by residents or institutions and services they could pay for, with imports in DAC countries, direct investment (including at the average level for 1978 or 1979. Country- reinvested earnings), and purchases of se- group averages are weighted by country im- curities of international organizations active ports of goods and services in current dollars. in development. Net bilateral flows exclude un- allocated bilateral flows and all disbursements TABLE 18. DEBT AND DEBT SERVICE to multilateral institutions. Grants, including the value of technical co- Debt service is defined in the technical notes operation and assistance, are gifts in money for Table 17. Note, however, that debt service or in kind for which no repayment is re- in this table includes publicly guaranteed pri- quired. Both bilateral ODA and disbursements vate debt. All values are in current dollars. from multilateral institutions are included in Table 23. TABLES 19, 20, AND 21. EXTERNAL DEBT AND The country-group averages in Table 22 are LOANS weighted by country population for net re- The loans referred to in all three tables are source flow and ODA per capita. ODA as a medium- and long-term loans whose matur- percentage of GNP iS weighted by country ities exceed one year. Data for private non- GNP, and ODA as a percentage of gross do- guaranteed loans are subject to greater error mestic investment is weighted by country than those for public and publicly guaranteed GDP. loans. TABLE 24. FooD AID IMPORTS Concessional terms refer to loans which con- tain a grant element of at least 25 percent. Food aid includes cereals only, and is ex- The grant element is the grant equivalent ex- pressed in metric tons grain equivalent. The pressed as a percentage of the face value of data are for the fiscal year, extending from the commitment. In turn the grant equivalent July 1st of the preceding calendar year and is the face value of the commitment minus ending on June 30th of the current calendar the discounted present value of the future year. The country-group averages for per cap- flow of repayments of principal and interest ita food aid are weighted by country popu- payments. lation for the appropriate year. TABLES 22 AND 23. FOREIGN ASSISTANCE TABLES 25, 26, AND 27. GROWTH OF AND OTHER RESOURCES AGRICULTURE, CROP PRODUCTION, AND YIELDS Official development assistance (ODA) consists of loans and grants made at concessional finan- Food includes commodities that are consid- 192 ered edible and contain nutrients. Nonfood next. Thus, the corresponding rates of growth comprises all inedible and/or nonnutritive ag- pertain only to those subcategories of dairy ricultural commodities. Accordingly, coffee imports for which data are reported, and and tea are classified as nonfood because, must be interpreted with caution. All data are although edible, they have virtually no nu- for the calendar year. tritive value. (Note that the definition of food used here is not consistent with that used in TABLE 31. DOMESTIC TERMS OF TRADE OF Tables 8 and 9, where all beverages, regard- EXPORT CROPS less of nutritive value, are considered as food items.) Barter terms of trade are defined as the ratio Yield is defined as the volume of production between the weighted average official prices of an agricultural commodity per unit of land paid to producers for export crops and prices area, here expressed as kilograms per hectare. paid by them for certain consumer goods. The index of relative yields compares average Income terms of trade are defined as the ratio annual yields of selected crops in all devel- between producers' income derived from ex- oping countries and in Sub-Saharan Africa port crops and prices paid by them for certain with average annual worldwide yields for the consumer goods. three specified time periods. Developing countries is an FAO classification TABLE 32. PROCUREMENT AND DISTRIBUTION which includes all low- and middle-income OF AGRICULTURAL INPUTS countries except Albania, Greece, Israel, Por- Procurement and distribution activity is con- tugal, Romania, South Africa, Spain, and Yu- sidered private if more than 80 percent of it goslavia. is carried out by the private sector and gov- The countrv-group averages for average ernment if more than 80 percent is carried out annual growth rate of volume of agricultural production in Table 25 are weighted by the by the public sector. product of 1975 country GDP in current dollars and each country's average agricultural share TABLE 33. POPULATION GROWTH, PAST AND of total GDP for the period 1970-77. Country- group averages for the average annual growth The growth rates of population are period av- rate of total agricultural production per capita erages calculated from mid-year country pop- are weighted by 1979 country population. All ulations. The country-group averages are data are for the calendar year. weighted by country population in 1970. The projections of population for 1980 and TABLE 28. FERTILIZER CONSUMPTION 2000, and for the year in which population Figures represent nutrient tons of nitrogen, is projected to become stationary, were made phosphate, and potash. The fertilizer year for each country separately. Starting with in- begins on July 1st of the preceding calendar formation on total population, fertility rates, year and ends on June 30th of the current and mortality rates in base year 1978, param- year. Thus, fertilizer year 1979 extends from eters were projected to 1980 and thereafter July 1978 to June 1979. Country-group aver- for each subsequent five-year interval on the ages for average annual growth rate of fertil- basis of generalized assumptions until the izer consumption are weighted by each coun- population became stationary. For all of the try's average annual fertilizer consumption. projections in this table, it was assumed that international migration would have no effect. TABLES 29 AND 30. AGRICULTURAL IMPORTS The net reproduction rate (NRR) indicates the AND EXPORTS number of daughters that a newborn girl is projected to bear during her lifetime, assum- Subcategories of dairy imports are sporadic- ing fixed age-specific fertility rates and a fixed ally reported from one time period to the set of mortality rates. An NRR of 1 indicates 193 that fertility is at replacement level: that is, year. For countries with reliable death regis- child-bearing women, on the average, bear tration, these rates refer to a variety of years, only enough daughters to replace themselves generally not more than two years distant in the population. A population continues to from those specified. For other countries, the grow after replacement-level fertility has been rates were derived from the appropriate reached because higher birth rates of the past Coale-Demeny Model life tables which cor- will have produced a relatively high propor- respond to the expectation of life at birth for tion of women in, or still to enter, the repro- 1960 and 1978. The country-group averages ductive ages. The time taken for a country's in this table are weighted by country popu- population to become stationary after reach- lation. ing replacement-level fertility thus depends on its age structure and previous fertility pat- TABLE 35. LABOR FORCE tems. A stationary population is one in which age- The population of working age refers to the pop- and sex-specific mortality rates have not ulation between 15 and 64 years of age. The changed over a long period, while age-spe- country-group averages are weighted by cific fertility rates have simultaneously re- country population. mained at replacement level (NRR = 1). In The labor force comprises economically ac- such a population, the birth rate is constant tive persons, including members of the armed and equal to the death rate, the age structure forces and the unemployed, but excluding is also constant, and the growth rate is zero. housewives, students, and economically in- The estimates of the hypothetical size of active groups. Agriculture, industry, and serv- the stationary population, the assumed year ices are defined in the same manner as in of reaching replacement-level fertility, and Table 2. Most of the 1979 estimates are geo- the year of reaching a stationary population metric extrapolations of International Labour are speculative. They should not be regarded as Office (ILO) estimates for 1960 and 1970. The predictions. They are included to provide a country-group averages are weighted by summary indication of the long-run implica- country labor force. tions of recent trends on the basis of highly The labor force growth rates were derived stylized assumptions. A fuller description of from the World Bank's population projections the methods and assumptions used to cal- and ILO data on activity rates. The country- culate these estimates is available from the group averages for 1960-70 and 1970-80 are Population and Human Resources Division weighted by country labor force in 1970, and of the World Bank. those for 1980-2000, by projections of country labor force in 1980. TABLE 34. DEMOGRAPHIC AND FERTILrrY- The application of ILO activity rates to the RELATED INDICATORS Bank's latest population estimates may be in- appropriate for some countries in which there The crude birthi and death rates indicate the num- have been important changes in levels of un- ber of live births and deaths per thousand employment or underemployment, interna- population in a year. The total fertility rate rep- tional and internal migration, or both. The resents the number of children that would be labor force projections for 1980-2000 should born per woman, if she were to live out her thus be treated with caution. child-beaning years and bear children at each age in accordance with prevailing age-specific TABLE 36. URBANIZATION fertility rates. Life expectancy at birth is defined in the technical notes for Table 1. Because the estimates in this table are based The child death rate is defined as the number on the individual national definitions for of deaths of children aged 1-4 per thousand "urban," cross-country comparisons should children in the same age group in a given be interpreted with caution. The country- 194 group averages for urban population as a per- in the ratios given. For countries with uni- centage of total population are weighted by versal primary education, the gross enroll- country population; the other country-group ment ratios may exceed 100 percent because averages in this table are weighted by country some pupils may be below or above the of- urban population. ficial primary-school age. The data on number enrolled in secondary TABLE 37. HEALTH-RELATED INDICATORS school were calculated in the same manner, with secondary-school age generally consid- Nusesta nude gauate, practical, and .as ered to be 12-17 years. The adult literacy rate personnel vary and the data, while generally Th data in the technical notes for Table 1. not~~~~~~~~~~ ~ moeta w er dsatfo hs The data in this table refer to a variety of not more than two years distant from those specified, are for a variety of years. Conse- years, generally not more than two years dis- quently the data for populatioin per physician tant from those specified. Country-group av- and nurse are not strictly comparable be- erages are weighted by country population. tween countries. The percentage of total population with access TABLES 39 AND 40. CENTRAL GOVERNMENT to safe water is the proportion of persons with BUDGETARY OPERATIONS, TAXES, AND reasonable access to safe water, which is de- EXPENDITURE fined as including treated surface water and Fiscal data for state, provincial, and local gov- such untreated but uncontaminated water as eret r nvial.Cneunl,od thatfro boehols, prigs,and sanitary ernments are unavailable. Consequently, only that from boreholes, springs, central government data are presented. This wells. may seriously understate or distort the sta- The daily calorie supply per capita was calcu- lated by dividing the calorie equivalent of the tistical portrayal of the allocation of resources food supplies in a country by its population. for various purposes, especially in large coun- Food supplies comprise domestic production, tries where lower levels of government have imports less exports, and changes in stocks; considerable autonomy and are responsible they exclude animal feed, seeds for use in for a large number of social functions. Central agriculture, and food lost in processing and government expenditure and revenue covers that ag,. .ulture, and food lost in processing and by all govemment departments, offices, es- distribution. The daily calorie requirement p.er tablishments, and other bodies that are agen- capita refers to the calories needed to sustain a person at normal levels of activity and cies or instruments of the central authority of health, taking into account age and sex dis- a country. The operation of public enterprises tributions, average body weights, and envi- (public entities which sell their output) ap- ronmental temperatures. pears only to the extent that their net profits The country-group averages in this table are transferred to the treasury. Consequently, are weighted by country populagion. expenditure excludes the transactions of gov- ernment marketing organizations. Certain TABLE 38. EDUCATION foreign aid flows, such as parastatal borrow- ing, are omitted, while technical assistance is The data on number enrolled in primary school also likely to be excluded. Expenditure in- refer to estimates of total, male, and female cludes subsidies and transfers. enrollment of students of all ages in primary Net lending (disbursements of loans and school; they are expressed as percentages of purchase of equities less amortization) in- the total, male, or female populations of pri- dudes only government transactions under- mary-school age to give "gross primary en- taken for public policy rather than for man- rollment ratios." Although primary-school agement of government liquidity or for earning age is generally considered to be 6-11 years, a retum. Most government transactions in- the differences in country practices in the volving changes in government debt are, ages and duration of schooling are reflected therefore, included in financing. 195 Data for Table 40 are not more than two TABLE 42. FUNCTIONAL DISTRIBUTION OF years distant from those specified. INCREASE IN GOVERNMENT EXPENDITURE, 1972-78 TABLE 41. CENTRAL GOVERNMENT FUNCTIONAL EXPENDITURE The functional distribution of increase in central government expenditure is calculated by ex- Central government expenditure is explained in pressing the difference between the amounts the technical notes for Tables 39 and 40. The spent in the final and base years under each various functional classifications of central functional category as a percentage of the government expenditure are defined in the change in total expenditure between the two [MF Government Finance Statistics Yearbook. years. The time span covered is 1972-77 for It must be emphasized that the data pre- Ethiopia, Nigeria, Somalia, and Swaziland. sented, especially those for education and Other deviations from the standard time pe- health, are not comparable across countries riod are for Burundi (1973-77), Chad (1972-76), for a number of reasons. In many countries, Ivory Coast (1975-78), Mauritius and Rwanda private health and education services are sub- (1973-78), and Tanzania (1972-79). The var- stantial; in others, public services represent ious functional classifications of central gov- the major component of total expenditure. emient expenditure and problems with in- Considerable caution should therefore be ex- tercountry comparability are discussed in the ercised in using the data for cross-country technical notes for Table 41. comparisons. Data for Burundi, Cameroon, Ethiopia, Mali, TABLE 43. MILITARY EXPENDITURE Mauritius, Niger, Somalia, Swaziland, Togo, Upper Volta, and Zaire refer to 1977. Data for Armed forces refer to active-duty military per- Liberia and Tanzania are for 1979. Owing to sonnel including paramilitary forces if those adjustment items and unallocated transac- forces resemble regular units in their orga- tions, the sum of components in Botswana, nization, equipment, training, or mission. Cameroon, Chad, Liberia, Mali, Niger, Reserve forces are not included. Military ex- Rwanda, Togo and Zambia falls short of, Or penditure data in this table differ from central exceeds, 100 percent. government defense expenditure in Table 41. The primary source for military expenditure data here is the Agency for International De- velopment, and it is their definitions that are used. Country-group averages are weighted by country population for the armed forces per thousand population and by 1970 country GDP for the percentage share of military expend- iture in GDP. 196 Bil-Iiography of Data Sources Basic Production Yearbook. Rome: UN Food and Agriculture Organization (FAO), various Indicators issues. (Table 1) Statistical Yearbook, 1978-79. Paris: United Nations Educational, Scientific, and Cultural Organization (Unesco), 1980. United Nations population tapes. 1980 World Bank Atlas. Washington, D.C.: World Bank, 1980. World Bank data files. World Bank, World Development Report 1981. New York: Oxford University Press, 1981. Production A System of National Accounts. New York: UN Department of International (Tables 2-6) Economic and Social Affairs, 1968. (Tables 2, 3) World Bank data files. World Bank, World Development Report 1981. New York: Oxford University Press, 1981. (Tables 2-6) World Energy Supplies, 1950-74, 1972-76 and 1973-78. UN Statistical Papers, Series J, nos. 19, 21, and 22. New York: UN Department of International Economic and Social Affairs, 1974, 1978, and 1979. (Table 6) Trade Commodity Trade and Price Trends. Washington, D.C.: World Bank, August 1980. (Tables 7-16) (Tables 10, 14, 15) Direction of Trade. Washington, D.C.: International Monetary Fund (IMF), various issues. (Tables 7, 12) Financing Mineral Projects in Developing Countries: A United Nations Study. London: Mining Journal Books, 1979. Handbook of International Trade and Development Statistics. New York: United Nations Conference on Trade and Development (UNCrAD), various issues. (Tables 7, 13, 14) "The Influence of Protectionism on Trade in Primary and Processed Commodities." New York: UNCTAD TD/B/C.11207/add 2, 14 August 1980. Processed. (Table 16) International Financial Statistics. Washington, D.C.: IMF, various issues. (Table 7) Monthly Bulletin of Statistics. New York: UN Department of International Economic and Social Affairs, various issues. "Price Prospects for Major Primary Commodities." Washington, D.C.: World Bank, January 1980. Processed. (Table 15) United Nations data files. (Tables 8, 9) World Bank data files. (Tables 7-9, 12) World Bank, World Development Report 1981. New York: Oxford University Press, 1981: World Bank, World Tables. The Second Edition [1980]. Baltimore, Md.: Johns Hopkins University Press, 1980. Yearbook of International Trade Statistics. New York: UN Department of International Economic and Social Affairs, various issues. (Tables 8, 9, 11) 197 Aid, Debt, Development Cooperation. Paris: Organization for Economic Cooperation and and Capital Development (OECD), statistical annexes to various issues. (Tables 22, 23) Flows Food Aid Bulletin. Rome: FAO, no. 4, October 1980. (Table 24) (Tables 17-24) Geographical Distribution of Financial Flows to Developing Countries, 1969-76 and 1970-79. Paris: OECD, 1977 and 1980. (Tables 22, 23) International Monetary Fund data files. (Table 17) World Bank Debt Reporting System. (Tables 18-21) World Bank, World Development Report 1981. New York: Oxford University Press, 1981. Agriculture "Agricultural Pricing Policy in Sub-Saharan Africa in the 1970s." Washington, (Tables 25-32) D.C.: World Bank, December 1980. Processed. (Table 31) Fertilizer Yearbook. Rome: FAO, 1978. (Table 28) Food and Agriculture Organization data files and Production Yearbook tapes. (Tables 25-27) Trade Yearbook. Rome: FAO, various issues. (Tables 29, 30) World Bank, World Development Report 1981. New York: Oxford University Press, 1981. Social Coale, Ansley J., and Paul Demeny, Regional Model Life Tables and Stable Indicators Populations. Princeton, N.J.: Princeton University Press, 1966. (Table 34) (Tables 33-38) Demographic Yearbook. New York: UN Department of International Economnic and Social Affairs, various issues. (Tables 34, 36) Food and Agriculture Organization data files. (Table 37) Labour Force Estimates and Projections, 1950-2000. 2nd ed. Geneva: International Labour Office (ILO), 1977. (Table 35) Population Council data files. (Tables 33, 34) Statistical Yearbook. Paris: Unesco, various issues. (Table 38) United Nations Population Division data files. (Tables 35, 36) United States Bureau of the Census data files. (Tables 33, 34) World Bank data files. (Tables 33-36) World Bank, World Development Report 1981. New York: Oxford University Press, 1981. (Tables 33-38) World Health Organization data files. (Table 37) World Population Trends and Prospects by Country, 1950-2000: Summary Report of the 1978 Assessment. New York: UN Department of International Economic and Social Affairs, 1979. (Tables 33, 34) Fiscal Data Chelliah, Raja, "Trends in Taxation in Developing Countries." IMF Staff Papers, (Tables 39-43) July 1971. (Table 40) Government Finance Statistics Yearbook. Washington, D.C.: IMF, vol. IV, 1980. (Tables 39-42) International Monetary Fund consultation reports, various years. (Tables 39-42) Revenue Statistics of OECD Member Countries, 1965-78. Paris: OECD, 1979. (Table 40) World Military Expenditures and Arms Transfers, 1969-78. Washington, D.C.: United States Arms Control and Disarmament Agency, 1980. (Table 43) 198 PUBLIC ADMINISTRATION AND DEVELOPMENT PADEDR 14(5) 421-518 (1994) (Public adm. dev.) ISSN 0271-2075 *, CONTENTS ning, Research, and Practice VOLUME 14, ISSUE No. 5 December 1994 Managing organisations in the Third World: a case study of management control in a Bangladeshi jute mill Z Hoque, N. A. Siddiquee and T Hopper............................. ........... 421 Financing growth in the 1990s: new challenges for OECS countries K C Simmonds . 439 Local government decentralization and the health sector in Tanzania L Gilson, P. Kilima and M. Tanner . 451 Market-oriented approaches to capacity building in Africa P. Demongeot ................................................................. 479 Dr. G. Glentworth The consolidation of adjustment: implications for public administration Finance, Management and Administration R Batley. .489 Advisory Department (FMAIAD) of the Professional Developments: Intemational Association of Schools and Institutes of ODA. London, UKE Administration (IASIA): Working Group VI-Managing the transition from the Raja G. Gomez centrally-planned economy. Report on July 1994 Hong Kong meeting Secretariat, Commonwealth Parliamentary P. Colis................................................... 506 Association, London, UK Book Reviews S09 Dr. Stephen Hickey Chief Executive, Abstracts Sectio .513 Civil Service College, Ascot, U.K Book Review Index to Volume 14 .517 Professor David Murray Volume Contents. i The Open University, Milton Keynes, UK Professor Gelase Mutahaba Commonwealth Secretarial, London, UK Professor Brian Smith Department of Political Science & Social Policy, The University of Dundee, UK Sir Kenneth Stowe GCB, CBO Lol?don Professor AlU D. Yahaya Secretary-General. African Association for Public Administration and Management (AAPAM). Nairobi. Kenya Abstracts Section Edftor: Mr. C. Reeve Librarian, Public Policy Library, University of Birmingham, Edgbaston, Birmingham B15 2TT. U. K lms L-so. Cinestor. West Sn,-oe P19 IUD, Engiand ITolephooe to p.b)sood q-trtrly, pila ane sp-saS aste by John Wiley & Sorj ,ptlon p- U S S;: 5o Sp-ctal mtes for RIPA memben and t5\t:dtJe Isnc.. 2D.s % : ' :- , xl .- F!y-e ,sod D Beleg.oo. clo ,-I *o,r-dstc In.. 2158 N WA f2nd W>;;~ ~ ~ -,.-.--------- .. - - .-;N* .. . S. . . -. . - - TABLE OF CONTENTS VOLUMEI NUMBER,X 2 Alcatel Subsidiaries Plow Fertile Russian Switching Markets 4 PTN Builds Rival Transmission Network 6 MMT Searches for Niche DEPARTMENTS 8 PRIVATIZATION AND I0 SATELLITE NEWS REGULATION Morsviazputnik Installing Inmarsat Land Consortia Formed to Vie for Czech Earth Station in Moscow Privatization Matav Pushing to Exceed Privatization 1 0 CARRIER STRATEGiES Targets SOxSO Working Group Begins Difficult Task of Crafting Business Plan 8 WIRELESS WATCH SPT Preparing Final Backbone Transm is- Italtel Launches Second GSM Network in sion Project Tyumen Region IFC Takes Stake in Sprint RP Tel ekom Westel 900 Reaches Half of Hungary After Ericsson Wins Supply Contract in Hungary Six Months Chelyabinsk Sells Bonds to Pay for Phones US West Takes Stake in Bulgarian GSM Telia International Expands in the Baltics j COMPETITIVE ANALYSIS LuxCel Wins Lithuanian Paging License Siemens Wins Russian Terrestrial Portion Philips to Supply MARS in Latv'ia & Russia of ITUR 9 PRIVATE NETWORKS 11 TENDERS AM& ~~~~~Vladimir Telecommunications Center Matav Tenders Radio Local Loop Opens Matav Tenders Erzsebet Local Loop Kazakhs tan to Award NVationial Baniking Kyrgyz Microwave Tender Network Contract Ii:. September 1994 Pyramid Research, Inc. Vol. 9, No. 9 TELCOS LOOK SKYWARD FOR RURAL Mexico: Early Advocate of VSAT for PSTN SOLUTIONS The government of Sonora, Mexico, in con- by Sheila Marcelo junction with Mexico's Secretaria de Comunica- ciones y Transportes (SCT), was an early rural Developing countries are beginning to look VSAT advocate, turning to a VSAT network to seriously at VSAT systems as an alternative to provide telephony services in remote areas. Cut terrestrial telecom networks in rural areas. In over in March 1991, the Frequency Division Multiple addition to providing high-quality digital links across Access (FDMA)/single-channel-per-carrier (SCPC) difficult terrain, satellites offer advantages - dis- tance-insensitivity, flexibility, and ease of deploy- ment - over fiber optic and traditional metallic wire solutions. Still, few VSAT telephony networks currently exist in developing countries, largely due to their traditionally high price tag - the VSAT O4ET ~Jnetworks that have been installed mainly serve data: communication needs. But, several developmenits fW A are improving the outlook for rural VSATs. In addition to costly ground station equipment, high satellite construction and launching costs, as . C ANM M N U , l IL well as great demand for geosynchronous orbital L ILILPQ IE slots, have put transponder space at a premium. A~~A1~ REIG1 Leasing a full transponder can range in cost from $2 million annually in the competitive U.S. environment to $1 0 million in Brazil's recently liberalized market. AI AM RC /A B8A But, new digital compression techniques and ~ ~ hG1 advances in access technologies, in concert with falling prices for ground station equipment, have ARG /W L AT improved the efficiency of satellite systems, making them more affordable for developing countries. Satellite operators are relying on frequency reuse ATR4E OEINS RfFfG 1 techniques - spatial reuse and polarization - to double the number of transponders aboard each spacecraft. Whereas earlier satellites typically N-I carried 12 transponders, most present-day satellites carry between 24 and 48 transponders. Telecommnunicalions Developtmtent Report is a monthIly strate,y letter that anlyzes market, t,eclnologies. and competition in the regi;ons of Asia, Africa, Latin America, the Middle E-ast, Eastern Europe, and the NIS Republics. Copyright 0 1994 PYRAM%ID RE-SEARCH, INC. ISSN #0889-907X. All rights reserved. Reproduction in anyftormw,thout the express written permissionof the publisheris forbidden. Informationand statistical data contained herein havebeen obtained from sources which webelieve tobe reliable, butare int nowaywarrant-das toaccuracyorcompleteness. Publisher& Managing Editor.William W. Ambrose. Editor. Stuart Melnitsky. Contributors:Jonathan Dower,\ Mark Kelsey, Achmad Chadran,Stephen Dalla Betta,Joshui Levenberg. Linda Barrabee, Ayesha Ahmad, Carole Chapelier. Charles Willikms, Alexaindra Rehak, Laura Rozen. Kiendel Burritt, Nicole Caso,John Wang,Shelia Marceio, and M4eredith lersily Suobscriptiont price: US $795.00 annually 112 issues). Inquiries: Telecommunications Development Report. Pyramid Research, 14 Arrow Street. Cambridge, MA~ 02138 USA. Tel: 1617) 868-4725. Fax- 16171 868.5574 E-mail: pvrainid4pvr.ccm O MM UC U:ATION : < ;--VHEOPMENlh {T LRep or Ug November 1994 Pyramid Research, Inc. Vol. 9, No. 11 PRIVATIZATION TAKES BACK SEAT TO However, despite the acknowledged suc- "LIBERALIZATION" IN LATIN AMERICA cesses, privatization is not considered a panacea for telecom development in all Latin American by Stpe Dall Bettadvlpen nal ai Aeia by Stephen Dalla Betta countries. Even now, privatization often faces stiff For nearly a decade,aLatin America has led the opposition from unions and their supporters in government. And, privatization remains anathema developing world with its frequent privatizations to many governments that consider telecom a *through strategic foreign investment. Today,yg privathrou stegics provideign basinveservient. T , Anational strategic asset that should not be sold to privatized telcos provide basic services in Argen- the highest foreign bidder. tina, Chile, Mexico, Peru, Puerto Rico, and Venezu- ela, while carriers are currently on the block in other countries, such as Nicaragua and Bolivia. However, despite the overwhelming successes of these privatizations, Latin American countries are now pursuing other initiatives in their telecom -ONTE S sectors, such as deregulation of basic services without privatization and build-operate-transfer aBa TAKES A (BOT) financing schemes. These initiatives are 'N tIx AMEICA. creating opportunities for private services operators ' BBB ' B. 7 and equipment vendors alike. R'' Why Liberalization, Not Privatization? CELLUL M x , " The privatizations that have already taken a . place in Latin America's telecom sector have, with few exceptions, proven successful. In fact, the LTIAMrCIA BEN results of these privatization efforts have been 8Rl£ g' g . I.. ' . aa a.' 'a a'" B touted in nearly every Latin American country and . throughout the world. Concrete indicators - main- PIAMD EEA . line growth, network investment, the introduction of EFING , new technologies, and increased coverage at sharply reduced prices - demonstrate a marked EASTERN EUROPE N RIEFTN- -1a difference in the pre- and post-privatization periods (see "Latin America Privatization: The Results Are gB In," TDR, July 1993). Telecommunications Development Report is a monthly strategy letter that analyzes markets, technologies, and comnpetition in the regions of Asia. Afnca, Latin America. the Middle East, Eastem Europc, and the NIS Republics. Copyright ©D 1994 PYRAMID RESEARCH. INC. ISSN #0889-907X. All nghts reserved. Reproduction in any form without the express written permission of the publisher is forbidden. Information and statistical data containied herein havo been obtained from sources which we believe to be reliable, but are in no way warranted as to accurary or completeness. Publisher & Managing Editor: Widiam W. Ambrose. Editor: Stuart Melnitsky. Contributors: Jonathan Dower, Mark Kelsey, Achmad Chadran. Stephen Dafla Betta, Joshua Levenberg, Linda Barrabee, Ayesha Ahmad, Carolc Chapeer. Chardes Williams, Alexandra Rehak, Laura Rozen. Kicndel Burrirt. Nicolc Caso. John Wang, Sheila Marcelo, Meredith Persdv, Todd Allmendinger. and Alex Kabanovsky. Subscription price: US 5795.00 annuallv (12 issues). Inquiries: Telecomimunications r,,.,, R,-, P-id P--.mi6 R A A - -1 a,A, IA r l.-. i.yA_ t 'I,1'A T.A. ,Al 57 QA2_A_71'r . (I7< iA 17) RfR i7. t C,1 ,-, nvr-com -- -- - -- -- :E COuvuimCAT-10 _ _ from Washington * Brussels o Tokyo A biweekly news ser , trade, and networks Volume 6, No. 1 SLP036077 January 6, 1995 C 0 N T E N T S European Commission Proposes European Survey ...... 3To Ease CATV/Telecom Limits EU Telecom Survey ................ 8 he European Commission adopted a draft directive Dec. 21 proposing to - lift restrictions on che use of cable TV networks and permit them to carry Satellite Services ................9 all telecommunications services that have been liberalized in the European Union. The commission said the directive, whose chief proponents were Competition Survey of the Americas .................... Policy Commissioner Karel van Miert and Telecommunications Commissioner Martin Bangemann, aims to open up low-cost capaciry for carriage of new multi- Japan Telecom Survey ............... 16 media services. ('"8 Emerging Services .18 The commission intends to present its proposals next year to EU member states and the European Parliament and consult wich other interesced parties before U.S. Regulatory Scene .19......... ...... formally adopting it. Legally, the proposal is framed as an amendment to the 1990 services directive that has led to liberalization of telecom services not reserved to News in Brief national monopoly providers - i.e., public switched telephony. Under this regu- Personnel .20 latory policy initiative, the major EU countries have committed to open their public switched service markets to competition by 1998. Contracts, Ventures . 21 The 1990 services directive was issued under article 90 of the EU Treaty, which permits the commission to adopt measures without gaining the approvals of Tariffi, Se-rvices ................23 member states. Such procedures are available only when member states fail to fulfill an undertaking, and they apply only to public bodies that have "special or DEAR SUBSCRIBER exclusive rights." - With this first issue ofthe newyer The commission stressed that the new directive "does not affect the member Telecommunicatios Restiseo ternaw onar states' rights to maintain monopolies in provision [of] voice telephony until 1998, Telecommuaz onspor Inta7wional as the directive concerns only the provision of 'non-reserved' services." It also un- introduces a new graphic format. The derlined its "intention to present its proposals... .to both the Council [ofEU miin- aim of the redesign is to offer subscrib- isters] and the Parliament, and to proceed in close cooperacion with them, wirh ers added value with a deaner, brighter utmost respect for transparency." look, while continuing TR['s in-depth coverage of international telecom issues Furthermore, the commission noted chac "the revenues of the [telecom opera- and events. No editorial space is lost as tors] are not chreatened by a transfer of customers [to cable TV operators], since a result of the changes, and we hope you they are, for the most part, not even providing these new services yet." will find the publication easier both to In a scatement on the draft directive, che commission said that "liberalizing scan and to read. We appreciate your access to cable infrascructure should permit a lowering of costs and a significant support during TRl's first five years. increase in the amount of capacicy available for new services. Alongside this, it (continued on next page) have distinct comparative advantage in ex- traders an increased role in these markets. port crop production. An export-sacrificing Some observers object to proposals for more policy of self-reliance would therefore have competitive marketing arrangements on the costs in terms of income (see Box G). And a grounds that rural African markets function policv aiming at food security at the price of imperfectly, that traders would therefore ex- lessened emphasis on exports has a further ploit farmers, and that indigenous traders are pitfall: most methods of intensification imply still few in number in parts of the continent, increased use of inputs such as fertilizers, in- so that trade in foodstuffs might once again secticides, and fuel for pumping (in irrigation come to be dominated by nonnationals. But schemes), i.e., they rely heavily on imported many recent studies suggest that African food inputs. Thus, agricultural production under markets are reasonably competitive, that trader these known methods of intensified cultiva- profits are rarely "excessive" and that farmers tion becomes more vulnerable to external dis- are usually well protected against "exploita- equilibria. If the pursuit of food self-sufficiency tion" by market information and the availa- diverts resources from export crops to food bility of alternative points of sale.25 But even crops, declining export earnings may lead to if this were not so, governments can more balance-of-payments problems jeopardizing efficiently protect farmers by making markets the self-sufficiency objective itself. Sudan and more competitive through better information, Tanzania are countries that have, in recent roads, and marketing facilities than by acting years, deliberately sacrificed export expan- as substitutes for traders. sion for the sake of increasing food produc- In any case, it is important to recall that in tion (see box on Tanzanian exports in Chapter a large number of African countries, food 4). Their present balance-of-payments crises, markets continue to operate, as they have in as severe as those of some mineral exporters, the past, without much public control. Per- are partly related to that policy. formance is generally impressive; in Nigeria, private trade suppliers two very large cities (Lagos and lbadan) and many more towns of EFooo CROPS AND INPUT SUPPLY 100,000-500,000 people. In Mali, despite the While many details regarding food price pol- uncertainties of public policy, private trade in icies can only be assessed in the context of the mid-1970s supplied two thirds of the ce- individual country situations, two principles reals consumed in the Sixth Region, the most are central. First, food imports should be sub- remote part of the country. ject to duties, so that the import price reflects Indeed, even the most casual visitor to a at least the true cost of foreign exchange. market town in Africa has to come away im- Otherwise, low-price imports will continue to pressed by the range of goods and services replace domestic production, with negative available for sale, their variety and quality effects on rural income and growth. Second, there should be a gradual freeing of domestic foodeshould mket, to eragl greater cfdompeti- 25. See Henry Hayes, Marketing and Storage of Food Grains food markets, to encourage greater competi- in Nigeria. Samaru Miscellaneous Paper no. 50 (Samaru, tion. This would in most cases merely rec- Nigeria: Institute for Agricultural Research, Ahmodu Bello ognize the existing reality, which is that, University, 1979) and IDEr-CEGOS, Pre-etude de la commerciali- whatever the legal or formal situation regard- sation des produits vivriers au Cameroun (Republique Unie du Cameroun: Ministere de l'Agriculture, avril 1980). See also ing public monopoly, the main part of the Van Roy Southworth, William 0. Jones, and Scott R. Pear- cereals trade goes through private channels son, "Food Crop Marketing in Atebubu District, Ghana" in and, consequently, the great majority of con- Food Research Institute Studies, vol. 17, no. 2 (Palo Alto. Cal- ifornia: Stanford University Press, 1979), J.T. Mukui (ed.), sumers already pay "free market" prices. "Price and Marketing Controls in Kenya," Institute of De- With regard to food marketing and input velopment Studies Occasional Paper no. 32 (Nairobi, Kenya: supplv, the proposal to allow a fuller degree University of Nairobi, 1979); and Guy Nicolas, "Processus d'approvisionnement vivrier d'une ville de savane: Maradi of competition means encouraging coopera- (Niger)," Travaux et documents de geographie tropicale, no. 7, tive actions by farmers and allowing private decembre 1972. 64 group averages for urban population as a per- in the ratios given. For countries with uni- centage of total population are weighted by versal primary education, the gross enroll- country population; the other country-group ment ratios may exceed 100 percent because averages in this table are weighted by country some pupils may be below or above the of- urban population. ficial prirmary-school age. The data on number enrolled in secondary TABLE 37. HEALTH-RELATED INDICATORS school were calculated in the same manner, with secondary-school age generally consid- Nsestan incudses.Countrydefinits oe p r ical, ered to be 12-17 years. The adult literacy rate sistant nurses. Country definitons of medical i eie ntetcnclntsfrTbe1 personnel vary and the data, while generally eIS defined m the techmcal notes for Table 1. not~ ~ mor thntoyasdsatfo hs The data in this table refer to a variety of not more than two years distant from those specified, are for a variety of years. Conse- years, generaly not more than two years dis- quently the data for populatiorn per physician tant from those specified. Country-group av- and nurse are not strictly comparable be- erages are weighted by country population. tween countries. The percentage of total population with access TABLES 39 AND 40. CENTRAL GOVERNMENT to safe water is the proportion of persons with BUDGETARY OPERATIONS, TAXES, AND reasonable access to safe water, which is de- EXPENDITURE fined as including treated surface water and Fiscal data for state, provincial, and local gov- such untreated but uncontaminated water as that from boreholes, springs, and sanitary centrernments are unavailable. Consequently, only wells. central government data are presented. Ths The daily calorie supply per capita was calcu- may seriously understate or distort the sta- lated . dividing the calorie equivalt of te tistical portrayal of the allocation of resources fated bypdividing corie equivatiof. for various purposes, especially in large coun- food supplies in a country by its population. tries where lower levels of government have Food supplies comprise domestic production, imports less exports, and changes in stocks; considerable autonomy and are responsible thev exclude animal feed, seeds for use in for a large number of social functions. Central agriculture, and food lost in processing and government expenditure and revenue covers that distribultioe,an. T oda calortie proesuirnt ad by all government departments, offices, es- cpistarestion The diycalories need red et sta tablishments, and other bodies that are agen- capita p eferson at noralolevs ofeded ac stivit n cies or instruments of the central authority of hal takingat intorm accounto age ivity and sexa country. The operation of public enterprises tu, aage body we and se vis- (public entities which sell their output) ap- ronmutntas, ae,verage b pears only to the extent that their net profits are transferred to the treasury. Consequently, The country-group averages in this table expenditure excludes the transactions of gov- are weighted by country population. ernment marketing organizations. Certain TABLE 38. EDUCATION foreign aid flows, such as parastatal borrow- ing, are omitted, while technical assistance is The data on number enrolled in primary school also likely to be excluded. Expenditure in- refer to estimates of total, male, and female cludes subsidies and transfers. enrollment of students of all ages in primary Net lending (disbursements of loans and school; they are expressed as percentages of purchase of equities less amortization) in- the total, male, or female populations of pri- cludes only government transactions under- mary-school age to give "gross primary en- taken for public policy rather than for man- rollment ratios." Although primary-school agement of government liquidity or for earning age is generally considered to be 6-11 years, a return. Most government transactions in- the differences in country practices in the volving changes in government debt are, ages and duration of schooling are reflected therefore, included in financing. 195 ered edible and contain nutrients. Nonfood next. Thus, the corresponding rates of growth comprises all inedible and/or nonnutritive ag- pertain only to those subcategories of dairy ricultural commodities. Accordingly, coffee imports for which data are reported, and and tea are classified as nonfood because, must be interpreted with caution. All data are although edible, they have virtually no nu- for the calendar year. tritive value. (Note that the definition of food used here is not consistent with that used in TABLE 31. DOMESTIC TERMS OF TRADE OF Tables 8 and 9, where all beverages, regard- EXPORT CROPS less of nutritive value, are considered as food items.) Barter terms of trade are defined as the ratio Yield is defined as the volume of production between the weighted average official prices of an agricultural commodity per unit of land paid to producers for export crops and prices area, here expressed as kilograms per hectare. paid by them for certain consumer goods. The index of relative yields compares average Income terms of trade are defined as the ratio annual yields of selected crops in all devel- between producers' income derived from ex- oping countries and in Sub-Saharan Africa port crops and prices paid by them for certain with average annual worldwide yields for the consumer goods. three specified time periods. Developing countries is an FAO classification TABLE 32. PROCUREMENT AND DISTRIBUTION which includes all low- and middle-income OF AGRICULTURAL INPUTS countries except Albania, Greece, Israel, Por- Procurement and distribution activity is con- tugal, Romania, South Africa, Spain, and Yu- sidered private if more than 80 percent of it goslavia. is carried out by the private sector and gov- The country-group averages for average ernment if more than 80 percent is carried out annual growth rate of volume of agricultural by the public sector. production in Table 25 are weighted by the product of 1975 country GDP in current dollars TABLE 33. POPULATION GROWTH, PAST A and each country's average agricultural share of total cDP for the period 197G-77. Country- PROJECTED group averages for the average annual growth The growth rates of population are period av- rate of total agricultural production per capita erages calculated from mid-year country pop- are weighted by 1979 country population. All ulations. The country-group averages are data are for the calendar year. weighted by country population in 1970. The projections of population for 1980 and TABLE 28. FERTILIZER CONSUMPTION 2000, and for the year in which population Figures represent nutrient tons of nitrogen, is projected to become stationary, were made phosphate, and potash. The fertilizer year for each country separately. Starting with in- begins on July 1st of the preceding calendar formation on total population, fertility rates, year and ends on June 30th of the current and mortality rates in base year 1978, param- year. Thus, fertilizer year 1979 extends from eters were projected to 1980 and thereafter July 1978 to June 1979. Country-group aver- for each subsequent five-year interval on the ages for average annual growth rate of fertil- basis of generalized assumptions until the izer consumption are weighted by each coun- population became stationary. For all of the try's average annual fertilizer consumption. projections in this table, it was assumed that international migration would have no effect. TABLES 29 AND 30. AGRICULTURAL IMPORTS The net reproduction rate (NRR) indicates the AND EXPORTS number of daughters that a newborn girl is projected to bear during her lifetime, assum- Subcategories of dairy imports are sporadic- ing fixed age-specific fertility rates and a fixed ally reported from one time period to the set of mortality rates. An NRR of 1 indicates 193 Bibliog,raphy of Data Sources Basic Production Yearbook. Rome: UN Food and Agriculture Organization (FAO), various Indicators issues. (Table 1) Statistical Yearbook, 1978-79. Paris: United Nations Educational, Scientific, and Cultural Organization (Unesco), 1980. United Nations population tapes. 1980 World Bank Atlas. Washington, D.C.: World Bank, 1980. World Bank data files. World Bank, World Development Report 1981. New York: Oxford University Press, 1981. Production A System of National Accounts. New York: UN Department of International (Tables 2-6) Economic and Social Affairs, 1968. (Tables 2, 3) World Bank data files. World Bank, World Development Report 1981. New York: Oxford University Press, 1981. (Tables 2-6) World Energy Supplies, 1950-74, 1972-76 and 1973-78. UN Statistical Papers, Series J, nos. 19, 21, and 22. New York: UN Department of International Economic and Social Affairs, 1974, 1978, and 1979. (Table 6) Trade Commodity Trade and Price Trends. Washington, D.C.: World Bank, August 1980. (Tables 7-16) (Tables 10, 14, 15) Direction of Trade. Washington, D.C.: International Monetary Fund (IMF), various issues. (Tables 7, 12) Financing Mineral Projects in Developing Countries: A United Nations Study. London: Mining Journal Books, 1979. Handbook of International Trade and Development Statistics. New York: United Nations Conference on Trade and Development (UNCrAD), various issues. (Tables 7, 13, 14) "The Influence of Protectionism on Trade in Primary and Processed Commodities." New York: UNCTAD TD/B/C.1/207/add 2, 14 August 1980. Processed. (Table 16) International Financial Statistics. Washington, D.C.: IMF, various issues. (Table 7) Monthly Bulletin of Statistics. New York: UN Department of International Economic and Social Affairs, various issues. "Price Prospects for Major Primary Commodities." Washington, D.C.: World Bank, January 1980. Processed. (Table 15) United Nations data files. (Tables 8, 9) World Bank data files. (Tables 7-9, 12) World Bank, World Development Report 1981. New York: Oxford University Press, 1981. World Bank, World Tables. The Second Edition [1980]. Baltimore, Md.: Johns Hopkins University Press, 1980. Yearbook of International Trade Statistics. New York: UN Departrnent of International Economic and Social Affairs, various issues. (Tables 8, 9, 11) 197 Table 5. Structure of Demand Distribution of gross domestic product (percent) Exports of goods Public Private Gross domestic Gross domestic and nonfactor consumption consumption investment saving services Resource balance 1960 1979o 1960 1979, 1960 1979a 1960 1979' 1960 1979' 1960 1979 Low-income countries 10 w 15 w 81 w 80 w 11 w 15 w 9 w 6 w 16 w 16 w -2w -9w Low-income semiarid 11w 18w 84w 78w 12w 23w 7w 4w 13w 22w -6w -19w I. Chad 13 18 82 96 11 13 5 - 14 23 33 -6 - 27 2. Somalia 8 19* 89 79* 10 16* 3 2* 11 12* - 7 - 14* 3. Mali 12 23 79 82 14 15 9 - 5 12 16 - 5 -20 4. Upper Volta 10 14 94 89 10 24 -4 -3 9 15 -14 -27 5. Gambia*' 20 26 72 83 13 22 8 -9 59 65 -5 -31 6. Niger 9 9 79 72 13 28 12 19 9 25 -1 -9 7. Mauntania 39 47 .. 51 14 38 .. -37 Low-income other 10 w 15 w 80 w 80 w 11 w 13 w 10 w 6 w 17 w 15 w -1w -8w 8. Ethiopia 8 17 81 87 12 10 11 -4 9 10 -1 -14 9. Guinea-Bissau . b 102 32 -2 10 -34 10 Burundi 3 16 92 80 6 12 5 4 13 13 -1 -8 11. Malawi 16 17 88 70 10 29 -4 13 21 21 -14 -16 12. Rwanda 10 16 82 72 6 19 8 12 12 25 2 -7 13. Benin 16 12 75 87 15 21 9 1 12 27 -6 -20 14. Mozambique 11 15 81 85 10 10 8 (.) 14 13 -2 -10 15. Sierra Leone . 18 . 78 15 . 4 . 24 . -11 16. Tanzania 9 16 72 76 14 21 19 8 31 14 5 -13 17. Zaire 18 b 61 88 12 9 21 12 55 30 9 3 13. Guinea .. 16 70 15 14 24 -1 19. Central African Republic 19 20 72 72 20 20 9 8 23 18 -11 -12 20. Madagascar 20 17 75 73 11 22 5 10 12 17 -6 -12 21. Uganda 9 b 75 96 11 4 16 4 26 4 5 (.) 22. Lesotho 17 16 108 143 2 29 -25 -59 12 21 -27 -88 23. Togo 8 15 88 74 11 39 4 11 19 32 -7 -28 24. Sudan 6 11 85 84 9 14 9 5 12 9 (.) -9 Middle-income oil importers 11 w 15 w 67 w 68 w 22 w 19 w 21 w 16 w 37 w 27 w 0 w -4w 25. Kenva 11 20 72 65 20 22 17 15 31 26 -3 -7 26. Ghana 10 9 73 86 24 5 17 5 28 12 -7 (.) 27. Senegal 17 b 68 98 16 21 15 2 40 34 - 1 -19 28. Zimbabwe 11 13* 67 63 23 15' 22 24* - 1 5* 29. Liberia 7 15 58 62 28 27 35 23 39 53 7 -4 30. Zambia 11 27 48 45 25 21 41 28 56 45 16 7 31. Cameroon .. 10 .. 80 25 10 25 .. -15 32. Swaziland- 18 22 54 64 13 28 29 14 47 77 16 -13 33. Botswana" 15 25 88 63 8 43 -3 12 23 47 -12 -31 34. Maurinus" 15 14 79 61 30 38 6 25 32 50 -24 -13 35. Ivory Coast 10 17 73 56 15 31 17 27 37 35 2 -4 Middle-income oil exporters 7 w 11 w 84 w 57 w 15 w 30 w 10 w 31 w 17 w 26 w -6w 1 w 36. Angola 9 26 77 56 12 9 14 18 20 43 2 9 37. Congo 23 30 98 58 45 22 -21 12 21 -66 -10 38. Nigeria 6 10 87 58 13 31 7 32 15 25 - 6 1 39. Gabon- 10 12 40 36 50 30 50 53 32 1 Sub-Saharan Africa 10 w 13 w 77 w 65 w 15 w 23 w 13 w 20 w 22 w 23 w -3w -3w All low-income countries 9w 11w 78w 66w 18w 26w 16w 23w 7w 11w -2w -3w All middle-income countries 11 w 13 w 70 w 62 w 21 w 26 w 19 w 25 w 16 w 20 w -2w -1 w Industrialized countries 15 w 17 w 63 w 61 w 21 w 23 w 22 w 22 w 12 w 19 w 1 w -1 w a. Figures marked with an * are for 1978. b. Public consumption is included in private consumption. 147 IBRD 16041 20 0 0 - - -- 200 - 400 SEPTEMBER 1981 North At lan tic Ocean d;0\ 40' ~~~~~~~~~~~~~~~~~~~~~~~~~~400- ISIA 20, h At/aiterrtieon Oceao MAU. - --..r-.. TROrPI Of CAPRI N V 20'~ ~ ~ ~~~~~~~~O 0020 l 20 ~ ~~~~~~~~ ~~~ 1000h 2000ti 3000n KM.~_ __ _ SIERR AFRCANO0° This map hats been prepared by the Worid Bank s staff eoclusi'.elY for the convenience of the readers of the report to which it is attached. The denominations used and the boundairies shown on this map do not imply, on the part of the World Bank and its affiliates, any judgment on the legal status of any terrtory or any endorsement or acceptance of such boundaries. 200 00 ~200 4Q0 Table 28. Fertilizer Consumption Average annual consumption Average annual growth rate, (thousands of 1962-66b 1969-71b metric tons) to to 1977-78b 1969-71 1977-78 Low-income countries 219.42 t 12.0 w 6.8 w Low-income semiarid 31.18 t 31.0 w 22.7 w 1. Chad 6.57 42.7 20.0 2. Somalia .. 22.7 3. Mali 13.00 47.6 19.7 4. Upper Volta 6.65 31.3 45.3 5. Gambia 2.95 22.5 37.8 6. Niger 0.81 18.4 30.7 7. Mauritania 1.20 .. 24.8 Low-income other 188.24 t 11.4 w 5.9 w 8. Ethiopia 27.99 23.6 26.0 9. Guinea-Bissau .. . 10. Burundi 0.65 . 10.9 11. Malawi 22.42 11.0 14.1 12. Rwanda 0.31 .. 8.4 13. Benin 1.50 40.6 -13.5 14. Mozambique 12.85 4.8 9.3 15. Sierra Leone 2.04 42.0 0.5 16. Tanzania 34.59 20.5 15.7 17. Zaire 10.74 25.5 14.0 18. Guinea 1.45 0.5 - 6.2 19. Central African Republic 2.20 27.8 1.9 20. Madagascar 9.60 22.5 - 2.2 21. Uganda 2.13 8.7 - 11.8 22. Lesotho 1.40 -0.8 19.1 23. Togo 2.37 .. 31.9 24. Sudan 56.00 8.8 5.3 Middle-income oil importers 403.25 t 27.0 w 10.7 w 25. Kenya 52.71 16.6 3.3 26. Ghana 25.05 5.2 40.7 27. Senegal 45.05 - 0.8 23.7 28. Zimbabwe 113.50 9.4 2.5 29. Liberia 4.69 52.5 14.0 30. Zambia 67.20 15.9 14.9 31. Cameroon 22.22 21.2 3.5 32. Swaziland 5.20 4.3 -0.5 33. Botswana 2.05 1.9 3.7 34. Mauritius 23.03 -1.1 0.7 35. Ivory Coast 42.55 9.5 14.4 Middle-income oil exporters 95.80 t 29.9 w 12.3 w 36. Angola 16.05 34.8 7.0 37. Congo 2.38 30.1 -11.3 38. Nigeria 76.50 26.2 33.0 39. Gabon 0.87 . Sub-Saharan Africa 718.47 t 23.1 w 9.7 w Developing countries 16,381.45 t 14.6 w' 10.2 w' World 97,469.10 t 9.0 w' 5.7 w' a. Growth rates for groups of countries were calculated on the basis of those countries for which data on all three sub-periods were available. b. See technical notes. c. Nonweighted mean. 170 Table 12. Destination of Merchandise Exports Percentage share of merchandise exports Sub-Saharan Centrally Industralized African Other developing planned Capital-surplus market economies countries countries economies oil exporters 1960 1979 1960 1979 1960 1979 1960 1979 1960 1979 Low-income countries 76 w 64 w 6 w 8 w 16 w 22 w 1 w 3 w 1 w 4 w Low-income semiarid 76 w 71 w 18 w 6 w 6 w 11 w O w O w O w 13 w 1. Chad 73 30 27 13 0 52 0 0 5 2. Somalia 85 18 0 1 15 2 0 (.) (.) 80 3. Mali 93 68 7 15 0 17 0 (.) (.) (.) 4. Upper Volta 4 75 96 9 0 16 0 .. 0 0 5. Gambia 97 93 3 1 0 6 0 (.) 0 0 6. Niger 74 97 26 1 0 0 0 0 2 7. Mauritania 89 88 11 2 0 9 0 . . 0 1 Low-income other 76 w 63 w 5 w 8 w 17 w 23 w 0 w 3 w 1 w 3 w 8. Ethiopia 69 72 4 (.) 20 11 1 7 6 10 9. Guinea-Bissau' .. 29 32 22 .. 38 . 1 . 0 10. Burundi .. 89 I. 1 .. 9 . 1 0 11. Malawi .. 84 .. 12 .. 4 1 12. Rwanda .. 80 .. 4 .. 16 I. 13. Benin 90 89 8 2 0 8 2 1 0 (-) 14. Mozambique 29 43 5 4 66 45 (.) 1 (.) 7 15. Sierra Leone 99 98 1 1 0 1 0 0 (.) 16. Tanzania 74 57 4 4 21 36 1 2 0 1 17. Zaire 89 64 5 26 6 10 (.) (.) (.) (.) 18. Guinea 63 69 10 3 9 26 18 (.) 2 19. Central African Rep. 83 78 9 2 8 20 0 (.) 0 (.) 20. Madagascar 79 67 18 4 2 29 1 (.) (.) (.) 21. Uganda 62 67 7 3 31 27 0 1 0 2 22. Lesotho .. . .. . 23. Togo 74 67 26 8 0 17 0 8 0 24. Sudan 59 36 2 (.) 27 45 8 9 4 10 Middle-income oil importers 88w 75w 3w 9w 7w 13w 3w 3w Ow Ow 25. Kenya 77 63 7 21 15 15 0 (.) (.) 1 26. Ghana 88 70 2 2 3 15 7 13 (.) (.) 27. Senegal 89 59 4 27 7 14 0 (-) 0 (.) 28. Zimbabwe . . . . . 29. Liberia 100 86 0 (.) (.) 14 0 (.) 0 (.) 30. Zambia . 82 . 2 .. 16 . (.) (.) 31. Cameroon 93 84 3 6 3 8 1 2 (.) (.) 32. Swaziland . . 33. Botswana . . ... . . 34. Mauritius" 97 95 2 4 1 1 0 0 0 0 35. lvorv Coast 84 78 3 6 13 11 0 5 0 (-) Middle-income oil exporters 89w 84w 2w 3w 8w 13w 1w Ow Ow (.)w 36. Angola 64 33 7 (.) 27 66 2 0 0 1 37. Congo 93 72 (.) 1 7 27 0 (.) 0 (.) 38. Nigeria 95 87 1 2 3 11 1 (.) 0 (.) 39. Gabon" 87 60 6 8 7 32 0 0 0 (.) Sub-Saharan Africa 82 w 78 w 4 w 5 w 12 w 15 w 2 w 2 w (.)w 1 w All low-income countries 51 w 61 w 29 w, 29 w, 19 w 5 w 1w 5 w All middle-income countries 68 w 67 w 24 w, 26 w, 8 w 4 w (w 3 w Industrialized countries 67 w 69 w 30 w 24 w, 3 w 3 w (.) w 4 w a. Includes exports destined for Sub-Saharan Africa. 154