9777 Education in Sub-Saharan Africa Policies for Adjustment, Revitalization, and Expansion -x7ksJJe t N ii 1 ~s Education in Sub-Saharan Africa A World Bank Policy Study Education in Sub-Saharan Africa Policies for Adjustment, Revitalization, and Expansion The World Bank Washington, D.C. Copyright (o 1988 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 First printing January 1988 The judgments expressed in this study do not necessarily reflect the views of the World Bank's Board of Executive Directors or of the governments that they represent. The maps used in this document are solely for the convenience of the reader and do not imply the expression of any opin- ion whatsoever on the part of the World Bank or its affili- ates concerning the legal status of any country, territory, city, area, or of its authorities, or concerning the delimita- tion of its boundaries or national affiliarion. Library of Congress Cataloging-in-Publication Data Education in Sub-Saharan Africa. (World Bank policy study) Bibliography: p. 1. Education and state-Africa, Sub-Saharan. 2. Education-Economic aspects-Africa, Sub-Saharan. 1. World Bank. 11. Series. LC95.A357E38 1988 379.67 87-34650 ISBN 0-8213-0996-X Foreword Without education, development will not occur. Only proving quality and further expanding the education an educated people can command the skills necessary systems, however, will certainly require an incremen- for sustainable economic growth and for a better tal flow of resources. Thus African countries will need quality of life. Recognizing this, African governments to strike a balance between demands for education have placed heavy emphasis on expanding educational and the scarcity of resources, and they will need to opportunities from primary school through university develop country-specific, comprehensive, and inter- in the two or three decades since their independence. nally consistent sets of policies along three dimen- The number of students enrolled in African institu- sions: adjustment to current demographic and fiscal tions at all levels has more than quintupled since 1960. realities, revitalization of the existing educational in- Indeed, the spread of educational access may be the frastructure to restore quality, and selective expansion single most significant development achievement of to meet further demands. Africa. The analysis and recommendations contained in The key role of education in the development this study should contribute to this educational plan- process is the reason, too, why the World Bank has ning process. To this end, the study attempts to put so much emphasis on supporting educational diagnose the problems of erosion of quality and recent expansion and improvement in Sub-Saharan Africa. stagnation of enrollment and to offer a set of policy Even so, education in Sub-Saharan Africa is in crisis responses commensurate with the severity of these today. Rapid population growth has resulted in more problems. In doing so, it does not prescribe one set of children than ever seeking places in schools already educational policies for ali of Sub-Saharan Africa, as pressed for resources because of the financial crises of the nature and scope of both the issues and the the 1980s. A lower proportion of children and young responses to them differ markedly among African people are able to obtain places in educational insti- countries. Nor does it propose specific educational tutions as a result, and the quality of education has investments. Instead, the study presents a framework dropped as classrooms have become overcrowded and within which countries may formulate strategies tai- teaching materials increasingly scarce. lored to their own needs and circumstances. Governments cannot be expected to increase sub- The study has been prepared after very broad stantially the resources they devote to education. Al- consultation both in Africa and in the industrialized ready, many Sub-Saharan African countries allocate countries. In particular, its preparation has benefited over 20 percent of the government budget to educa- enormously from discussions at two conferences of tion. Further increases would cut too deeply into other African ministers and permanent secretaries of educa- pressing demands for public funds. Measures for im- tion planning and finance, convened for this purpose v in Côte d'lvoire and Ethiopia. The comments of many purpose. But most important, we hope this study will experienced and knowledgeable people helped im- provide insights for our member governments in Afri- measurably to shape and refine both the tone and the ca as they develop their own country-specific plans for substance of this study on a number of complex issues. educational reform, revitalization, and expansion. This study will guide the World Bank's lending and technical assistance to Sub-Saharan African education Barber B. Conable over the next several years. lt should serve to provide President a common ground for other donors as well, to expand their assistance to education in Africa, and to increase The World Bank the effectiveness of international assistance for this December 11, 1987 vi Contents Glossary and Acronyms ix Preface xi Summary 1 Educational Development in Africa 1 A Policy Framework 2 Formulation and Implementation of National Programs 3 Policy Options by Levei of Education 4 The Role of the International Donor Community 6 Why Meet the Challenge? The Expected Benefits of Education 6 Part 1. The Policy Context 1. The Remarkable Progress of African Education 11 Before Independence 11 Advances after 1960 12 Expenditure on Education 14 2. Education and the External Environment 18 The Demographic Challenge 18 Macroeconomic Adjustment and Fiscal Austerity 20 Investment in Education 21 3. The Stagnation of Enrollment and Decline in Quality 28 Enrollment Stagnation 28 Declining Quality of Education 31 Part II. Policy Options for African Governments 4. The Foundation: Primary Education 39 vii Measures for Improving Quality 40 The Containment of Unit Costs 46 Mobilizing Resources for Primary Education 51 5. The Consolidation of Competence: Secondary Education and Training 54 Meeting the Demand for Expansion by Reducing Unit Costs 56 Equity Issues: Increased Participation by Females 61 Training for Vocational Competence 62 Financing Secondary Education and Training 65 6. Preparation for Responsibility: Higher Education 68 The Challenge and the Promise 69 Issues in Higher Education 70 A Program for Structural Adjustment of African Higher Education 77 7. Using Resources Well: The Mandate for Education Managers 81 Improving Organizational Structure 82 Improving Information: Testing, Statistical, and Accounting Systems 85 Strengthening Analytical Capacity 86 Development of Managerial Staff 87 Priorities and Resource Requirements 89 Part III. An Agenda for Action 8. Policy Packages for Educational Development 93 Adjustment 94 Revitalization: Restoring Quality 97 Selective Expansion 98 Policy Design and Implementation 100 9. International Assistance for African Educational Development 101 Sources of Aid and Its Recent Use 101 The Comparative Advantage of Aid: Past and Present 105 New Structures to Support Policy Design and Implementation 107 Future Amounts and Targeted Areas of Aid to African Education 109 A Call to Action 112 Bibliography 113 Appendix 119 Maps following page 185 Notes on the Data Most of the discussion and ali of the statistics about Africa in this study refer to just thirty-nine countries south of the Sahara, for which the terms Africa and Sub-Saharan Africa are used interchangeably. The appendix lists and provides comparative information on these countries. "Dollars" means U.S. dollars. "Billion" means one thousand million. viii Glossary and Acronyms Correspondence education. A type of distance edu- home economics, agriculture) at the lower secondary catíon through which students receive textbooks for levei to provide a prevocational orientation and to individual study in their homes and supplementary develop a positive attitude toward work. The second mass media broadcasts on the subject matter. Each includes a general academic stream plus one or more textbook includes exercises to be completed and specialized occupational streams, usualiy at the up- mailed to a postal tutor, who grades each exercise per secondary level, that are suited to economic and provides an individual evaluation to the student. conditions in the surrounding area. Curriculum. A set of courses in a field of study, often Enrollment ratio. School enrollment, both public and constituting an area of specialization at the higher private, as a percentage of a given age group in the levels of education. population. The gross enroliment ratio is the total number of students enrolled at a given education level Distance education, or distance teaching. An educa- divided by the population of the age group for that tion delivery system that uses a variety of media and levei. This ratio may include students who are younger a system of feedback to teach people who are unable or older than the age expected at that level. The net to attend traditional schools. Distance education enrollment ratio is calculated by using only that part of usually combines the use of media broadcasts, print- the total number of students enrolled that corresponds ed materials, and some kind of face-to-face study. to the specific age groups defined for that level. Distance teaching programs can range from in-school programs, in which broadcasts supplement learning Equivalency program. A teaching program that pro- activities in the classroom, to out-of-schooi pro- vides opportunities for education to students who grams, such as correspondence lessons in which stu- would otherwise be unable to attend formal schoois dents may never meet their tutors and may have little and that emphasizes the acquisition of knowledge or no contact with the regular education system. rather than the place where the knowledge is ac- quired. Equivalency programs exist for all leveis of Diversified secondary education. Education at the schooling. secondary leve] that introduces practical or occupa- tional subjects into an otherwise completely academ- External examination. An examination set by an ic program and is thus intended to meet the expected independent organization and administered to a large needs of school leavers. Two models are prevalent. number of students from different schools to allow The first introduces practical subjects (industrial arts, comparison of results across schools. ix Extramural or external degree or certificate. A degree the higher-level skilled worker. Instruction is typical- or certificate awarded by a university to students in ly in preemployment training institutions such as equivalency programs who have studied the curricu- polytechnics; administration is commonly the re- lum on their own and have not attended formal sponsibility of ministries of education, labor, or em- classes. ployment, or comes under the authority of specific industries. Courses of study in technical education Incomplete school. A school that does not have the usually include a component of general education. full number of grades for the levei of education provided. Training. Instruction in job-related skills to prepare students for direct entry into a trade or occupation. Nonformal education. Education and training for Such instruction can take place in training centers, out-of-school youths and adults in classes, courses, through apprenticeships, or at the place of employ- or activities intended to promote learning but not ment (on-the-job training). constituting part of the formal school system and not leading to formal qualifications, such as diplomas or Vocational education. Training in craft or trade skills specific trade standards. Nonformai education typi- for the semiskilled worker. Instruction typically oc- cally concentrates on short programs of a few months curs in schools and comes under the direction of in duration. ministries of education. Nongovernmental or pri vate voluntary organization. World Bank. An international lending institution A nonprofit organization with private membership comprising the International Bank for Reconstruc- that provides development assistance. Such organiza- tion and Development, which lends to governments tions include foundations, lay and religious aid asso- of member developing countries on nonconcessional ciations, and nongovernmental cooperatives. terms for projects and other purposes intended to stimulate economic growth, and the International Postsecondary, tertiary, or higher-level education. Development Association, which lends for the same Education that requires, as a minimum condition of purposes but on concessional terms and only to the entry, the successful completion of education at the poorest member countries. secondary level or proof of equivalent knowledge or experience. Instruction is given in various types of Acronyms institutions, including universities, vocational and technical training institutes, and teacher training in- ADB African Development Bank Group stitutes. ECA United Nations Economic Commission for Africa Primary or basic education. The first level of educa- GNP Gross national product tion, in which students follow a common curriculum. IBRD International Bank for Reconstruction and Primary education offers students instruction in pri- Development mary or elementary schools that are part of the IDA International Development Association formal education system. These schools span grades IEA International Association for the Evalua- 1-4, 1-5, 1-7, or 1-8 and teach communication, tion of Educational Achievement mathematics, and science. Basic education generally ILO International Labour Organisation refers to instruction in literacy and numeracy skills NGO Nongovernmental organization for out-of-school youths and adults. ODA Overseas Development Administration OECD Organisation for Economic Co-operation Secondary education. Secondary education requires and Development at least four years of primary preparation for entry. OPEC Organization of Petroleum Exporting Students may follow a general academic program Countries that typically leads to admission to a postsecondary SIDA Swedish International Development Au- institution, or they may study technical and vocation- thority ai or agricultural curriculums to prepare for direct UNDP United Nations Development Programme entry into a trade or occupation. This level may span Unesco United Nations Educational, Scientific, and any subset of grades 5-12. Cultural Organization UNICEF United Nations Childrens Fund Technical education. Training in specialist skills for USAID U.S. Agency for International Development x Preface Since their independence, the nations of Sub-Saha- tion policy for Africa and noted the very hopeful ran Africa have invested heavily in education. The recent trends in government policy. Clearly, the 1980s achievements in the sector have been impressive both mark an important transition as Africa's economic absolutely and in relation to other sectors and other and demographic policies set the context for produc- countries at other times. In many African countries, tive investment. however, enrollments have stagnated recently, and the Any discussion of policies and priorities for a region quality of education has apparently declined. as vast and diverse as Sub-Saharan Africa naturally These reversals have occurred in an environment of runs the risks of overstating commonalities and under- unprecedented population growth, mounting fiscal stating differences. Readers will fully appreciate the austerity, and often tenuous political and administra- enormous diversity that exists among African cooin- tive institutions. Each of these factors has hurt educa- tries along several dimensions-economic, political, tion in the region, and the ensuing deterioration in institutional, cultural, linguistic, and educational. And educational services has made it more difficult to solve the extent of a country's own internal diversity should the region's economic and social problems. To break not be underestimated. this cycle of eroding prospects for the people of the Most of the discussion and all of the Africa-wide region, policies need to be identified that will renew statistics in this study refer to just thirty-nine countries progress in Africa's education. The role of human south of the Sahara. (They are listed in the statistical skills in development is critical. appendix, which provides full comparative informa- The return to investment in human skills depends, tion on them.) Whenever the term "Africa" or "Afri- nevertheless, on the macroeconomic policy environ- can" is used, it refers only to these countries and their ment. The United Nations and the World Bank have residents. The many averages cited throughout the spelled out the essential ingredients of this environ- paper, although they reflect the thirty-nine most pop- ment. And most African governments have initiated ulous African countries, apply specifically to none. the necessary economic reforms-with some good Individual countries, country blocs, and language results. In 1986 real incomes per capita rose in low- groups will thus provide exceptions to all conclusions. income Africa for the first time in the 1980s. To sustain In the end, national authorities must decide the poli- this growth in personal income, however, there must cies and priorities for educational investments and be a moderation of the region's unprecedented pop- tailor those investments to specific conditions, needs, ulation growth. The World Bank, in Population and aspirations. For these reasons, there is no attempt Growth and Policies in Sub-Sabaran Africa (1986), here to prescribe an education policy for the conti- summarized the widely accepted elements of popula- nent-that would be inappropriate and futile, Instead, xi the focus is on generalizations-on addressing trends Plan of Action, adopted by African Heads of State and issues that, because of their importance in many under the auspices of the Organization of African Afric.', countries, assume importance for the whole Unity (OAU) in April 1980, analyzes Africa's problems continient. and puts forward a comprehensive plan for addressing This study-on investment in education and train- them (OAU, Lagos Plan of Action for the Economic ing-is the first in a series that the World Bank is Development of Africa, 1980-2000, 1981). In accord- preparing to stimulate discussion of sectoral policies ance with the policies set forth in this plan, the OAU for Africa in the 1990s and beyond. It has three main met again and agreed to a set of activities and priori- objectives. The first is to identify and describe com- ties; these were spelled out in a second document mon problems and issues of educational development (OAU, Africa's Priority Programme for Economic Re- in Africa. The second is to provide leaders in each covery, 1986-1990, 1985). Subsequently, the U.N. country with comparative data and analytical tools for General Assembly adopted a set of resolutions pertain- developing their own policies and priorities. The third ing to African development (United Nations, Pro- is to suggest specific policy directions for considera- gramme of Action for African Economic Recovery tion by national education authorities and by donors. and Development, 1986-1990, 1986). Concurrently The key word here is '-consideration." This study is with these efforts, the World Bank has published its about diverse and variably applicable policies for own series of special reports on macroeconomic issues African educational development, not about a mono- and development requirements for Africa. The first in lithic and universally applicable policy. It will meet its this series, Accelerated Development in Sub-Sabaran objectives if it helps initiate serious reflection and Africa: An Agenda for Action, was released in 1981; debate in Africa on future directions for the sector. the fourth and most recent, Financing Adjustment For broader analysis than could be included here, with Growth in Sub-Sabaran Africa, 1986-90, ap- several important documents are available. The Lagos peared in 1986. Education in Sub-Sabaran Africa was prepared by Peter noted in the bibliography. Comments on various drafts R. Moock, task leader, and Ralph W. Harbison under of the study, by World Bank staff and numerous review- the general direction of Aklilu Habte and the immediate ers in Africa, Europe, and North America, helped to supervision of Dean T. Jamison. Janet Leno drafted shape and refine the analyses and recommendations. many of the boxes and provided writing and editorial Finally, thanks to financial assistance from the Norwe- assistance throughout. Rosemary Bellew had principal gian Ministry of Development Cooperation, the study responsibility for preparing the appendix. Birger Fred- benefited greatly from discussion of an early draft by riksen and John Middleton wrote the initial drafts of African policymakers at two international meetings held chapters 4 and 7, respectively. Others who made impor- in Ethiopia and Côte d'lvoire in early 1987, and it was tant contributions to this effort include Wadi Haddad, the topic of a conference of international donors to Price Gittinger, Kenneth King (University of Edinburgh), education held in France in early 1988. and Adriaan Verspoor. The study was discussed by the World Bank's Board Moreover, a series of background papers was commis- of Executive Directors on October 15, 1987. sioned in preparing rhe study; authors and titles are xii Summary African societies have a long and rich history of Progress since Independence education and training. Indigenous education among all groups remains an important transmitter of cul- Between 1960 and 1983 the number of students tural identity from one generation to the next. In enrolled in African institutions at all levels quin- addition, Christianity and Islam have for centuries tupled to about 63 million. Enrollments increased had a pervasive influence on education, community about 9 percent annually during the 1970s, double life, and perceptions in many parts of the region. the rate in Asia and triple that in Latin America. At In the colonial era, missionaries and metropolitan the primary level, the gross enrollment ratio rose governments opened up a network of Western-type from 36 percent in 1960 to 75 percent in 1983. At the schools in Africa. The administration of education tertiary level, the number of students enrolled in systems was dominated, however, by expatriates, as African institutions had reached 437,000 by 1983, was teaching beyond the primary level. Moreover, growing from just 21,000 in 1960. The substantial access to education was quite limited, especially in expansion of education since independence has in- the thinly populated areas of French West Africa. By creased the rarticipation of some groups who had 1960, the gross primary enrollment ratio in all of Sub- previously had little or no access to formal education. Saharan Africa was still only 36 percent, about half This massive educational expansion has substan- the levels then found in Asia and Latin America. tially improved the human capital stock. The esti- Many countries-including The Gambia, Côte mated average educational attainment of working- d'lvoire, and Senegal in West Africa and Tanzania age men and women in the median African country and Somalia in East Africa-had literacy rates below increased from less than half a year in 1960 to more 10 percent at the time of independence. than three years in the early 1980s. The adult literacy rate in the median country rose from about 9 percent Educational Development in Africa to 42 percent. The education systems inherited by the African na- lhe Current Challenge tions at the time of independence were thus quite inadequate to meet the needs of the new countries for The advances since the early 1960s are now seriously self-governance and rapid economic growth. From threatened-in part by circumstances outside educa- this low starting point, the progress achieved in tion. Africa's explosive population growth greatly African education has been spectacular. Quantitative increases the number of children seeking access to expansion has been particularly impressive. schools and increases the number of potential illiter- 1 ates. Between 1970 and 1980 Africa's population African countries, the first will not be obtainable grew at 2.9 percent a year, a full percentage point without the second. higher than the worldwide rate. Between 1980 and the end of the century, Africa's population is pro- A Policy Framework jected to grow at 3.2 percent a year, its primary- and secondary-school-age population at 3.3 percent. If Hard decisions on education policy should not be the growth of educational places is to keep pace postponed. In most African countries the cost would with the growth of the school-age population, more be continued stagnation of enrollment and decline in schools, teachers, books, and other inputs are re- quality through the 1990s. This study urgently re- quired each year. This requirement comes at a time commends that each African nation now embrace the when economic decline has necessitated significant task of formulating and implementing an internally cutbacks in public spending. Public spending on coherent set of policies that reflects the nation's education in Africa has dropped from $10 billion in unique history and aspirations and that effectively 1980 to $8.9 billion in 1983. These fundamental facts addresses its own recently exacerbated problems in sharply constrain the options open to policymakers the education and training sector. Although the par- and have serious implications for African education ticulars of the policy packages can be expected to policy. vary from one country to the next, every country- The main educational issues in Africa today are specific package needs to contain, in varying propor- the stagnation of enrollments and the erosion of tions, three distinct dimensions: adjustment, revital- quality. Although total enrollments in Sub-Saharan ization, and selective expansion. Africa grew at an average annual rate of 6.5 percent Although undoubtedly painful and politically dif- during 1960-70 and 8.9 percent during 1970-80, the ficult, adjustment policies will alleviate the burden of rate of increase plummeted to 4.2 percent in the first education and training on public budgets. Measures threc years of the 1980s. The slowing of enrollment for revitalization and expansion, however, will cer- growth affected all levels of education but was most tainly require additional resources. Thus, in the con- evident at the primary level, where the rate of growth text of ongoing austerity in Africa, resolute move- fell from 8.4 percent a year during 1970-80 to 2.9 ment toward adjustment is a necessary condition for percent in 1980-83. If the population of primary- implementing forward-looking policies on the other school-age children increases at the projected average two dimensions. Moreover, if new policies are in fact annual rate of 3.3 percent, a 2.9 percent increase in to be implemented, management practices will need enroilments will not even keep pace. And as long as to be improved. enrollments stagnate, current inequalities in access to education are not likely to be eliminated. Male- Adjustment female differentials remain a particularly serious problem in most African countries, especially past Adjustment to current demographic and fiscal reali- the primary level. ties, though it will be difficult, is essential if the Complicating the problems of stagnating enroll- disruptive effects of these external factors are to be rnents are the low levels and recent erosions of minimized in the years ahead. Adjustment will take educational quality. Cognitive achievement among two main forms: African students is low by world standards, and there Diversifying sources of finance. A necessary part is some suggestion of further decline recently. Much of country-specific policy packages, diversification of the evidence is indirect: supplies of key inputs can be achíeved through increased cost sharing in (especially books and other learning materials) are public education and through increased official critically low, and the use of these inputs has declined tolerance and encouragement of nongovernmental in relation to the use of teachers' time and of physical suppliers of educational services. For many African facilities. Less is known about the output-the per- countries, increased user charges in public educa- formance of students. But in the few cross-national tion will be inevitable, but this policy should be studies that have been conducted, academic achieve- directed especially at the tertiary level, where more ment in Africa has been sufficiently poor to be a cause than a third of public expenditure in the typical for serious concern. country now covers student welfare costs, as dis- Addressing these issues of stagnation and low tinct from pedagogical costs; for most African quality will require additional resources. Equally countries, the scope for further cost sharing in important, it will require profound changes in edu- primary education is negligible or nonexistent. cational policy for many countries. Indeed, for most * Unit cost containment. In the adjustment process, 2 the containment of unit costs will be just as impor- educational services that shift more of the burden tant as, and in many countries probably more for learning onto the students themselves. Now is important than, policies to diversify sources of the time to begin planning such programs and finance. The most promising areas for containing developing the correspondence materiais, radio costs are utilization of teachers, construction stan- programs, examinations systems, and other sup- dards, and the tendency of students to repeat port that will be needed. grades or drop out of school. Training. Training for those who have entered the labor force must be increased. This training should Revitalization serve both school leavers and those who have had no exposure to formal schooling, and it should be Revitalization of the existing educational infrastruc- designed to ensure that individuals can acquire the ture is the second dimension of a properly conceived necessary job-related skills and renew these skills educational strategy. Renewed emphasis on funda- during their working lifetime in response to chang- mentals is needed to take maximum advantage of the ing market conditions. current capacity of education and training systems. * Research and postgraduate education. Expansion Three kinds of measures are necessary for the resto- of Africa's capacity to produce its own intellectual ration of quality: talent to fill the highest scientific and technical * A renewed commitment to academic standards, jobs-in educational establishments, in govern- principally by strengthening examination systems. ment, and in the private sector-is a critical matter * Restoration of an efficient mix of inputs in educa- to be addressed in building for Africa's future. tion. A minimum package of textbooks and other Here, as with programs for distance education, learning materials is usually the most pressing economies of scale are likely to be important, and need. these will be difficult to achieve fully within a Greater investment in the operation and mainte- national context except, perhaps, in a few of nance of physical plant and equipment, and Africa's largest and wealthiest countries. The greater expenditure on other inputs that would pressing need is for Africa to develop, probably increase the utilization of these capital assets. with the support of the international donor com- munity, regional and subregional approaches to Selectve Expansion these particular goals. The selective expansion of educational services is the Formulation and Implementation third dimension of any complete strategy for educa- of National Programs tional development. Measures in this area, viable only after measures of adjustment and revitalization For most African countries, the formulation of a have begun to take hold, will concentrate in four comprehensive and coherent educational develop- areas; success in all will depend upon a general effort ment program, derived from a balanced package of to safeguard the quality of instructional staff at all policies for adjustment, revitalization, and selective leveis: expansion, will be a new experience. Each country * Renewed progress toward the long-term goal of will organize for the task in its own way. In many universal primary education. Expanding access to countries, however, a fruitful approach to policy primary education should remain a high priority design might be expected to include the following: in most African countries. To maintain the high establishing a national commission to oversee the economic and social returns that have accrued to work; constituting a technical staff to support the this investment in the past, however, parallel ef- commission; for both, drawing upon the nation's forts are required to combat the incidence of best political judgment and analytical talent, from disease and malnutrition among young children. the ministries of finance and planning as well as * Distance education programs. At the secondary education, and from institutions of tertiary education level, and later on at the tertiary, expansion of and research; building a national consensus through enrol!ments in selected subjects and streams will public debate of the emerging findings and recom- be necessary in most countries as soon as appro- mendations; and taking advantage of the experience priate measures of adjustment and revitalization of other African countries in developing the nation's have been put in place. To accommodate these own educational development program. Budgetary increases in postprimary education, most countries flows would have to be sufficient to cover not only will need to consider alternative ways of delivering the personnel costs of the national commission but 3 also its operating expenses for travel, communica- education by changing the input mix to include more tsons, publications, and specialized contractual ser- textbooks and learning materials. At the secondary vices (such as data collection and processing, expert level there is far more scope for containing unit technical consultants, targeted research, and analy- costs-partly through making fuller use of availabie sis). resources and partly through switching to cheaper Although the careful elaboration of educational ways of providing services. Higher education poses a development programs is essential, African capacity set of problems uniquely its own. Rapid expansion for implementation will ultimately determine the has left in its wake an abundance of institutions, effectiveness of the programs. Improvement in edu- programs, and graduates that are often of low quality cation management is a necessary concomitant to and dubious relevance. Modest consolidation, an policy reform and must be given immediate and adjustment measure that would lower unit costs, and continuing attention. increased cost sharing are the first steps on the road Management at the national level may be signifi- to higher educatíon's revitalization, which must be cantly improved by delegating various administrative regarded as a prerequisite for any further expansion functions. Some functions, however, must appro- of the subsector. priately remain with the central ministry, and the performance of these functions will need to be im- Pnrmary Education proved. But the toughest ch, ilenge to improving management lies closer to the classroom, at the level During the 1960s and 1970s some analysts warned of individual schools and districts. African policy- that Sub-Saharan Africa's preoccupation with the makers should consider how, with adequate safe- quantity of education would lead to a serious deteri- guards against abuses, schools and the local com- oration in its quality. In the countries where educa- munities they serve can be given more authority in tional standards have deteriorated the most, the the acquisition and utilization of the resources essen- choice between expansion and quality is no longer an tial to effective classroom teaching and learning. either-or choice. Without some basic revitalizing In addition, central ministries must tend more ser- inputs, particularly textbooks and instructional ma- iously to the development of their own managerial terials, almost no learning can be expected to occur. capacity, especiaily in monitoring performance and Ensuring the availability of essential inputs is a policy planning and analysis. Improvements are needed prerequisite both for quality and for expansion. Be- in examination systems (which was mentioned above yond this minimal level, however, are the questions with reference both to academic standards and to of finding for each country an appropriate balance of distance education), in the nature and timely availabil- quality and quantity-and of identifying efficient ity of statistical and financial accounting information, approaches to enhance quality and expand enroll- and in the number and qualifications of staff engaged ments, and of financing both improvements. in full-time analytical work. Incentives in many minis- A review of possible measures for improving the tries of education are insufficient to attract, motivate, quality of primary education yields two broad con- and retain able staff. Governments committed not only clusions: to the formulation of an educational development First, the safest investment in educational quality program but also to its expeditious implementation will in most countries is to make sure that there are need to be imaginative in addressing the issue of enough books and supplies. These materials are incentives at all leveis of the education system. effective in raising test scores and, almost invari- ably, are underfunded currently relative to teach- Policy Options by Level of Education ers' salaries. External aid might be used to address this problem in the short run, especially where The mix of adjustment, revitalization, and expansion foreign exchange is a governing constraint. Other components that is appropriate in light of country- possibilities for improving quality are found in specific conditions and goals can be expected to school feeding and health programs, intensive use differ. Within a country, the mix will differ also of radio, in-service education of teachers in subject among levels of education. matter, and stronger systems of inspection and At the primary level in most African countries, supervision. there is only limited scope for adjustment in the form Second, some investments are not likely to have a of either lowered unit costs or increased cost sharing. noticeable effect on primary school quality des- There is, however, good potential for improving the pite their potentially high costs. These invest- quality of output-that is, for revitalizing primary ments include reducing class size (within the 4 range of 25 to 50), providing primary teachers and how to make the transition from programs and with more than a general secondary education, subjects that have broad vocational relevance (lan- providing teachers with more than minimal expo- guage, mathematics, and science) to programs and sure to pedagogical theory, constructing high- subjects that will prepare individuals for specific jobs quality buildings, and introducing televisions or or clusters of jobs. International experience shows computers into classrooms. that a strong general education, which schools can Even with no quality improvements (and assuming provide efficiently, greatly enhances an individual's no improvements in efficiency), the resources devoted future trainability. It also shows that job-specific to the primary level would need to increase more training is very important. Such training usually is than 3 percent a year just to keep pace with popula- most efficiently provided after initial job decisions tion growth. There is, however, little likelihood of have been made and in institutions under, or strongly significantly reducing unit recurrent costs at this influenced by, the ultimate employer. Occupation- level, especially if countries hope also to improve the specific and job-specific training need not provide quality of education. There is greater scope for individuals with degrees or credentials. reducing unit capital costs, but even so it is unlikely Because of the high costs and tenuous vocational that overall per pupil costs can be reduced very much relevance of much school-based training for specific at the primary level. Thus the further growth of jobs and occupations, there is an urgent need to primary education will require additional resources. establish industrial training centers and to encourage (through incentives and technical assistance) local Secondary Education and Training enterprises to offer skill development programs and other types of on-the-job training. Governments in- Can adjustment measures at the secondary level terested in laying the groundwork for a more techni- generate sufficient savings to provide the necessary cally oriented economy should place heavier empha- books and supplies and, at the same time, significant- sis on general mathematics and scientific skills in the ly expand capacity? Substantial economies are possi- secondary and postsecondary curriculum. These pro- ble in the operation of regular schools. Boarding grams are relatively inexpensive and are generally costs can also be reduced. And, most significant, more conducive to economic growth than is in- there can be economies from the creation of distance school vocational education. education systems that combine radio and corres- pondence techniques and reduce (but do not elimi- Higher Education nate) the amount of face-to-face interaction with qualified teachers. Such systems would extend sec- Preparing and supporting people in positions of re- ondary education of reasonable quality to many sponsibility-in government, business, and the pro- more communities than could be reached in any fessions-is the central and essential role of the other way for the same price. continent's universities. In numbers at least, the uni- Beyond this general support for expanding access versities have risen impressively to this challenge. to secondary education, African governments should Enrollments grew from 21,000 in 1960 to more than give serious consideration to policies designed to 430,000 in 1983. remedy existing inequalities in school participation. Higher education's contribution to development in Females are the largest group underrepresented in Africa is being threatened, however, by four interre- postprimary education in Africa. Small, community- lated weaknesses. First, higher education is now based schools (whether or not they rely on distance producing relatively too many graduates of programs education) will tend to attract girls more readily than of dubious quality and relevance and generating too larger schools located in urban centers and at greater little new knowledge and direct development sup- average distances from homes. Although smaller con- port. Second, the quality of these outputs shows ventional schools may imply higher unit costs, this is unmistakable signs in many countries of having de- not necessarily true if larger schools tend to include teriorated so much that the fundamental effectiveness boarding facilities while smaller schools do not. of the institutions is also in doubt. Third, the costs Increasing the number of female teachers may also of higher education are needlessly high. Fourth, the attract more girls, especially in Islamic areas. pattern of financing higher education is socially ine- Another set of issues to be addressed mainly at the quitable and economically inefficient. secondary level concerns the relevance of the curri- Wherever the foregoing diagnosis of weaknesses in culum to the needs of individuals and societies. The higher education can be confirmed, policy reform main questions that many countries face are when should seek four objectives: (a) to improve quality; 5 (b) to increase efficiency; (c) to change the output cialized technical expertise without direct financial mix, which may imply smaller enrollments in certain or political ties to any government or international fields of study; and (d) to relieve the burden on public donor. African governments could call on this sources of financing by increasing the participation expertise for help in formulating policies at the of beneficiaries and their families. But quality im- outset, and in monitoring, evaluating, and correct- provements, the first objective, will cost money. ing them during implementation. Thus, implementing adjustment policies to achieve Appropriate mechanisms do not now exist for the other three objectives will, almost everywhere in meeting these three interrelated needs for the im- Africa, be a prerequisite for freeing the resources provement of policy development. The requirement needed to achieve the first. is for expeditious action to develop them. Any donor initiative in this area that would take more than a The Role of the International Donor Community year to get started would be an inadequate response to the needs of African governments. This study argues that adjustment measures are need- Beyond its assistance in policy design, the interna- ed to alleviate the burden of education and training tional community should help finance the implemen- on public budgets in Africa. The "savings" generated tation of sound programs. These programs will typ- by lower unit costs, increased cost sharing, and ically require more resources, for a longer period, greater tolerance of the private provision of educa- than can be mobilized internally. Countries that have tional services can be used to help fund the necessary demonstrated their willingness to address policy is- revitalization and ultimate expansion of the sector. sues should have access to increased, longer-term, Regrettably, all such savings from adjustment and more flexible international aid. To the extent measures will not be sufficient, in most countries, to that a country's policy package involves thoroughgo- cover the substantial resources needed to revitalize ing reform, there are likely to be substantial one-time and build African education to the extent essential transition costs to a new and more sustainable policy for future development. International aid will remain regime. a critical determinant of the pace of progress of The international commitment to the reform pro- education in the region. However, the rapid evolu- gram must be seen from the beginning as continuing, tion of African needs, as summarized in the three a characteristic that has been missing in the past. In dimensional framework for policy reform, demands addition, the sum total of aid to different levels corresponding changes in the organization and na- of education and different expenditure categories ture (and not just an increase in the level) of interna- should reflect, at least in very rough terms, the tional aid for African educational development. priorities given to these leveis and categories in the The pressing requirement is for aid in support of national program. In recent years, only 7 percent of policy reform. The international donor community international aid to African education has been used should quickly offer three related kinds of support for for primary education, compared with 16 percent for the design of national policy. general secondary education, 33 percent for voca- • The first form of planning support is simple: seed tional and technical education (including teacher money to cover both the local and foreign costs of training), and 34 percent for higher education. In developing policies and improving management. terms of expenditure categories, only 11 percent of The willingness of the international donors to bear aid has been used to support operational costs (local a part of these extraordinary expenses, perhaps on salaries, consumable supplies, and instructional ma- a matching basis, would provide an important teriais). The donor organizations, both individually incentive for African governments to review their and as a group, need to review these allocations to policies for the sector. make sure that they are consistent with the programs * The second kind of support that the international being formulated by African governments for the community should provide is ready access to the development of education and training. ongoing experience of other countries in formulat- ing and implementing policy reform. Intensive Why Meet the Challenge? collaboration among countries, so that they share The Expected Benefits of Education their accumulated experience widely, should pay high dividends as countries grapple with common Greater investment in education can, at this time in issues. Africa's history, be expected to yield broad economic * Third, the international donor community should benefits. These benefits include higher incomes and establish and finance a source of high-quality spe- lower fertility. The research evidence to this effect is 6 compelling. A caveat, however, is in order. The Men and women with more education, in addition to studies examining the welfare benefits of education having fewer children, tend to live healthier and are based necessarily on education as provided at longer lives. And numerous studies have shown that some historical point in the past. To the extent that parents' education affects children's survival and the quality of education has declined recently and is enhances their physical and cognitive development. allowed to deteriorate further, new investments in the The benefits of education go far beyond those for quantity of education may not yield returns commen- income and fertility, however. The rapid transition in surate with those in the past. Hence the strong Africa from colonial status to self-government to emphasis in this study that quality be enhanced participation in the international arena was possible through revitalization as a prerequisite and comple- only because African educational systems produced ment to further expansion. people to replace expatriates at all leveis. The nur- Assessments of the !abor market returns to past turing of leaders who can address the increasingly investments in education have consistently found complex tasks of nation-building is a continuing rates of return above 10 percent and sometimes responsibility of African education. In addition, the above 20 percent-rates that compare favorably with stock of human capital in Africa will determine those in most sectors in Africa today. A recent study whether Africans can harness the universal explosion on the long-term impact of educational investments of scientific and technical knowledge for the region's on development in thirty-one African countries cor- benefit-or whether Africa will fali farther and farth- roborates the microeconomic findings of education's er behind the world's industrial nations. Above all, high returns. education is a basic right, an end in itself, an intrinsic Increased investment in the quality and quantity of part of life and development. When all the benefits education can also be expected to reduce fertility. In of education are considered, the case for revitaliza- general, there is a strong negative relationship be- tion and expansion of schooling and training in tween how much education a woman receives and Africa is compelling, even in this period of unusal the number of children she bears during her lifetime. scarcity. 7 The Policy Context The Remarkable Progress of African Education Around the time that most countries of Sub-Saharan period. Especially in northeastern Africa and the Nile Africa gained independence from colonial rule, the Basin, Christianity has thrived for more than 1,500 region lagged far behind the rest of the world on years. In about the year 450, the Ethiopian Christian nearly every indicator of Western-style educational Church, a prime example, established a comprehen- development. Efforts since then have been truly dra- sive system of education that provided an underpin- matic, especially during the 1960s and 1970s, even ning for Ethiopian cultural, spiritual, literary, scien- though they did not cdose the education gap. The tific, and artistic life. record of this period is a tribute to the determination A third major antecedent to the colonial period is of African leaders and the sacrifices of African par- the influence of Islam on African education. Arab ents in their quest to provide a better standard of culture and language were adopted in much of North living for their children's generation. Africa, and the Islamic faith also won converts in the Sahelian zone, along the coast of East Africa, and in Before Independence much of the Horn of Africa. Both formal and non- formal school systems were established to teach the African societies have a long and rich history of ethics and theology of Islam; they included a small educational traditions. Indigenous education was of- number of elite centers of excellence such as the ones fered by all ethnic and linguistic groups and remains at Tombouctou in Mali and at Lamu on the east an important transmitter of cultural identity from coast. Designed to impart skills and knowledge with- one generation to the next. It aims to instili in in the religious realm, the Islamic education system children the attitudes and skills appropriate for male emphasizes reading and recitation in Arabic. and female social roles, emphasizing the duties and The Western colonial period in Sub-Saharan Afri- privileges derived from cultural values. Imparted ca began with the arrival of the Portuguese in the through language and example at home as well as in fifteenth century and ended only recently. Of the formal lessons and rituals outside the home, indigen- African countries covered in this study, Ethiopia and ous education responds to the concrete problems of Liberia alone have been sovereign states for longer local communities. It prepares political leaders and than thirty years. All of the other countries achieved ordinary farmers, and it engenders a sense of citizen- sovereign status in the relatively short period be- ship in the people of the community. tween 1957 (Ghana) and 1980 (Zimbabwe). Colonial Africa's early Christian heritage represents a sec- precedents are still much in evidence in most of ond important element of education in the region, Africa, and they sometimes constrain the degree to with roots extending back long before the colonial which governments are free to initiate new policies. 11 The principal suppliers of Western-style education unequal educational participation frequently tran- before independence were the colonial governments scended colonial boundaries. Participation patterns themselves and the African missions of the Roman in northern Nigeria, for example, had less in com- Catholic and various Protestant churches. The divi- mon with those in the south of the same British sion of administrative and financial responsibilities territory than with those in the north of neighboring between government and church differed from one French Cameroon. Such within-country differences colonial regime to the next. The British, for example, and between-country similarities remain evident to- were generally more tolerant of religious and local day. community autonomy than were the French. Transition rates from one educational level to the In their quest for converts and literate African next were low in 1960, and dropout rates were high. subjects, the missionaries and colonial governments As a result, enrollment pyramids were typicaily very opened up a network of schools in the region. Many narrow at the top. Only 6 percent of all Sub-Saharan were of a high standard. Yet the curricula were based enrollments in 1960 were at the secondary level, and for the most part on overseas models and reflected tertiary education was virtually nonexistent until the little in the way of African content. The administra- very end of the colonial period. The gross enrollment tion of "modern" education systems in Africa was ratio at the secondary level in Africa was 3 percent dominated by expatriates, as was teaching beyond in 1960, compared with 14 percent in Latin America the primary level. and 21 percent in Asia. The ratio at the tertiary level Nevertheless, the economic changes that the colon- was 1 to 500, about one-sixtieth those then found in ial powers set in motion in Africa helped create a Asia and Latin America. According to Unesco fig- demand for Western-style education that, in many ures, at tbe time of independence there were only 90 areas, seemed nearly insatiable. Education became African university graduates in all of Ghana, 72 in the vehicle for moving, within one generation, from Sierra Leone, and 29 in Malawi. When Botswana peasantry and poverty to the topmost ranks of socie- became independent in 1966, 96 percent of higher- ty. This fact of modern-day life escaped few African level posts in the country were filled by expatriates. parents looking for ways to promote a better future for their children. Advances after 1960 Access to education was quite limited, however, especially in the thinly populated areas of French The systems of education inherited by the African West Africa. In 1960, the gross primary enrollment nations at the time of independence were altogether ratio in all of Sub-Saharan Africa was still only 36 inadequate to meet the needs of the countries for self- percent. This was about half the levels then found in governance and rapid economic growth. From this Asia (67 percent) and Latin America (73 percent). low starting point, the progress achieved in African The enrollment ratio was 38 percent in the Franco- education has been remarkable. phone territories (50 percent in the Belgian colonies Quantitative expansion has been particularly im- and just 31 percent in the French) and 40 percent pressive. Between 1960 and 1983 the number of in the Anglophone. Many countries, including The students enrolled in African institutions at all leveis Gambia, Côte d'Ivoire, and Senegal in West Africa quintupled to about 63 million students (appendix and Tanzania and Somalia in East Africa, had over table C-2). Enrollments increased about 9 percent 90 percent illiteracy at the time of independence. annually between 1970 and 1980, double the rate in There were also significant differences in educa- Asia and triple that in Latin America. The substantial tional access and participation within colonial terri- expansion of education after independence has in- tories-between urban and rural populations, males creased the participation of some groups who had and females, and members of different ethnic or had little or no access to formal education. religious groups. Such patterns stemmed from a va- Primary school enrollments increased the most in riety of causes. Different African peoples were re- absolute terms, growing from approximately 11.9 garded and treated differently by colonial adminis- million pupils in 1960 to 51.3 million pupils in 1983 trators; the costs of providing education differed, (see table 1-1). The gross primary enrollment ratio certainly between urban and rural areas; some pop- rose from 36 percent to 75 percent over this period. ulation groups were more responsive than others to In only three of the region's thirty-nine countries educational opportunities; and most Africans, re- (Congo, Lesotho, and Mauritius) was the enrollment sponding to the incentives imposed by patrilineal ratio higher than 80 percent in 1960; by 1983, sixteen customs, preferred education for their sons to educa- countries had achieved this milestone (see maps 5 and tion for their daughters. As a result, problems of 6 in the appendix). 12 Table 1-1. School Enrollments and Enrollment Ratios in Sub-Saharan Africa, 1960 and 1983 Levei 1960 1983 Primary education Enrollment (thousands) 11,900 51,300 Gross enrollment ratio (percent) 36 75 Secondary education Enrollment (thousands) 800 11,100 Gross enrollment ratio (percent) 3 20 Higher education Enrollment (thousands) 21 437 Gross enrollment ratio (percent) 0.2 1.4 Total enrollments (thousands) 12,700 62,900 Source: Appendix table C-2. In fact, twelve of the thirty-nine countries now much higher initially, enrollments had increased by have gross primary enrollment ratios equal to or only a factor of fifteen. For every student enrolled in greater than 100 percent (appendix table A-7). Enroll- higher education in the Francophone countries in ment ratios in excess of 100 percent arise from the 1960, there were four enrolled in the Anglophone definition of the gross enrollment ratio, which divides countries; by 1983, the ratio was only about 1 to 1.5. the enrollment by the total (male and female) popu- The building of schools and training of teachers to lation of school age. The official age range differs accommodate the additional students throughout the from country to country depending on the structure region were mammoth achievements. Between 1960 of the education system. The gross enrollment ratio and 1983 the number of primary schools in Sub- is an accurate measure of the school system's capac- Saharan Africa increased from about 73,000 to ity relative to the school-age population. By including roughly 162,000, and the number of primary school students who are outside the official age range, how- teachers, from 310,000 to more than 1.3 million. ever, it overstates the percentage of this population Although the average pupil-teacher ratio remained that is actually enrolled. The net ratio, which ex- roughly the same during this period (approximately cludes overage and underage children, is conceptual- 39 to 1), the average primary school size increased ly superior, but generally this information is not from 162 pupils in 1960 to 317 in 1983, almost available. (Problems with gross enrollment statistics doubling. are further discussed in the technical notes for appen- The number of teachers employed at the secondary dix tables A-7 to A-9.) levei increased eightfold, going from about 46,000 in Enrollments at the tertiary level increased the most 1960 to about 373,000 in 1983. This understates the in relative terms, especially between 1960 and 1970. increase in African teachers, since many expatriate This happened because most African nations were teachers were replaced during this period. At the emphasizing higher education in an effort to alleviate tertiary level, the number of institutions in Sub- manpower shortages and also because enrollments Saharan Africa has more than tripled since 1960 and had started from a low base. In 1960 there were only today exceeds 80. (See appendix table A-10; the about 21,000 university students in Africa (approxi- numbers of schools and teachers in the preceding two mately one in 500 of the age group) and a few paragraphs reflect reasonable assumptions about thousand studying in foreign universities; by 1983, missing data in the table.) 437,000 (seven per 500) were enrolled in African This massive educational expansion, which began institutions, and a further 100,000 Africans were in some countries in the 1950s and intensified every- studying abroad. The relative increase in tertiary where after independence, has built up the stock of enrollments was particularly dramatic in French- human capital considerably. Using data from coun- speaking Africa. In the group of eighteen Franco- tries that conducted censuses about 1980, it is possi- phone countries, there were forty times more stu- ble to examine the changing educational attainment dents enrolled in higher-level institutions in 1983 of successive cohorts of individuals who were of than in 1960. By way of contrast, in the sixteen school age at different times during the thirty years Anglophone countries, where enrollments were beginning in 1950. For the few countries that have 13 collected data, the proportion of males who reached ing school. Approximately a quarter of ali African adulthood without having attended school declined countries have launched programs to eradicate adult from 31 percent in the early 1950s to 22 percent in illiteracy. the late 1970s in six Anglophone countries, and from Other forms of adult and nonformai education 59 percent to 42 percent in four Francophone coun- have also been strengthened in the first two decades tries. For the same countries and the same period, the of independence. Many universities now have depar~- proportion of males who completed primary school ments and institutes dedicated to the production and rose from 47 percent to 58 percent in the Anglophone study of adult education, and nongovernmental bod- countries, and from 16 percent to 30 percent in the ies concerned with functional literacy and income Francophone. Estimated mean years of schooling for generation in urban and rural areas have prolifer- the male population (over age 15) increased from 4.6 ated. to 5.4 years in the Anglophone countries and from In addition to the remarkable quantitative achieve- 2.0 to 3.4 years in the Francophone. ments, many other significant changes have occurred An alternative to using census data for estimating in African education over the past quarter century. stocks of education is to add up past enroliments. In Especially at the primary level, Africanization of the conjunction with demographic assumptions, this al- curriculum has been widespread. In nearly every part lows estimates to be made, country by country, of the of Sub-Saharan Africa, texts have been adapted and educational attainment of the working-age popu- new texts written so that basic skills are now taught lation. Table 1-2 reports estimates of educational with reference to African customs, the local environ- stocks produced in this fashion. The estimated aver- ment, and the area's own history. Twenty-one of the age educational attainment of working-age men and thirty-nine countries in the region now officially women in the median African country increased from begin instruction in one or more African languages, 1.0 years in 1970 to 3.3 years in 1983. The figures in rather than asking children to use a European lan- the table underline the wide disparities that remain guage as the medium of instruction from their first in this variabie, especially between the Sahelian (low- day in primary school. These transformations have income semiarid) countries and the rest of Africa. made the African classroom far more conducive to Table 1-2 also shows how literacy rates have risen learning and have enabled children to acquire cogni- since 1960. In the median African country, the per- tive skills more quickly. centage of adults reportedly able to read and write has increased by a factor of nearly five. This progress Expenditure on Education reflects both the growth of formal education and, in many countries, the success of programs to promote The amount of resources allocated to education literacy among adults and young people not attend- reveals the high degree of commitment of African Table 1-2. Indicators of Educational Progress Median Median estimated literacy rate number of years of (percent) schooling of working-age population Latest av'ailable Country group 1965 1983 1960 year Economic status Low-income semiarid 0.1 0.9 2 15 Low-income other 0.5 2.9 10 41 Middle-income oil importers 1.3 4.2 19 72 Middle-income oil exporters 0.7 3.6 16 56 Linguistic Francophone 0.5 2.4 7 40 Anglophone 1.2 3.4 18 58 Sub-Saharan Africa 0.5 3.3 9 42 Source: Appendix tables C-3 and C-4. 14 nations to educational development and explains the $6.3 billion, up 66 percent over 1970. By 1980 it was significant advances that have occurred in the sector nearly $10.0 billion, up another 58 percent. Between since 1960. 1980 and 1983 public expenditure fel] somewhat, down approximately $1.1 billion, to $8.9 billion Public Domestic Expenditure on Education in constant 1983 dollars. This probably overstates somewhat the decline in real education expenditure In 1970 the thirty-nine countries of Sub-Saharan since, in most African countries in recent years, Africa allocated approximately $3.8 billion of public teachers' salaries (which constitute approximately 90 d dínestic expenditure to the education sector. This is percent, 70 percent, and 50 percent of recurrent a World Bank estimate of total (capital plus recur- expenditure in primary, secondary, and tertiary edu- rent) expenditure (see box 1-1), expressed in constant cation, respectively) have not gone up as rapidly as 1983 U.S. dollars. By 1975 this figure had reached prices in general as reflected in the GDP deflator. Box 1-1. Aggregate Public Domestic Expenditure Indexes of Total Enrollment and Expenditure on Education in Sub-Saharan Africa on Education, 1970-83 The discussion of public education expenditure in the Ildex text is based on information provided by Unesco and 280 presented in appendix tables A-14 to A-23. Unesco local- currency expenditure figures were inflated to reflect 1983 _. prices using country-specific GDP deflators and then con- verted to U.S. doilars using official 1983 exchange rates. Total expenlditure Because, for any given year, expenditure data have been \ / reported for only a subset of countries, the total of public ...... domestic education expenditure for ali thirty-nine Afri- 200 - r s s s nan - A:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. . .... .. . .. . can countries is not known. For example, 1980 expendi- s . r tures are known for only twenty-nine of the thirty-nine _...... ..... countries (see appendix table A-14). . :: Total enroiImen In an attempt to bypass this data limitation, the tabie .... in this box reports estimated totals of pubIic education expenditure in 1970, 1975, 1980, and 1983 for the thirty- nine countries as a group. These aggregates were arrived at by taking all countries for which actual data were ..". availabie for the particular year and estimating, for this R0 R rrtt exr.n....re subset, the average relationship between expenditure and certain other measures including per capita income and enroliment at each of the three education levels. The ......--.... estimated relationship was then used to impute expendi- ............................ ................................... ture figures for the remaining countries for which actual .............. . values were not reported. These estimated values were ::-.:: xc ir then added to the sum of reported values to arrive at the . .......itai................. .......... aggregate values given in the table. 0 1970 197.5 1980 1983 Public Domestic Expenditure on Education, 197083 Item 1970 1975 1980 1983 Actual data available (see appendix table A-14) Number of countries 27 26 29 28 Percentage of Sub-Saharan African population 60 77 89 75 Estimated expenditure for ali 39 Sub-Saharan African countries (millions of 1983 doilars) Capital 459 1,767 1,316 865 Recurrent 3,329 4,516 8,636 8,032 Total 3,788 6,283 9,952 8,897 15 Although data are not available for 1984 and 1985, Another way to look at public expenditure on expenditures may have resumed growth after 1983, education is as a percentage of total public expendi- when the economic crisis was at its peak. ture. On this indicator, the African countries have Since the use of official exchange rates may give a clearly treated education generously in the past. The false picture of expenditure over time and may also median African country allocated 17.6 percent of the distort comparisons between regions or countries, an budget to education in 1970, 17.4 percent in 1975, alternative measure of public expenditure on educa- 18.5 percent in 1980, and 15.3 percent in 1983. In tion is useful. Expenditure as a percentage of national terms of weighted means, as distinct from medi- income is an especially meaningful indicator of a ans, education's share went from above 16 percent government's effort in the area of education. Figure throughout the 1970s to below 12 percent in 1983. 1-1 shows public domestic expenditure on education This more precipitous drop in education's share of as a percentage of national income in Africa and expenditure after 1980 reflects, to a very large extent, elsewhere in four years between 1970 and 1983. The what was happening in a single large country- figure shows that Sub-Saharan Africa has been allo- Nigeria (see appendix table A-14). At the lower end cating a larger share of total income to education of the spectrum, Malawi, Nigeria, and Somalia all than have the developing countries in general, but spent below 10 percent of the central government still a smaller share than developed countries have budget on education in 1983. At the upper end, managed to do. The figure also corroborates the education's share was above 20 percent in Burkina decline in government expenditure on education af- Faso, Côte d'Ivoire, Mali, Niger, Rwanda, and Togo. ter 1980. With regard to the distribution of expenditure across levels of education, the recurrent expenditure on primary education was between 43 percent and 49 _percent of total recurrent expenditure in the median African country from 1970 to 1983. The share going Figure 1-1. Percentage of National Income Spent to secondary education was between 25 percent and on Public Education in Sub-Sabaran, Developing, 31 percent, and the share going to tertiary, between and Industrial Countries, 1970-83 13 percent and 18 percent. The "unspecified" catego- ry remained nearly constant at about 10 percent (see Pcrcciit appendix table A-16); as a residual, it cannot be taken 7 very seriously but would presumably include most allocations for central administrative services and for adult education in ministries of education. 6 ----------------------------- Pnvate Spending on Education ,,,--' Induistrial coulntries The focus in the appendix tabies on government expenditures reflects the relatively better data avail- able on public, as distinct from private, sources of educational finance. Although information on the total of private spending is not available in most /Sub-Saliaraii Africa countries, it is clear that families and nongovernmen- tal organizations bear a significant and growing por- 4 tion of the financial burden of education in much of _- - .-Africa. It has been estimated, for example, that private expenditure accounted for 14 percent of total countries national spending on education between 1975 and Devcloping coiltries 1980 in the Sudan, 23 percent in Tanzania, 31 percent 3 ,/ in Zimbabwe, 48 percent in Sierra Leone, and 53 percent in Ghana. These figures tend to be larger today than they were in the early years after independence. 2 1 1 Government-collected fees (which in official statistics 1970 1975 1980 1983 are often counted as part of "public" expenditure on education) have been imposed or increased recently 16 in many African countries. In addition to these fees, A-24 to A-27). In 1981-83, the total of bilateral, there are significant private outlays that are not multilateral, and private voluntary organization aid channeled through government and are difficult to to African education granted through central govern- quantify. They include most fees paid to private ment ministries of education (and not through other education institutions; the privately borne costs of departments of government, nor directly to individ- such items as transportation, school uniforms, text- uals) was estimated to be about $915 million a year. books, and supplies; and, especially important at the This figure includes about 3 percent from the IBRD of primary leve], family and community outlays, either the World Bank Group and from the African Devel- in cash or in kind, for the construction or repair of opment Bank that was in the form of market-rate school buildings in public education. Commitment of loans. The remaining 97 percent can be counted student time is, of course, another valuable input to as concessional aid to African education. Nearly a the education system, but this study does not attempt quarter of the total was in the form of scholarships to account for it. for African students to study abroad; a significant During the colonial period, much of Western-style portion of aid to education (about 44 percent) was in education in Africa was organized and, to a significant the form of technical assistance. extent, financed by missionary groups that often relied African governments received an estimated addi- heavily on community self-help. After independence, tional $394 million of external aid each year to governments assumed responsibility for most such finance project-related training in sectors other than education in nearly every African country, Lesotho education. In addition, when African students study being a major exception to this rule. During this abroad, the host countries incur indirect costs- period, wherever school fees had been charged, they nonexplicit subsidies that students, including foreign were often reduced or eliminated. More recently, students, receive at most universities. A very rough however, in response to growing fiscal and demograph- estimate of the total of such subsidies to African ic pressures, African leaders have condoned a reversal students who were studying abroad between 1980 of this postindependence trend. In the late 1970s and and 1982 is $245 million annually. 1980s students, their families, and communities have The total of all categories of external aid to Afri- been asked to contribute ever larger amounts toward can education and training in the early 1980s was the costs of education. Often, to avoid having to collect about $1.6 billion annually. This is a significant school fees, governments have referred to these sub- amount, nearly $4 per capita, but there is no evidence scriptions by another name, such as development fund to suggest that external aid has increased in recent contributions or activity payments. years to compensate for the decline in African gov- ernment expenditures. To the contrary, the total of Foreign Aid Flows to African Education all aid to Africa for all purposes was smaller in 1983 than the 1980-82 average, and smaller in 1984 than Foreign aid also accounts for a significant part of in 1983. Further discussion of external aid to educa- total spending on African education (appendix tables tion in Africa appears in chapter 9. 17 ~,th Educa tion and the Extemal En vironmnt lhe impressive gains recently won in African educa- fertility rates have remained, for the most part, at the tion are now seríously threatened by círcumstances sarne hígh leveis that prevailed historically. In 1960 outside the sector, including Africa's explosive pop- the crude death rate in Sub-Saharan Africa ,wsv~ ulation growth, which swells the numnber of potential twenty-five per thousand and the crude birth rate jilliterates on the continent, and Africa's recent eco- forty-nine per thousand. By 1983, owing to signifi- nomic decline, which has necessitated significant cant improvements in health care, the death rate had cutbacks in public spending. Although economic fallen to seventeen per thousand, but the birth rate prospects may have brightened somewhat since 1983 remained practically unchanged. Bírth rates actuaIly and population growth has begun to slow in a increased over this period in a handful of countries. handful of countries, for most of Sub-Saharan Africa Africa's rapid population growth creates serious current economic and demographic realities under- problems for education. For the growth of education- mine both the quantitative and qualitative education- ai places to keep pace with the growth of school-age al advances achieved since independence. Retrogres- children, more schools, teachers, books, and other sion in education will, in turn, make the solution of inputs are required each year. Moreover, because as these problems more difficult. And so the pattern is populations grow the number of school-age children repeated, and it is destined to continue until extraor- increases more rapidly than the number of working dinary efforts are made to interrupt this cycle of adults, the burden of supporting an expanded educa- deteriorating prospects. tion system falis on an adult population that is shrinking as a proportion of the overall population. The Demographic Chailenge Sub-Saharan Africa today has the youngest popula- tion of any region in the world. One in three persons Between 1970 and 1980, while the world's population is of primary or secondary school age in Africa (see wvas growing at an average annual rate of 1.9 percent, figure 2-1>, as compared with just one in five persons Àfrica's population grew at 2.9 percent, one and a ín Latin America and in Asia, and one ín six persons half times the world's rate. Between 1980 and the end in the industrial countries. of the century, Africa's population is prolected to By the year 2000 Africa's primary and secondary grow even faster, at about 3.2 percent a year, while school-age population is likely to reach 220 million, growth rates are pro,ected to decline for the other which is 90 million, or 70 percent, more than the principal regions of the world. number in 1984 (see figure 2-2). This massive explo- Africa's high population growth rate reflects the sion in the potential demand for educational services rapid decline in mortality rates in recent years while is virtually inevitab-e, given the number of peopie 1 8 $250 per secondary school pupil), the recurrent costs Figure 2-1. Population by Age, 1984 of just the first two leveis of education would go up by approximately $4.0 billion, becoming $9.4 billion by the year 2000 (see table 2-2). This amount is already more than the total allocated by African Ax : jge,BS 6X-11 . . Ae,< governments to all three levels of education and their ............. g %central administration in 1983. ........... »In addition to meeting the swollen recurrent costs, /,Ages l7-lX'\\ ::::// \ government budgets would have to finance school 1X /,''/''@>/ Ages o- buildings and equipment-both to meet the needs of new students and to replace (at an estimated 2 percent a year) outmoded facilities. Assuming very modest capital costs of $150 per primary place and \sL/Ages 19 anx'e $1,250 per secondary place (this best-practice scena- rio involves capital costs well below actual average capital costs in Africa today), the required investment in facilities would be another $1.4 billion annually. The total of recurrent and capital expenditures for primary and secondary education alone would come to nearly $11 billion annually. This is about $2 billion already born who wili be of child-bearing age during more than the total spent on all aspects of education the period. This inevitability of growth is of funda- mental importance. It sharply constrains African planners' options, and its implications color all of the analysis presented here. Figure 2-2. Growth of School-Age Populations, For example, there were 51.3 million primary and 1965-80 and Projected to 2000 11.1 million secondary school places in Africa in 1983. These figures would have to reach 90.7 million Milliolis and 19.7 million, respectively, by the year 2000 just 140 - 1965 to maintain participation rates at their 1983 levels (see table 2-1). The goal of universal primary educa- El 1984 tion, which African education planners at one time 1 -2000 hoped could be achieved even by 1980, unfortunately 120- from the perspective of the mid-1980s still seems very far away. For the region to have all primary-school- age children enrolled by the year 2000 would require 100 - that 131.8 million primary school places be made available, a 157 percent increase in just seventeen years. The 80.5 million additional places required are 80 - more than all of the places that exist now after many decades of educational development. The investment in the construction of new class- 60 - rooms and the recurrent costs necessary to support such massive expansion in education, while not alto- gether unreasonabie if the achievements of the 1960s 40 - and 1970s are simply extrapolated forward, unfor- ::: W _ z tunately do not seem very likely in the context of long-run austerity beginning in the 1970s. Table 2-1 shows that to maintain gross enrollment ratios at their 1983 levels would require the addition of 39.4 d ::: -.-::: milion primary and 8.6 million secondary school places by the turn of the century. Assuming that per o pupil expenditures will remain at their 1983 leveis in constant dollar terms (roughly $50 per primary and 19 Table 2-1. School-Age Population and School Enrollment, 1983 and Projected to 2000 Primary Secondary Change, Change, 1983-2000 1983-2000 Indicator and date Millions (percent) Millions (percent) School-age population 1983 68.5 n.a. 55.6 n.a. 2000 120.9 76 98.6 77 School enroliment 1983 51.3 n.a. 11.1 n.a. 2000 Assuming enrollment ratios as in 1983 90.7 77 19.7 77 Assuming fuil enrollment and no repetition or dropout 120.9 136 98.6 788 Assuming full enrollment and repetition and dropout as in 1983 131.8 157 105.5 850 n.a. Not applicable. Source: Appendix tables A-1, A-2, A-7, A-8, A-12, and B-4. in 1983. In sum, the demographic tide requires mas- Economic stagnation, combined with rapid popu- sive efforts just to stay even in terms of enrollment lation growth, has meant a decline in living stan- ratios. dards, a fatal decline for many individuals in those Ironically, in addition to explosive population countries worst hit. Overall, income per capita fell growth, much of Africa suffers from another de- nearly 4 percent annually between 1980 and 1984; it mographic condition-low population density. The is lower in the region today than it was twenty years combination of rapid growth and low density may ago. The external public and publicly guaranteed seem strange since, given enough time, the first will debt increased elevenfold between 1970 and 1984, eventually eliminate the second. in the meantime, and debt service payments more than tripled as a however, low population density often implies high percentage of GNP, from 1.2 percent to 4.4 percent. unit costs in education, especially in rural areas, Central government expenditures increased while where economies of scale in the provision of educa- revenues remained about the same, and fiscal deficits tion are precluded given present methods for deliver- on the continent rose to about 10 percent of GNP. ing education. At the present, Africa's population Because continuation of these trends spelled eco- density is about nineteen people per square kilome- nomic disaster, African governments recognized that ter, compared with a worldwide density of about macroeconomic policy reforms were an absolute ne- thirty-six per square kilometer. cessity. In this regard, they have made substantial progress since 1980, along the general lines pres- Macroeconomic Adjustment and Fiscal Austerity cribed at the United Nations Special Session on Africa in May 1986. Thus many governments have The late 1970s ushered in an economic crisis that reconsidered their exchange rate policies, instituted accelerated after 1980 and has left most African wage and salary reforms, and begun to eliminate economies in serious disarray. Agriculture, which price distortions that penalized farmers. They have accounts for the largest share of goods produced also reduced public spending, including spending on (about a third of total production), was hurt by the education (chapter 3). drought that affected much of the region, by a It is expected that these structural reforms will marked deterioration in terms of trade (declining gradually encourage higher investment and lead to agricultural prices combined with rising energy increased consumption levels. Moreover, in 1986 the prices), and by the continuation of national policies drought in Africa abated, and there was an improve- that discriminate against the sector. Moreover, the ment in the region's terms of trade. In that year low- investment rate in Sub-Saharan Africa fell from more income Africa registered a growth in per capita in- that 18 percent of income in the 1970s to less than 15 come for the first time in the 1980s; figures for 1987 percent in 1983 and is currently the lowest of any are expected to show some further improvement. In developing region, thus threatening to undermine sum, there may be room again for cautious optimism Africa's long-term productive capacity. with regard to the economic environment and long- 20 Table 2-2. Enrollment and Recurrent Expenditure, 1983 and Projected to 2000 Assuming Enrollment Ratios as in 1983 Indictor and date Primary Secondary Total Enroliment (millions) 1983 51.3 11.1 62.4 2000 90.7 19.7 110.4 Recurrent expenditure (billions of 1983 dollars) 1983 2.6 2.8 5.4 2000 4.5 4.9 9.4 Required average annual capital investment, 1983-2000 (billions of 1983 dollars) 0.5 0.9 1.4 term growth prospects of the region, but this rel- ported by several broad and politically compelling ative optimism should by no means give rise to explanations of why African countries have invested complacency. so heavily in education since independence and why they should endeavor to protect and expand their Investment in Education commitment to the sector in the future. First, the essence of sovereignty is the control of nationais over Extensive experience from Africa and elsewhere pro- the destiny of their country. The transition from vides strong evidence that increased investment in colonial dependence to self-government and to active education and training at this stage in Africa's history independent participation in the international arena can yield broad economic benefits-including higher was possible only because African educational sys- incomes and lower fertility, both of critical impor- tems were able to produce indigenous personnel at tance in the African context. Although direct evi- all levels to replace expatriate rule. The qualitative dence for developing areas and especially for Sub- nurturing and quantitative regeneration of that lead- Saharan Africa is less extensive and rich than for the ership elite, so that it can address the increasingly industrial countries, enough information exists to complex challenges of nation-building in the future, conclude that the direction of the relationship be- is a continuing requirement for the effective mainte- tween education and various indicators of economic nance of sovereignty and, as such, a fundamental well-being is the same everywhere and that the pos- responsibility of African systems of education. itive relationship holds true for both formal general Second, the rate and extent of the growth of education and for training. The evidence suggests, human capital in African countries, attained through however, that the impact of education is somewhat the improvement and expansion of education at all larger in developing countries, because of its relative levels, will ultimately determine whether the univer- scarcity there. sal explosion of scientific and technical knowiedge Education cannot, in itself, bring about economic can be harnessed for the benefit of the region, or growth, but the evidence indicates it to be a vital whether Africa will be left behind and denied the factor. It provides the fertile ground without which enormous benefits accruing to technological change. other development initiatives will not take root. Finally, education is everywhere a "merit good," a Education accelerates the growth process; it is an basic human right, an end in itself; indeed, education essential complement to other factors. Thus, for is an intrinsic element of the development process. It example, although education is associated with in- is unthinkable that African governments, and their creases in agricultural productivity, the impact of international partners, would permit a decline in the education is found to be much greater in environ- fraction of the population having access to educa- ments that are already modernizing. Likewise, invest- tion's many benefits. And yet that is the specter that ments in the provision of family planning services now menaces many African countries. have been shown to have greater effect if the women The economic evidence indicating large payoffs to using the service (or potentially using it) are better education is of three principal sorts, each providing educated. corroboration for the others. The first two rely on The economic rationale for enhanced activity in individual-level data; the last consists of macroeco- African education, elaborated below, may be sup- nomic evidence. The first one examines the relation 21 between an individual's education level and his or her substantial difference (nine percentage points) ob- productivity in the labor market as a wage employee served in the past between primary and secondary or self-employed worker. The second links the ed- education may have completely disappeared, which ucation of individuais to important outcomes of would reflect the declining relative scarcity value of household behavior such as fertility rates or child primary education (see box 2-1). Even so, the Kenya survival. Finally, macroeconomic evidence relates the study confirms that rates of return to both levels of growth rates of national economies to prior invest- education, after adjusting for other factors, remain ments in education and can control for other factors substantial (around 12 percent). presumed to influence growth. The macroeco- The relatively lower rate of return to education at nomic evidence is an important complement to the the tertiary level should not be interpreted to mean individual-level data, particularly the data on educa- that high-level skills are not important in Africa. tion and earnings, in that it provides evidence to What the findings reflect most of all are the extremely assess the view, occasionally advanced, that educa- high unit costs of tertiary education. In addition, tion improves the lot of individuals by reallocating most studies ascribe all costs at this level to teaching societal resources but does not increase the overall and none to other outputs of the tertiary system, such flow of resources. The remaining pages of this chap- as research. If the costs of tertiary education were to ter summarize evidence of all three types. be prorated and the value of the research and other outputs of the education were to be assessed broadly Labor Productivity to include some that may accrue to society in general, rather than only to the individuals who receive the Economists have assessed the economic value of education, then the rate of return to tertiary educa- education by observing differences in the earnings of tion would certainly appear higher. workers with different levels of education, having Many of the early rate of return studies relied on controlled for other differences that exist between the earnings data for workers employed in the formal groups, and then comparing the adjusted earnings wage sector. To the extent that the number of pri- differences with the costs of the education. The mary school leavers able to find employment in the indicator that summarizes the information on costs formal sector has declined recently in relation to the and benefits is the social rate of return. In technical number of uneducated people so employed, and to terms, the rate of return is the discount rate that the extent that earnings in the formal sector are equates the present value of the economic costs and "protected" and thus remain higher than elsewhere benefits of an investment. Private rates of return to in the economy, the estimated rates of return might education are calculated using after-tax earnings have been inflated. The perceived validity of such differentials and only those educational costs that are benefit-cost studies has been considerably enhanced, actually borne by students and their families. Social therefore, by a number of recent studies of educa- rates of return, which are more useful for public tion's impact, including some done in Africa, that are policy, are based on before-tax earnings differentials based on information about seif-employed workers. and on education's full resource costs. The full costs Since the self-employed do not receive wages as such, of education equal the sum of all private costs plus many of these studies have sought to estimate the any subsidies given. value of education by looking at the value of what A recent survey of cost-benefit studies conducted more-educated individuals produce. Most of the re- in sixteen African countries suggests average social search has focused on agriculture, especially on crop rates of return to investment in African education of production. the following magnitudes: primary, 26 percent; sec- A recent review of eighteen studies of farmer ondary, 17 percent; and tertiary, 13 percent. AI- education and farm productivity in thirteen countries though these rates of return certainly seem attractive concluded that farmers who have completed four ín relation to many other forms of investment, they years of education produce, on average, about 8 may overstate the actual current returns, because the percent more farm output than farmers who have not data on which many of the estimates are based are gone to school, controlling for differences in the use out of date. The cross-sectional earnings data used in of physical inputs. Moreover, the percentage increase these studies are, on average, ten years old, and there in output associated with four years of education is has been considerable educational expansion over found to be about 10 percent in "modernizing" the decade since then. environments (indicated by such factors as the avai]- A more recent study from Kenya indicates that ability of new crop varieties and reasonable incen- rates of return have fallen off somewhat and that the tives); in more traditional environments, where tech- 22 Box 2-1. Rates of Return to Primary and Secondary Education in Kenya Perhaps the most influential fact to emerge from twenity- primary school completer is fortunlate to get evCIl a five ycars of studying rates of return in dcveloping meniial blue-collar job in the wage sector, and his chance countries concerns thc relative rcturns to primary and of obtaining a white-collar position is virtually nil. secondary schooling. Most studies, including most of Thc process hy which successive cohorts of workers at thosc conducted in Africa, have reported substantially a particular cducation level enter less skilled jobs is called higher rates of return at the primary level. The implica- "filtcring down." A reccnt stmdy took this process intio tion generally drawn is that top priority should be givcn account in caicuilating average and marginal ratcs of to primary education as a form of investment in human returin iti Kenya. The study found that the rate of rerurni resources. Recent studies from Kenya throw some new to primary schooling is highly seilsitive to the distinction light on this issue. betwccn average and marginal rates of return, whercas Returns to education are measured by the gaps in the rate of return to secondary schooling is llot. The wages, and presumably productivity, between workers average return to primary schooling, as convenltionlally with different levels of education. lt is often assumed that measured, is 17 percent; the marginal return is 12 per- the average wage of labor with a given level of education cent. The marginal return is lower for two reasons. At measures the wage received by the marginal (most re- the primary level there are both substantial filtering cently recruited) worker with that levei of schooling. down and large differences in wages by occupation, However, the average may not always indicate the mar- whereas for the uneducated there is less scope for filter- ginal wage. ing down and wage differences by occupation are smail. Because of rapid expansion in the education system, For secondary education, by contrast, the average and the labor market conditions faced by those just leaving marginal rates of return are both 13 percent. Because the school now are very different from the conditions faced degree of filtering down of primary and secondary school by earlier cohorts when primary school leavers were in completers is similar, the differential in earnings betwcen short supply. For those entering the labor market a the two groups is little affected. In Kenya moving from thc generation ago, a primary school certificate was a pass- average to the marginal concept erases the usually report- port to a white-collar job, and typically those who ed difference in rates of return: at the margin, the rate of obtained jobs at that time remain in them today. But, return to secondary education (13 percent) about equals because the education system has expanded, today's the rate of return to primary education (12 percent). nology and opportunity are changing only slowly, ment, is often cited as the primary reason for the there is little payoff to education. Only one African failure of particular projects to meet their objectives. country, Kenya, was included among the thirteen More generally, the education level of the entire countries in the review, but the results there were population involved in a project-farmers who consistent with the general findings. A recent study would be using new inputs made available under a of this kind based on data from South Asia identifies project or women who would be visitinlg a new health numeracy and literacy among the critical cognitive or family planning clinic-is often a key element in skills through which education's effect on farmer determining whether the project has a significant productivity is mediated. impact. Investment in education, in sum, increases One may conclude from all of the available evi- the return to investments in other sectors. dence that rates of return to investment at all levels There are other ways that education may enhance of education in Africa today compare favorably with worker productivity. It has been shown, for example, realized rates in most other sectors, including some that education increases the propensity of individuais sectors for which the World Bank and other develop- to migrate. This tendency will increase economic ment agencies have been providing relatively large growth to the extent that individuals shift their jobs, amounts of money in recent years. The Bank's expe- as they will naturally seek to do, from less-productive rience with project lending reinforces what research- to more-productive sectors. Moreover, employed ur- ers have concluded regarding the value of education. ban migrants typically remit a part of their earnings The probability that a development project in any to their families back home, and this resource flow sector will be successful has been observed to in- has been shown to result in increased investment in crease as a function of the human capacity available Africa's rural areas. to plan, implement, and benefit from it. The absence Education may also, under certain conditions, be of important skills, particularly in project manage- expected to fulfill equity goals. Since it was first 23 Box 2-2. Education and the Decline of Mortality: Research from Ghana, Nigeria, and Sudan A number of survey and census analyses have detected occupation. The analysis concluded that the single most an inverse relationship between a child's chances of important influence on child survival is the levei of the survival and the mother's levei of education. Data from mother's education. In lbadan the child mortality .indcx the 1960 ctisus of Ghana, for example, reveal that the for women with some primary schooling was 68 percent rate of child mortality is almost twice as high for mothers of that recorded for women with no schooling, and the with no edtucation as for mothers with an elementary index for women with more than primary schooling was education, anld early four times higher for mothers with 39 percent of that for women with no schiooling. In no education as for those with secondary schooling. The southwestern Nigeria the figures were almost the same, patterns are much the same for children in urban and 68 percent and 41 percent, respectively. rural areas. Although the father's education was also found to he A more comprehensive study was conducted in Niger- significant, it was less important than the mother's in ia as part of the 1973 Changing African Family Project explaining differences in child mortality. Family inconie, Survey. One component of this study was a probability too, was found to be of little importance, after the effects sample of 6,606 Yoruba women between the ages of of education were taken into account. Although child fifteen and fifty-nine in the city of Ibadan. The second mortality was higher in polygamous than in monoga- component consisted of a probability sample of 1,499 mous homes, the effect of a mother's education to the Yoruba women over age seventeen living in southwestern secondary leve] was at least a 50 percent reduction in Nigeria. Analysis of these data allowed for an examina- mortality in both polygamous and monogamous bomes. tion of rural-urban differences, which serve as a reason- Once other factors were controlled for, child survival able proxy for differences in access to modern health was found to be higher among parents who practiced services. birth control, which might be explained by the greater The study considered child survival in relation to a care accorded children in smaller families. variety of variables, including the quality of medical Results in the Sudan from the same Changing African services at childbirth, the parents' practice of birth con- Family Project Survey confirmed the findings on the trol, and the family's income as measured by the father's importance of mother's education to chi!d survival. introduced in Africa, Western-style education has Lower Fertility and Other Benefits been a vehicle by which able children of poor families have managed to move to higher leveIs in society's The more education a woman receives, the fewer occupational and income structure. As access in- children she is likely to bear. Of course, in areas creases, the additional expenditures on education where income is very low fertility may actually in- will flow increasingly to the disadvantaged elements crease with the first few years of schooling given te of society-the poor, those in rural areas, and girls. women, because of education's impact on health and The study in box 2-1 compares the experience in fecundity. Nevertheless, research on the determinants Kenya, where secondary education was allowed to of family size in Africa and elsewhere indicates con- expand rapidly during the 1970s as a function of vincingly that raising the educational attainment of demand, with the experience in Tanzania, where the wornen ultimately reduces fertility. This longer-term expansion of secondary education was relatively con- behavioral change occurs through education's direct strained. Although both Kenya and Tanzania ex- effects on individual decisions to have children and pressed concern over the widely dispersed wage through its indirect effects on children's survival and structures that existed, Kenya's policy resulted in women's employment opportunities. substantially greater compression in the distribution From the household perspective it is desirable that, of earnings over the period than occurred in Tanza- through education, women gain the power to control nia. In addition, the larger size of the secondary the number and spacing of the children they bring system has meant greater access for the children of into the world. Moreover, in light >f the economic the poor and uneducated in Kenya than in Tanzania. problems associated with Africa's unprecedented Education yields the recipients other important population growth rate, the social benefits of educa- benefits that are not (immediately) reflected in the tion's impact on fertility-related behavior can be form of increased earnings or a more equal distribu- expected to be high. tion of earnings. These benefits have to do with In addition to having fewer children, more-educat- fertility, health, and individual fulfillment. ed men and women are found to live healthier and 24 longer lives. One recent multination study demon- example), measurement of income differentials by strated that a difference of one percentage point in education level may understate education's impact. the national literacy rate is associated with a two- To address these concerns, macroeconomic evidence year gain in life expectancy, controlling for per capita on education and economic growth complements the income and food energy consumption. extensive microeconomic literature. The education of parents, particularly mothers, Macroeconomic analyses attempt to explain differ- has been shown to affect the physical and cognitive ences in growth rates and other development out- development of children. This benefit occurs, in part, comes among countries in terms of differences, through education's effect on family income and, in among other things, in their patterns of investment part, through its effects on parents' knowledge and in education. The World Bank's World Development use of good health and nutrition practices. Further- Report 1980 undertook an analysis of this sort and more, intellectual skills acquired through one's own concluded the following: (a) increases in literacy education tend to "rub off" on one's children. Studies contribute both to increased investment and (given have shown that children of more-educated parents the level of investment) to increases in output per are more likely to be enrolled in school and, once worker; (b) literacy, as well as nutrition and income, enrolled, are more successful and continue higher up affects life expectancy; and (c) variations in life the education ladder. expectancy, literacy, income, and the strength of There is, following from education's effects on family planning programs explain most of the varia- nutrition and health, a consistent and strongly posi- tion in fertility rates across countries. tive relationship between parental education and Although the World Development Report's find- child survival. A review of research in this area ings on the development impact of investment in concludes that the addition of one year of mothers' education are quite clear, the analysis was undertak- schooling reduces child mortality by nine deaths per en on a sample of all developing countries, and thousand live births. Box 2-2 summarizes new re- somewhat different results might pertain to Africa. search from Africa on the impact of maternal educa- More recent analyses suggest, however, that levels of tion on child survival. investment in education are, if anything, more impor- tant in explaining differences in growth rates in Economic Growth Africa than elsewhere. Box 2-3 summarizes the re- sults of an analysis, undertaken in conjunction with Microeconomic evidence of the sort just discussed the preparation of this paper, of the contribution of strongly suggests the importance of education in education to the growth of thirty-one African coun- determining economic growth. Yet some concerns tries over several decades. The analysis concludes, in may remain. Perhaps education's high rates of return short, that investments made in education have con- are accounted for, indirectly, by reductions in the tributed significantly to growth in GDP; indeed, per- earnings of the less well educated. Or, conversely, to haps 30 percent of GDP growth has resulted from the extent that the better educated enhance the pro- education investments. The aggregate evidence thus ductivity of those around them (through their entre- corroborates the microeconomic findings of educa- preneurial activities or technical contributions, for tion's high returns. Box 2-3. Education and Economic Growth in Sub-Saharan Africa Between 1965 and the early 1980s Africa invested heavily contributed to recent economic growth. Such an analysis in education. The growth of primary school enrollments, was undertaken, using data from the appendix and other which rose from 12 million in 1960 to over 50 million in World Bank sources. The analysis used a "production 1983, is one measure of this investment. The average function" that relates the GDP of each country in each time investment in education in Sub-Saharan Africa in recent period (five intervals between 1965 and 1983 were used) to years has been between 4 and 5 percent of national a number of factors potentially important in determining income, plus direct parental contributions and the poten- GDP. These included the size of the working-age population, tial eamings forgone by students while they attend school. the area of land under cultivation, the available physical In order to guide future allocations to the education capital stock, and the available human capital stock as sector,;it is important to evaluate the extent to which measured by the number of years that members of the African governments' past investments in education have (Box continues on the following page.) 25 Box 2-3 (continued) working-age population attended secondary school. A mea- amount of land per worker grew more slowly than GDP sure of years of primary school attendance was also avail- per worker, but that physical capital and, particularly, able, but since the two education variables are highly human capital stocks grew much more rapidly. colinear, one was dropped from the analysis. The final two columns show how these input growth These data permitted use of standard econometric rates accounted for the 1.8 percent growth of GDP per methods to estimate the parameters of the production worker. Increases in the stock of human capital account- function that relates GDP to the availability of the deter- ed for 0.55 of the 1.8 percentage points, or 31 percent of mining variables. Specifically, a Cobb-Douglas produc- total GDP growth. This is an impressive fraction, indeed, tion function was estimated by examining how changes in and clearly demonstrates the importance of investment input availability resulted in changes in GDP. With an in education in explaining GDP growth in Africa. The estimated production function and knowledge about most nearly analogous figure from Edward Denison's growth in the availability of the input factors, it is work on economic growth in the United States between possible to account for the growth in overall GDP by 1929 and 1976 is 26 percent. Growth in the availability disaggregating it into components that reflect the rate of physical capital accounted for 71 percent of the total of growth of each input's availability and the relative growth. A "residual" of unaccounted-for growth was importance of that input for determining output, as actually negative in this analysis, resulting in a decrease assessed from the production function. of 0.1 percentage points a year in the level of growth that The table below reports the results of this analysis. would have been expected on the basis of growth in the The first column of the table lists and defines the vari- measured inputs. This residual (often interpreted as ables used in the analysis; the second indicates their technical and institutional improvement) is positive in growth during 1965-83. For example, for the thirty-one studies undertaken with data from the United States and countries included in the analysis GDP grew at an average other industrial countries and often accounts for a sub- rate of 4.3 percent a year during the period, and the size stantial fraction of total growth. That it has been nega- of the working-age population grew at a rate of 2.5 tive during this period in Africa may be consistent with percent a year, so that the growth of GDP per member of impressionistic accounts of recent sharp technological the working-age population was, on average, about 1.8 and institutional deterioration. percent a year. The third column indicates that the Contributors to GDP Growth in Sample Countries, 1956-83 Contribution to Annual growth of GDP per growth rate (percent) member of working- Per member age population ofworking-age Percentage Percent Variable Absolute population points of total GDP 4.3 1.8 n.a. n.a. Labor (size of working-age population) 2.5 n.a. n.a. n.a. Capital (physical stock) 7.8 5.3 1.25 69 Land (area under cultivation) 3.6 1.1 0.09 5 Human capital (years of secondary education in working-age population) 18.3 15.8 0.55 31 Residual n.a. n.a. -0.09 -5 n.a. Nor applicable. Economic and demographic pressures will, clearly, combined with development assistance from abroad, constrain the capacity of societies-and, even more, will create an environment in which sustained of governments-to invest as fully as would be desir- growth can resume. Indeed, economic statistics able in the education levels of their future popula- emerging since the mid-1980s have reflected a return tions. Over a period of time, however, governmental to growth. To take full advantage of an improving policy reform and individual responses to reform, economic policy environment, sustained infrastruc- 26 tural investments must be made, both public and tries should try to increase both public and private private. The evidence is powerful that education is a expenditures on education. Government expendi- key area for such investment. African governments ture, currently averaging between 4 and 5 percent of and peoples have successfully implemented massive national income in the region, should be gradually educational investments, and evaluation of the con- increased in most countries, particularly those that sequences shows the returns to have been high. We now dedicate significantly less than this proportion come, then, to the first central recommendation of to the sector or that lag far behind on key indicators this report. of educational development. Such increases in public spending are warranted, however, only in conjunc- tion with efforts at all leveis to improve education's Recommendation 1 internal efficiency and financial viability. The broad array of measures appropriate for these objectives Because of the unusually favorable rates of return will typically include greater nongovernmental prov- and the feasibility of project implementation, educa- ision of education services, thereby raising private tion is prominent among the economic sectors in resources dedicated to the sector at least in propor- Africa that call for greater investment. To pro- tion to the increase in public resources. mote long-term development most African coun- 27 The Stagnation of Enrollment and Dedline ii> Quality Aithough African nations have made enormous prog- remain universal primary enrollment, which ulti- ress in education as described in chapter 1, much mately obviates the need for adult education of this remains to be done. Africa still lags behind other kind. developing regions with regard to most indicators With respect to the qualitative dimension of edu- of educational development. Moreover, the external cational output, although the evidence is limited, it factors described in chapter 2 provide an inhospitable nonetheless points convincingly to the conclusion environment for eliminating the education and train- that the performance in Africa is unsatisfactory and ing gap. has declined recently. Indeed, because of the invidious combination of rapid population growth and economic stagnation, Enrollment Stagnation the gap between Sub-Saharan Africa and the rest of the world appears to be widening; certainly it is no Between 1960 and 1970 total enrollments in Sub- longer closing as in recent decades. Unless steps are Saharan Africa grew at an average an.ual rate of 6.5 taken to address the serious problems in education, percent. Between 1970 and 1980 this rate rose to 8.9 this gap wiil in time become a gulf. Given the vital percent. During the first three years of the 1980s, links between education and a people's prospects for however, the rate of increase plummeted to 4.2 per- economic growth and development (see chapter 2), cent. (Comparative data more recent than 1983 were this cannot be allowed to happen. not available at the time of this study, but there is This chapter describes the alarming state of edu- evidence of some economic recovery since 1983 that cation in Africa auong the two principal dimensions could have led to some recovery in enrollments.) of educational output: quantity and quality. On the Although the rate of increase in enrollment de- quantitative side, school enroiaments have fallen as clined at ali leveis of education, the drop was most a proportion of school-age populations in many pronounced at the first leveI, where it feli from 8.4 African countries; absolute declines have occurred percent