53878 AFRICA HUMAN DEVELOPMENT SERIES Developing Post-Primary Education in Sub-Saharan Africa Assessing the Financial Sustainability of Alternative Pathways Alain Mingat, Blandine Ledoux, and Ramahatra Rakotomalala THE WORLD BANK Developing Post-Primary Education in Sub-Saharan Africa ASSESSING THE FINANCIAL SUSTAINABILITY OF ALTERNATIVE PATHWAYS AFRICA HUMAN DEVELOPMENT SERIES Developing Post-Primary Education in Sub-Saharan Africa ASSESSING THE FINANCIAL SUSTAINABILITY OF ALTERNATIVE PATHWAYS AFRICA HUMAN DEVELOPMENT SERIES Alain Mingat, Blandine Ledoux, and Ramahatra Rakotomalala Note: The analysis and conclusions presented in this document are those of the authors alone. They do not necessarily reflect the official position of the AFD or its partner institutions or that of the World Bank, its affiliated organizations, and the members of its Board of Executive Directors and the countries they represent. Any errors are the sole responsibility of the authors. © 2010 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 13 12 11 10 This volume is a product of the staff of the International Bank for Reconstruction and Devel- opment / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. ISBN: 978-0-8213-8183-0 eISBN: 978-0-8213-8236-3 DOI: 10.1596/978-0-8213-8183-0 Cover photo: Ramahatra Rakotomalala Library of Congress Cataloging-in-Publication Data Developing post-primary education in Sub-Saharan Africa: assessing the financial sustainability of alternative pathways. p. cm. Includes bibliographical references. 1. Education, Secondary--Economic aspects--Africa, Sub-Saharan. 2. Education, Secondary--Africa, Sub-Saharan--Finance. 3. Sustainable development--Africa, Sub-Saharan. LC67.58.A357D48 2010 373.67--dc22 2009052116 Table of Contents Foreword ix Acknowledgments xiii Abbreviations xv Executive Summary xvii CHAPTER 1. INTRODUCTION 1 Notes 2 CHAPTER 2. THE POLICY CONTEXT 3 CHAPTER 3. METHODOLOGY AND DATA 7 Notes 8 CHAPTER 4. THE PRESSURE OF RISING ENROLLMENT 9 Enrollment Growth in Primary Education 11 Impact of Primary School Completion on Lower Secondary Education 12 Notes 15 CHAPTER 5. COVERAGE OF POST-PRIMARY EDUCATION 17 Simulation Assumptions in Student Flow Management 18 Choosing Options for Student Flow Management 20 Projected Enrollment in 2020 under Alternative Options for Student Flow Management 21 Notes 23 v vi · Table of Contents CHAPTER 6. THE QUALITY AND REACH OF EDUCATIONAL SERVICES AND RELATED UNIT COSTS 25 Current Patterns of Per-Student Public Spending by Level of Education 25 Comparative Perspectives on Cost Drivers in Secondary Education 27 Benchmarks for Cost Drivers in Secondary Education 30 Cost Assumptions for Technical and Vocational Education and Training 34 Cost Assumptions for Higher Education 35 Treatment of Capital Costs in the Model 36 Making the Indicative Framework More Flexible 37 Notes 41 CHAPTER 7. EVALUATING THE FINANCING GAP 45 Benchmarks Affecting Volume of Government Resources for Education 45 Benchmarks for the Public-Private Division of Responsibility for Education 46 Notes 49 CHAPTER 8. SIMULATION RESULTS 51 A Recapitulation of the Key Assumptions 52 Simulations of Aggregate Costs and Funding Gaps in 2020 53 Diversity Across the Sample of 33 SSA Countries 58 Notes 60 CHAPTER 9. CONCLUSIONS 61 APPENDIX A. RELATION BETWEEN TEACHER SALARIES AND COUNTRY INCOME LEVEL 65 APPENDIX B. BASIC DATA FOR 33 SSA COUNTRIES 67 APPENDIX C. SCHEMATIC SUMMARY OF SCENARIOS 81 APPENDIX D. AGGREGATE SIMULATION RESULTS 85 APPENDIX E. SIMULATION RESULTS BY COUNTRY 91 Table of Contents · vii REFERENCES 191 INDEX 193 TABLES 4.1 Educational Coverage in Sub-Saharan Africa by Level of Education, Circa 2005 10 4.2 Primary School Completion Rates and Number of Pupils Completing the Cycle, Circa 2005 and Projected in 2020 11 4.3 Potential Growth in Lower Secondary Enrollments by Country 12 4.4 Potential Logistic Difficulties for Countries to Achieve Two Lower Secondary Education Coverage Goals by 2020 14 5.1 Simulation Results: Student Enrollments by Level of Education in 2020 for 33 Low-Income SSA Countries 22 6.1 Public Spending per Student by Level of Education, Sub-Saharan Africa and Other Country Groups, Circa 2005 26 6.2 Characteristics and Costs of School Organization in Secondary Education 28 6.3 Comparison of Coverage and Organization of Secondary Education, Sub-Saharan Africa and Other Country Groups, Circa 2005 29 6.4 Indicative Framework for the Organization of Services in Secondary Education 31 6.5 Indicative Framework for the Organization of Services in TVET and Tertiary Education 36 8.1 Aggregate Annual Public Spending on Education and Financing Gap in 2020 for 33 Low-Income SSA Countries 54 B.1 Quantitative Coverage of Education and Growth in Primary School Leavers 68 B.2 Potential Growth in Lower Secondary Enrollment by Country 70 B.3 School Organization in Lower Secondary Education, Circa 2005 72 B.4 School Organization in Upper Secondary Education, Circa 2005 74 B.5 School Organization in Primary and Tertiary Education, Circa 2005 76 B.6 Population Growth Rates Used in Simulations 78 C.1 Options in Simulation Model for Coverage, Level of Per-Student Spending, and Share of Private Financing in 2020 82 D.1 Student Enrollment by Level of Education in 2020 for 33 Low-Income SSA Countries, Assuming 0% Real Growth of GDP per Capita per Year 86 viii · Table of Contents D.2 Student Enrollment by Level of Education in 2020 for 33 Low-Income SSA Countries, Assuming 4% Real Growth of GDP per Capita per Year 87 D.3 Aggregate Annual Public Spending on Education and Financing Gap in 2020 for 33 SSA Countries, Assuming 0% Real Growth of GDP per Capita per Year 88 D.4 Aggregate Annual Public Spending on Education and Financing Gap in 2020 for 33 SSA Countries, Assuming 4% Real Growth of GDP per Capita per Year 89 Foreword any African countries are eager to sharpen their competitive edge M in today's global economy. Attaining long-term sustainable growth through a diversified and competitive economy, improv- ing governance, empowering people, and mitigating environmental risks are some of the crucial challenges they face in the coming years. Because these challenges cannot be met without qualified human resources, investments in education must deliver quality basic education for all and produce the skilled professionals required to support economic growth strategies. Most African countries have achieved remarkable progress toward universal primary school completion over the last decade, averaging a 15 percentage-point gain between 2000 and 2007. In the coming years, primary school enrollments are expected to continue expanding as domestic and international efforts to advance the Education Millennium Development Goals (MDGs) bear fruit. Success on this front will make a major contribution to Africa's renaissance and economic competitiveness. The surge in primary school completion has had a knock-on impact on secondary (general and vocational) and tertiary education. The social demand for post-primary education is growing rapidly as increasing num- bers of primary school leavers seek to continue their schooling. There are signs, however, that the trend is putting significant pressure on the educa- tion system, often straining its capacity to deliver services that meet mini- mal quality standards. In some countries, unemployment among educated youth is an increasing concern, given the still-limited absorptive capacity of the labor market. These emerging problems raise questions about the social and financial sustainability of current policies and motivate the need for thoughtful reflection on the trade-offs that may be required to reconcile competing priorities amidst constrained resources. ix x · Foreword African countries differ widely in terms of pattern of enrollments, unit costs, student flow management, and resource mobilization. The inten- sity and nature of the challenges they face are therefore quite diverse, which means that policies must be tailored to each country's particular circumstances. Nonetheless, a comparative analysis of post-primary edu- cation across the 33 Sub-Saharan African (SSA) low-income countries with added reference to other regions and also to middle-income countries, can offer a useful lens for assessing needs and constraints and for drawing lessons from the diversity of education across countries. To provide this comparative lens, the Agence Française de Développe- ment (AFD) and the World Bank (WB) embarked on a joint analytical effort. A research team was set up in 2007, comprising Professor Alain Mingat from the University of Bourgogne, Blandine Ledoux (AFD), and Ramahatra Rakotomalala (WB), to prepare the analysis under the direction of Jean-Claude Balmes (AFD) and Jee-Peng Tan (WB), with assistance from Kirsten Majgaard (WB). Their effort builds on the earlier work that helped define the Fast Track Initiative (FTI) indicative framework for primary edu- cation and has produced a new tool that is both more comprehensive, covering all levels of the system, and more sensitive to the conditions in low-income countries. The findings of the study were first presented at the 2008 Biennale of the Association for the Development of Education in Africa (ADEA) in Maputo. Subsequently, the results were shared with members of the Catalytic Fund Committee of the Education for All Fast Track Initiative (EFA-FTI) at their September 2008 meeting in Paris, and also with participants at the July 2009 Conference for African Ministers of Finance and of Education that was organized by the World Bank, the African Development Bank, and ADEA. Among the study's most compelling findings is the scale of the pres- sure on the education system in most African countries. If current trends persist in the next decade, enrollments at the primary and secondary lev- els are projected to multiply by a factor between 1.8 and 14. The cost of service delivery in secondary and tertiary education, as reflected in public spending per student, is on average three times as high in Africa as in other regions of the world. While there is great variability across countries, a consideration of the projected size of the enrollments and cost patterns leads to the stark conclusion that a laissez-faire approach is likely to pro- duce a downward spiral of continuous deterioration in education quality. Given the diversity across African countries, this study offers no generic policy fix. Rather, it is intended as an analytical tool for national leaders and their development partners to inform discussion and debate about alterna- tive options in light of country circumstances. The simulation scenarios it Foreword · xi presents therefore serve an illustrative function, to draw attention to options such as raising the share of education in the national budget, reforming the service delivery arrangements to manage costs, diversifying the student flow beyond lower secondary education, enlarging the role of private funding, particularly in post-primary education, and so on. Helpfully, the study captures the nature of the choices by presenting alternative packages of poli- cies, using them to clarify the affordability of what the authors characterize as "spartan" and "generous" choices. The simulation model's flexibility can be exploited to adapt the package of policies to suit national contexts. This feature is perhaps one of the study's most valuable contributions. We hope that this joint work sponsored by our respective organiza- tions will prove useful to all--governments, experts and funding agencies alike--and that it will contribute to the design of holistic and sustainable education policies to help advance educational development in Africa in the coming years. Michel Jacquier Yaw Ansu Deputy CEO Director of Human Development AFD Africa Region, World Bank Acknowledgments his study was prepared by a joint team from the Agence Française de T Développement and the World Bank that includes Alain Mingat (IREDU­CNRS and the University of Bourgogne), Blandine Ledoux (AFD), and Ramahatra Rakotomalala (WB). Kirsten Majgaard (WB) assisted the team in finalizing the report. The authors thank the country teams from the six Sub-Saharan African countries (Republic of Congo, Ghana, Guinea, Niger, Sierra Leone, and Tanzania) who participated in two multicountry workshops conducted in early 2008 to test and adapt the simulation model. The workshops improved the simulation model in several important ways. Many other individuals contributed to the study. Jean-Claude Balmes (AFD), Jee-Peng Tan (WB), and Yaw Ansu (WB) provided overall leadership and guidance. Further, the study benefited from constructive feedback from Sajitha Bashir, Jacob Bregman, Barbara Bruns, Paul Coustere, William Experton, Birger Fredriksen, Jean-Pierre Jarousse, Thomas Melonio, Aidan Mulkeen, Steven Obeegadoo, Bob Prouty, Michel Welmond, and many others. Marjorie Leach translated the draft report. The authors also acknowledge the generous financial support of the Agence Française de Développement and the Norwegian Post-Primary Education Fund. xiii Abbreviations AFD Agence Française de Développement ADEA Association for the Development of Education in Africa BREDA UNESCO Regional Office for Education EdStats World Bank Education Statistics EFA Education for All Enrollm. enrollment ES enrollment scenario FTI Fast Track Initiative GDP gross domestic product GER gross enrollment ratio GNI gross national income IBRD International Bank for Reconstruction and Development IDA International Development Association IMF International Monetary Fund MDG Millennium Development Goal P primary S1 lower secondary S2 upper secondary SSA Sub-Saharan Africa--usually used to indicate the 33 SSA countries in the study TVET technical and vocational education and training UIS UNESCO Institute for Statistics UN United Nations UNESCO United Nations Educational, Scientific, and Cultural Organization Note: All dollar amounts are U.S. dollars (US$) unless otherwise indicated. xv Executive Summary RECENT ACCOMPLISHMENTS AND PROSPECTS FOR GROWTH IN COVERAGE AND ENROLLMENTS Primary school completion rates in Sub-Saharan Africa (SSA) have been rising rapidly in recent years. Across the 33 low-income SSA countries that are the focus of this study, they have climbed from an average of 43 per- cent in 1999 to 53 percent in 2005, a rate that exceeds the increase of the 1990s. As a result of this salutary trend, the absolute number of primary school completers has been rising steadily and will continue to do so as SSA countries press on toward the goal of universal primary school com- pletion. For example, if all SSA countries attained a primary completion rate of 95 percent by 2020, the 33 SSA countries would have a total of 22.2 million primary school completers in 2020, compared with 9.4 mil- lion around 2005, an increase of a factor of 2.4 over the period. In most SSA countries, the projected increase will inevitably put pres- sure on all levels of the education system, but particularly at the secondary and tertiary levels. The intensity of the pressure is likely to vary from coun- try to country, depending on factors such as demography, the speed of progress toward universal primary completion, and changes in transition rates between various cycles of education. For example, if the transition rate between primary and lower secondary education rises to 100 percent, projected enrollments in lower secondary education in 2020 as a multiple of enrollments around the mid-2000s would be around 2 times in Lesotho, Togo, and Zimbabwe; close to 4 times in Guinea and Mali; around 9 times in Burundi, Tanzania, and Uganda; and more than 11 times in Mozambique and Niger. xvii xviii · Executive Summary SIMULATION MODEL TO ASSESS POLICY OPTIONS Given the scale of the projected enrollment increases, countries in SSA should find it timely to take stock of the implications for policy develop- ment. A key challenge is to ensure that the education system continues to develop in an efficient, equitable, and fiscally sustainable manner, even as it expands to accommodate the rising numbers seeking a place in sec- ondary and tertiary education. Policy makers therefore face a difficult task: They must grapple with the inherent tensions between choosing outcomes that are desirable (for example, meeting the educational aspirations of all youths and responding to employers' needs for skilled labor) and those that are feasible in a given time frame and financially sustainable over the medium term. They also confront choices that must achieve a sustainable balance between national goals for economic and social development and the social demands of individuals--two impulses that often pull in opposite directions. Further- more, in choosing among the alternatives, policy makers will quickly real- ize that actions taken at one level of education cannot be isolated from those taken either upstream or downstream. Each subsector is part of the same systemic architecture and depends on the same overall resource enve- lope, whether funded by domestic or foreign sources. A holistic approach is thus highly desirable and indeed unavoidable. To help clarify the policy options, this study constructed a sectorwide education simulation model to simulate the financial implications of alternative policy choices. Although the model pays specific attention to post-primary education and training (which here comprises lower and upper secondary education, technical and vocational training, and higher education), it incorporates policies in all the subsectors simultaneously in the simulations. The results thus provide a basis for assessing alternative options not only in each subsystem but also for the system as a whole. The results are an aggregation of simulations specific to each of the 33 low-income SSA countries in the study. The findings provide a rich basis for policy dialogue within each country and between the countries and their development partners. The simulation model is designed to address fundamental questions such as (1) How much domestic resources can countries themselves rea- sonably mobilize for education? (2) How much foreign aid can the donor nations realistically contribute to the sector? (3) What are acceptable levels of dependency on external aid in recipient and donor countries alike? and (4) What reforms in education sector policies--in terms of the pattern of coverage by level and type of education and of service delivery Executive Summary · xix arrangements--will be required to put the education system on a sus- tainable path? In today's climate of global economic slowdown, such analysis has become especially pertinent. Three sets of policy parameters give the model its basic structure: enrollments, cost drivers, and financing arrangements. To illustrate the implications of alternative choices, the study focused on a few specific options within these parameters. While the simulation tool can be used to explore additional scenarios pertinent to country-specific circum- stances, the generic scenarios presented below already provide a useful overview of the challenges that most SSA countries are likely to face in finding a sustainable approach for developing post-primary education. OPTIONS PERTAINING TO ENROLLMENT The simulations explored five enrollment scenarios (ES), the first two of which are distinctly different from the next three in terms of student flow management. In particular, the first two scenarios (ES-1 and ES-2) envision progress toward universal or nearly universal primary completion by 2020 and wide coverage in the lower secondary cycle, followed by transition into tertiary education at current rates of transition from the previous cycle. In addition, the scenarios envision some diversification of educa- tional options in the form of school-to-work life skills training and formal technical and vocational education and training. The other set of three enrollment scenarios (ES-3, ES-4, and ES-5) envision similar goals for coverage in primary and lower secondary education as in ES-1 and ES-2, but explicitly calibrates enrollments in tertiary education (and hence of upper secondary education) to expected employment opportunities in the formal sector of the economy. Because this sector is expected to remain relatively small in the foreseeable future, coverage at the upper levels of the system is expected to remain limited. At the same time, the scenarios envision greater provision of school-to-work transition options intended to relieve the pressure on the formal education system, particularly at the upper sec- ondary and tertiary levels. The last simulation scenario, ES-5, corresponds to the option of combining the primary and lower secondary cycle into a basic education cycle. This option is of interest mainly because it reflects an organizational strategy to manage the cost of service delivery. OPTIONS PERTAINING TO COST DRIVERS The second set of options relates to policies on service delivery that affect the cost of services. In primary education, these factors include variables xx · Executive Summary such as the pupil-teacher ratio, average teacher salaries, and spending on non-teacher inputs. This study uses the same values for these parameters as the FTI benchmarks, except for the poorest countries, where the value for teacher salaries is adjusted for a country's income level. In secondary education, two sets of parameter values are used in the simulations, one to mirror relatively "generous" (that is, more expensive) service delivery arrangements in terms of class size, teacher salaries, teaching loads, instructional time, and spending on non-teacher inputs, and the other to mirror relatively "spartan" (that is, more economical) arrangements. The parameters relating to life skills training, technical and vocational educa- tion and training (TVET), and tertiary education are specified directly in terms of spending per student, and the values for these parameters are linked proportionally to the level of spending per student in secondary school. In addition, for tertiary education, the model differentiates spend- ing per student by field of study and computes overall average spending per student by weighting the field-specific spending by the corresponding distribution of enrollment by field of study. The simulation model also includes parameters for capital costs for the construction of additional classrooms in primary and secondary schools to accommodate growing student enrollments in these cycles. The model computes capital costs simply as a function of the choice of enrollment scenario (ES-1 to ES-5), the class size, and a standard unit construction cost applied to all the 33 countries. Capital costs for tertiary education and TVET are excluded for lack of a reasonable basis for estab- lishing benchmarks that might apply to the whole sample of countries in the study. OPTIONS PERTAINING TO PUBLIC AND PRIVATE RESOURCES FOR EDUCATION The third set of assumptions relates to resource mobilization, both in terms of government allocation of public budgets and the distribution between publicly funded and privately funded enrollments. For resource mobilization, the parameter values are the same as those in the Educa- tion for All­Fast Track Initiative (EFA-FTI) framework, while those for the public-private shares are chosen to reflect policy choices that take into account concerns about equity and efficiency. Thus, the values for the share of privately funded services are set at a modest level of about 10 percent for primary and lower secondary education and related school-to-work transition programs, but at a much higher rate of about 40 percent for courses of study beyond lower secondary education. These Executive Summary · xxi figures take into consideration the high social impact of primary and lower secondary education, which justifies public spending, as well as the high private rate of return of tertiary education, which would imply a greater scope for private funding. USING MORE FLEXIBLE PARAMETER VALUES FOR THE SIMULATIONS In the indicative framework of the EFA-FTI, only one set of benchmarks is used regardless of country context. While this standard approach can facilitate an analysis involving a large number of countries, it tends to reduce the persuasiveness of the results for particular countries. The FTI benchmark of 3.5 times the per capita gross domestic product (GDP) for teacher salaries, for example, has been singled out as being particularly inappropriate in the poorest countries. Where the per capita GDP is $100 or less, for example, the benchmark implies a level of pay that often falls below the poverty line. In the present study, important features have been incorporated to overcome these flaws, including linking to the per capita GDP the assumed parameter values for teacher salary and also adjusting for the distribution of the population between urban and rural areas. The latter variable affects costs to the extent that services are more costly to deliver in rural than urban settings, as well as to the extent that demand- side interventions may be required in rural areas to enroll the remaining pockets of children who have not attended school. SIMULATION RESULTS The simulations demonstrate that funding gaps are highly sensitive to the choice of policies in all three areas captured by the model, that is, cover- age or enrollment, service delivery, and financing arrangements. In a scenario that combines the most expansive policy for coverage (ES- 1), the more input-intensive of two possible assumptions for service deliv- ery, and a 20 percent share of education in overall government spending (and assuming a real GDP per capita growth of 2 percent per year), the resulting aggregate gap in recurrent funding in 2020 for the 33 countries would amount to $29.1 billion a year for post-primary education and $3.1 billion a year for primary education (for a systemwide total of $32.2 billion a year). In addition, these countries would have a funding gap for capital investments in primary and secondary education of about $2.8 billion a year. Filling the gaps with external funds would imply that 68 percent of total aggregate expenditures would rely on outside sources, which is a very high rate of dependency, possibly an unrealistic one. With xxii · Executive Summary slightly less ambitious goals for coverage (that is, a transition rate between primary and lower secondary education of 80 percent instead of 100 per- cent), as captured by scenario ES-2, the gap in recurrent expenditure in 2020 would be $22.7 billion a year for post-primary education and $3.1 billion for primary education (for a systemwide total of $25.8 billion a year), and the gap in capital spending on primary and secondary edu- cation would be $2.5 billion a year. Correspondingly, the level of aid dependency would fall only slightly, to 63 percent, which is clearly not fundamentally distinct from the first scenario in terms of the fiscal sus- tainability of the policies. The next three enrollment scenarios, ES-3 to ES-5, mark a clear-cut departure from the previous two in that they reflect much more selective policies for coverage in upper secondary and tertiary education. However, even in the most restrictive of these scenarios, ES-5, the recurrent funding gap is projected at $3.4 billion a year for post-basic education and $5.2 billion a year for basic education, (a 9-year cycle in ES-5. This gives a systemwide total of $8.6 billion a year), while the gap in capital funding for basic and secondary education is projected at $2.6 billion a year. The dependency ratio would be 40 percent if the policies for service delivery were to favor the more input-intensive options and the share of education in total government spending were limited to 20 percent. With more eco- nomical options for service delivery, an increase in the share of govern- ment financing to at least 23 percent, and restrained expansion of upper secondary and tertiary education, the systemwide recurrent funding gap could be brought down to $6.5 billion a year and the dependency ratio would then fall below 35 percent. CONCLUSIONS This study finds that, in all the policy scenarios considered, the develop- ment of post-primary education in the 33 SSA countries would require sizable amounts of external aid to fill funding gaps. If a dependency rate of 35 percent represents a sort of upper bound that most SSA countries and their donor partners prefer not to exceed, some clear conclusions can be distilled from the simulation results. First, policies that envision mas- sification of the first cycle of secondary education combined with conti- nuity of student flow at subsequent levels would generally be untenable. Staying within the 35 percent threshold will typically require policies that regulate student flow more actively. Second, between the generous and spartan options for service delivery, the latter will probably be the choice that would allow most SSA countries to expand post-primary education Executive Summary · xxiii without exceeding the 35 percent threshold for dependency on external funding. Third, for most SSA countries, an allocation of 23 percent of total public spending for the education sector may be required to avoid an untenable dependency on external aid to finance educational devel- opment. Competing needs from other sectors may make it difficult to attain this level of spending everywhere, but a share lower than 20 percent would in most of the sample countries jeopardizes achievement of uni- versal basic education along with a reasonable pace of development in post-basic education. While the broad conclusions drawn from the simulations point to the common challenges faced by the 33 SSA countries included in this study, the degree of their relevance will depend on each country's initial conditions and political context. Given the diversity in the pressures on enrollment, the cost of service delivery, and the potential for resource mobilization, policies will clearly need to be tailored accordingly. For this reason, the results reported in this study are merely indicative and must be complemented by additional country-specific analysis. Such analysis can be prepared using the tool developed for this study or other similar models that permit an assessment using a sectorwide framework in which key policy choices in the main subsectors of the education sys- tem are given explicit consideration. CHAPTER 1 Introduction ll countries in Sub-Saharan Africa (SSA) face the prospect of a A substantial increase in the number of primary school completers in the coming years. Although initial conditions vary widely from country to country, this increase will inevitably intensify pressure on the education system, particularly at the secondary and tertiary levels. African countries may thus find it timely to align their education policies and strategies to the emerging challenges. A key goal is to ensure that the education system continues to develop in an efficient, equitable, and fis- cally sustainable manner even as it expands to accommodate the rising numbers seeking a place in secondary and tertiary education. In evaluating the policy options, decision makers cannot easily ignore the inherent tensions involved, particularly between outcomes that are desirable (such as meeting the educational aspirations of all youths and responding to employers' needs for skilled labor) and those that are fea- sible in a given time frame and financially sustainable over the medium term. Decision makers must also balance national goals for economic and social development with the social demands of individuals--two impulses that often pull in opposite directions. Furthermore, in choosing among the alternatives, policy makers will quickly realize that actions taken at one level of education cannot be isolated from those taken either upstream or downstream. A holistic approach is required simply because each subsector is part of the same systemic architecture and depends on the same overall resource envelope, whether funded by domestic or for- eign sources (see, for example, Hoppers and Obeegadoo 2008). Policy makers thus face a difficult task. This study seeks to clarify the options available to them, focusing in particular on post-primary educa- tion and training, which is defined here as comprising the two subcycles of secondary education, technical and vocational training and higher 1 2 · Developing Post-Primary Education in Sub-Saharan Africa education. It examines policy options in each of these subsystems as well as the package of choices taken as a whole. The results rely on a simula- tion model and are based on data for 33 low-income1 SSA countries.2 The rest of this report is organized as follows. Chapter 2 elaborates the policy context for education development in SSA. Chapter 3 explains the methodology and data sources. Chapter 4 examines the challenges and constraints posed by the sheer volume of increases in enrollments in post- primary education with which most education systems in SSA must grapple in the coming years. Taking these constraints into account, the report evaluates the scope for policy development from three perspec- tives in the subsequent chapters: the coverage of education systems (chapter 5), the quality and cost of service delivery (chapter 6), and the division of financing by public and private sources (chapter 7). The fis- cal implications of plausible policy packages that SSA countries might consider3 are assessed in chapter 8. Chapter 9 seems up the general con- clusions of the report. NOTES 1. Low-income countries are here defined as countries eligible for lending from the International Development Association (IDA). These countries had a 2006 gross national income (GNI) per capita income of less than $1,065 per year. The threshold changes every year as do the countries' per capita incomes, so there may be small changes from year to year in this group. 2. This group of countries is the same as that used in Bruns, Mingat, and Rakotomalala (2003), the study that estimated the cost and financing gaps to universalize primary school completion by 2015 and provided the analytical foundations for the indicative framework of the EFA-FTI launched in 2002. 3. This study complements that of Lewin (2008), and the overall conclusions of the two studies are consistent. CHAPTER 2 The Policy Context frican countries can take pride today in their unmistakable A progress toward the Education for All (EFA) goals for 2015 that were agreed upon at the Dakar 2000 World Education Forum. The Dakar Forum galvanized action leading to increases in domestic and external funding for education in developing countries, particularly in Sub-Saharan Africa (SSA) and to the creation of a new partnership arrangement with donors, the EFA Fast Track Initiative (EFA-FTI), to help accelerate progress toward universal primary education. Economic con- ditions in most SSA countries have also been improving rapidly as growth rates accelerated, from an average of 3.1 percent in 2000 to 6.1 percent by 2007. Against this favorable economic backdrop and the heightened focus on education, SSA countries' primary school com- pletion rate rose by 15 percentage points, on average, between 2000 and 2007 (UN 2009), a remarkable feat considering the continent's rapid population growth. Enrollments at the other levels also rose by impressive margins, albeit from much lower initial levels of coverage. These salutary trends notwithstanding, a realization is emerging that universalizing primary education will not be sufficient to help sustain SSA's rapid economic growth. The World Bank's Africa Action Plan (World Bank, forthcoming) notes "the need to equip the workforce in African countries with the skills, particularly in science, technology and innova- tion, to help businesses compete successfully in the global economy." The United Nations (2009) affirms the importance of broadening the scope of concern beyond primary education to encompass the full set of EFA goals. The need to strengthen post-primary education, including technical and vocational education and training, was also emphasized at various recent international meetings, including the 2008 Biennale of the Association of for the Development of Education in Africa (ADEA); the 2009 United 3 4 · Developing Post-Primary Education in Sub-Saharan Africa Nations Educational, Scientific, and Cultural Organization (UNESCO) World Conference on Higher Education and the 2009 Conference for Min- isters of Finance and of Education hosted by the World Bank, the African Development Bank, and ADEA to discuss options for sustaining the educa- tion momentum in Africa amid the current global financial crisis. The challenge of developing post-primary education is not one of demand. In fact, enrollments have grown so rapidly that they outstrip funding availability, and the resulting imbalance has seriously eroded the quality of services (Brossard and Foko 2007) and crowded out funding for tertiary education and research (World Bank, forthcoming). These prob- lems are starkly obvious in countries such as Burkina Faso, Cameroon, Côte d'Ivoire, Guinea, Kenya, Mali, Mauritius, and Togo, where higher education enrollments have grown three times faster than public spending on this subsector between 1991 and 2006 (World Bank, forthcoming). In countries where overall spending on education already absorbs a sizable share of total public spending, the need for a sectorwide strategy to ration- alize spending across the whole sector and improve its efficiency is proba- bly unavoidable and may be very beneficial. In all settings, such a holistic approach may help to create a virtuous cycle by ensuring that scarce resources are used to equip the workforce with skills that can help sustain and accelerate economic growth, so that growth in turn can generate the resources for further deepening the investments in education. A holistic approach that takes into account all subsectors of the edu- cation and training system has become even more relevant in the context of the current global economic downturn. Although data are not yet avail- able to fully assess its impact, the financial crisis is likely to make it more difficult to mobilize funding for education from domestic as well as exter- nal sources (UN 2009). At the same time, preliminary figures suggest that foreign direct investment and remittances to SSA countries have already started falling (IMF 2009). Despite continued expressions of donor com- mitments to the Millennium Development Goals (MDGs), the emergence of new global priorities--including climate change, food security, and pandemics--and a greater concern about countries' macroeconomic sta- bility may weaken rather than strengthen the education sector's claim on external aid in the coming years (AFD 2009). Because resources are fungi- ble, the amount of aid earmarked for education may be less important than the amounts that governments budget for education, including how they allocate the budget across subsectors. In a resource-constrained envi- ronment, the challenge is to find the best outcome for the sector as a whole. The analytical approach proposed in this book--centering on a unified and internally coherent simulation model that incorporates policy The Policy Context · 5 choices relating to quantity and quality for all key subsectors--is a tool designed to help in this regard. The results presented here are by no means comprehensive; they are intended instead to illustrate the nature of the policy choices and tradeoffs that policy makers face. Ideally, the tool would be adapted to country-specific contexts and used to generate timely inputs relevant to dialogue at the country level. CHAPTER 3 Methodology and Data ountries in Sub-Saharan Africa (SSA) are highly diverse in the many C dimensions that describe an education system. These dimensions include, for example, the structure and organization of the educa- tion system, the initial coverage of the different levels of education, the volume of domestic resources (public and private) mobilized for the sec- tor, the degree of dependence on external aid, the way services are deliv- ered, and the pattern of spending per student across types and levels of education (World Bank, forthcoming). In view of the diversity of these dimensions, the analytical approach in this study entails building a sim- ulation model around a common structural framework that also leaves room for adaptation to country-specific conditions.1 In essence, the sim- ulation results are an aggregation of country-specific results generated on the basis of data for each of the SSA countries included in the study. As an analytical tool, the simulation model can easily compute any number of permutations of policies. The focus here, however, is on a lim- ited number of simulations whose results are intended to illustrate key tradeoffs, to identify the scope of possible options, to point out policy choices that are promising and thus worthy of further exploration, to quantify the financing gap implied in each policy scenario, and to evalu- ate the overall coherence and sustainability of subsector policies when considered as a package for the whole system. The base-year data for the simulations have been culled from various sources. These sources include the large number of education country sta- tus reports2 prepared thus far by the World Bank, often in collaboration with the Pôle de Dakar, United Nations Educational, Scientific, and Cul- tural Organization (UNESCO) Regional Office for Education (BREDA); public expenditure reviews prepared by the World Bank; data used in pre- vious education sector simulation models of the World Bank and those of 7 8 · Developing Post-Primary Education in Sub-Saharan Africa UNESCO; data supplied by UNESCO Institute of Statistics (UIS);3 and institutional databases maintained by the World Bank (such as EdStats4) and the International Monetary Fund (for macroeconomic and public finance indicators). Because of the large number of countries involved and because each variable in the simulation model must share the same definition across countries, it proved impossible to compile the requisite base-year data for the same base-year for all countries. Thus, depending on data availability, the base year fell between 2000 and 2005. The time horizon for the simulation is fixed at 2020 for all countries. NOTES 1. See the background volume for more details of the specification of the parameters of the simulation model and how the country-specific features are embedded in the model. 2. Most education country status reports can be found at either http:// www.worldbank.org/afr/education or http://go.worldbank.org/4Z2L1R94O0. 3. UIS data on education are available online at http://www.uis.unesco.org. 4. EdStats can be accessed online at http://www.worldbank.org/education/ edstats. CHAPTER 4 The Pressure of Rising Enrollment o contextualize the discussion, the data in table 4.1 show the current T status of enrollment coverage by level of education on average for the 33 countries included in the study. For illustrative purposes, the table also includes data for a few specific countries.1 Three features are noteworthy in the table: · The SSA region remains far from achieving universal primary school completion. · The coverage of secondary education is comparatively low in SSA. · The variation in education coverage across countries in SSA is very large at all levels of education. Primary school completion is still far from being universal in SSA, averaging only 52 percent across the 33 countries around the mid-2000s. Although completion rates have risen much more rapidly since 1999 than in the past--at 1.7 percentage points a year compared to only 0.1 points a year during 1990­99--continuation at this rate implies that SSA's pri- mary school completion rate in 2015 would be only 68 percent instead of the Millennium Development Goal of 100 percent by that date (World Bank, forthcoming). Coverage of post-primary education in SSA is also modest in absolute terms as well as compared with such education in low-income countries in other parts of the world. Barely two-thirds of primary school leavers make the transition to lower secondary education, compared to about 80 percent in low-income countries in Asia.2 While only 33 and 14 percent of the rel- evant age groups enroll in the lower and upper secondary cycles, respec- tively, in low-income SSA, the corresponding figures are around 55 and 30 percent in low-income Asian countries. Coverage in tertiary education averages 323 students per 100,000 inhabitants in low-income SSA, com- pared to about 900 per 100,000 inhabitants in low-income Asian countries. 9 10 Table 4.1 Educational Coverage in Sub-Saharan Africa by Level of Education, Circa 2005 Primary Lower secondary Upper secondary TVET Tertiary Primary completion Transition rate Gross intake Transition rate Gross intake Share of secondary Students per Country (base year) rate (%) P­S1 (%) rate (%) S1­S2 (%) rate (%) enrollment (%) 100,000 inhabitants Burkina Faso (2004) 31 58 18 44 4.2 9 201 Cameroon (2003) 60 55 33 65 18.7 23 494 Mali (2004) 42 81 33 40 8.4 10 284 Niger (2002) 22 66 14 40 2.2 3 51 Nigeria (2005) 76 52 40 83 28.3 1 1,188 Rwanda (2003) 46 35 16 78 10.7 11 252 Tanzania (2002) 60 28 16 30 2.0 <1 84 Zambia (2005) 73 62 45 41 16.0 2 218 Average of 33 countries 52.2 63.1 33.1 58.2 13.7 8.8 323 Source: Appendix table B.1. Note: P: primary, S1: lower secondary, S2: upper secondary. The Pressure of Rising Enrollment · 11 Significant variation in educational coverage occurs across SSA coun- tries at all levels. The primary school completion rate varies, for example, from 22 percent in Niger to nearly 80 percent in Nigeria, while the share of the age group with access to lower secondary education varies from 16 percent in Rwanda to 45 percent in Zambia. The diversity in coverage is even greater in higher education, with the number of students per 100,000 inhabitants ranging from around 50 in Niger to nearly 1,200 in Nigeria. The great diversity across countries implies that, while regional averages are a convenient summary statistic, they are an inadequate basis for generalizations about education in Sub-Saharan Africa. ENROLLMENT GROWTH IN PRIMARY EDUCATION The pressure on post-primary education depends on demographic factors3 and the pace of progress toward universal primary completion. Because ini- tial primary school completion rates are still relatively low and fertility rates remain high in most SSA countries (Lam 2007), the projected number of primary school completers in 2020 is expected to be sizable. For example, if all SSA countries attained a primary completion rate of 95 percent by 2020, the 33 SSA countries in this study would have a total of 22.2 million primary school completers in 2020, compared to 9.4 million around 2005, an increase by a factor of 2.4 over the period (table 4.2). Table 4.2 Primary School Completion Rates and Number of Pupils Completing the Cycle, Circa 2005 and Projected in 2020 Number of students completing Primary completion the primary cycle (thousands) rate (%) Ratio 2020/ Country (base year) Base year 2020 Base year 2020 base year Burkina Faso (2004) 31 95 106 514 4.8 Cameroon (2003) 60 95 231 439 1.9 Mali (2004) 42 95 144 533 3.7 Niger (2002) 22 95 69 574 8.3 Nigeria (2005) 76 95 2,595 4,135 1.6 Rwanda (2003) 46 95 112 295 2.6 Tanzania (2002) 60 95 555 1,096 2.0 Zambia (2005) 73 95 227 367 1.6 Avg. 33 countries 52.2 95.0 9,355 22,235 3.0a­ 2.4b Source: Appendix table B.1. Note: a. Unweighted average b. average weighted by school-age population in the different countries. 12 · Developing Post-Primary Education in Sub-Saharan Africa The country-specific projections in the table reveal a mixed picture, with the projected increase over the period ranging from a factor of no more than 2 in such countries as Cameroon, Nigeria, and Zambia, to a factor of about 5 in Burkina Faso, to a factor greater than 8 in Niger. These data suggest that the effort required to universalize primary school com- pletion will differ considerably from country to country; correspondingly, the pressure on lower secondary education in the next 15 years or so will also be more intense in some countries than others. IMPACT OF PRIMARY SCHOOL COMPLETION ON LOWER SECONDARY EDUCATION The expanding pool of primary school completers will inevitably put pres- sure on subsequent levels of education, beginning with lower secondary education. How intense the pressure will be depends on the rate of tran- sition between the primary and lower secondary cycles. Table 4.3 offers some perspective in this regard. In the initial year (around 2005), there were 14.9 million students enrolled in lower secondary education in our sample of 33 SSA countries. If primary school completion rates rose to Table 4.3 Potential Growth in Lower Secondary Enrollments by Country 2020 Base-year transition rate Base year maintained 100% transition Gross Gross intake rate Lower intake rate Lower Ratio Enrollm. Ratio Transition to lower secondary to lower secondary 2020/ in lower 2020/base Country rate P-S1 secondary enrollm. secondary enrollm. base year secondary year (base year) % (%) (thousands) (%) (thousands) enrollm. (thousands) enrollm. Burkina Faso (2004) 58 18 231 55 1,078 4.7 1,859 8.1 Cameroon (2003) 56 32 560 52 945 1.7 1,718 3.1 Mali (2004) 81 33 347 77 1,186 3.4 1,474 4.2 Niger (2002) 66 14 148 63 1,370 9.3 2,075 14.1 Nigeria (2005) 52 40 3,706 50 6,212 1.7 11,850 3.2 Rwanda (2003) 35 16 117 34 290 2.5 821 7.0 Tanzania (2002) 28 16 454 27 1,194 2.6 4,265 9.4 Zambia (2005) 62 45 271 59 437 1.6 705 2.6 Avg. 33 countries 63.1 33.1 14,909 59.9 37,228 3.5*­ 2.5** 62,934 6.1a­ 4.2b Source: Appendix table B.2 Note: a. Unweighted average b. average weighted by school-age population in the different countries. The Pressure of Rising Enrollment · 13 95 percent by 2020 and the rates of primary-to-lower-secondary transition were maintained at current levels in each country, the number of students in lower secondary education would be 37.2 million by 2020--an increase by a factor of 2.5 over current levels. If all primary school com- pleters were allowed to continue on to lower secondary education, enrollments in the lower secondary cycle would rise to 62.9 million by 2020, an increase by a factor of 4.2 over the period. As expected, SSA countries face diverse prospects in terms of pressure on lower secondary education. With unchanged transition rates between primary and lower secondary education, the projected number of stu- dents in lower secondary education in 2020 relative to the number in the base year varies from less than 1.5 times in Lesotho and Zimbabwe; to around 3.0 times in Mali, Rwanda, and Tanzania; to more than 6.0 times in Mozambique; and to as high as 9.0 times in Niger. If the transition rate were raised to 100 percent, the corresponding multiples would be much higher and more diverse across countries: around 2.0 times in Lesotho, Togo, and Zimbabwe; close to 4.0 times in Guinea and Mali; around 9.0 times in Burundi, Tanzania, and Uganda; and more than 11.0 times in Mozambique and Niger. Some of the 33 SSA countries will be able to cope better than others with the projected increase in enrollment in lower secondary education. The following cutoff points provide one way to separate the countries into three groups according to the logistical challenges4 posed by the projected increase in lower secondary enrollments in 2020: · A ratio no higher than 2.0 relative to base year enrollments indicates that the projected increase is probably feasible to handle. · A ratio between 2.0 and 3.5 indicates that it presents a real challenge. · A ratio exceeding 3.5 indicates that it is poses great difficulties. These benchmarks provide the typology shown in table 4.4. Only five countries in the sample--Republic of Congo, Ghana, Lesotho, Togo, and Zimbabwe--fall into the first group, because both primary school com- pletion rates and transition rates between the primary and secondary cycles are already high in the base year. The large majority of the countries fall into the third category, among them Burkina Faso, Burundi, Chad, Central African Republic, Guinea-Bissau, Madagascar, Mali, Mauritania, Mozambique, Niger, and Uganda. For these countries, even the goal of maintaining the current rate of transition between primary and lower secondary education--in the face of progress toward universal primary school completion by 2020--presents practically insurmountable logis- tical challenges. 14 · Developing Post-Primary Education in Sub-Saharan Africa Table 4.4 Potential Logistic Difficulties for Countries to Achieve Two Lower Secondary Education Coverage Goals by 2020 Logistic aspect of achieving near universal lower secondary education Fairly easy Fairly difficult Very difficult Logistic aspect of Fairly easy Congo, Rep. Cameroon Malawi achieving the goal of Ghana Côte d'Ivoire maintaining the current Lesotho Nigeria primary-secondary Togo Zambia transition rate Zimbabwe Fairly difficult Benin Ethiopia Congo, Dem. Rep. Guinea Eritrea Rwanda The Gambia Senegal Sierra Leone Tanzania Very difficult Burkina Faso Burundi Chad Central African Republic Guinea-Bissau Madagascar Mali Mauritania Mozambique Niger Uganda Source: Authors' construction based on data in appendix table B.2. Note: Kenya and Sudan are not included because they have 8-year primary cycles and no lower secondary cycle. The logistical scale of expanding the number of lower secondary places is obviously only one aspect of challenges to developing appro- priate policies for post-primary education. Decision makers face simi- lar concerns about numbers at the other levels as well. Beyond the numbers, the financial implications of alternative paths for expansion under various options for organizing provision of services also matter in policy development. The discussion herein examines the options in greater detail, using the simulation model described previously to quantify their implications. The simulation assumptions correspond to policy choices that fall under three main rubrics: · coverage of the education system by level and type of education · the quality and organization of services · the division between public and private financing of services. The Pressure of Rising Enrollment · 15 These assumptions are explained in later chapters, followed by an analysis of the financial implications of various combinations of the assumptions, providing a sectorwide perspective on the policy options.5 NOTES 1. Data on all 33 countries can be found in appendix table 1. 2. Afghanistan, Bangladesh, Cambodia, the Democratic Republic of Korea, Lao PDR, Myanmar, Nepal, Pakistan, Papua New Guinea, Solomon Islands, and Vietnam. 3. Appendix table B.6 provides the rates of demographic growth used in the projections. 4. These include building additional classrooms or recruiting and training new teachers. 5. Although the simulation model can compute any number of policy combi- nations, only selected scenarios are presented, sufficient to illustrate the range of possible options without overwhelming the reader. CHAPTER 5 Coverage of Post-Primary Education etermining the quantitative coverage of the system is a logical first D step in shaping policies for developing post-primary education. Two approaches are available to guide the choices. One approach assumes a model of continuity between cycles of education. It corre- sponds to the idea that the curriculum at each level of schooling is meant to equip students for the next level and that parents want their children to go as far as possible in their formal education. In this somewhat laissez- faire picture, the number of entrants to a given level of education--hence the corresponding extent of coverage--is deduced simply by applying the current (or an assumed target) rate of transition between the two cycles concerned to the number of students enrolled in the last grade of the previous cycle. The second approach assumes a model of disconti- nuity in the pattern of student flow. It responds to the fact that, in many Sub-Saharan African (SSA) countries, employment opportunities in the formal or modern sector of the economy are limited and likely to remain so in the foreseeable future. Also, in most of these countries, the labor mar- ket is already saturated with large numbers of unemployed secondary and tertiary graduates. In this setting, taking proactive measures to regulate and diversify the pattern of student flow would help prevent an unsustainable buildup of frustration among jobless, educated youth as well as improve efficiency in the use of public funds (Brossard and Foko 2007). In this chapter, we use the simulation model to compare the implica- tions for enrollment growth corresponding to various degrees of continu- ity (or, conversely, discontinuity) in the management of student flow. The simulation results are presented following an explanation of the simula- tion assumptions and a discussion of the factors that may inform the decisions of policymakers. 17 18 · Developing Post-Primary Education in Sub-Saharan Africa SIMULATION ASSUMPTIONS IN STUDENT FLOW MANAGEMENT As in most simulation models, our model allows for choices pertain- ing to enrollments at the following specific levels and types of educa- tional options: · primary education · lower secondary education · upper secondary education · school-to-work transition programs · technical and vocational education and training (TVET) · tertiary education. The addition of school-to-work transition programs is an important feature in the simulation model in that it makes it possible to simulate the implications of explicit strategies for diversifying educational options at both the post-primary and post-secondary levels.1 Five scenarios of student flow are used to illustrate the range of enroll- ment growth by level and type of program, with 2020 as the end date for the simulations. In brief, their distinguishing assumptions are as follows:2 · Scenario ES-1 corresponds to a situation in each of the 33 SSA coun- tries where (1) the primary completion rate rises from the level in the base year to 95 percent by 2020, (2) all primary school completers go on to lower secondary school and 10 percent among them enroll in TVET programs, (3) the transition rate from lower secondary to the upper secondary cycle remains as in the base year and 15 percent of the upper secondary students enroll in TVET programs, (4) half of the students who complete lower secondary education exit to the labor market via a school-to-work transition program, and (5) the transition rate between upper secondary and tertiary education remains the same as in the base year. · Scenario ES-2 is comparable to scenario ES-1 in all respects, except that only 80 percent rather than 100 percent of the primary school completers continue on to the lower secondary cycle and 15 percent among them enroll in TVET programs, and half of the remaining pri- mary school completers exit to the labor market via a school-to-work transition program. · Scenario ES-3 marks a shift from the previous two scenarios by making a distinct separation between the lower levels of the system (primary and lower secondary education) and the upper levels (upper secondary and tertiary education). TVET and school-to-work Coverage of Post-Primary Education · 19 transition programs both have lower and upper level components, as in the other scenarios. The simulation first derives the target enrollment in tertiary education from a regression analysis of the relation between coverage at this level of education and the level of graduate unemployment3; for the purpose of this scenario, the tar- get rate of unemployment is set at 25 percent. In turn, the number of students in upper secondary education is computed simply by multiplying the cohort size in tertiary education by a factor of 2.5, the latter an optional value chosen to simulate a reasonable number of pupils entering the labor market at this stage. In addition, the scenario assumes that 25 percent of upper secondary students are enrolled in TVET programs. For the lower levels, the assumptions are the same as in ES-2, implying near universal access to schooling up to the lower secondary cycle. Reconciling this implication with the restricted enrollments in the upper secondary cycle means that a proportion of lower secondary school leavers will not go beyond this level of schooling. The simulation assumes that all of these school leavers will exit to the labor market via a school-to-work tran- sition program. · Scenario ES-4 is similar to ES-3, except that it is more restrictive on three counts: (1) the share of primary school leavers transitioning to the lower secondary cycle declines from 80 percent to 65 percent; (2) the share of lower secondary school leavers who benefit from school-to-work transition programs falls from 100 percent to 50 percent; and (3) the number of places in tertiary education is made consistent with an unemployment rate of 20 percent instead of 25 percent (imply- ing fewer places at this level of education than in scenario ES-3). · Scenario ES-5 introduces the option of altering the structural organ- ization of primary and secondary schooling. Keeping unchanged the total duration of these cycles, the simulation allows for a 9-year cycle of basic education, followed by a single cycle of secondary education of the relevant residual duration. Two reasons motivate consideration of this scenario: (1) the complementary nature of primary and lower secondary schooling and the fact that most SSA governments have expressed a desire to universalize basic educa- tion, and (2) the greater scope for managing the cost of service delivery, as will be explored in a later chapter. For the simulation results presented below, the calculations assume that completion rates in the 9-year basic education cycle would reach 80 percent by 2020. The parameters for upper secondary and tertiary educa- tion are the same as in ES-4. There would no longer be a lower 20 · Developing Post-Primary Education in Sub-Saharan Africa secondary TVET sector because it would be absorbed into the 9-year basic cycle. CHOOSING OPTIONS FOR STUDENT FLOW MANAGEMENT The five simulation scenarios range from models of continuity (scenarios ES-1 and ES-2) for managing student flow, to models of discontinuity (ES-3 and ES-4) and to a model of structural change (ES-5). As education systems in SSA countries mature and as the countries make progress toward universal primary schooling, the last three options become increasingly pertinent as issues of fiscal sustainability come to the fore. Each option assumes that explicit choices are made about the placement of selection points beyond primary education, about cohort sizes and selection criteria and mechanisms at each gateway, and about diversifying the pathways for school-to-work transitions. The advantage of this approach is that explicit choices in these dimensions can bring a strategic, coherent, and proactive perspective to policy design for the whole sector. In considering the relevance of the various options, three factors are particularly useful to bear in mind: · The structural duality of SSA economies. In most of the SSA coun- tries, a predominant share of the working population--as much as 90 percent, on average--is employed in agriculture and the infor- mal sector (Fields 2007; World Bank, forthcoming; Walther 2006). The percentage employed in the formal or modern sector is modest and has remained unchanged between 1990 and 2003 at about 10 percent of the workforce. The prospects of a significant and rapid shift in this dual structure are modest, at least in the short run, implying that the number of modern sector jobs is likely to grow only slowly. · The orientation of educational options at system's upper levels. In most SSA countries, the options are limited and are typically geared toward preparing graduates for modern sector jobs. For the relatively few who actually secure such jobs, their advanced training may confer significant value; the rest of the school leavers or graduating class must settle for other jobs in the informal sector where such training may be less valued (Kingdon and Soderbom 2008; Fasih 2008). · The social returns on investment in education in SSA countries. The pattern typically suggests that investments at the lower levels of the system (up to lower secondary education) are generally more efficient at generating social benefits than investments at the higher levels. Coverage of Post-Primary Education · 21 These patterns are particularly striking for girls' education, as house- hold data from a large number of SSA countries reveal (Cohen and Bloom 2006; World Bank, forthcoming). They suggest that as a mother's educational attainment rises beyond lower secondary edu- cation, the marginal gain per dollar invested tends to diminish sharply in terms of the impact on social welfare, as reflected in such indicators as incidence of poverty, ability to read as an adult, health and childcare behaviors, and child mortality. Taking these factors into consideration suggests a strategy for man- aging student flow that recognizes the possible necessity--indeed the desirability--of distinguishing between the lower and upper levels of the education system and of diversifying the curricula to facilitate young people's transition to the world of work (Walther and Filipiak 2008; World Bank 2008, forthcoming). At the lower levels of the system, the aim would be to universalize access, while at the upper levels the size of enrollments would be geared toward the absorptive capacity of the labor market. The five simulation scenarios correspond to varying degrees of distinction between the cycles: the sharper the distinction, the more selective the transition between the lower and upper levels of the system. The implications for enrollments in 2020 under each of the simulation scenarios are presented below. These results set the stage for analyzing, in the final chapter of this paper, the fiscal implications of the various policy options. PROJECTED ENROLLMENT IN 2020 UNDER ALTERNATIVE OPTIONS FOR STUDENT FLOW MANAGEMENT The aggregate picture of 2020 projected enrollment, based on the sim- ulation results for each of the 33 SSA countries in the sample, is shown in table 5.1. For primary education, the first four simulation scenarios-- ES-1 to ES-4--assume a primary school completion rate of 95 percent by 2020. Thus, in all four scenarios, the aggregate number of primary school pupils is expected to rise from 100.4 million in the base year, to 161.0 million by 2020, an increase by a factor of 1.6. Under scenario ES-5, where basic education is extended to 9 years and the completion rate is assumed to be 80 percent, total enrollment is projected to reach 211.6 million in 2020, an increase by a multiple of 1.85 over base-year enrollments in the primary and lower secondary cycles. These results suggest that the base of the system will continue to experience very substantial growth in numbers in the coming decade. 22 Table 5.1 Simulation Results: Student Enrollments by Level of Education in 2020 for 33 Low-Income SSA Countries million Lower secondary Upper secondary School-to- School-to- Transition work work Coverage rate P­S1 Transition transition transition Year scenario (%) beyond S1 Primary General TVET program General TVET program Tertiary Base year 63 ­ 100.4 13.8 n/a n/a 7.2 n/a n/a 3.0 2020 ES-1 100 56.6 5.7 0.0 25.8 3.5 2.7 13.8 "Continuity" ES-2 80 42.8 6.9 2.0 20.4 2.7 2.1 10.8 161.0 ES-3 80 42.8 6.9 2.0 5.8 1.5 9.5 3.1 "Discontinuity" ES-4 65 32.7 7.5 3.4 5.4 0.8 3.8 2.5 9 years of ES-5 "Discontinuity" 211.6 0.0 0.0 5.8 0.8 7.5 2.5 primary Notes: Rate of real growth of per capita GDP: 2% per year. n/a: not available. P: primary, S1: lower secondary, S2: upper secondary. Appendix E provides the corresponding simulation results for each of the 33 countries. Appendix D provides the aggregate simulation results corresponding to real per capita GDP growth of 0% and 2% per year. Coverage of Post-Primary Education · 23 Projections of enrollments in 2020 for the other levels and types of programs vary considerably under the various scenarios. Not surprisingly, they are highest under ES-1 for all segments of the post-primary system, an outcome fully consistent with the scenario's assumptions of universal transition from the primary to the lower secondary cycle and of perpetu- ation of current rates of transition beyond the lower secondary cycle. The most striking result is the difference in the projected enrollments between two sets of simulations, one set consisting of ES-1 and ES-2, and the other ES-3, ES-4, and ES-5. To recall, the first set corresponds to a model of con- tinuity in student flow management, while the second set corresponds to a model of discontinuity in which progression beyond the lower second- ary cycle is calibrated to the absorptive capacity of the modern sector of the labor market. In the first set of projections, enrollments in 2020 in ter- tiary education, for example, would rise by a factor of 3.6 and 4.6 relative to the base year; in the second set, the increase would be by a factor of between 1.1 and 1.5. To gain an overall perspective on the growth in enrollments, consider the projected aggregate for all segments of the system beyond the primary cycle. In the base year, the aggregate enrollment was 24.4 million for the 33 SSA countries in the sample. By 2020, the number could reach 59 to 108 million, an increase of 2.4 to 4.5 times the enrollments in the base year. Thus, even under the most selective scenario (ES-4), the increase is significant and will pose very demanding logistical challenges. In ES-5, where a 9-year basic education cycle is introduced to integrate primary and lower secondary edu- cation into a single cycle, the projected enrollments are comparable to those in scenario ES-4. The main difference between these two scenarios relates to costs, a topic considered in greater detail in later chapters. NOTES 1. The characteristics of these programs are discussed in a later chapter. For the purpose of the current chapter, the focus is on the number of students that might be involved. 2. Appendix table C.1 provides a schematic summary of the five coverage options. 3. The regression is documented in World Bank, forthcoming. This feature in the model implies that in scenarios where enrollments in tertiary education are linked to the level of per capital GDP and the unemployment rate, the projections will be sensitive to the assumed rate of economic growth. This expectation would apply to scenarios ES-3 as well as ES-4 and ES-5, but not to ES-1 and ES-2. Simi- larly, since enrollments in upper secondary education are derived from the pro- jected enrollments in tertiary education, this feature in the simulation model also affects the number of students projected in upper secondary education. CHAPTER 6 The Quality and Reach of Educational Services and Related Unit Costs igh-quality educational services are a desirable goal, but they are H costly to provide. In countries facing severe financial constraints, these services inevitably would be available to relatively few people. Therefore, for the system as a whole, compromises must be struck between providing educational services that meet acceptable standards for quality and reaching as many of the target populations as desired. The challenge is to determine what those standards are, bearing in mind that what one country deems "acceptable" may not qualify as such in another country. For the purpose of this report, the focus is on the quality of services from the perspective of inputs, given its direct link to costs and resource needs. The larger issue of how well resources are transformed into learning outcomes is also important--perhaps critically so, given that enhancing student learn- ing is a key goal of educational policy--but it requires a separate analysis not attempted here.1 The discussion in this chapter first documents the diversity in the delivery of educational services across SSA countries. It then proposes reference benchmarks for the simulation model in terms of "acceptable" standards for the quality of services from an input perspective. The discussion also explains two features embedded in the simulation model, one to respond to the fact that some of the simulation parameters (such as teacher salaries) are sensitive to a country's level of per capita gross domestic product (GDP), and the second to allow for a pattern of rising marginal costs as educational services are extended to rural populations. CURRENT PATTERNS OF PER-STUDENT PUBLIC SPENDING BY LEVEL OF EDUCATION Table 6.1 shows average per-student public spending in our sample of 33 low-income Sub-Saharan African (SSA) countries; for comparison, it also includes the relevant data for three other groups of developing 25 26 · Developing Post-Primary Education in Sub-Saharan Africa Table 6.1 Public Spending per Student by Level of Education, Sub-Saharan Africa and Other Country Groups, Circa 2005 Per-student spending in percentage of GDP per capita Secondary Tertiary Lower Upper Ratio Ratio Primary secondary secondary Average secondary/ Tertiary tertiary/ (%) (%) (%) (%) primary (%) primary Low-income countriesa Sub-Saharan Africa 11 33 60 40 3.7 370 34 Other regions 11 ­ ­ 12 1.1 53 5 Middle-income countriesb Sub-Saharan Africa 15 ­ ­ 27 1.8 200 13 Other regions 14 ­ ­ 15 1.1 34 2 Sources: Data for low-income SSA are from the database compiled for this study. Data for other countries and regions are from UIS and EdStats. Note: See also appendix table B.3­appendix table B.5 for more details about school organization and per-student spending. a. Countries eligible for lending from the International Development Association (IDA). b. Countries eligible for lending from the International Bank for Reconstruction and Development (IBRD). Dash = data not available. countries in Africa and elsewhere. Relative to per capita GDP, the level of public spending per student in SSA countries averages 11 percent for pri- mary education, 40 percent for secondary education, and 370 percent for tertiary education. For middle-income SSA countries, the corresponding figures are 15 percent, 27 percent, and 200 percent; and for low-income countries outside SSA, they are 11 percent, 12 percent, and 53 percent. These comparisons are a first indication that the cost structure of post- primary education in SSA countries is likely to pose a serious impediment to its expansion and development. A closer look at the data on public spending per student reveals the following noteworthy features: · In primary education, the figure for SSA countries is comparable to the averages for the other country groups, generally falling in the range of 11 to 15 percent of per capita GDP. · In secondary education, SSA countries' average in upper secondary is nearly twice as high as in lower secondary: 60 percent of per capita GDP compared to 33 percent.2 The average for the two cycles, at 40 percent of per capita GDP, is more than 3 times the average for low-income countries in other parts of the developing world. Across SSA countries, public spending per student in secondary education is, on average, about 3.7 as high as in primary education. The Quality and Reach of Educational Services and Related Unit Costs · 27 · In tertiary education, SSA countries' average is substantially greater than the average for lower levels in the system, and it is also much higher than the averages in other country groups. On average, SSA countries spend 34 times as much on each student in tertiary education as on a pupil in primary school; the corresponding ratio is 5 times in low-income coun- tries outside SSA and 13 times in middle-income SSA countries. The generally high levels of public spending per student in secondary and especially in tertiary education set African education systems apart from those of other countries. They are one reason the coverage of post- primary education in the region remains relatively modest today and why expanding it will pose significant challenges in the coming years. To relieve this constraint on expansion, one must understand the underlying drivers of SSA countries' high level of spending per student. Because the bulk of expansion in the coming decade will be in secondary education, understanding the cost drivers in secondary education is especially impor- tant. This issue is considered in greater detail below. COMPARATIVE PERSPECTIVES ON COST DRIVERS IN SECONDARY EDUCATION How much teachers are paid, how their instructional time is used, and how much is spent on complementary inputs are important drivers of per- student spending. These variables are therefore the main loci of policy decision making. To evaluate the options in this regard, a key challenge is to identify appropriate benchmarks that would correspond to reasonable input standards for service delivery. To explore these issues, one may find it helpful to consider first the data on the key cost drivers in table 6.2, which, for reasons of space, shows only the average for all the 33 SSA countries in the sample and the corresponding figures for selected countries in the sample.3 A striking fea- ture is that the regional average masks highly diverse arrangements for service delivery. The average number of students per class can vary from fewer than 20 to more than 80; the average teacher's pay relative to the per capita GDP can go from a multiple of 2 to a multiple of more than 9; and spending on inputs other than teachers as a share of total recurrent spend- ing can range from 15 to 60 percent. The great diversity across countries suggests that no fixed organizational arrangements can be implied by such generic terms as "lower secondary" or "upper secondary" education; alternatively stated, countries do enjoy substantial scope in choosing how services are organized and delivered. 28 · Developing Post-Primary Education in Sub-Saharan Africa Table 6.2 Characteristics and Costs of School Organization in Secondary Education Average Share of teacher recurrent Per-student salary Teachers expenditure spending (in multiples Students Students per other than (in multiples of per teacher per class class (ratio for teacher of GDP/capita) (1) (2) (2)/(1)) salaries (%) GDP/capita) Lower secondary Burkina Faso 9.3 50.0 75.0 1.5 60.4 0.47 Burundi 9.3 20.5 42.4 2.1 29.1 0.64 Cameroon 6.5 31.1 40.3 1.3 34.3 0.32 Ethiopia 8.1 48.3 67.8 1.4 18.6 0.21 Guinea 2.9 47.6 88.9 1.9 44.0 0.11 Malawi 7.7 26.3 50.0 1.9 40.0 0.49 Togo 8.7 54.2 87.9 1.6 13.6 0.19 Uganda 7.4 6.8 15.2 2.2 25.1 1.45 Average of 33 countries 6.0 35.5 53.5 1.6 38.0 0.33 Upper secondary Burkina Faso 13.0 39.0 52.0 1.3 48.0 0.64 Burundi 11.0 16.4 32.4 2.0 72.0 2.40 Cameroon 6.8 29.1 36.4 1.3 36.5 0.37 Ethiopia 11.9 50.3 81.7 1.6 40.5 0.40 Guinea 2.9 52.3 95.0 1.8 41.0 0.09 Malawi 7.7 26.3 50.0 1.9 40.0 0.49 Togo 9.0 33.3 62.4 1.9 15.9 0.32 Uganda 7.4 6.8 15.2 2.2 25.1 1.45 Average of 33 countries 7.4 27.1 45.7 1.8 39.9 0.60 Source: See appendix table B.3 and appendix table B.4. Given the considerable spread in the way secondary education is organized, the future development of secondary education is likely to pose fewer problems in some countries than others. In particular, the gov- ernment's financial burden would be more manageable where current organizational choices already favor modest levels of per-student spend- ing. Yet lower costs in and of themselves are not necessarily optimal if this implies accepting enormous class sizes for teaching and learning, The Quality and Reach of Educational Services and Related Unit Costs · 29 possibly 80 or more students in each class. An important challenge for policy makers, therefore, is to identify suitable benchmarks for the cost drivers in order to set the future development of the subsector on a sus- tainable path while assuring adequate schooling conditions. The experi- ences of other developing countries that have achieved sustained growth of their secondary school systems provide some inspiration in this regard. Table 6.3 contains pertinent comparative data on the main cost drivers. It shows that with regard to student-teacher ratios, the average for SSA (31.4) is comparable to that for low-income countries in other world regions (29.8), but exceeds by a large margin the ratio of about 20 in middle-income countries in Africa and elsewhere. With regard to teacher salaries, the difference between SSA and other countries is par- ticularly striking. Salaries in SSA countries average about 6.0 times the per capita GDP in the lower secondary cycle and 7.4 times in the upper secondary cycle (implying an average of 6.5 times for the two cycles taken together). The corresponding figure is 3.8 in middle-income African countries, 2.8 in low-income non-African countries, and 2.4 in middle-income non-African countries. Thus, when salaries are expressed Table 6.3 Comparison of Coverage and Organization of Secondary Education, Sub-Saharan Africa and Other Country Groups, Circa 2005 Determinants of level of per-student spending Private Average Context sector teacher Per-student GER GDP per enrollm. salary spending capita Lower Upper as share Share of (in multiples (in multiples (USD secondary secondary of total repeaters Students/ of GDP of GDP per 2003) (%) (%) (%) (%) teacher per capita) capita) Low-income countriesa Sub-Saharan Africa 320 32.7 13.8 19.8 13.2 31.4 6.5 0.40 Other regions 734 73.5 46.3 38.6 4.9 29.8 2.8 0.12 Middle-income countriesb Sub-Saharan Africa 3,182 81.6 57.1 21.2 8.8 21.0 3.8 0.27 Other regions 2,477 88.0 58.5 22.4 5.3 19.1 2.4 0.15 Source: Data for low-income SSA are from the database compiled for this study. Data for other countries and regions are from UIS and EdStats. a. Countries eligible for lending from International Development Association (IDA). b. Countries eligible for lending from International Bank for Reconstruction and Development (IBRD). 30 · Developing Post-Primary Education in Sub-Saharan Africa in units of per capita GDP, the average for low-income African countries is 2.4 times as high as that in low-income countries elsewhere in the world. These comparisons of the cost-related variables provide a basis for delimiting the potential values of the cost parameters in the simula- tion model. BENCHMARKS FOR COST DRIVERS IN SECONDARY EDUCATION Benchmarks are useful in the simulation exercise for two main reasons. First, as an alternative to the current values of the cost-drivers, they bring a forward-looking orientation to the evaluation of policy options for cost management. Second, the benchmarks are highly relevant for the 33 low- income SSA countries, given that most, if not all, of them will require external aid to complement domestic resources to attain their educational goals. In this context, having a reference framework helps focus specifi- cally on issues of cost efficiency, thereby providing donor agencies one yardstick, among others, for ensuring that taxpayer money is not being wasted in the education sector. At the same time, the benchmarks pro- vide a basis for treating countries equitably when evaluating appropriate levels of aid and for minimizing the risk of allocating too much to coun- tries with inefficient service delivery arrangements at the expense of countries with more efficient arrangements. The foregoing considerations motivate an approach similar to that associated with the Education for All Fast Track Initiative (EFA-FTI), which involves specification of indicative benchmarks for an array of cost-sensitive parameters relating to the organization of educational services.4 To recall, the main purpose of the indicative framework is to provide an objective reference for assessing alternative policy options for the education sector. Its main use is to help focus attention on the most critical structural issues that may impede the sector's future development. As such, the indicative framework is an analytical device, meant as an aid for policy development rather than as a tool for identifying specific conditions to impose on aid-receiving countries. Because initial conditions and fiscal constraints on the education sec- tor differ widely across the 33 SSA countries in this study, it may be appealing to argue against measuring all the countries against a single ref- erence framework. Yet the diversity across countries is also a strong argu- ment for the use of precisely such a framework, in order to systematize the policy discussions. To reconcile these two perspectives, two indicative frameworks for service delivery in secondary education are developed for The Quality and Reach of Educational Services and Related Unit Costs · 31 the simulations. The first might be termed a "generous" or liberal frame- work, with parameter values that are consistent with relatively favorable conditions for schooling. The second might be called a "spartan" or cost- conscious framework, in which the parameter values correspond to more economical (but still acceptable) schooling conditions. By negotiating between these two sets of benchmarks, the simulation exercise for each country can be customized to its specific context. Naturally, the choice of parameter values has different cost and budget implications. To characterize these two indicative frameworks, table 6.4 shows the corresponding values of the key variables in the simulation model. They are organized and discussed as follows: (1) the organization of service delivery, as defined by a set of three variables; (2) the level of teacher salaries; (3) the share of spending on non-teacher inputs; and (4) the rate of grade repetition. ORGANIZATION OF SERVICE DELIVERY The three pertinent variables are: average class size or the number of stu- dents per instructional group, weekly instructional hours received by stu- dents, and weekly hours of teaching performed by teachers. These three variables bear an algebraic relation to the student-teacher ratio.5 But because they correspond to policy levers that decision makers can act upon more directly than the student-teacher ratio, the simulation model Table 6.4 Indicative Framework for the Organization of Services in Secondary Education Lower secondary Upper secondary "Generous" "Spartan" "Generous" "Spartan" Organization of service delivery Students per class 40 50 40 45 Hours of instruction per week (students) 28 25 30 27 Teaching hours per week (teachers) 18 21 16 18 Students per teacher 25.7 42.0 21.3 30.0 Average teacher salary (multiples of GDP/capita) 5.0 4.6 6.5 5.6 Expenditure other than teacher salary (%) 45 38 50 42 Share of repeaters (%) 10 10 10 10 Per-student spending (in percent of GDP per capita) 35.4 17.7 60.9 32.2 Source: Authors. Note: This parameter describes the "basic" level of per-student spending before any adjustment for level of wealth, geographic dimension, or stimulation of demand. 32 · Developing Post-Primary Education in Sub-Saharan Africa seeks inputs of values for these variables rather than for the student- teacher ratio. The model uses the following values of these three variables: · In the lower secondary cycle, the values for the "generous" scenario are as follows: 40 students per class (better than the current SSA aver- age of 54); 28 hours of instruction per week; and 18 hours of teaching per week. With these parameter values, the implied student teacher ratio is 25.7 (compared with the base year SSA average of 35.5). In the "spartan" scenario, the parameter values are: 50 students per class (still better than the current average); 25 hours of instruction per week; and 21 hours of teaching per week. The student-teacher ratio would then be 42.0, which is slightly higher than the SSA average for the lower secondary cycle (35.5) but better than the SSA average of about 46 for the primary cycle.6 The assumed instructional hours and teach- ing loads in both scenarios are well within the ranges observed among SSA countries and are reasonable in absolute terms. · In the upper secondary cycle, the parameter values shown in table 6.4 are generally more resource-intensive than in the previous cycle, implying student-teacher ratios of 21.3 under the "generous" scenario and 30.0 under the "spartan" scenario, compared with the current SSA average of 27.1. The assumed class sizes in these scenarios are 40 and 45, respectively, relative to the SSA average of 45.7 in the base year. As in the scenarios for the lower secondary cycle, the assumptions about instructional hours and teaching loads are within the scope of current practices in the sample of 33 countries. LEVEL OF TEACHER SALARIES As mentioned, this variable presents a daunting challenge in most SSA countries because of its cost implications. The annual cost of teacher salaries in the lower secondary cycle averages 6.0 times the per capita GDP, while in the upper secondary cycle, it is 7.4 times, giving an average of 6.5 for the two cycles taken together. For comparison, the annual salary cost of lower secondary teachers is only 2.8 times in low-income non-African countries and 2.4 times in middle-income non-African countries. Adjust- ing the 2.8 and 2.4 ratios to be consistent with income levels in SSA yields a teacher salary cost of around 3.6 times per capita GDP.7 Current sec- ondary school teacher salaries in SSA are about 80 percent higher than this level. Thus, even with this adjustment for income differences, the absolute value of the implied salary would still be too low (ie, too far from the cur- rent situation). To be a realistic policy option in the coming years. The Quality and Reach of Educational Services and Related Unit Costs · 33 Another approach for choosing values for this variable is to relate it to salaries in primary education. Under the indicative framework of the EFA-FTI, salaries for primary school teachers are benchmarked at 3.5 times the per capita GDP. Assuming that lower secondary school teachers have three more years of schooling than teachers in primary schools, and assuming that the private rate of return to each additional year of schooling is 12 percent--a very reasonable rate--the salary of lower sec- ondary school teachers would be about 40 percent higher, or 4.9 times the per capita GDP. Similarly, if upper secondary school teachers have five more years of schooling than primary school teachers, their salary would be 76 percent higher, or 6.2 times the per capita GDP.8 Thus, for the sim- ulation exercise, the assumptions about teacher salaries are as follows: in the "generous" scenario, the values are 5 times the per capita GDP for lower secondary and 6.5 times for upper secondary; in the "spartan" scenario, the corresponding ratios are 4.6 and 5.6, respectively. RECURRENT EXPENDITURE EXCLUDING THE SALARIES OF TEACHERS9 The choice of values for this variable in the simulation scenarios is based on the range observed across the 33 SSA countries. On average, expendi- ture on inputs other than teacher salaries accounts for about 38 percent of total recurrent spending in the lower secondary cycle and 40 percent in the upper secondary cycle. The variance is large around these averages. For the "generous" scenario in the simulations, the assumed value of this vari- able is 45 percent in lower secondary education and 50 percent in the upper secondary cycle. In the "spartan" scenario, the shares are, respec- tively, 38 and 42 percent. REPETITION RATE This variable does not affect spending per student but does have budget implications in that the higher its value, the more students remain in the system and the greater the budget required to maintain operations. In recent years, repetition rates in SSA countries have been rising and cur- rently average about 13 percent in both cycles of secondary schooling in the 33 SSA countries. The rates are around 5 percent in non-African coun- tries and less than 9 percent in middle-income African countries. For sim- plicity, the simulation exercise assumes a rate of 10 percent for both cycles of secondary schooling for both the "generous" and "spartan" scenarios. The level of per-student spending implied by the combinations of pol- icy choices discussed above appears in the last row of the table. Under the 34 · Developing Post-Primary Education in Sub-Saharan Africa "generous" scenario, per-student spending amounts to an average of 0.354 times the per capita GDP in the lower secondary cycle, and an average of 0.609 times in the upper secondary cycle. The corresponding averages for the "spartan" scenario are 0.177 and 0.322 times, respec- tively. These results suggest that, while schooling conditions are inevitably less comfortable under the "spartan" scenario, the disadvantage is not dramatic, as reflected in the differences between the two scenarios in the values of the variables pertaining to organizational arrangements, the level of teacher salaries and the share of spending on inputs other than teacher salaries.10 Yet the combined impact of making cost-conscious choices on all three dimensions is substantial: per-student spending under the "spartan" scenario is, on average, only about half as high as that under the "generous" scenario.11 COST ASSUMPTIONS FOR TECHNICAL AND VOCATIONAL EDUCATION AND TRAINING The cost assumptions for technical and vocational education and training (TVET) and higher education (discussed in the next chapter) do not use the in-depth composite approach adopted for lower and upper secondary education. Instead, for simplicity, the cost assumptions are linked through proportional ratio to the cost assumptions in secondary educa- tion, and to costs in other countries. With regard to TVET, the simulation model makes a distinction between two very distinct types of activities: (1) formal courses, typi- cally institution based, that are designed as an alternative to the aca- demic or general stream of studies and (2) life skills training, usually of short duration, that are offered at the end of the primary or lower sec- ondary cycle, mainly to help facilitate young people's transition into the world of work. The latter type of TVET activities serves a dual purpose: (1) to equip young people (most of whom will work in the informal sector of the economy) with practical skills at the start of their working life in hopes of enhancing their labor market productivity and (2) to relieve pressure on the upper levels of the system by offering a positive alternative to young people who might otherwise seek to continue their studies in the formal system. For the purpose of this simulation exercise, only one set of cost param- eters for TVET is maintained. This analytical strategy simplifies the pres- entation without compromising the main conclusions of the exercise. According to available information (which pertains mainly to the first type of courses discussed above), per-student spending on TVET courses, The Quality and Reach of Educational Services and Related Unit Costs · 35 averaged over the different specializations, is typically about three times as high as spending per student in the general stream. On the basis of this relationship, the simulations assume the following cost structure for for- mal TVET courses: 1.2 times the per capita GDP for courses offered at the lower secondary level and 1.7 times the per capita GDP for those offered at the upper secondary level. With regard to the school-to-work transition type of TVET training, cost data are very limited, not least because few countries in fact offer such training at present. For the purpose of the simulation exercise, we made the following cost assumptions: 0.6 times the per capita GDP for courses offered at the end of primary schooling and 0.8 times per capita GDP for those offered at the end of the lower secondary cycle. COST ASSUMPTIONS FOR HIGHER EDUCATION The choice of cost benchmarks for the 33 SSA countries encounters two immediate difficulties: · SSA's much higher costs relative to the per capita GDP than else- where. For the 33 SSA countries, per-student spending in higher education averages 3.7 times the per capita GDP, compared with 0.54 times in low-income non-African countries and 2.0 times in middle-income SSA countries. These comparisons suggest the pres- ence of Africa-specific cost drivers. While SSA's high costs are nei- ther inevitable nor desirable, it would nonetheless be unrealistic to ignore them entirely in deciding on an appropriate benchmark for the simulation exercise. · Highly variable levels of public spending per student across coun- tries. Among the 33 SSA countries, those with the lowest levels of spending per student probably provide insufficient funding to assure services of a reasonable quality, while those with the highest levels probably have systems that are wasteful. The data for the 33 countries therefore provide an uncertain basis for choosing an appropriate benchmark. In light of the foregoing observations, a reasonable compromise would be to take the average per-student expenditure in middle-income African countries--that is, 2.0 times the per capita GDP--as a benchmark at which all countries would converge by 2020. This average includes operational costs as well as spending on student welfare services. In addition to benchmarking the average level of spending per stu- dent, the simulation model also differentiates spending by field of study. 36 · Developing Post-Primary Education in Sub-Saharan Africa For the purpose of this study, three major fields are distinguished: humanities and social sciences; natural sciences; and professional studies12. Using available cross-country data, the simulation model assumes the following distribution of enrollment across these three fields: 60 percent, 15 percent, and 25 percent, respectively (table 6.5). Combining this information with available data for the 33 countries on relative cost structures, the simulation assumes the following cost parameters by field of study (in units of per capita GDP): 1.5 times for the humanities and social sciences, 2.4 times the natural sciences, and 2.9 times for professional studies. These assumptions are consistent with the overall average spending of 2.0 times the per capita GDP. TREATMENT OF CAPITAL COSTS IN THE MODEL The simulation model also calculates the capital costs in the different scenarios, but only those associated with the construction of classrooms in primary and secondary schools to accommodate growing student enrollments in these cycles. It does not include other capital or devel- opment costs, such as those for preschool, teacher training, or higher education. In the model, the capital costs are simply a function of the choice of enrollment scenario (ES-1 to ES-5), the class size, and the unit construction cost. For class size, a benchmark of 40 students per class is used for primary education; in lower and upper secondary school, a class size of 40 is used in the "generous" option for school organization, while the "spartan" option raises this standard to 50 for lower and 45 for upper secondary schools.13 Unit construction costs per classroom (furnished and equipped) are $12,000 for primary schools and $16,000 in both lower and upper secondary schools. Table 6.5 Indicative Framework for the Organization of Services in TVET and Tertiary Education Lower secondary Upper secondary Tertiary School-to- School-to- work work Humanities transition transition and social Professional TVET program TVET program sciences Sciences studies Total Student distribution ­ ­ ­ ­ 60% 15% 25% 100% Unit cost (in multiples of GDP/capita) 1.2 0.6 1.7 0.8 1.5 2.4 2.9 2.0 Source: Authors' construction. The Quality and Reach of Educational Services and Related Unit Costs · 37 MAKING THE INDICATIVE FRAMEWORK MORE FLEXIBLE In the context of the Education for All Fast Track Initiative (EFA-FTI), an indicative framework was used to evaluate the cost of attaining the Millennium Development Goal of universalizing primary school com- pletion. Experience with this framework suggests that, while the appli- cation of standard benchmarks helped in rationalizing the analysis across a large number of countries, some flexibility in the benchmarks, particularly to take account of differences in countries' level of eco- nomic development, might have made the exercise even more useful. The use of a single set of benchmarks proved particularly problematic with regard to teacher salaries--which for primary school teachers were set at an average of 3.5 times the per capita GDP in the FTI framework-- and was singled out as particularly problematic in the poorest countries. Where the per capita GDP was $100 or lower, it implied absolute pay levels that in fact fell below the poverty line in many of the countries. Although this study follows the same practice as the FTI indicative framework in relying on benchmarks--on the argument that it provides a fair basis for evaluating funding gaps for possible external support-- important features have been incorporated to make the framework more flexible. In particular, on the cost side, the simulation model allows for adjustments for differences across countries in per capita GDP as well as in population distributions between urban and rural areas, which could affect the cost of service delivery and also require demand-side interven- tions to enroll the remaining pockets of out-of-school children.14 These features of flexibility are briefly explained below. ADJUSTING THE BENCHMARKS FOR TEACHER SALARIES WHERE INCOMES ARE ESPECIALLY HIGH OR LOW In the simulation exercise, teacher salaries are expressed in units of per capita GDP to avoid the complication of separate assumptions about price inflation over the long time frame of the simulations and also to allow for salary increases as a country grows richer. Analysis of available data on teacher salaries suggests that when expressed in per capita units, this variable bears an inverse relationship to the per capita GDP, even though in absolute terms its value tends to be smaller the poorer the country (see appendix A for further discussion of this point and its impli- cations). This statistical relationship has been used in this simulation exercise to adapt the assumed value of teacher salaries to a country's per capita GDP. 38 · Developing Post-Primary Education in Sub-Saharan Africa As a result, in the model's primary education module, the value of salaries is benchmarked at 3.0 times per capita GDP for countries with GDP per capita above $600, 3.5 times per capita GDP for countries with GDP in the $300­600 range, and among the latter group, the benchmark is raised to 3.6, 3.9, and 4.2 times the per capita GDP, respectively, when the per capita GDP is $250, $150, and $100 (in 2000 prices). With regard to secondary school teacher salaries, the assumed values for the poorest countries (that is, per capita incomes no higher than $250) are raised by the same proportion above the standard benchmarks attached to the other countries.15 Similarly, for countries with incomes above $600, lower secondary school teacher salaries (as a multiple of per capita GDP) are lowered as compared with the standard benchmarks.16 With regard to TVET and higher education, the simulation model does not specify teacher salaries as a separate variable. Instead, it simply uses spending per pupil as an aggregate variable to reflect the cost of service delivery. For TVET, this variable is pegged to the corresponding variable for secondary education, which therefore implies a built-in adjustment for the poorest countries through the mechanism explained above. For tertiary education, the standard benchmark for per-student spending is 2.0 times the per capita GDP. For the poorest countries, a higher ratio is used that takes into consideration the fact that a large share of the expen- diture at this level of education typically goes for inputs that are priced in regional markets (such as academic staff) or even procured in foreign cur- rency and therefore relatively more expensive for the poorest countries (for example, books, computers, and laboratory consumables).17,18 ADJUSTING THE BENCHMARKS FOR CURRENT EXPENDITURE, EXCLUDING TEACHER SALARIES One difficulty in choosing appropriate benchmarks is that this variable covers expenditure denominated in local currency (for example, consum- ables and salaries of administrative staff) as well as those bought with for- eign currency (for example, textbooks). Because this variable is expressed in the simulation model as a share of total unit cost, it is effectively pegged to the level of teacher salaries. Since the latter variable, when expressed in absolute terms, is lower the poorer the country, the implica- tion is that spending on non-salary inputs would also be smaller in absolute terms the lower the income of the country. For the poorest countries, then, the amount thus derived may not be realistic for purchas- ing inputs on international markets that are required to deliver services of reasonable quality. Thus, for these countries, the amount for non-salary The Quality and Reach of Educational Services and Related Unit Costs · 39 spending is benchmarked at the same absolute amount as countries with a per capita GDP of $300 (in 2000 U.S. dollars). ADJUSTING FOR THE INCREASING MARGINAL COSTS TO REACH RURAL POPULATIONS Because universal coverage in primary education, and, increasingly, in lower secondary education, are common goals in many SSA countries, the simulation exercise takes explicit account of the rising marginal costs of reaching rural populations, given that most of the out-of-school popula- tion resides in rural areas.19 The following are two observations about the pattern of enrollments in urban and rural areas in the 33 country sample: · The urban share of the school-age population currently averages about 30 percent in the sample, but its range stretches from less than 20 percent in Burundi and Burkina Faso to about 50 percent in Repub- lic of Congo and The Gambia. · The coverage of the education system is generally much lower in rural than in urban areas at both the primary and lower secondary levels. In lower secondary education, for example, the average gross enrollment ratio is 66.5 percent in urban areas, but only 22.2 percent in rural areas. The pattern of coverage shows the usual wide variation across countries.20 These observations and further analysis of the available data imply that among the current out-of-school primary school-age population, about 80 percent on average live in rural areas. The same analysis for the lower secondary cycle suggests an even higher share of 85 percent. While the share does vary around the regional average, the reality is that in all the sample countries, rural residents make up the bulk of the relevant out- of-school populations. If the goal is to universalize primary and lower secondary schooling, the simulation model must incorporate, in a more explicit manner, the cost of servicing and reaching rural populations. For primary schooling, analysis of economies of scale in service delivery for 16 of the sample countries suggests that per-pupil costs in a school with 100 pupils is on average about 35 percent higher than those in a school with 300 pupils; and it is on average about 25 percent higher in village schools than in city schools. For lower secondary education, economies of scale are more pro- nounced, with per-student costs in a school with 120 students exceeding by 70 percent those in a school of 400 students. These differences in unit spending among schools of different sizes reflect the impact of fixed costs 40 · Developing Post-Primary Education in Sub-Saharan Africa and the use of specialized teachers who, because of fragmented curricula, cannot fully fill their weekly teaching load in small rural schools. As before, there is substantial variation around these regional averages. In lower secondary education, the difference between schools of 120 and 400 students could be a factor of just 1.15 in some countries to as much as 2.40 in others.21 PRIMARY EDUCATION MODULE For the primary education module in the simulation model, we retained the pupil-teacher ratio at the EFA-FTI indicative benchmark of 40 to 1. A key consideration in doing so is that, given the already large system size in many countries, there should be sufficient room around this average to accommodate, in a pragmatic manner, (slightly) lower ratios in rural or sparsely populated areas while tolerating (slightly) higher ratios in more densely populated areas. By contrast, the simula- tion exercise raises the benchmark for teacher salaries beyond the FTI indicative value of 3.5 times the per capita GDP. Experience suggests that attracting the remaining out-of-school population typically requires posting teachers to remote rural schools. In practice, teachers may not take up their posts or may fail to turn up regularly to teach their classes without additional incentives. Faced with this difficulty, several SSA coun- tries (for example, The Gambia, Lesotho) have indeed taken measures to address the problem. These measures must obviously target only localities where teacher placement is an issue, and the incentives they provide must be sufficiently large to attract the target teachers. Taking into consideration the experience of countries that have dealt with this challenge, we adapted the benchmarks for salaries by assuming that 15 percent of the teachers in rural areas would require a salary supplement and that the supplement would increase their salaries by an average of 20 percent beyond the standard benchmark. Finally, the simulation exercise makes one other adjustment to take into account the challenges of reaching the remaining out-of-school population as coverage progresses toward the 100 percent mark. Because the relevant population typically consists of the poorest and most tradition-bound segments of society, special efforts to reach them with customized services might be required. The simulation model makes allowance for this possibility by assuming that 10 percent of the rural pop- ulation falls in this category and that extending primary school coverage to them would entail services that cost 30 percent more than the standard benchmark for per-pupil spending for the rest of the population. The Quality and Reach of Educational Services and Related Unit Costs · 41 LOWER SECONDARY MODULE For the lower secondary module in the simulation model, the simulation benchmarks were adjusted only to take account of economies of scale in service delivery; this approach focuses on the source of the rising marginal costs to extend services to rural areas at this level of schooling. To keep costs under control, countries will need to carefully manage two aspects of service delivery simultaneously: (1) the administrative arrangements and pedagogical support functions in small (rural) schools and (2) the curriculum design and the capacity of teachers to teach multiple subjects. While countries clearly differ in their capacity to manage these factors, a key tenet in this study is that some approaches are clearly more efficient and equitable than others and that these should be promoted as the pre- ferred arrangements for expanding coverage in rural areas. Without going into detail on how best to organize educational services in small schools, analysis of existing practices suggests that it is possible, through careful management of staff resources (both teaching and administrative staff), to limit the cost per student in small (rural) schools to no more than 1.3 times that in larger urban schools. For each country, the cost simulations are based on the application of this marginal cost benchmark to the esti- mated shares of small schools at each point in time. NOTES 1. Student learning is the subject of a large and continuing research agenda in developed and developing countries alike. The available findings suggest that while schools cannot function without a minimal level of resources, the relation between resources and learning outcomes is weak at best. Important factors that affect performance include the behaviors of key personnel involved in service delivery and the organizational, governance, and management arrangements in the system (Hanushek and Wößmann 2007). 2. For countries outside Sub-Saharan Africa, most of the available data comes from the UNESCO Institute for Statistics. Data from this source do not always make a distinction between the two cycles of secondary education. 3. See appendix table B.3 and appendix table B.4 for data for all 33 SSA countries. 4. The EFA-FTI indicative framework provides benchmarks for seven indica- tors: government spending on education (about 20 percent of budget), spend- ing on primary education (about 50 percent of education budget), teacher salary (about 3.5 times GDP per capita), pupil-teacher ratio (about 40:1), non- teacher salary spending (33 percent of recurrent spending), average repetition rate (10 percent or lower), and annual hours of instruction (850 or more). See Bruns, Mingat, and Rakotomalala (2003) for details on how these bench- marks were derived. 42 · Developing Post-Primary Education in Sub-Saharan Africa 5. Student-teacher ratio = class size × (teachers' weekly hours of teaching duties/ students' weekly hours instructional time). 6. See appendix table B.5 for data on school organization in primary and ter- tiary education. 7. Based on a regression between teacher salaries and GDP per capita. The average GDP per capita in low-income SSA countries was $320 in 2003 compared with $734 in low-income countries in other regions. 8. It can also be observed that, in the current situation, a lower secondary school teacher earns 40 percent more on average than a primary school teacher, while the corresponding figure is 70 percent for an upper secondary teacher. Applying these coefficients to the EFA-FTI indicative framework value for primary education, we arrive at a teacher salary level of 4.9 and 6 times the GDP/capita for the two secondary cycles, respectively. These figures are close to those obtained with the other method. 9. This expenditure concerns textbooks, educational materials for students and teachers, water and electricity, upkeep of school buildings and furniture, in- service teacher training, expenditure on administrative and pedagogical support personnel in schools, central and decentralized departments, as well as associated operating expenditures. 10. The effective possibilities of distance education will probably also have to be used. 11. It should also be noted that the indicative framework describes average sit- uations (this is what is important financially), but that its implementation can incorporate relevant modulations around the averages in order to take local or regional specificities into account within each country. 12. Preparing for jobs as professions such as doctor, lawyer, journalist, librar- ian, or engineer. 13. See also table 6.4. 14. One other feature that might have been added to increase the flexibility of the simulation model pertains to its resource mobilization module. In par- ticular, it would appear appropriate to link a country's ability to mobilize domestic resources through tax revenues to its prospects for economic growth. This expectation is based on the observation that as economies grow and become more formalized, their tax base and capacity to raise revenues also improve. This tendency is borne out by the experience of most countries. How- ever, among the non-oil-producing SSA countries with a per capita GDP below $1,000, the relation is weak at best. A positive and significant relation between per capita GDP and fiscal capacity exists only for more economically mature countries. Thus, in the simulation exercise no adjustment is made to the bench- marks for variables relating to resource mobilization. 15. Appendix A documents the regression behind these assumptions. 16. Appendix E provides the adjusted teacher salaries and unit costs used for each country in the simulations. 17. Per-student spending by field of study is assumed to retain the same rela- tive structure as in table 6.5. 18. Appendix E provides the tertiary unit costs used in the simulations for each country. The Quality and Reach of Educational Services and Related Unit Costs · 43 19. For upper secondary education, the assumption is that schools will con- tinue to be located mainly in urban areas. This assumption does not apply to lower secondary schools, which must remain close to the populations concerned, particularly with a view to a significant increase in schooling coverage. 20. See, for example, Mingat and Ndem (2008), which work analyses the schooling differences between urban and rural children (but also between girls and boys, rich and poor) at different levels of education in more than 30 countries in the region (schooling profiles), and the procedures followed by more than 20 countries to organize lower secondary education, with very dif- ferent economies of scale structures from one country to another. 21. Differences are due partly to the number of administrative personnel (whether 1 or 3 in a 120-student lower secondary school), and partly (1) to the degree of teacher polyvalence in different subjects and (2) to organization of curriculum content. CHAPTER 7 Evaluating the Financing Gap key aim of the simulation exercise is to determine the amount of A external aid that Sub-Saharan African (SSA) countries might require to attain their education goals. The simulation model includes all levels and types of education to provide a holistic, sectorwide picture, but the focus is on the post-primary system levels. The indicative bench- marks discussed in the foregoing chapters relate to the cost implications of countries' policy choices that affect the future size of enrollments by level and type of education and training and that influence service deliv- ery arrangements. In this chapter, we consider the benchmarks for vari- ables that determine the mobilization of domestic public spending, as well as the public-private division of responsibility for financing. The gap between aggregate costs and the resources that the country is able to mobilize provides a basis for estimating the resource gaps that external financing might contribute to fill. BENCHMARKS AFFECTING VOLUME OF GOVERNMENT RESOURCES FOR EDUCATION The relevant variables determining government resources allocated to education include the following: · aggregate tax revenues as a percentage of gross domestic product (GDP) · the education sector's share of overall government spending · the share of primary education in government spending on education. Because the benchmarks for these variables in the Education For All­Fast Track Initiative (EFA-FTI) indicative framework remain relevant, we retain them for the present exercise. In particular, for the first variable, the bench- marks are: 14, 16, or 18 percent of GDP, depending on the country's per 45 46 · Developing Post-Primary Education in Sub-Saharan Africa capita GDP.1 For the second variable, three benchmarks were used (shares of 20, 23, or 26 percent), each corresponding to a different simulation scenario. The model is structured such that if the share of education in overall government spending exceeds 20 percent (reflecting a high prior- ity for education), all of the resources mobilized beyond this benchmark are channeled to the post-primary levels. With regard to the third variable, the EFA-FTI benchmark of 50 percent was retained for the standard 6-year primary cycle; in countries where the cycle is longer than 6 years, the share is adjusted proportionally. Finally, in view of the whole sector perspective of this exercise, the model includes a variable to capture government spending on preschool education. The benchmark for this variable is set at the common reference level of 5 per- cent of total government spending on education. Taken together, the fore- going parameter values imply that at least 45 percent of overall government spending on education is effectively targeted in most countries to the post-primary levels. BENCHMARKS FOR THE PUBLIC-PRIVATE DIVISION OF RESPONSIBILITY FOR EDUCATION In all countries, there will always be some families that prefer and are able to afford private education for their children. In SSA, the share of privately financed enrollments averages about 10 percent at the primary level, 20 percent in the general stream of secondary education, 35 percent in the technical stream at the secondary level, and 17 percent in tertiary education.2 As with many other aspects of education in SSA, there is great diversity across the 33 countries in our sample: in lower secondary schooling, for example, the share of enrollments in private schools ranges from 5 percent or less in Lesotho, Sierra Leone, and Zambia, to more than 45 percent in Rwanda and Uganda. The diversity across countries makes it difficult to propose a com- mon set of benchmarks for this simulation exercise. Two basic princi- ples, equity and economic efficiency, were used to determine the choice of benchmarks. The first implies that primary schooling that is sup- posed to be universal should be free of charge to students because some parents cannot afford even minimal charges for schooling, so any cost would be sufficient to keep some children away, particularly girls and those from the poorest households. The second implies that individuals should share in the cost of their education if the benefits accrue mostly to the individuals themselves rather than to their communities. Because privately captured benefits generally rise with the level of education, there Evaluating the Financing Gap · 47 is a strong argument to favor the lower levels of education for public financing. The fact that social benefits from basic education are significant (for example, health gains, fertility behavior, and literacy) further strength- ens this argument. Based on these two principles and the cycle-specific considerations below, we chose the following benchmarks for the public-private division of responsibility for education in the simulation. PRIMARY AND LOWER SECONDARY EDUCATION For primary and lower secondary education, we retained the EFA-FTI indicative benchmark of 10 percent for the share of enrollments that are privately financed. This value simply leaves room for a small percentage in each population to exercise their choice for a different schooling serv- ice, one that is paid for privately. For lower secondary education, this assumption means that countries will move from the current share of some 20 percent private financing to 10 percent by 2020. The current high share reflects the combined impact of several fac- tors: the rise of private schooling to cope with the rapid increase in demand associated with the progress in universalizing primary educa- tion, the relative neglect of secondary education as countries prioritize the development of primary education, and the fact that secondary education is currently concentrated in urban areas where the market for private services is more developed. These conditions are unlikely to be sustained, however, as the coverage of lower secondary education is scaled up. Increasing shares of the school population will reside in rural areas and come from poorer strata of society; correspondingly, we expect the role of public sector provision to increase. Given that the goal is universal (or nearly universal) coverage by 2020, we have retained the same benchmark for private enrollment of 10 percent as the FTI benchmark for primary education. UPPER SECONDARY EDUCATION For the second cycle of general secondary education, the situation is struc- turally very different. Whereas a case can be made for progressively uni- versalizing coverage in lower secondary education on equity and social grounds, the development of upper secondary education needs to be aligned more explicitly to preparing youth for work in the modern sector of the economy, either directly after this cycle or after further preparation at the tertiary level. Given the small size of the modern sector in most 48 · Developing Post-Primary Education in Sub-Saharan Africa SSA countries and its slow growth in the foreseeable future, an appropri- ate strategy for the development of upper secondary education is to put a greater emphasis on the quality of skills imparted to students who can benefit from the training than on the sheer number of students enrolled. Indeed, it is at this juncture of the education system that the issues of stu- dent flow regulation and the diversification of students' schooling careers come into sharp focus. At this level in the system, substantial private financing is both more rel- evant as a mechanism for managing student flow and more justified by the usually high private returns to schooling. Public funding remains rele- vant, however, especially when targeted to support able students from modest backgrounds who might otherwise not be able to attend. Thus, for the simulations, we chose a benchmark of 40 percent by 2020 for the share of students that attend privately financed schools. Raising the share to this level from the current average of 20 percent would obviously require further thinking about the best way to make the transition toward a larger role for private financing at this level in the system (Walther 2005; World Bank, forthcoming). SCHOOL-TO-WORK TRANSITION PROGRAMS These programs are included in the structure of the simulation model as specific efforts that target young people who, at the end of primary or lower secondary education, discontinue their formal schooling. Such pro- grams typically involve short-duration activities that seek to equip school leavers with life skills for the world of work, particularly in the informal economy where most of them can expect to earn their livelihoods. The programs are ideally implemented by the relevant professional bodies with government collaboration in the form of financing, rather than by direct involvement in service delivery. The programs serve two important functions: helping improve labor productivity in the informal sector and providing a socially acceptable way to relieve pressure on the upper levels of the formal schooling system. Because the clientele of these programs is likely to be predominantly from disadvantaged groups, for the purpose of the simulation exercise it was assumed that these programs would be funded in full by the government. TECHNICAL AND VOCATIONAL EDUCATION AND TRAINING (TVET) TVET in the simulation model implies formal courses of study that typi- cally last two or three years in an institutional setting (either regular Evaluating the Financing Gap · 49 schools or specialized centers). They are in effect an alternative to the general stream in secondary education, and the curriculum may include work-based attachments to enhance student learning. The private sector plays an important role in providing this type of service, currently account- ing for about 35 percent of enrollments, on average, among the 33 SSA countries in our sample. Experience suggests that private sector involve- ment, both in terms of financing and provision, helps ensure that train- ing content is responsive to labor market needs, that training costs are kept in check, and that networks with prospective employers are developed to facilitate trainees' transition into working life. The simulation model benchmarks the share of private sector enrollment at 40 percent, with the expectation that the current role of the private sector would not only be maintained but indeed grow slightly stronger. Clearly, each country's strategy for attaining this benchmark will require additional discussion to define the most appropriate and effective strategy for implementation. ACCESS TO TERTIARY EDUCATION In higher education, many SSA countries have experienced sharp enroll- ment increases over the past decade with hardly any regulation of the student flow between the two cycles of secondary education and of access to tertiary education. Because aggregate public spending has not kept pace with the massive enrollment growth, conditions of service delivery have significantly deteriorated as stagnant budgets are stretched to cover more and more students. At the same time, large numbers of stu- dents have begun enrolling in private institutions, even though most of them have the option of entering virtually tuition-free public universi- ties. Many of the private institutions are in fact relatively new, having sprung into existence to cater to strong student demand. It is estimated that the private sector currently accounts for at least 18 percent of stu- dents, on average, in the 33 SSA countries. For the same reasons dis- cussed above for TVET, we envisaged a further enlargement of the private sector's role and therefore benchmarked its share of enrollment to rise to 40 percent by 2020. This increase is plausible in light of current devel- opments in tertiary education and the success of many institutions in tapping private contributions as a source of income for their operations. NOTES 1. The simulations use 14 percent for countries with GDP per capita less than $300, 16 percent when the GDP per capita is in the $300­600 range, and 18 percent for countries with GDP per capita above $600. 50 · Developing Post-Primary Education in Sub-Saharan Africa 2. These shares are approximate and are likely to be underestimated, in part because the coverage of official statistics tends to be more complete for institutions in the public sector. Data are particularly sparse for private sector institutions offering technical education and training and post-secondary education. CHAPTER 8 Simulation Results he simulation assumptions discussed in Chapters 5 to 7 regarding T coverage, service delivery arrangements, and financing options, as well as the introduction of flexibility in setting country-specific benchmarks in light of per capita gross domestic product (GDP) levels and other conditions, have explicit implications for cost and financing gaps. These implications are presented below to illustrate potential com- binations of strategic policy options for holistic development of education in Sub-Saharan Africa (SSA). This indispensable exercise is not a substitute for policy making but a tool that provides policy makers with critical information to evaluate the range of possibilities and consider the trade- offs that might be required to achieve their countries' education goals. An important feature of the simulations is that, although they focus on the implications of policies in post-primary education, the scope of the model is in fact sectorwide, a feature that helps put the policies in proper context. We focus here on the aggregate results for all 33 SSA countries in the sample, which are a summation of country-specific computations (see the results reported by country in appendix E). Finally, to draw atten- tion to the policy aspects of the exercise, although the simulations gener- ate many results, only key results captured by the following variables are reported below in this chapter: · the projected aggregate costs in 2020 to attain enrollment goals in pri- mary, secondary, and tertiary education while diversifying the options for managing student flow · the financing gap between the projected costs and what SSA countries might reasonably be able to mobilize from their own resources · the implied dependency on external resources to attain the assumed goals given the policies in service delivery and domestic financing. 51 52 · Developing Post-Primary Education in Sub-Saharan Africa A RECAPITULATION OF THE KEY ASSUMPTIONS A summary of key assumptions discussed earlier follows. The first set per- tains to policies on education coverage. Five enrollment scenarios (ES) were explored, the first two distinctively different from the next three in terms of student flow management. In particular, the first two scenarios (ES-1 and ES-2) envision progress toward universal or nearly universal primary completion by 2020 and wide coverage in lower secondary cycle, followed by transition into tertiary education at current rates of transition from the previous cycle. In addition, the scenarios envision some diversifi- cation of educational options in the form of school-to-work life skills training and formal technical and vocational education and training. The other set of three enrollment scenarios (ES-3, ES-4, and ES-5) envision similar goals for coverage in primary and lower secondary education, but explicitly fix the size of enrollments in tertiary education (and hence of upper secondary education) in relation to expected employment oppor- tunities in the formal sector of the economy. Because this sector is expected to remain relatively small in the foreseeable future, coverage at the upper levels of the system is expected also to remain limited. At the same time, the scenarios envision greater provision of school-to-work transition options to relieve the pressure on the formal education system, particularly at the upper secondary and tertiary levels. The last option con- sidered is to combine the primary and lower secondary cycle into a basic cycle, which is relevant mainly because it reflects an organizational strat- egy to manage the cost of service delivery. The second set of assumptions relates to policies about service delivery, which, as explained above, affect the cost of service delivery. In primary education, the same benchmarks as in the Education for All­Fast Track Ini- tiative (EFA-FTI) framework were used for the pupil-teacher ratios, average teacher salaries, and spending on non-teacher inputs. The benchmarks on teacher salaries were adjusted upward for the poorest countries, as explained earlier. In secondary education, two sets of benchmarks were used, one to mirror relatively "generous" (that is, more expensive) service delivery conditions in terms of class size, teacher salaries, teaching loads, instruction time, and spending on non-teacher inputs), and the other to mirror relatively "spartan" (that is, more economical) conditions. The benchmarks for life skills training, technical and vocational education and training (TVET), and tertiary education are specified simply in terms of spending per student, and these are linked proportionally to the levels at the secondary level. In addition, for tertiary education, the model dif- ferentiates spending per student by field of study and computes average Simulation Results · 53 spending according to assumptions about the corresponding distribution of enrollments. It also recalls that the model computes the capital costs of classroom construction for primary and secondary schools. Finally, the third set of assumptions relates to resource mobilization, both in terms of government allocation of public budgets and the distri- bution between publicly funded and privately funded enrollments. For resource mobilization, the benchmarks are the same as those in the EFA- FTI framework, while those for the public-private shares are chosen to reflect concerns about equity and efficiency. Thus, the values for the share of privately funded services are benchmarked at a modest level of about 10 percent for primary and lower secondary education and related school- to-work transition programs, but at the much higher rate of about 40 per- cent for study beyond lower secondary education. SIMULATIONS OF AGGREGATE COSTS AND FUNDING GAPS IN 2020 In the following, simulation results are shown for three different rates of growth in the per capita GDP between the base year and 2020. This vari- able affects the results in five ways. First, it influences the fiscal resources available for education. Second, because teachers' salaries are expressed as multiples of per capita GDP, the variable also affects the cost of service delivery. Third, for the poorest countries, the benchmarks for teachers' salaries have been adjusted to take into account the countries' current and projected levels of per capita GDP. Fourth, for some simulation sce- narios, the enrollments in upper secondary and tertiary education are linked to the per capita GDP. Fifth--and perhaps most important-- higher rates of per capita GDP growth imply higher absolute values in the level of spending per student and therefore larger aggregate costs expressed in absolute terms. For the purpose of this exercise, the assumed rates are: (1) 0 percent a year, which implies real GDP growing at the same rate as the population; (2) 2 percent a year, which implies that it would grow faster than the population by 2 percentage points; and (3) 4 percent a year, which corresponds more or less to the experience in SSA in the last decade, with real GDP growth averaging about 6.5 percent a year and population growth averaging 2.5 percent a year. Table 8.1 summarizes the projected aggregate costs of service delivery in 2020, financing gaps, and dependency on external support under vari- ous policy scenarios, on the assumption that the real per capita GDP grows at a real rate of 2 percent a year. The results for higher and lower rates of growth are shown in Appendix D. For primary education, the financing 54 Table 8.1 Aggregate Annual Public Spending on Education and Financing Gap in 2020 for 33 Low-Income SSA Countries 2005 US$ billion unless otherwise indicated Aggregate Annual recurrent financing gap annual public in post-primary education under Total financing gap as education Annual recurrent Annual Policy scenario 3 scenarios for domestic share of total public education spending financing gap in capital resource mobilization spending (%) Per-student spending (recurrent primary or basic financing Coverage at post-primary levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 51.6 3.1 29.1 26.5 23.9 2.8 68 63 58 "Spartan" 42.6 20.4 17.8 15.2 2.6 61 55 49 ES-2 "Generous" 44.8 22.7 20.0 17.4 2.5 63 57 51 "Spartan" 37.9 16.0 13.4 10.8 2.3 56 49 42 ES-3 "Generous" 36.7 14.7 12.1 9.5 2.3 55 48 40 "Spartan" 31.2 9.5 6.8 4.2 2.1 47 38 30 ES-4 "Generous" 31.5 9.7 7.2 5.0 2.1 47 39 31 "Spartan" 27.2 5.6 3.2 1.1 1.9 39 29 20 ES-5 "Generous" 27.6 5.2 3.4 1.3 0.4 2.6 40 31 21 "Spartan" 26.9 2.8 0.8 0.3 2.6 38 29 19 Notes: Rate of real growth of per capita GDP: 2 percent per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 33% or less. Simulation Results · 55 gap is, as expected, the same for different combinations of policies for the post-basic levels regarding coverage or enrollment (ES-1 to ES-4) and serv- ice delivery: $3.1 billion a year in 2020 for the 33 countries in the sample. With a structural change to attain universal coverage of nine years of basic education (ES-5), the gap would increase to $5.2 billion a year--an increase that would be compensated for by smaller gaps at the post-basic levels, as explained below. At the post-primary levels, the funding gaps vary dramatically depend- ing on the choice of policies. Gaps occur when the share of education in overall government spending is pegged at 20 percent (and the per capita GDP is growing at 2 percent a year). With the most expansive policy for cov- erage (ES-1) and the more generous assumptions for service delivery, the aggregate gap in recurrent funding in 2020 is estimated to be $29.1 billion a year for post-primary education and $3.1 billion a year for primary edu- cation (for a systemwide total of $32.2 billion a year). In addition, the 33 SSA countries would have a funding gap for capital investments in pri- mary and secondary education of about $2.8 billion a year. Filling the gaps with external funds implies that these countries would depend on external funding sources for 68 percent of their expenditure on educa- tion. Both the absolute amounts and the rate of dependency are very high, possibly unrealistically so. More spartan choices in service delivery would reduce the post-primary recurrent funding gap significantly--to $20.4 billion a year--but its size and its implication for aid dependency (61 percent) remain challenging and very likely beyond thresholds that are feasible or even desirable. With slightly less ambitious goals for the transition between primary and lower secondary education (that is, 80 percent instead of 100 percent), as captured by scenario ES-2, the recurrent funding gap in 2020 with gen- erous policies for service delivery would amount to $22.7 billion a year for post-primary education and $3.1 billion a year for primary education, for a systemwide total of $25.8 billion a year, and the gap in capital spending on primary and secondary education would be $2.5 billion a year. Correspondingly, the level of aid dependency would fall only slightly to 63 percent (and with spartan service delivery policies to 56 per- cent). These figures are certainly lower than those for the scenario ES-1, but they are not fundamentally distinct in terms of the fiscal sustainability of the policies. The next three enrollment scenarios--ES-3 to ES-5--mark a clear-cut departure from the previous two in that they reflect much more selective policies for coverage in upper secondary and tertiary education. Under scenario ES-3, the size of enrollments in tertiary education is benchmarked 56 · Developing Post-Primary Education in Sub-Saharan Africa to tolerate a graduate unemployment rate of 25 percent, while the transi- tion rate between primary and lower secondary education is retained at 80 percent. When service delivery policies are generous, the recurrent financ- ing gap in 2020 would be $14.7 billion a year for post-primary education and $3.1 billion a year for primary education, and the gap in capital funding is projected at $2.3 billion a year. The corresponding dependency ratio would then be 55 percent. With spartan service delivery policies, the post-primary recurrent gap would fall to $9.5 billion a year and the dependency rate to 47 percent. Under scenario ES-4, the assumed coverage in upper secondary and tertiary education is lowered to be consistent with a graduate unemploy- ment rate of 20 percent instead of 25 percent, and the transition rate between primary and lower secondary education is reduced to 65 percent. The recurrent funding gap would then be $9.7 billion a year for post- primary education under generous service delivery policies and $5.6 billion a year under the spartan policies. The corresponding aid dependency rate would be 47 and 39 percent, respectively. Finally, under scenario ES-5--which involves a structural change to provide 9 years of basic schooling and enrollments in upper secondary and tertiary education that are consistent with a graduate unemployment rate of 20 percent--the recurrent funding gap in post-basic education would be $3.4 billion a year with generous service delivery policies and $2.8 billion a year with spartan policies. As indicated above, this sharp fall compensates for the increased gap for basic education--$5.2 billion com- pared with $3.1 billion in the previous four enrollment scenarios. The gap in capital funding for basic and secondary education would be $2.6 billion a year, and the dependency on external support would fall to 40 percent with the generous service delivery policies and 38 percent with spartan policies. Considering all the simulations obtained thus far, it is clear that with an education sector share of 20 percent in overall government recurrent spending and the assumed levels of private financing (which are relatively high, particularly in upper secondary and tertiary education), the implied funding gaps are massive under all ten combinations of policies in cover- age and in service delivery. The prospects for mobilizing the required vol- ume of external funding appear, at best, dim. If we accept as an upper limit that external funding exceeding 50 percent of what countries them- selves spend would be untenable, then the maximum dependency ratio would be 33 percent.1 By this somewhat normative but reasonable meas- ure, none of the simulation scenarios considered thus far would meet the mark of acceptability. In view of this assessment, two further lines of Simulation Results · 57 exploration seem useful to consider in the simulation exercise: increase the budget allocation to the education sector and reduce the goal level of coverage or of the resource-intensity of service delivery. Raising the budget share of education to 23 percent (ES-2) does indeed reduce the funding gap, as the relevant columns in table 8.1 show. However, accepting a cutoff dependency rate of 33 percent means that only the three cells shaded in grey in the table meet this criterion. They corre- spond to the following combinations of coverage and service delivery: (1) nine years of basic schooling (ES-5), with generous or spartan service delivery policies and (2) relatively tight controls on enrollments in upper secondary and tertiary education (ES-4) combined with spartan service delivery options. With a budget share of 26 percent, two additional policy combinations meet the criterion for acceptability: (1) ES-4 com- bined with generous service delivery policies and (2) slightly more ambitious goals for coverage in post-primary education (ES-3), but only with spartan choices for service delivery. Among the total of eight scenarios that imply a dependency rate no higher than 33 percent, all involve a model of discontinuity in student flow management in which enrollments in upper secondary and tertiary education are aligned to employment prospects for graduates. Even so, the dependency rate for these scenarios ranges from 19 percent to 31 percent--probably man- ageable but nonetheless still high. The above figures assume that the per capita GDP would grow at an average of 2 percent a year between the base year and 2020. The results corresponding to per capita GDP growth rates of 0 and 4 percent a year are shown in appendix D. These results suggest that, in terms of depend- ency rates, using the threshold of 33 percent would confine the acceptable policy options to the same ones as those discussed above. This means that to achieve holistic development of education, it would be difficult for SSA countries to avoid the following: · issues relating to budget priority for education · proactive management of student flow, particularly in upper second- ary and tertiary education, including options for diversifying the path- ways for students to exit the system at lower levels into the workforce · service delivery arrangements that would help keep a lid on costs. However, the pace of growth in the per capita GDP affects the absolute size of the funding gaps (in international or national currency). As explained earlier, with a slower growth in per capita GDP than the 2 percent assumed thus far, these gaps would be smaller than the ones shown in the table 8.1; conversely, with a faster rate of growth, the gaps 58 · Developing Post-Primary Education in Sub-Saharan Africa would be bigger. The funding gaps, while sizable under all scenarios, do vary significantly according to the projected rate of growth in the per capita GDP. Consider, for example, the scenario involving the ES-3 option for coverage, the spartan option for service delivery, and an allo- cation of 20 percent of government budget for education. The recurrent financing gap in 2020 would be $6.3 billion a year for post-primary edu- cation if the per capita GDP stagnated between the base year and 2020, $9.5 billion a year if it grew at 2 percent a year, and $16.4 billion a year if it grew at 4 percent over the period. Clearly, while sustained growth in the aid-receiving countries is clearly the ultimate objective, the simu- lation results confirm that, as countries grow richer, the cost of support- ing their education programs, and hence the effort to mobilize external resources, will grow correspondingly higher. DIVERSITY ACROSS THE SAMPLE OF 33 SSA COUNTRIES The analysis above pertains to the aggregate picture for all 33 countries in our sample, based on the set of country-specific simulations shown in appendix E. This approach establishes general guidelines and overall orders of magnitude of financing gaps and dependency rates. It clearly needs to be supplemented with more detailed analysis in which the simulation assumptions are tailored to the specific conditions each country faces, especially because these conditions tend to vary widely across countries. Even without such second-stage analysis, the findings obtained thus far can help frame policy dialogue on the development of education in SSA. Consider first the results for scenarios where the per capita GDP is growing at 2 percent a year. In this setting, only 9 of the 33 countries can envisage a dependency rate below 35 percent with policies in coverage cor- responding to ES-1 and ES-2.2 These countries are Cameroon, Republic of Congo, The Gambia, Ghana, Guinea, Kenya, Lesotho, Mauritania, and Nigeria. Five of these countries--Republic of Congo, The Gambia, Kenya, Lesotho, and Mauritania--are projected to have the lowest dependency ratios. The remaining four--Cameroon, Ghana, Guinea, and Nigeria-- can maintain their dependency ratio below 35 percent while pursuing goals in coverage corresponding to ES-2 only if they also adopt spartan benchmarks for service delivery and allocate 26 percent of total govern- ment spending to the education sector. For 8 of the 33 countries, none of the combinations of policies for coverage, for service delivery, and for domestic resource mobilization leads to a level of dependency on external financing that falls at or below Simulation Results · 59 35 percent; this result holds true for all three rates of growth in the per capita GDP. These countries are Burundi, the Democratic Republic of Congo, Eritrea, Ethiopia, Guinea-Bissau, Malawi, Niger, and Sierra Leone. The results for Burkina Faso and Chad are very close to those of the pre- vious eight countries. There is only one scenario leading to a level of dependency on external financing that falls at or below 35 percent; it corresponds to ES-5 combined with spartan options for service delivery and a 26 percent share of public spending allocated to education. For these ten countries, new simulation scenarios would be helpful to clarify the policy options that might warrant consideration aside from those included here. As the spartan options for service delivery are already quite basic, it may not be desirable, as far as ensuring minimum condi- tions in schools, to depart too far from the benchmarks used in this exer- cise. Accepting this argument leaves room for maneuvering only in terms of reducing the ambition of goals for coverage by 2020 or pushing back the horizon for achieving these goals or of raising the share of public spending allocated to the education sector beyond the 26 percent prior- ity assumed in the most optimistic scenario. The prospect for such an increase poses very difficult intersectoral tradeoffs, however, since gov- ernments must also consider the needs of other sectors, such as health and agriculture. For the remaining countries, the simulations present a more mixed picture. With a 4 percent per capita GDP growth between the base year and 2020, the rate of dependency on external funding would be man- ageable in Benin, Côte d'Ivoire, Madagascar, Mozambique, Senegal, Sudan, Tanzania, Togo, and Zimbabwe if the countries choose policies for coverage that correspond to ES-4 combined with spartan policies for service delivery and if they allocate a share of 23 percent of overall government spending to the sector. These policy options in Central African Republic, Mali, Rwanda, Uganda, and Zambia would also imply manageable dependency rates if the budget share of educa- tion rose to 26 percent. While the foregoing results provide some insight into the nature of the policy challenges facing SSA countries, they are only a first step in policy development. In particular, some of the benchmarks, particularly those that define how educational services are organized, how much teachers are paid, and how much is spent on non-teaching inputs, cannot be detached from each country's historical, social, and political contexts. For example, implementing a move toward the benchmarks represented by spartan choices in service delivery may entail unrealistic decreases in the level of teacher salaries in some countries; and changing the teaching 60 · Developing Post-Primary Education in Sub-Saharan Africa loads and instructional time for students may require reforms in curriculum and teacher training practices that may take time and effort to organize.3 In short, therefore, revisions and adjustments will be needed to make the simulation exercise pertinent for policy development in each of the 33 countries. The results, and more importantly, the simulation tool, rep- resent a public good that is available to countries and to the development community, one that can be used as an aid for decision making and policy dialogue. While the tool can generate any number of policy sce- narios, the challenge is to identify the combination of options that are sustainable in the medium term in terms of the requirements for national resources and external aid. NOTES 1. As shown in the next chapter, a slightly higher cutoff of 35 percent has been used to evaluate the acceptability of simulation scenarios in individual countries; see also appendix E. 2. At present, external aid in SSA is, on average, about 16 percent of total pub- lic spending on education, ranging from less than 10 percent in such countries as the Republic of Congo, Lesotho, and Nigeria, to around 60 percent in Burundi, Mozambique, or Tanzania. Excluding Nigeria (a country with a large population and one that is not as dependent on aid as other SSA countries), the average for the remaining 32 other countries rises to 35 percent. It is indeed difficult to estab- lish an undisputable benchmark to decide if this level of external funding is high or low and to evaluate whether it can (must) be increased, and if so, to what level. For the purpose of this report, we use a cutoff rate of 35 percent. 3. For example, a salary level representing about 5 times the GDP per capita is anticipated in lower secondary education while it represents about 9 times the GDP per capita in Togo and Mozambique at present. CHAPTER 9 Conclusions he development of post-primary education in Sub-Saharan Africa (SSA) T is an emerging challenge in light of the progress that many of these countries have made in advancing the goal of universal primary school completion. For the purpose of this report, we constructed a sec- torwide simulation model to characterize key policy scenarios and their implications regarding future enrollments and funding requirements. The model takes a sectorwide approach even though the focus is on poli- cies at the post-primary levels, and it examines the prospects for funding under alternative scenarios for income growth. The results bring to light key issues for policy dialogue, not only within each country, but also between the 33 SSA countries included in this study and their develop- ment partners. At heart, the issues pertain to fundamental questions: How much domestic resources can countries themselves reasonably mobilize for education? How much foreign aid can the donor nations real- istically contribute for the sector? What are acceptable levels of depend- ency on external aid in recipient and donor countries alike? What reforms in education sector policies--in terms of the pattern of coverage by level and type of education and of service delivery arrangements--will be required to put the education system on a sustainable path of development to meet national development goals? Although a technical study such as the present one cannot provide firm answers to these questions, its value lies in setting out the policy scenarios and evaluating their logistical and financial implications. In today's climate of global economic slowdown, such analysis has become especially pertinent. Seven important conclusions emerge from the simulations, assuming that a dependency rate of 35 percent represents a threshold that few SSA 61 62 · Developing Post-Primary Education in Sub-Saharan Africa countries and their donor partners would find acceptable to surpass. These conclusions are as follows: · Policies that envision a significant expansion of coverage in the first cycle of secondary education combined with a model of continuity in student flow at subsequent levels generally fail the test of financial sus- tainability. Passing this test will typically require policies that regulate student flow more actively, so as to keep graduate unemployment rates at acceptable levels--a desirable goal in and of itself. Such policies include efforts to diversify curriculum options at the end of the first cycle of secondary education and also at the end of primary schooling as a temporary measure as long as lower secondary schooling is still not universal. · Between the generous and spartan options for service delivery, the lat- ter will probably be the more relevant one in most SSA countries. Although the practical feasibility of policies will depend on country context and implementation constraints, cost-conscious choices will often be unavoidable for countries to put the development of post- primary education on a financially sustainable path. · To avoid an untenable dependency on external aid to finance educa- tional development, SSA countries must typically exceed the 20 percent reference benchmark for the share of government spending allocated to the education sector. A minimum share of 23 percent seems relevant in most of the countries. · The option of restructuring primary and lower secondary education into a single nine-year cycle of basic education is attractive in that aid dependency is lowest for this scenario, particularly when the policy is combined with spartan policies for service delivery. The advantage is not dramatic, however, because its main impact is to shift the dis- tribution of the overall funding gap between basic education and post-basic education rather than reduce the absolute size of the gap itself. One drawback is that structural changes are disruptive and may cause misalignment between national and international quali- fication standards. · While the 33 SSA countries in the sample share common goals and challenges, their specific circumstances in terms of initial conditions and political context are extremely varied. In light of the diversity across countries, the results of this simulation exercise are indeed indicative and must therefore be complemented by additional country-specific analysis. Such analysis can be prepared using either the tool developed for this study or other similar models. Conclusions · 63 · Faster rates of growth in the per capita incomes of SSA countries can be expected to facilitate the development of their education systems, in part by increasing the affordability of educational inputs that are purchased with domestic funds. Yet with income growth, the cost of services in absolute terms also rises. As a result, the absolute size of funding gaps also tends to increase even as the dependency rate falls. · Finally, in all the policy scenarios considered in this paper, the devel- opment of post-primary development in the 33 SSA countries will require sizable volumes of external aid to fill funding gaps. The absolute amounts involved are as relevant as the level of aid depend- ency that they imply. Both issues belong on the agenda for policy dia- logue within countries as well as between SSA countries and their development partners. APPENDIX A Relation Between Teacher Salaries and Country Income Level 1. As economies grow, average teacher salaries increase in absolute dollar terms, but tend to decline in relation to countries' gross domestic product (GDP) per capita (for an analysis of teacher salary trends in Sub-Saharan Africa (SSA) countries, see Mingat 2004). Based on a cross-country data set for Sub-Saharan Africa, the authors have esti- mated a statistical relation between primary teacher salaries, expressed in multiples of GDP per capita, and GDP per capita itself. The constant in the regression relation was subsequently adjusted so that it gives a salary of 3.5 times GDP per capita (indicative benchmark of the EFA- FTI) for a country with a GDP per capita of $350 (close to the average for SSA). The final relation is: Average primary teacher salary (in multiples of GDP per capita) = 4.2698 - 0.0022 ×; GDP per capita (in 2003 prices) 2. This relation is used in the simulation model to project teacher salaries for 2020 as a function of the level of income expected in that year. The relation is only used to calculate teacher salaries for countries with a GDP per capita less than $600, however. For countries with incomes above this benchmark, the simulations use a primary teacher salary of 3.0 times the GDP per capita for all (see table on page 64). As shown in the table, although salaries decline with higher per capita incomes, when measured relative to per capita GDP, they still grow when meas- ured in constant U.S. dollars. 65 66 · Developing Post-Primary Education in Sub-Saharan Africa GDP per capita in 2020 (2003 prices) 100 150 200 350 500 600 Primary teacher salary in 2020: in multiples of GDP per capita 4.05 3.94 3.83 3.5 3.17 3.00 annual salary in US$ (2003 prices) 405 591 766 1,225 1,585 1,800 Note: Appendix E lists the teacher salaries used in the simulations for each of the 33 countries. 3. A similar adjustment is made for secondary school teacher salaries by raising (or lowering) the average salary by the same proportion above (below) the average benchmark as for primary education. APPENDIX B Basic Data for 33 SSA Countries 67 68 Table B.1 Quantitative Coverage of Education and Growth in Primary School Leavers Primary completers Primary Lower secondary Upper secondary TVET Tertiary Number of students completing the cycle Ratio Primary Transition Transition Share of Students completion rate P-S1 Gross intake rate S1-S2 Gross intake secondary per 100 000 (thousands) 2020/base Country (base year) rate (%) (%) rate (%) (%) rate (%) enrollm. (%) inhabitants Base year 2020 year Benin (2004) 49.0 73.0 36.0 58.0 14.0 8.6 521 101.6 302.9 3.0 Burkina Faso (2004) 31.0 58.0 18.0 44.0 4.2 8.9 201 106.1 513.8 4.8 Burundi (2004) 33.1 52.2 17.3 44.3 3.5 9.7 228 65.5 318.1 4.9 Cameroon (2003) 57.6 56.3 32.4 64.5 18.7 23.4 494 231.3 438.6 1.9 Central African Rep. (2004) 30.0 56.4 16.8 78.9 5.4 6.4 233 30.8 116.4 3.8 Chad (2004) 37.8 72.0 26.0 80.0 10.0 1.6 117 91.0 384.5 4.2 Congo, Dem. Rep. (2005) 49.6 39.0 21.0 88.0 15.0 46.1 235 727.9 2,354.2 3.2 Congo, Rep. (2005) 72.0 79.2 57.1 37.1 16.2 24.3 337 74.4 165.5 2.2 Côte d'Ivoire (2000) 49.0 63.1 30.9 45.9 12.3 8.4 629 243.1 514.3 2.1 Eritrea (2005) 51.3 74.5 38.2 68.8 22.3 0.1 150 58.4 162.8 2.8 Ethiopia (2002) 53.0 78.7 42.0 77.5 14.5 1.6 138 1,050.4 2,640.5 2.5 Gambia, The (2001) 60.0 74.1 44.8 56.5 21.0 0.0 140 19.2 44.0 2.3 Ghana (2001) 66.0 98.8 65.0 40.0 22.4 1.6 354 340.3 589.1 1.7 Guinea (2005) 54.0 75.0 38.0 93.0 19.0 3.5 257 128.4 316.9 2.5 Guinea-Bissau (2002) 40.0 83.7 33.8 80.9 16.9 3.2 79 14.4 64.5 4.5 Kenya (2005) 70.0 47.0 30.0 16.8 279 584.6 1,177.0 2.0 Lesotho (2005) 69.4 76.0 51.0 84.0 24.0 1.4 380 33.4 38.4 1.1 Madagascar (2003) 37.0 65.0 28.5 52.4 7.7 10.3 205 168.2 623.2 3.7 Malawi (2002) 73.0 30.1 22.0 67.9 14.3 1.7 32 201.0 428.7 2.1 Mali (2004) 41.6 80.5 33.4 40.2 8.4 10.2 284 143.8 533.4 3.7 Mauritania (2004) 51.0 61.7 29.0 94.7 18.0 3.2 295 36.0 106.2 2.9 Mozambique (2001) 23.0 54.5 12.4 43.9 2.9 32.6 83 101.6 598.2 5.9 Niger (2002) 21.4 66.0 14.1 40.0 2.2 3.0 50 69.0 573.5 8.3 Nigeria (2005) 76.3 52.4 40.0 83.0 28.3 0.8 1,188 2,595.4 4,135.0 1.6 Rwanda (2003) 46.0 35.3 16.0 77.5 10.7 10.6 252 111.6 295.2 2.6 Senegal (2003) 49.0 54.0 27.0 62.9 11.0 6.6 496 143.0 357.0 2.5 Sierra Leone (2004) 55.0 63.0 35.0 39.2 10.0 15.4 269 68.1 183.0 2.7 Sudan (2003) 37.0 ­ ­ 64.0 24.0 11.2 1,052 293.4 974.5 3.3 Tanzania (2002) 60.0 28.0 16.0 30.0 2.0 0.3 84 555.4 1,096.0 2.0 Togo (2005) 73.0 80.0 58.4 40.8 13.9 4.1 485 113.7 201.5 1.8 Uganda (2002) 51.0 44.6 22.6 40.0 7.1 7.1 292 344.4 1,312.7 3.8 Zambia (2005) 73.0 62.0 45.0 41.0 16.0 1.5 218 227.1 367.2 1.6 Zimbabwe (2003) 80.2 70.0 58.0 13.0 6.2 4.8 592 282.3 307.8 1.1 Avg. 33 countries 52.2 63.1 33.1 58.2 13.7 8.8 323 9,354.9 22,234.5 3.0*­ 2.4** Source: Database compiled from Country Status Reports on Education, Public Expenditure Reviews, previous education sector simulation models, and UIS and EdStats databases. Population data used to calculate enrollment ratios are from United Nations. * nonweighted average; ** average weighted by school-age population in the different countries. Dash = data not available. 69 70 Table B.2 Potential Growth in Lower Secondary Enrollment by Country 2020 Base year Base year transition rate maintained 100% transition Lower Lower Enrollm. in Gross intake secondary Gross intake secondary lower Ratio Transition rate rate to lower enrollm. rate to lower enrollm. Ratio 2020/base secondary 2020/base year Country (base year) P-S1 (%) secondary (%) (thousands) secondary (%) (thousands) year enrollm. (thousands) enrollm. Benin (2004) 73.0 36.0 290.7 69.4 811.7 2.8 1,111.9 3.8 Burkina Faso (2004) 58.0 18.0 230.6 55.1 1,078.2 4.7 1,859.0 8.1 Burundi (2004) 52.2 17.3 121.8 49.6 550.9 4.5 1,055.4 8.7 Cameroon (2003) 56.3 32.4 559.8 52.3 944.8 1.7 1,717.7 3.1 Central African Rep. (2004) 56.4 16.8 57.8 53.6 251.0 4.3 445.1 7.7 Chad (2004) 72.0 26.0 211.3 68.4 974.1 4.6 1,352.8 6.4 Congo, Dem. Rep. (2005) 39.0 21.0 615.5 37.1 1,682.5 2.7 4,314.0 7.0 Congo, Rep. (2005) 79.2 57.1 259.0 75.2 466.6 1.8 589.1 2.3 Côte d'Ivoire (2000) 63.1 30.9 643.4 59.9 1,253.0 1.9 1,985.7 3.1 Eritrea (2005) 74.5 38.2 137.1 70.7 344.2 2.5 462.2 3.4 Ethiopia (2002) 78.7 42.0 2,579.2 74.8 7,782.5 3.0 9,888.8 3.8 Gambia, The (2001) 74.1 44.8 38.4 70.4 95.6 2.5 129.0 3.4 Ghana (2001) 98.8 65.0 941.3 93.9 1,703.9 1.8 1,724.6 1.8 Guinea (2005) 75.0 38.0 317.8 71.3 882.2 2.8 1,176.2 3.7 Guinea-Bissau (2002) 83.7 33.8 34.7 79.5 145.9 4.2 174.3 5.0 Kenya (2005) ­ ­ ­ ­ ­ ­ ­ ­ Lesotho (2005) 76.0 51.0 65.4 72.2 87.5 1.3 115.2 1.8 Madagascar (2003) 65.0 28.5 422.2 61.8 1,533.1 3.6 2,358.5 5.6 Malawi (2002) 30.1 22.0 117.4 28.6 249.7 2.1 828.6 7.1 Mali (2004) 80.5 33.4 346.8 76.5 1,186.2 3.4 1,473.6 4.2 Mauritania (2004) 61.7 29.0 72.4 58.6 243.1 3.4 394.1 5.4 Mozambique (2001) 54.5 12.4 154.7 51.8 940.1 6.1 1,724.9 11.1 Niger (2002) 66.0 14.1 147.5 62.7 1,369.6 9.3 2,075.2 14.1 Nigeria (2005) 52.4 40.0 3,706.1 49.8 6,212.1 1.7 11,849.5 3.2 Rwanda (2003) 35.3 16.0 117.3 33.5 289.8 2.5 820.8 7.0 Senegal (2003) 54.0 27.0 279.2 51.3 735.1 2.6 1,361.3 4.9 Sierra Leone (2004) 63.0 35.0 125.1 59.9 320.9 2.6 509.4 4.1 Sudan (2003) ­ ­ ­ ­ ­ ­ ­ ­ Tanzania (2002) 28.0 16.0 453.5 26.6 1,194.1 2.6 4,264.5 9.4 Togo (2005) 80.0 58.4 359.5 76.0 612.2 1.7 765.3 2.1 Uganda (2002) 44.6 22.6 507.7 42.4 2,003.8 3.9 4,492.9 8.8 Zambia (2005) 62.0 45.0 271.4 58.9 436.9 1.6 704.7 2.6 Zimbabwe (2003) 70.0 58.0 724.3 66.5 846.6 1.2 1,209.4 1.7 Average 33 countries 63.1 33.1 14,909.0 59.9 37,227.7 3.5*­2.5** 62,933.9 6.1*­ 4.2** Source: Database compiled from Country Status Reports on Education, Public Expenditure Reviews, previous education sector simulation models, and UIS and EdStats databases. Population data used to calculate enrollment ratios are from United Nations. * nonweighted average; ** average weighted by school-age population in the different countries. Dash = data not available. 71 72 Table B.3 School Organization in Lower Secondary Education, Circa 2005 Lower secondary Average Share of recurrent Per-student teacher salary Students per Students per Teachers per expenditure other spending (in multiples of teacher class class than for teacher (in multiples of Country (base year) GDP/capita) (1) (2) (ratio (2)/(1)) salaries (%) GDP/capita) Benin (2004) 3.0 29.6 52.6 1.8 63.9 0.28 Burkina Faso (2004) 9.3 50.0 75.0 1.5 60.4 0.47 Burundi (2004) 9.3 20.5 42.4 2.1 29.1 0.64 Cameroon (2003) 6.5 31.1 40.3 1.3 34.3 0.32 Central African Rep. (2004) 4.8 56.7 78.0 1.4 44.1 0.15 Chad (2004) 4.4 40.6 59.7 1.5 44.4 0.19 Congo, Dem. Rep. (2005) 2.4 16.8 28.5 1.7 25.9 0.19 Congo, Rep. (2005) 2.3 63.8 86.0 1.3 70.2 0.12 Côte d'Ivoire (2000) 7.7 37.9 64.4 1.7 40.4 0.34 Eritrea (2005) 9.9 52.8 69.5 1.3 25.0 0.25 Ethiopia (2002) 8.1 48.3 67.8 1.4 18.6 0.21 Gambia, The (2001) 6.5 29.7 49.6 1.7 15.5 0.26 Ghana (2001) 3.9 18.5 18.5 1.0 29.7 0.30 Guinea (2005) 2.9 47.6 88.9 1.9 44.0 0.11 Guinea-Bissau (2002) 2.2 23.3 31.3 1.3 32.0 0.14 Kenya (2005) ­ ­ ­ ­ ­ ­ Lesotho (2005) 9.0 26.7 50.0 1.9 27.0 0.46 Madagascar (2003) 4.3 27.2 44.0 1.6 41.7 0.27 Malawi (2002) 7.7 26.3 50.0 1.9 40.0 0.49 Mali (2004) 6.8 46.1 73.0 1.6 44.1 0.26 Mauritania (2004) 3.7 27.3 54.5 2.0 56.0 0.31 Mozambique (2001) 9.4 47.5 50.5 1.1 37.9 0.32 Niger (2002) 8.5 39.7 49.2 1.2 56.3 0.49 Nigeria (2005) 7.2 55.3 61.0 1.1 35.1 0.20 Rwanda (2003) 5.9 28.7 46.6 1.6 59.8 0.51 Senegal (2003) 6.2 39.0 60.7 1.6 27.5 0.22 Sierra Leone (2004) 5.9 29.4 49.0 1.7 30.1 0.29 Sudan (2003) ­ ­ ­ ­ ­ ­ Tanzania (2002) 5.2 22.5 35.5 1.6 47.0 0.44 Togo (2005) 8.7 54.2 87.9 1.6 13.6 0.19 Uganda (2002) 7.4 6.8 15.2 2.2 25.1 1.45 Zambia (2005) 3.7 32.3 42.0 1.3 31.7 0.17 Zimbabwe (2003) 4.1 24.1 36.2 1.5 26.0 0.23 Average 33 countries 6.0 35.5 53.5 1.6 38.0 0.33 Source: Database compiled from Country Status Reports on Education, Public Expenditure Reviews, previous education sector simulation models, and UIS and EdStats databases. Dash = data not available. 73 74 Table B.4 School Organization in Upper Secondary Education, Circa 2005 Upper secondary Share of recurrent Per-student Average teacher Students per Students Teachers expenditure other spending salary (in multiples teacher per class per class than for teacher (in multiples of Country (base year) of GDP/capita) (1) (2) (ratio (2)/(1)) salaries (%) GDP/capita) Benin (2004) 5.2 20.4 46.0 2.3 63.9 0.71 Burkina Faso (2004) 13.0 39.0 52.0 1.3 48.0 0.64 Burundi (2004) 11.0 16.4 32.4 2.0 72.0 2.40 Cameroon (2003) 6.8 29.1 36.4 1.3 36.5 0.37 Central African Rep. (2004) 5.1 37.1 65.8 1.8 44.4 0.25 Chad (2004) 6.8 54.4 59.2 1.1 50.0 0.25 Congo, Dem. Rep. (2005) 2.4 16.8 28.5 1.7 25.9 0.19 Congo, Rep. (2005) 2.7 23.0 44.5 1.9 67.3 0.36 Côte d'Ivoire (2000) 8.9 23.5 39.9 1.7 40.4 0.63 Eritrea (2005) 11.8 45.0 70.3 1.6 25.0 0.35 Ethiopia (2002) 11.9 50.3 81.7 1.6 40.5 0.40 Gambia, The (2001) 6.5 17.3 34.8 2.0 20.6 0.47 Ghana (2001) 5.9 19.1 32.5 1.7 63.7 0.85 Guinea (2005) 2.9 52.3 95.0 1.8 41.0 0.09 Guinea-Bissau (2002) 2.2 23.3 31.3 1.3 32.0 0.14 Kenya (2005) 10.6 20.9 40.0 1.9 25.4 0.68 Lesotho (2005) 9.0 17.4 32.6 1.9 27.0 0.71 Madagascar (2003) 7.7 19.0 38.7 2.0 44.6 0.73 Malawi (2002) 7.7 26.3 50.0 1.9 40.0 0.49 Mali (2004) 8.3 23.1 39.6 1.7 69.3 1.17 Mauritania (2004) 4.8 20.3 42.8 2.1 44.7 0.42 Mozambique (2001) 23.8 26.4 45.7 1.7 34.7 1.38 Niger (2002) 10.2 13.0 33.2 2.6 49.9 1.57 Nigeria (2005) 7.2 44.9 52.9 1.2 35.8 0.25 Rwanda (2003) 6.4 21.8 40.1 1.8 54.0 0.63 Senegal (2003) 7.1 25.7 48.0 1.9 23.4 0.36 Sierra Leone (2004) 5.9 27.0 42.0 1.6 28.2 0.30 Sudan (2003) 3.4 26.7 53.5 2.0 28.6 0.18 Tanzania (2002) 5.2 22.5 35.5 1.6 47.0 0.44 Togo (2005) 9.0 33.3 62.4 1.9 15.9 0.32 Uganda (2002) 7.4 6.8 15.2 2.2 25.1 1.45 Zambia (2005) 4.7 25.3 45.0 1.8 26.7 0.25 Zimbabwe (2003) 4.1 24.1 41.0 1.7 26.0 0.23 Average 33 countries 7.4 27.1 45.7 1.8 39.9 0.60 Source: Database compiled from Country Status Reports on Education, Public Expenditure Reviews, previous education sector simulation models, and UIS and EdStats databases. 75 76 Table B.5 School Organization in Primary and Tertiary Education, Circa 2005 Primary Tertiary Share of recurrent Per-student Average teacher expenditure other Per-student spending spending salary (in multiples Students than for teacher (in multiples of (in multiples of Country (base year) of GDP/capita) per teacher salaries (%) GDP/capita) GDP/capita) Benin (2004) 4.2 52.0 39.5 0.11 1.6 Burkina Faso (2004) 6.4 52.8 33.8 0.18 2.4 Burundi (2004) 6.8 51.7 13.2 0.15 8.3 Cameroon (2003) 3.9 63.7 30.8 0.07 0.8 Central African Rep. (2004) 7.0 91.6 34.3 0.07 2.4 Chad (2004) 5.4 70.3 37.8 0.05 3.3 Congo, Dem. Rep. (2005) 2.2 37.7 26.0 0.05 2.2 Congo, Rep. (2005) 1.8 75.7 68.0 0.05 2.0 Côte d'Ivoire (2000) 4.8 42.6 25.0 0.15 1.1 Eritrea (2005) 3.9 43.7 25.3 0.12 4.3 Ethiopia (2002) 6.8 73.9 16.0 0.11 11.2 Gambia, The (2001) 4.5 36.5 21.8 0.16 3.8 Ghana (2001) 3.9 32.9 28.1 0.16 3.7 Guinea (2005) 1.7 51.3 44.2 0.06 1.5 Guinea-Bissau (2002) 1.9 37.5 31.0 0.07 1.2 Kenya (2005) 5.3 39.7 17.9 0.16 2.6 Lesotho (2005) 4.4 46.0 46.6 0.18 7.9 Madagascar (2003) 4.4 57.7 42.8 0.09 2.1 Malawi (2002) 4.2 58.4 15.0 0.08 14.9 Mali (2004) 6.0 63.5 37.8 0.11 1.9 Mauritania (2004) 3.3 44.2 31.8 0.11 0.9 Mozambique (2001) 3.9 55.0 22.6 0.09 6.6 Niger (2002) 5.5 42.9 35.0 0.20 5.7 Nigeria (2005) 4.9 51.7 34.2 0.14 1.1 Rwanda (2003) 3.9 60.3 22.2 0.08 7.9 Senegal (2003) 4.6 50.8 37.6 0.15 2.8 Sierra Leone (2004) 4.2 61.0 30.9 0.09 3.4 Sudan (2003) 2.2 36.0 22.5 0.08 1.1 Tanzania (2002) 3.8 46.2 34.0 0.12 5.3 Togo (2005) 6.2 33.6 12.8 0.10 1.3 Uganda (2002) 3.2 56.1 31.7 0.08 1.9 Zambia (2005) 3.1 57.2 28.2 0.07 2.9 Zimbabwe (2003) 4.1 39.0 21.0 0.13 2.3 Average 33 countries 4.3 51.9 30.3 0.11 3.7 Source: Database compiled from Country Status Reports on Education, Public Expenditure Reviews, previous education sector simulation models, and UIS and EdStats databases. 77 78 Table B.6 Population Growth Rates Used in Simulations Average annual growth in primary school-age population (age 6­11) Average annual growth rate Country (base year) 2000­2005 2005­2010 2010­2015 2015­2020 Average 2000­2020 in total population 2000­2020 Benin (2004) 2.6 2.6 2.6 2.1 2.5 2.9 Burkina Faso (2004) 2.6 2.6 2.9 2.6 2.7 3.0 Burundi (2004) 0.8 2.4 4.9 4.6 3.2 3.2 Cameroon (2003) 0.9 0.8 0.6 0.3 0.7 1.6 Central African Rep. (2004) 1.5 0.8 0.8 1.0 1.0 1.4 Chad (2004) 3.5 2.7 3.2 3.3 3.2 3.0 Congo, Dem. Rep. (2005) 2.6 3.4 3.6 3.0 3.2 3.0 Congo, Rep. (2005) 3.4 3.3 3.3 3.2 3.3 3.1 Côte d'Ivoire (2000) 0.9 0.9 0.9 0.8 0.9 1.7 Eritrea (2005) 4.3 3.1 2.3 1.6 2.8 3.1 Ethiopia (2002) 2.1 1.7 1.7 1.6 1.8 2.3 Gambia, The (2001) 2.7 2.2 1.1 0.5 1.6 2.3 Ghana (2001) 0.8 1.0 0.9 0.6 0.8 1.9 Guinea (2005) 2.0 2.0 2.3 2.0 2.1 2.3 Guinea-Bissau (2002) 3.5 3.2 3.4 3.0 3.3 3.0 Kenya (2005) 1.0 2.8 3.4 1.9 2.3 2.4 Lesotho (2005) ­1.4 ­1.1 ­0.5 0.0 ­0.8 ­0.2 Madagascar (2003) 3.1 2.1 1.5 1.4 2.0 2.5 Malawi (2002) 3.9 1.9 1.3 1.7 2.2 2.2 Mali (2004) 3.3 2.8 2.7 2.6 2.8 3.0 Mauritania (2004) 3.0 3.1 2.4 1.5 2.5 2.7 Mozambique (2001) 2.2 1.5 1.2 1.0 1.4 1.8 Niger (2002) 4.1 3.6 2.8 2.8 3.3 3.3 Nigeria (2005) 1.8 1.6 1.4 1.1 1.5 2.0 Rwanda (2003) 0.1 1.7 2.3 1.8 1.5 2.2 Senegal (2003) 1.4 1.6 1.4 0.9 1.3 2.2 Sierra Leone (2004) 4.2 2.4 2.5 2.2 2.8 2.7 Sudan (2003) 1.6 1.6 0.8 0.3 1.1 1.9 Tanzania (2002) 1.4 1.2 0.8 0.6 1.0 1.8 Togo (2005) 2.3 2.2 1.7 1.1 1.8 2.5 Uganda (2002) 3.4 3.7 4.3 4.1 3.9 3.7 Zambia (2005) 1.4 1.3 1.4 1.5 1.4 1.7 Zimbabwe (2003) ­0.9 ­0.9 0.1 0.5 ­0.3 0.6 Source: United Nations population projections, 2006 revision. 79 APPENDIX C Schematic Summary of Scenarios 81 82 Table C.1 Options in Simulation Model for Coverage, Level of Per-Student Spending, and Share of Private Financing in 2020 Lower secondary Upper secondary Share of Share of lower primary school sec. school Tertiary leavers not leavers not Unemployment enrolled in Upper sec. enrolled in cutoff rate Primary lower sec., enrollm. upper sec., used to Share Share Primary Transition Share who enter in multiples Share who enter determine enrolled enrolled in Coverage Duration completion rate enrolled school-to-work of tertiary enrolled school-to-work tertiary in sciences professional options (years) rate (%) P-S1 (%) in TVET (%) transition program enrollm. in TVET (%) transition program enrollment (%) (%) ES-1 95 100 10 15 50 ES-2 95 80 15 50 15 50 ES-3 95 80 15 50 2.5 25 100 25 15 25 ES-4 95 65 20 50 2.5 15 50 20 15 25 ES-5 9 95/80a 2.5 15 50 20 15 25 Lower secondary Upper secondary Tertiary School-to- School- work TVET to-work Humanities General TVET transition General upper transition & social Primary lower sec. lower sec. program upper sec. sec. program sciences Sciences Professional Level of per-student spending in % of per capita GDPb "Generous" 35.4 90 50 60.9 140 100 150 240 290 "Spartan" 13/15c 17.7 90 50 32.2 140 100 150 240 290 Share of enrollment in 10 10 10 0 40 40 0 40 40 40 privately-financed schools Note: Shaded cell indicates no change in the indicator between the base year and 2020. a. The first number indicates the completion rate of the first 6 years (or whatever the duration of the primary cycle) and the second the completion rate of the 9-year cycle. b. This parameter describes the "basic" level of per-student spending before any adjustment for level of wealth, geographic dimension, or stimulation of demand. Appendix E provides the country-specific values for per-student spending used in the simulations. c. The first number indicates the per-student spending for the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number reflects the per- student spending for the 9-year cycle. Unit cost of classroom construction (2005 US$) Primary 12,000 Lower secondary 16,000 Upper secondary 16,000 83 APPENDIX D Aggregate Simulation Results 85 86 Table D.1 Student Enrollment by Level of Education in 2020 for 33 Low-Income SSA Countries, Assuming 0% Real Growth of GDP per Capita per Year US$ million Lower secondary Upper secondary School-to-work School-to-work Coverage Transition Transition transition transition Year scenario rate P-S1 beyond S1 Primary General TVET program General TVET program Tertiary Base year 63% ­ 100.4 13.8 n/a n/a 7.2 n/a n/a 3.0 2020 ES-1 100% "Continuity" 161.0 56.6 5.7 0.0 25.8 3.5 2.7 13.8 ES-2 80% 42.8 6.9 2.0 20.4 2.7 2.1 10.8 ES-3 80% "Discontinuity" 42.8 6.9 2.0 4.9 1.3 9.9 2.6 ES-4 65% 32.7 7.5 3.4 4.7 0.7 3.9 2.2 ES-5 9 years of "Discontinuity" 211.6 0.0 0.0 5.1 0.7 7.7 2.2 primary Note: Rate of real growth of per capita GDP: 0% per year. Appendix E provides the corresponding simulation results for each of the 33 countries. Table D.2 Student Enrollment by Level of Education in 2020 for 33 Low-Income SSA Countries, Assuming 4% Real Growth of GDP per Capita per Year US$ million Lower secondary Upper secondary School-to-work School-to-work Coverage Transition Transition transition transition Year scenario rate P-S1 beyond S1 Primary General TVET program General TVET program Tertiary Base year 63% ­ 100.4 13.8 n/a n/a 7.2 n/a n/a 3.0 2020 ES-1 100% "Continuity" 161.0 56.6 5.7 0.0 25.8 3.5 2.7 13.8 ES-2 80% 42.8 6.9 2.0 20.4 2.7 2.1 10.8 ES-3 80% "Discontinuity" 42.8 6.9 2.0 8.1 2.2 8.4 4.4 ES-4 65% 32.7 7.5 3.4 7.1 1.0 3.5 3.3 ES-5 9 years of "Discontinuity" 211.6 0.0 0.0 7.8 1.0 7.2 3.3 primary Note: Rate of real growth of per capita GDP: 4% per year. Appendix E provides the corresponding simulation results for each of the 33 countries. 87 88 Table D.3 Aggregate Annual Public Spending on Education and Financing Gap in 2020 for 33 SSA Countries, Assuming 0% Real Growth of GDP per Capita per Year 2005 US$ billion Policy scenario Annual recurrent financing gap in Aggregate Annual recurrent post-primary education under 3 Total financing gap as share of Per-student annual financing gap Annual scenarios for domestic resource total public education spending at public education in primary capital mobilization spending (%) post-primary spending (recurrent or basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 38.8 2.6 21.3 19.4 17.5 2.8 69 64 59 "Spartan" 32.1 14.9 13.0 11.1 2.6 63 57 51 ES-2 "Generous" 33.7 16.6 14.7 12.8 2.5 64 59 53 "Spartan" 28.6 11.7 9.8 7.9 2.3 58 51 45 ES-3 "Generous" 26.9 10.0 8.1 6.2 2.3 55 48 41 "Spartan" 23.0 6.3 4.4 2.5 2.1 48 40 31 ES-4 "Generous" 23.4 6.7 4.9 3.2 2.1 49 40 32 "Spartan" 20.3 3.7 2.0 0.7 1.9 41 31 22 ES-5 "Generous" 21.2 4.8 2.0 0.5 0.2 2.6 43 34 25 "Spartan" 20.7 1.6 0.3 0.1 2.6 42 33 24 Notes: Rate of real growth of per capita GDP: 0% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 33% or less. Appendix E provides the corresponding simulation results for each of the 33 countries. Table D.4 Aggregate Annual Public Spending on Education and Financing Gap in 2020 for 33 SSA Countries, Assuming 4% Real Growth of GDP per Capita per Year 2005 US$ billion Policy scenario Annual recurrent financing gap in Aggregate Annual recurrent post-primary education under 3 Total financing gap as share of Per-student annual financing gap Annual scenarios for domestic resource total public education spending at public education in primary capital mobilization spending (%) post-primary spending (recurrent or basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 69.1 3.9 39.6 36.0 32.4 2.8 67 62 57 "Spartan" 57.1 27.9 24.3 20.7 2.6 60 54 48 ES-2 "Generous" 60.0 30.8 27.2 23.6 2.5 62 56 50 "Spartan" 50.8 21.9 18.3 14.7 2.3 55 48 41 ES-3 "Generous" 53.0 24.0 20.4 16.8 2.3 57 50 43 "Spartan" 45.1 16.4 12.8 9.2 2.1 50 42 34 ES-4 "Generous" 44.5 15.8 12.2 9.1 2.1 49 41 33 "Spartan" 38.4 9.8 6.4 3.3 2.0 41 31 22 ES-5 "Generous" 39.4 6.7 7.4 4.4 2.2 2.6 42 33 24 "Spartan" 37.9 6.0 3.1 1.6 2.6 40 30 21 Notes: Rate of real growth of per capita GDP: 4% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively; see also pages 43­44. Shaded cells: Dependency on external financing 33% or less. Appendix E provides the corresponding simulation results for each of the 33 countries. 89 APPENDIX E Simulation Results by Country BENIN BENIN: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 1,462,170 297,964 n/a n/a 62,569 n/a n/a 41,282 2020 ES-1 2,128,871 1,000,736 84,499 0 309,675 53,172 33,634 236,395 ES-2 2,128,871 756,111 101,399 29,556 233,977 40,174 25,412 178,609 ES-3 2,128,871 756,111 101,399 29,556 75,414 24,458 110,735 41,375 ES-4 2,128,871 578,203 109,849 51,724 70,233 12,059 41,357 33,999 ES-5 2,854,576 94,984 12,403 100,940 33,999 Note: Rate of real growth of per capita GDP: 2% per year. BENIN: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 109.6 38.4 n/a n/a 11.9 n/a n/a 6.7 505 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 24.5 1,859 51 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 18.5 1,405 48 ES-3 108.3 64.6 11.4 9.5 9.5 3.1 39.4 4.3 325 25 ES-4 108.3 49.4 12.4 16.6 8.8 1.5 14.7 3.5 267 20 ES-5 100.0 8.8 1.5 35.0 3.5 267 20 Note: Rate of real growth of per capita GDP: 2% per year. 91 92 BENIN: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 4.2 3.0 5.2 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2004 13 28 113 110 71 113 33 119 136 180 137 2020 "Generous" 13/15b 37 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. BENIN: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 788 444 401 359 43 66 61 55 "Spartan" 635 296 253 211 38 58 51 44 ES-2 "Generous" 673 336 293 251 37 60 54 47 "Spartan" 558 224 181 139 33 52 44 37 ES-3 "Generous" 570 235 193 150 34 53 45 38 "Spartan" 472 141 98 56 31 43 34 25 ES-4 "Generous" 490 159 116 74 30 45 36 28 "Spartan" 413 32 85 43 0 28 35 25 14 ES-5 "Generous" 405 48 5 ­37 36 34 23 13 "Spartan" 394 52 38 ­5 ­47 36 32 21 10 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 1 6 ­ 1 8 . Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively; see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 93 94 · Developing Post-Primary Education in Sub-Saharan Africa BURKINA FASO BURKINA FASO: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 1,120,726 217,104 n/a n/a 41,581 n/a n/a 24,975 2020 ES-1 3,654,662 1,673,066 141,539 0 511,464 87,819 55,891 406,654 ES-2 3,654,662 1,264,094 169,846 49,943 386,439 66,352 42,228 307,250 ES-3 3,654,662 1,264,094 169,846 49,943 107,953 35,012 191,647 59,590 ES-4 3,654,662 966,660 184,000 87,401 102,974 17,681 71,366 50,154 ES-5 4,863,415 139,487 18,240 171,073 50,154 Note: Rate of real growth of per capita GDP: 2% per year. BURKINA FASO: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 50.7 17.2 n/a n/a 5.0 n/a n/a 2.6 195 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 25.8 2,003 55 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 19.5 1,513 51 ES-3 108.3 64.6 11.4 9.5 8.2 2.7 41.0 3.8 293 25 ES-4 108.3 49.4 12.4 16.6 7.8 1.3 15.3 3.2 247 20 ES-5 100.0 7.8 1.3 35.5 3.2 247 20 Note: Rate of real growth of per capita GDP: 2% per year. BURKINA FASO: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 6.4 9.3 13.0 2020 "Generous" 3.0/3.2b 4.6 6.0 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2004 18 47 90 110 64 90 33 182 207 275 209 2020 "Generous" 14/16b 41 90 50 62 140 100 196 314 379 259 "Spartan" 20 90 50 32 140 100 196 314 379 259 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 95 96 BURKINA FASO: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 1,317 812 764 716 92 77 73 70 "Spartan" 1,081 585 537 488 84 72 67 63 ES-2 "Generous" 1,115 621 573 525 81 73 68 64 "Spartan" 937 449 401 353 75 67 62 57 ES-3 "Generous" 868 378 330 282 77 65 59 54 "Spartan" 719 235 187 139 71 58 51 44 ES-4 "Generous" 753 270 222 173 70 59 53 47 "Spartan" 636 108 158 110 62 65 52 45 37 ES-5 "Generous" 625 82 33 ­15 75 51 43 36 "Spartan" 610 163 68 19 ­29 74 50 42 34 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 97 BURUNDI BURUNDI: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 983,984 126,110 3,330 n/a 24,073 n/a n/a 24,975 2020 ES-1 2,321,824 949,892 81,266 0 268,965 46,182 30,589 216,360 ES-2 2,321,824 717,696 97,519 30,099 203,218 34,893 23,112 163,472 ES-3 2,321,824 717,696 97,519 30,099 48,271 15,656 108,959 27,732 ES-4 2,321,824 548,826 105,646 52,673 49,352 8,474 40,073 25,017 ES-5 2,998,656 67,565 8,911 96,924 25,017 Note: Rate of real growth of per capita GDP: 2% per year. BURUNDI: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 79.9 16.7 0.6 n/a 4.7 n/a n/a 2.7 221 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 26.7 1,764 61 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 20.2 1,333 58 ES-3 108.3 64.6 11.4 9.5 7.0 2.3 42.6 3.4 226 25 ES-4 108.3 49.4 12.4 16.6 7.1 1.2 15.7 3.1 204 20 ES-5 100.0 7.1 1.2 35.9 3.1 204 20 Note: Rate of real growth of per capita GDP: 2% per year. 98 BURUNDI: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 6.8 9.3 11.0 2020 "Generous" 4.0/4.2b 6.1 7.8 "Spartan" 5.6 6.8 Level of per-student spending in % of per capita GDPa 2004 15 64 220 110 240 220 33 615 702 932 707 2020 "Generous" 23/26b 69 90 50 109 140 100 331 530 640 438 "Spartan" 33 90 50 54 140 100 331 530 640 438 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. BURUNDI: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 293 163 156 150 50 86 84 81 "Spartan" 238 113 106 100 45 83 80 77 ES-2 "Generous" 247 124 117 111 43 83 81 78 "Spartan" 206 79 73 40 80 77 73 ES-3 "Generous" 195 73 67 60 41 79 75 72 "Spartan" 160 35 29 38 74 70 66 ES-4 "Generous" 172 55 48 42 37 76 72 68 "Spartan" 145 39 30 24 17 35 71 67 62 ES-5 "Generous" 154 14 8 1 41 73 69 64 "Spartan" 151 57 11 5 ­2 41 72 68 64 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively; see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 99 100 · Developing Post-Primary Education in Sub-Saharan Africa CAMEROON CAMEROON: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2003 2,533,204 496,365 n/a n/a 148,291 35,035 n/a 77,707 2020 ES-1 3,052,188 1,545,955 129,363 0 509,954 87,560 53,436 348,461 ES-2 3,052,188 1,168,055 155,236 43,785 385,299 66,156 40,374 263,282 ES-3 3,052,188 1,168,055 155,236 43,785 246,871 80,066 121,814 130,674 ES-4 3,052,188 893,218 168,172 76,624 204,656 35,140 48,386 95,584 ES-5 4,179,120 274,242 35,502 139,946 95,584 Note: Rate of real growth of per capita GDP: 2% per year. CAMEROON: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2003 99.6 31.8 n/a n/a 13.8 3.3 n/a 6.0 485 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 20.8 1,711 38 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 15.7 1,293 35 ES-3 108.3 64.6 11.4 9.5 18.8 6.1 27.3 7.8 642 25 ES-4 108.3 49.4 12.4 16.6 15.6 2.7 10.8 5.7 469 20 ES-5 100.0 15.6 2.7 31.1 5.7 469 20 Note: Rate of real growth of per capita GDP: 2% per year. CAMEROON: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2003 3.9 6.5 6.8 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2003 9 32 55 110 37 62 33 60 120 131 87 2020 "Generous" 13/14b 36 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 101 102 CAMEROON: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,263 1,354 1,224 1,093 54 63 55 52 "Spartan" 1,818 916 785 655 47 54 47 40 ES-2 "Generous" 1,918 1,019 888 757 45 57 50 43 "Spartan" 1,582 688 557 426 39 48 39 31 ES-3 "Generous" 1,771 874 743 612 43 53 46 39 "Spartan" 1,464 572 441 310 37 43 35 26 ES-4 "Generous" 1,465 574 443 312 36 43 35 26 "Spartan" 1,226 27 339 209 78 32 32 22 11 ES-5 "Generous" 1,224 279 149 18 50 32 22 11 "Spartan" 1,166 67 222 91 ­39 49 29 18 7 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 103 CENTRAL AFRICAN REPUBLIC CAMEROON: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 486,317 60,149 n/a n/a 18,581 n/a n/a 9,144 2020 ES-1 821,393 400,576 33,583 0 130,690 22,440 13,766 69,783 ES-2 821,393 302,657 40,299 11,517 98,743 16,954 10,401 52,725 ES-3 821,393 302,657 40,299 11,517 25,423 8,245 48,163 13,527 ES-4 821,393 231,444 43,657 20,155 24,695 4,240 17,895 11,594 ES-5 1,111,710 33,136 4,294 41,674 11,594 Note: Rate of real growth of per capita GDP: 2% per year. CENTRAL AFRICAN REPUBLIC: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 74.8 15.5 n/a n/a 7.0 n/a n/a 2.9 229 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 16.3 1,407 52 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 12.3 1,063 49 ES-3 108.3 64.6 11.4 9.5 7.6 2.5 41.9 3.2 273 25 ES-4 108.3 49.4 12.4 16.6 7.3 1.3 15.6 2.7 234 20 ES-5 100.0 7.3 1.3 35.8 2.7 234 20 Note: Rate of real growth of per capita GDP: 2% per year. 104 CENTRAL AFRICAN REPUBLIC: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 7.0 4.8 5.1 2020 "Generous" 3.3/3.4b 5.0 6.4 "Spartan" 4.6 5.5 Level of per-student spending in % of per capita GDPa 2004 12 15 59 110 25 59 33 134 200 238 170 2020 "Generous" 14/16b 41 90 50 62 140 100 160 255 308 211 "Spartan" 20 90 50 32 140 100 160 255 308 211 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. CENTRAL AFRICAN REPUBLIC: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 229 131 122 112 21 73 69 65 "Spartan" 183 87 77 68 19 67 62 56 ES-2 "Generous" 196 100 91 81 18 69 64 59 "Spartan" 161 67 57 48 17 62 56 50 ES-3 "Generous" 165 71 61 51 17 63 57 52 "Spartan" 136 43 34 24 16 55 48 41 ES-4 "Generous" 143 50 40 30 15 57 51 44 "Spartan" 120 17 28 19 9 14 49 41 33 ES-5 "Generous" 118 14 4 ­5 17 48 40 32 "Spartan" 115 26 11 2 ­8 17 47 39 30 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 105 106 · Developing Post-Primary Education in Sub-Saharan Africa CHAD CHAD: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 1,212,855 164,752 n/a n/a 54,212 n/a n/a 10,636 2020 ES-1 2,791,004 1,217,564 103,366 0 363,327 62,384 40,217 80,549 ES-2 2,791,004 919,937 124,039 37,128 274,513 47,134 30,387 60,859 ES-3 2,791,004 919,937 124,039 37,128 82,904 26,888 135,007 46,356 ES-4 2,791,004 703,481 134,376 64,974 77,976 13,388 50,361 38,471 ES-5 3,664,376 105,983 13,900 122,945 38,471 Note: Rate of real growth of per capita GDP: 2% per year. CHAD: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 76.9 18.7 n/a n/a 9.4 n/a n/a 1.6 113 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 7.3 541 36 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 5.5 409 31 ES-3 108.3 64.6 11.4 9.5 8.9 2.9 40.2 4.2 312 25 ES-4 108.3 49.4 12.4 16.6 8.3 1.4 15.0 3.5 259 20 ES-5 100.0 8.3 1.4 35.2 3.5 259 20 Note: Rate of real growth of per capita GDP: 2% per year. CHAD: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 5.4 4.4 6.8 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2004 12 19 82 110 25 82 33 80 180 430 182 2020 "Generous" 14/16b 41 90 50 62 140 100 202 323 390 267 "Spartan" 20 90 50 32 140 100 202 323 390 267 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 107 108 CHAD: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 877 460 420 380 67 71 66 62 "Spartan" 684 273 233 193 61 63 57 51 ES-2 "Generous" 763 354 314 274 59 67 61 56 "Spartan" 617 213 173 132 54 59 52 46 ES-3 "Generous" 724 318 278 238 56 65 59 54 "Spartan" 600 198 158 118 52 58 51 44 ES-4 "Generous" 628 227 187 147 51 60 53 47 "Spartan" 532 96 134 94 54 48 52 45 37 ES-5 "Generous" 525 71 31 ­9 55 52 44 36 "Spartan" 512 144 59 19 ­21 55 50 43 35 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 109 CONGO, DEMOCRATIC REPUBLIC CONGO, DEM. REP.: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 8,664,132 510,218 n/a n/a 530,964 n/a n/a 127,925 2020 ES-1 16,978,338 3,882,624 830,664 0 3,140,330 661,042 266,697 854,956 ES-2 16,978,338 2,933,538 996,797 227,635 2,372,694 499,454 201,504 645,967 ES-3 16,978,338 2,933,538 996,797 227,635 483,232 192,139 978,698 207,306 ES-4 16,978,338 2,243,294 1,079,864 398,361 491,453 103,451 362,749 186,029 ES-5 22,356,044 361,213 99,401 782,519 186,029 Note: Rate of real growth of per capita GDP: 2% per year. CONGO, DEM. REP.: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 92.6 18.7 n/a n/a 10.7 n/a n/a 3.1 229 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 12.9 950 53 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 9.8 718 49 ES-3 108.3 64.6 11.4 9.5 6.0 1.9 43.9 3.1 230 25 ES-4 108.3 49.4 12.4 16.6 6.1 1.0 16.3 2.8 207 20 ES-5 100.0 6.1 1.0 36.5 2.8 207 20 Note: Rate of real growth of per capita GDP: 2% per year. 110 CONGO, DEM. REP.: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 2.2 2.4 2.4 2020 "Generous" 3.9/4.1b 6.0 7.7 "Spartan" 5.5 6.6 Level of per-student spending in % of per capita GDPa 2005 8 19 140 110 19 160 33 164 187 248 188 2020 "Generous" 22/26b 69 90 50 111 140 100 353 565 681 467 "Spartan" 33 90 50 55 140 100 353 565 681 467 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. CONGO, DEM. REP.: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,256 1,237 1,174 1,112 262 82 80 77 "Spartan" 1,847 853 790 727 238 78 75 72 ES-2 "Generous" 1,941 957 895 832 227 80 76 73 "Spartan" 1,632 667 604 541 209 76 72 68 ES-3 "Generous" 1,520 559 496 433 205 74 70 66 "Spartan" 1,320 373 311 248 191 70 65 60 ES-4 "Generous" 1,366 422 359 297 188 71 66 62 "Spartan" 1,207 359 274 211 148 177 67 62 57 ES-5 "Generous" 1,356 144 81 18 271 71 66 61 "Spartan" 1,336 544 124 61 ­2 271 70 66 61 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 111 112 · Developing Post-Primary Education in Sub-Saharan Africa CONGO, REPUBLIC CONGO, REP.: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 564,457 170,421 n/a n/a 37,720 21,411 5,567 12,676 2020 ES-1 1,191,944 530,208 44,924 0 160,100 27,489 17,625 65,154 ES-2 1,191,944 400,601 53,909 16,017 120,964 20,770 13,316 49,228 ES-3 1,191,944 400,601 53,909 16,017 190,278 61,712 0 105,811 ES-4 1,191,944 306,342 58,401 28,031 145,443 24,973 0 71,363 ES-5 1,573,143 197,396 25,859 30,174 71,363 Note: Rate of real growth of per capita GDP: 2% per year. CONGO, REP.: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 85.7 46.6 n/a n/a 15.5 8.8 6.4 4.4 327 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 13.4 1,024 19 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 10.1 774 15 ES-3 108.3 64.6 11.4 9.5 46.2 15.0 0.0 21.8 1,664 25 ES-4 108.3 49.4 12.4 16.6 35.3 6.1 0.0 14.7 1,122 20 ES-5 100.0 35.3 6.1 19.8 14.7 1,122 20 Note: Rate of real growth of per capita GDP: 2% per year. CONGO, REP.: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 1.8 2.3 2.7 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2005 7 12 24 110 36 26 33 126 241 329 194 2020 "Generous" 12/14b 35 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 113 114 CONGO, REP.: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 1,075 569 500 431 31 59 53 46 "Spartan" 852 349 280 211 28 49 41 32 ES-2 "Generous" 930 428 359 290 27 53 45 38 "Spartan" 762 262 193 124 25 43 33 24 ES-3 "Generous" 1,198 694 625 556 28 63 58 52 "Spartan" 1,007 507 437 368 26 57 50 43 ES-4 "Generous" 951 451 382 313 25 54 47 39 "Spartan" 806 37 307 238 169 23 46 37 28 ES-5 "Generous" 843 321 252 183 28 48 40 32 "Spartan" 781 57 259 190 121 28 44 35 26 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 115 CÔTE D'IVOIRE CÔTE D'IVOIRE: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2000 1,916,047 393,089 n/a n/a 126,394 n/a n/a 110,472 2020 ES-1 3,587,016 1,787,143 149,711 0 586,097 100,633 61,564 731,005 ES-2 3,587,016 1,350,286 179,653 50,980 442,829 76,034 46,515 552,315 ES-3 3,587,016 1,350,286 179,653 50,980 254,701 82,606 153,178 135,146 ES-4 3,587,016 1,032,571 194,625 89,216 214,050 36,753 59,873 100,214 ES-5 4,886,999 287,007 37,169 165,727 100,214 Note: Rate of real growth of per capita GDP: 2% per year. CÔTE D'IVOIRE: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2000 66.7 22.0 n/a n/a 10.2 n/a n/a 7.5 619 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 37.9 3,132 47 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 28.7 2,366 44 ES-3 108.3 64.6 11.4 9.5 16.9 5.5 29.8 7.0 579 25 ES-4 108.3 49.4 12.4 16.6 14.2 2.4 11.6 5.2 429 20 ES-5 100.0 14.2 2.4 31.9 5.2 429 20 Note: Rate of real growth of per capita GDP: 2% per year. 116 CÔTE D'IVOIRE: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2000 4.8 7.7 8.9 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2000 15 34 110 110 63 110 33 83 94 125 95 2020 "Generous" 13/14b 36 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. CÔTE D'IVOIRE: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,985 1,935 1,777 1,619 62 66 61 56 "Spartan" 2,501 1,459 1,301 1,143 54 60 54 47 ES-2 "Generous" 2,486 1,446 1,288 1,130 52 60 53 47 "Spartan" 2,121 1,086 928 770 46 53 45 38 ES-3 "Generous" 1,894 857 699 541 50 47 39 30 "Spartan" 1,566 534 376 218 44 36 26 16 ES-4 "Generous" 1,579 547 389 231 43 37 27 17 "Spartan" 1,324 ­13 296 138 ­20 39 24 12 0 ES-5 "Generous" 1,306 249 91 ­67 49 23 11 ­1 "Spartan" 1,249 6 193 35 ­123 49 20 7 ­5 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 117 118 · Developing Post-Primary Education in Sub-Saharan Africa ERITREA ERITREA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 428,766 142,485 n/a n/a 92,156 n/a n/a 5,068 2020 ES-1 941,442 416,018 60,993 0 233,070 30,383 19,025 15,996 ES-2 941,442 314,324 73,192 15,990 176,097 22,956 14,375 12,086 ES-3 941,442 314,324 73,192 15,990 39,356 9,691 68,618 16,228 ES-4 941,442 240,366 79,291 27,982 39,267 5,119 25,454 14,287 ES-5 1,511,055 29,094 4,996 56,362 14,287 Note: Rate of real growth of per capita GDP: 2% per year. ERITREA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 73.9 46.1 n/a n/a 24.7 n/a n/a 1.5 120 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 3.1 243 25 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 2.3 184 9 ES-3 108.3 64.6 11.4 9.5 6.6 2.1 43.2 3.1 246 25 ES-4 108.3 49.4 12.4 16.6 6.5 1.1 16.0 2.7 217 20 ES-5 100.0 6.5 1.1 36.3 2.7 217 20 Note: Rate of real growth of per capita GDP: 2% per year. ERITREA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 3.9 9.9 11.8 2020 "Generous" 3.6/3.8b 5.5 7.1 "Spartan" 5.1 6.2 Level of per-student spending in % of per capita GDPa 2005 12 25 140 110 35 160 33 320 365 485 368 2020 "Generous" 18/22b 53 90 50 81 140 100 239 382 461 316 "Spartan" 26 90 50 42 140 100 239 382 461 316 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 119 120 ERITREA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 200 104 95 87 21 74 70 66 "Spartan" 151 57 49 41 19 66 60 55 ES-2 "Generous" 173 80 72 64 18 70 65 60 "Spartan" 136 45 37 29 16 62 56 50 ES-3 "Generous" 157 66 58 50 16 67 62 57 "Spartan" 130 41 33 24 15 60 54 48 ES-4 "Generous" 137 48 40 32 14 62 56 50 "Spartan" 116 23 28 20 12 13 55 48 41 ES-5 "Generous" 134 22 13 5 22 61 55 49 "Spartan" 131 38 20 11 3 22 61 54 48 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 121 ETHIOPIA ETHIOPIA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2002 5,954,992 2,396,728 n/a n/a 684,349 n/a n/a 101,829 2020 ES-1 12,200,467 8,899,888 988,876 0 3,728,775 165,024 302,207 567,591 ES-2 12,200,467 6,724,360 1,186,652 259,755 2,817,297 124,685 228,334 428,847 ES-3 12,200,467 6,724,360 1,186,652 259,755 603,370 50,440 1,098,916 247,023 ES-4 12,200,467 5,142,158 1,285,539 454,571 614,314 27,188 406,418 221,915 ES-5 24,145,693 455,971 26,640 899,173 221,915 Note: Rate of real growth of per capita GDP: 2% per year. ETHIOPIA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2002 69.1 31.0 n/a n/a 10.1 n/a n/a 1.8 135 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 6.6 527 44 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 5.0 398 39 ES-3 108.3 64.6 11.4 9.5 6.3 2.0 43.5 2.9 229 25 ES-4 108.3 49.4 12.4 16.6 6.4 1.1 16.1 2.6 206 20 ES-5 100.0 6.4 1.1 36.3 2.6 206 20 Note: Rate of real growth of per capita GDP: 2% per year. 122 ETHIOPIA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2002 6.8 8.1 11.9 2020 "Generous" 3.9/4.2b 6.0 7.7 "Spartan" 5.5 6.7 Level of per-student spending in % of per capita GDPa 2002 11 21 284 110 40 284 33 973 1,371 1,371 1,132 2020 "Generous" 22/27b 64 90 50 101 140 100 305 487 588 403 "Spartan" 31 90 50 51 140 100 305 487 588 403 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. ETHIOPIA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,280 1,276 1,205 1,133 337 80 77 74 "Spartan" 1,641 678 607 536 296 72 68 64 ES-2 "Generous" 1,917 969 898 826 281 76 73 69 "Spartan" 1,434 518 446 375 250 68 63 58 ES-3 "Generous" 1,679 761 689 618 252 73 69 65 "Spartan" 1,305 414 343 271 224 65 60 54 ES-4 "Generous" 1,451 565 493 422 220 69 64 59 "Spartan" 1,158 214 293 221 150 199 61 55 49 ES-5 "Generous" 1,488 303 231 160 346 70 65 60 "Spartan" 1,465 386 281 210 138 346 69 64 59 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 123 124 · Developing Post-Primary Education in Sub-Saharan Africa THE GAMBIA THE GAMBIA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2001 223,219 53,193 n/a n/a 26,842 n/a n/a 2,000 2020 ES-1 303,882 116,115 12,902 0 50,863 8,733 5,363 6,708 ES-2 303,882 87,732 15,482 4,374 38,430 6,598 4,052 5,068 ES-3 303,882 87,732 15,482 4,374 11,967 3,881 17,844 6,374 ES-4 303,882 67,089 16,772 7,654 11,287 1,938 6,643 5,305 ES-5 416,314 11,287 1,938 15,712 5,305 Note: Rate of real growth of per capita GDP: 2% per year. THE GAMBIA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2001 101.0 54.7 n/a n/a 30.1 n/a n/a 1.9 136 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 4.2 324 26 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 3.2 245 19 ES-3 108.3 64.6 11.4 9.5 9.1 3.0 39.8 4.0 308 25 ES-4 108.3 49.4 12.4 16.6 8.6 1.5 14.8 3.3 256 20 ES-5 100.0 8.6 1.5 35.1 3.3 256 20 Note: Rate of real growth of per capita GDP: 2% per year. THE GAMBIA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2001 4.5 6.5 6.5 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2001 16 26 140 110 47 160 33 284 324 431 327 2020 "Generous" 13/14b 36 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 125 126 THE GAMBIA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 81 36 30 24 4 51 44 36 "Spartan" 64 20 14 7 4 38 28 19 ES-2 "Generous" 71 27 21 15 3 44 36 27 "Spartan" 58 15 8 2 3 32 21 11 ES-3 "Generous" 69 26 19 13 3 43 34 25 "Spartan" 59 16 9 3 3 33 22 12 ES-4 "Generous" 59 16 10 4 3 34 23 13 "Spartan" 51 1 8 2 ­4 2 23 11 ­1 ES-5 "Generous" 51 6 ­1 ­7 4 23 11 ­1 "Spartan" 50 3 5 ­2 ­8 4 22 9 ­3 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 127 GHANA GHANA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2001 3,059,751 847,986 n/a n/a 219,450 10,525 n/a 75,000 2020 ES-1 4,090,999 1,552,114 172,457 0 683,918 117,429 71,756 244,256 ES-2 4,090,999 1,172,709 206,949 58,549 516,738 88,724 54,216 184,549 ES-3 4,090,999 1,172,709 206,949 58,549 155,473 50,424 241,139 82,400 ES-4 4,090,999 896,777 224,194 102,461 149,206 25,619 89,382 69,775 ES-5 5,591,645 149,206 25,619 210,715 69,775 Note: Rate of real growth of per capita GDP: 2% per year. GHANA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2001 93.0 54.4 n/a n/a 14.9 0.7 n/a 4.3 346 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 11.0 848 44 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 8.3 641 40 ES-3 108.3 64.6 11.4 9.5 8.8 3.1 40.2 3.7 286 25 ES-4 108.3 49.4 12.4 16.6 8.5 1.5 14.9 3.2 242 20 ES-5 100.0 8.5 1.5 35.2 3.2 242 20 Note: Rate of real growth of per capita GDP: 2% per year. 128 GHANA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2001 3.9 3.9 5.9 2020 "Generous" 3.1/3.3b 4.8 6.1 "Spartan" 4.3 5.3 Level of per-student spending in % of per capita GDPa 2001 16 30 140 110 83 160 33 277 316 419 318 2020 "Generous" 14/16b 40 90 50 63 140 100 191 305 368 252 "Spartan" 20 90 50 33 140 100 191 305 368 252 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. GHANA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 1,109 616 552 488 27 63 58 52 "Spartan" 892 400 336 272 26 54 47 40 ES-2 "Generous" 958 467 403 339 25 58 51 44 "Spartan" 794 304 240 176 24 49 41 33 ES-3 "Generous" 836 349 285 220 22 51 44 36 "Spartan" 708 220 156 92 22 43 34 24 ES-4 "Generous" 724 237 172 108 22 44 35 26 "Spartan" 623 60 135 71 7 22 35 25 14 ES-5 "Generous" 642 86 22 ­42 46 37 27 17 "Spartan" 628 103 72 8 ­56 46 35 25 15 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 129 130 · Developing Post-Primary Education in Sub-Saharan Africa GUINEA GUINEA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 1,474,640 306,211 n/a n/a 119,229 n/a n/a 23,107 2020 ES-1 2,246,231 1,058,597 89,194 0 332,355 57,066 35,819 73,429 ES-2 2,246,231 799,829 107,033 31,079 251,113 43,116 27,063 55,480 ES-3 2,246,231 799,829 107,033 31,079 67,703 21,958 123,932 36,858 ES-4 2,246,231 611,634 115,952 54,389 65,586 11,261 46,003 31,505 ES-5 3,012,329 88,522 11,539 109,020 31,505 Note: Rate of real growth of per capita GDP: 2% per year. GUINEA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 99.1 36.1 n/a n/a 21.5 n/a n/a 3.5 251 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 7.1 549 39 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 5.3 415 34 ES-3 108.3 64.6 11.4 9.5 7.9 2.6 41.4 3.5 276 25 ES-4 108.3 49.4 12.4 16.6 7.7 1.3 15.4 3.0 236 20 ES-5 100.0 7.7 1.3 35.6 3.0 236 20 Note: Rate of real growth of per capita GDP: 2% per year. GUINEA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 1.7 2.9 2.9 2020 "Generous" 3.2/3.4b 4.9 6.3 "Spartan" 4.5 5.5 Level of per-student spending in % of per capita GDPa 2005 6 11 66 110 9 66 33 114 130 172 131 2020 "Generous" 14/16b 40 90 50 59 140 100 151 242 291 200 "Spartan" 20 90 50 31 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 131 132 GUINEA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 554 272 241 211 54 65 60 54 "Spartan" 433 156 125 95 49 55 48 41 ES-2 "Generous" 481 206 175 144 47 60 53 47 "Spartan" 389 118 88 57 43 50 42 34 ES-3 "Generous" 447 175 145 114 44 57 50 43 "Spartan" 370 102 72 41 40 48 39 31 ES-4 "Generous" 387 119 89 58 39 50 42 34 "Spartan" 327 34 63 32 1 36 41 31 22 ES-5 "Generous" 322 31 1 ­30 44 40 30 21 "Spartan" 315 53 24 ­7 ­37 44 38 29 19 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 133 GUINEA-BISSAU GUINEA-BISSAU: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2002 206,875 29,023 n/a n/a 9,850 n/a n/a 1,384 2020 ES-1 465,147 156,913 11,844 0 43,143 10,901 7,044 5,820 ES-2 465,147 118,556 14,213 6,241 32,597 8,236 5,322 4,398 ES-3 465,147 118,556 14,213 6,241 7,272 3,471 25,415 5,998 ES-4 465,147 90,661 15,397 10,922 7,293 1,843 9,417 5,307 ES-5 612,890 7,293 1,843 21,328 5,307 Note: Rate of real growth of per capita GDP: 2% per year. GUINEA-BISSAU: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2002 80.4 27.0 n/a n/a 15.2 n/a n/a 1.3 90 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 3.2 235 24 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 2.4 177 5 ES-3 108.3 64.6 11.4 9.5 6.5 2.1 43.2 3.3 242 25 ES-4 108.3 49.4 12.4 16.6 6.6 1.1 16.0 2.9 214 20 ES-5 100.0 6.6 1.1 36.2 2.9 214 20 Note: Rate of real growth of per capita GDP: 2% per year. 134 GUINEA-BISSAU: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2002 1.9 2.2 2.2 2020 "Generous" 3.7/3.9b 5.7 7.3 "Spartan" 5.2 6.3 Level of per-student spending in % of per capita GDPa 2002 7 14 344 110 14 344 33 92 105 140 106 2020 "Generous" 17/20b 50 90 50 77 140 100 205 328 396 271 "Spartan" 25 90 50 40 140 100 205 328 396 271 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. GUINEA-BISSAU: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 60 26 23 20 7 71 67 62 "Spartan" 47 14 11 8 6 63 58 52 ES-2 "Generous" 53 19 17 14 6 67 62 57 "Spartan" 43 10 8 5 6 60 54 47 ES-3 "Generous" 52 18 16 13 6 66 61 56 "Spartan" 44 11 8 5 5 60 54 48 ES-4 "Generous" 45 13 10 7 5 62 56 50 "Spartan" 39 10 7 4 1 5 56 49 42 ES-5 "Generous" 43 4 1 ­2 7 59 53 46 "Spartan" 42 15 3 0 ­2 7 59 52 46 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 135 136 · Developing Post-Primary Education in Sub-Saharan Africa KENYA KENYA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 7,320,765 890,822 n/a n/a 91,541 2020 ES-1 11,227,719 2,350,418 106,670 198,607 257,229 ES-2 11,227,719 2,350,418 106,670 198,607 257,229 ES-3 11,227,719 391,154 33,531 995,764 166,960 ES-4 11,227,719 361,519 16,407 510,717 136,156 ES-5 11,515,684 265,964 15,806 405,964 136,156 Note: Rate of real growth of per capita GDP: 2% per year. KENYA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 104.7 27.1 n/a n/a 3.1 273 n/a 2020 ES-1 108.3 53.9 9.3 16.6 7.1 519 33 ES-2 108.3 53.9 9.3 16.6 7.1 519 33 ES-3 108.3 9.0 2.9 83.4 4.6 337 25 ES-4 108.3 8.3 1.4 42.8 3.7 275 20 ES-5 100.0 8.3 1.4 35.3 3.7 275 20 Note: Rate of real growth of per capita GDP: 2% per year. KENYA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 5.3 n/a 10.6 2020 "Generous" 3.0/3.1b n/a 5.9 "Spartan" n/a 5.1 Level of per-student spending in % of per capita GDPa 2005 16 n/a 140 110 68 140 33 193 221 293 222 2020 "Generous" 13/14b n/a 90 50 55 140 100 151 242 291 200 "Spartan" n/a 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 137 138 KENYA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,115 688 513 337 98 48 39 31 "Spartan" 1,841 418 243 68 94 40 30 21 ES-2 "Generous" 2,115 688 513 337 98 48 39 31 "Spartan" 1,841 418 243 68 94 40 30 21 ES-3 "Generous" 1,789 386 211 36 74 38 28 18 "Spartan" 1,745 341 166 ­9 74 36 26 16 ES-4 "Generous" 1,523 120 ­55 ­230 74 27 16 4 "Spartan" 1,482 220 79 ­97 ­272 74 25 13 2 ES-5 "Generous" 1,531 ­72 ­247 ­422 82 28 16 5 "Spartan" 1,501 412 ­102 ­277 ­452 82 26 14 3 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 139 LESOTHO LESOTHO: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 427,009 66,109 n/a n/a 22,009 560 n/a 6,800 2020 ES-1 316,163 103,667 7,680 0 31,341 5,381 4,837 9,829 ES-2 316,163 78,326 9,216 3,850 23,680 4,066 3,655 7,426 ES-3 316,163 78,326 9,216 3,850 10,512 3,409 14,054 8,195 ES-4 316,163 59,896 9,984 6,738 9,097 1,562 5,377 6,257 ES-5 372,682 13,652 1,564 13,563 6,257 Note: Rate of real growth of per capita GDP: 2% per year. LESOTHO: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 129.6 45.0 n/a n/a 22.7 0.6 n/a 3.9 379 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 6.1 572 28 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 4.6 432 23 ES-3 108.3 64.6 11.4 9.5 13.0 4.2 34.8 5.1 477 25 ES-4 108.3 49.4 12.4 16.6 11.3 1.9 13.3 3.9 364 20 ES-5 100.0 11.3 1.9 33.5 3.9 364 20 Note: Rate of real growth of per capita GDP: 2% per year. 140 LESOTHO: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 4.4 9.0 9.0 2020 "Generous" 3.0/3.1b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2005 18 46 140 110 71 160 33 588 671 891 676 2020 "Generous" 13/15b 37 90 50 55 140 100 151 242 291 200 "Spartan" 19 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. LESOTHO: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 121 57 48 39 1 53 46 38 "Spartan" 97 32 23 14 1 41 32 23 ES-2 "Generous" 107 42 33 25 1 47 38 30 "Spartan" 88 24 15 6 0 35 25 15 ES-3 "Generous" 110 46 37 28 1 48 40 32 "Spartan" 94 30 21 12 0 39 30 20 ES-4 "Generous" 93 30 21 12 0 39 30 20 "Spartan" 81 7 17 8 ­1 0 30 19 8 ES-5 "Generous" 78 6 ­3 ­12 0 27 16 4 "Spartan" 76 15 4 ­5 ­14 0 25 13 1 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44 . Shaded cells: Dependency on external financing 35% or less. 141 142 · Developing Post-Primary Education in Sub-Saharan Africa MADAGASCAR MADAGASCAR: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2003 3,253,389 344,935 n/a n/a 68,144 28,680 n/a 36,030 2020 ES-1 3,615,671 2,122,686 178,348 0 680,679 116,873 72,458 530,853 ES-2 3,615,671 1,603,807 214,017 61,418 514,291 88,304 54,746 401,089 ES-3 3,615,671 1,603,807 214,017 61,418 136,595 44,301 251,627 73,451 ES-4 3,615,671 1,226,441 231,852 107,482 132,251 22,708 93,468 62,748 ES-5 5,820,216 132,251 22,708 215,988 62,748 Note: Rate of real growth of per capita GDP: 2% per year. MADAGASCAR: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2003 129.7 20.0 n/a n/a 5.9 2.5 n/a 2.7 199 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 25.0 1,997 56 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 18.9 1,509 53 ES-3 108.3 64.6 11.4 9.5 7.8 2.5 41.6 3.5 276 25 ES-4 108.3 49.4 12.4 16.6 7.5 1.3 15.4 3.0 236 20 ES-5 100.0 7.5 1.3 35.7 3.0 236 20 Note: Rate of real growth of per capita GDP: 2% per year. MADAGASCAR: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2003 4.4 4.3 7.7 2020 "Generous" 3.2/3.4b 4.9 6.3 "Spartan" 4.5 5.5 Level of per-student spending in % of per capita GDPa 2003 13 27 140 110 73 160 33 193 295 295 234 2020 "Generous" 15/18b 43 90 50 65 140 100 187 299 361 248 "Spartan" 21 90 50 34 140 100 187 299 361 248 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 143 144 MADAGASCAR: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 1,358 856 802 748 77 75 71 67 "Spartan" 1,090 598 544 490 67 69 64 59 ES-2 "Generous" 1,135 645 592 538 63 70 65 61 "Spartan" 932 450 397 343 56 64 58 52 ES-3 "Generous" 869 385 331 277 58 61 55 49 "Spartan" 700 223 170 116 51 52 44 36 ES-4 "Generous" 740 264 211 157 50 54 47 40 "Spartan" 609 87 138 85 31 44 44 35 27 ES-5 "Generous" 663 129 75 22 69 49 41 33 "Spartan" 651 126 117 64 10 69 48 40 31 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44 . Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 145 MALAWI MALAWI: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2002 2,957,596 125,606 n/a n/a 85,318 n/a n/a 3,957 2020 ES-1 4,153,310 745,721 82,858 0 328,591 83,936 51,819 17,282 ES-2 4,153,310 563,434 99,429 42,316 248,269 63,419 39,152 13,058 ES-3 4,153,310 563,434 99,429 42,316 55,358 26,711 186,977 44,099 ES-4 4,153,310 430,861 107,715 74,054 55,170 14,093 69,384 38,779 ES-5 4,273,108 83,401 14,320 159,317 38,779 Note: Rate of real growth of per capita GDP: 2% per year. MALAWI: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2002 106.2 22.2 n/a n/a 16.1 n/a n/a 0.4 31 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 1.1 97 <10 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 0.8 73 <10 ES-3 108.3 64.6 11.4 9.5 6.5 2.1 43.2 2.8 248 25 ES-4 108.3 49.4 12.4 16.6 6.5 1.1 16.0 2.5 218 20 ES-5 100.0 6.5 1.1 36.3 2.5 218 20 Note: Rate of real growth of per capita GDP: 2% per year. 146 MALAWI: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2002 4.2 7.7 7.7 2020 "Generous" 3.6/3.7b 5.5 7.1 "Spartan" 5.0 6.1 Level of per-student spending in % of per capita GDPa 2002 8 48 355 110 48 355 33 1,108 1,264 1,679 1,274 2020 "Generous" 19/19b 54 90 50 84 140 100 258 413 498 341 "Spartan" 27 90 50 43 140 100 258 413 498 341 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. MALAWI: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 506 197 174 152 54 72 67 63 "Spartan" 421 116 93 70 50 66 60 55 ES-2 "Generous" 458 154 131 108 49 69 64 59 "Spartan" 393 92 70 47 47 63 58 52 ES-3 "Generous" 454 153 130 108 47 68 63 58 "Spartan" 405 106 83 61 44 64 59 53 ES-4 "Generous" 409 110 87 65 44 65 59 54 "Spartan" 370 110 73 51 28 42 61 55 49 ES-5 "Generous" 339 13 ­9 ­32 40 58 51 44 "Spartan" 333 142 7 ­16 ­38 40 57 50 43 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 147 148 · Developing Post-Primary Education in Sub-Saharan Africa MALI MALI: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 1,634,245 312,004 n/a n/a 75,955 n/a n/a 36,109 2020 ES-1 3,808,276 1,326,217 147,357 0 550,779 64,005 60,074 288,547 ES-2 3,808,276 1,002,031 176,829 51,931 416,144 48,359 45,389 218,013 ES-3 3,808,276 1,002,031 176,829 51,931 110,471 24,249 208,646 60,866 ES-4 3,808,276 766,259 191,565 90,879 105,578 12,269 77,784 51,326 ES-5 5,066,463 105,578 12,269 179,363 51,326 Note: Rate of real growth of per capita GDP: 2% per year. MALI: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 71.9 32.2 n/a n/a 8.7 n/a n/a 3.6 275 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 17.6 1,380 50 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 13.3 1,043 47 ES-3 108.3 64.6 11.4 9.5 7.8 2.5 41.6 3.7 291 25 ES-4 108.3 49.4 12.4 16.6 7.4 1.3 15.5 3.1 246 20 ES-5 100.0 7.4 1.3 35.7 3.1 246 20 Note: Rate of real growth of per capita GDP: 2% per year. MALI: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 6.0 6.8 8.3 2020 "Generous" 3.1/3.2b 4.7 6.0 "Spartan" 4.3 5.2 Level of per-student spending in % of per capita GDPa 2004 15 26 203 110 117 203 33 139 158 210 160 2020 "Generous" 14/16b 40 90 50 61 140 100 186 298 359 246 "Spartan" 20 90 50 32 140 100 186 298 359 246 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 149 150 MALI: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 1,095 596 547 498 87 72 67 63 "Spartan" 902 409 360 312 80 66 60 55 ES-2 "Generous" 948 458 409 361 77 68 62 57 "Spartan" 802 317 268 220 72 62 56 49 ES-3 "Generous" 794 308 260 211 73 61 55 49 "Spartan" 678 197 148 100 68 55 47 40 ES-4 "Generous" 697 217 168 120 67 56 49 42 "Spartan" 606 105 130 81 33 64 49 41 33 ES-5 "Generous" 613 68 19 ­30 78 50 42 34 "Spartan" 602 159 57 9 ­40 78 49 41 33 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 151 MAURITANIA MAURITANIA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 464,574 55,514 n/a n/a 39,143 n/a n/a 8,541 2020 ES-1 742,098 354,667 29,932 0 109,711 18,838 11,939 27,290 ES-2 742,098 267,971 35,919 10,412 82,893 14,233 9,020 20,619 ES-3 742,098 267,971 35,919 10,412 26,675 8,651 39,325 14,663 ES-4 742,098 204,919 38,912 18,220 24,802 4,258 14,696 12,029 ES-5 1,000,091 33,559 4,386 35,826 12,029 Note: Rate of real growth of per capita GDP: 2% per year. MAURITANIA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 100.5 21.2 n/a n/a 21.9 n/a n/a 3.9 287 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 8.1 610 36 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 6.1 461 32 ES-3 108.3 64.6 11.4 9.5 9.4 3.1 39.4 4.4 328 25 ES-4 108.3 49.4 12.4 16.6 8.8 1.5 14.7 3.6 269 20 ES-5 100.0 8.8 1.5 35.0 3.6 269 20 Note: Rate of real growth of per capita GDP: 2% per year. 152 MAURITANIA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 3.3 3.7 4.8 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2004 11 31 147 110 42 147 33 69 79 105 80 2020 "Generous" 13/15b 37 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. MAURITANIA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 244 122 107 92 15 61 54 48 "Spartan" 189 69 54 39 13 49 41 33 ES-2 "Generous" 212 92 77 62 12 55 48 40 "Spartan" 170 52 37 22 11 44 35 26 ES-3 "Generous" 203 85 70 55 12 53 45 38 "Spartan" 168 51 36 21 10 43 34 25 ES-4 "Generous" 174 57 42 27 10 45 36 28 "Spartan" 147 11 31 16 1 9 35 24 14 ES-5 "Generous" 143 17 2 ­13 12 33 22 12 "Spartan" 139 18 14 ­2 ­17 11 31 20 9 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 153 154 · Developing Post-Primary Education in Sub-Saharan Africa MOZAMBIQUE MOZAMBIQUE: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2001 2,949,260 167,792 n/a n/a 26,285 11,550 n/a 15,874 2020 ES-1 4,944,416 1,552,402 115,825 0 452,664 115,622 71,397 303,441 ES-2 4,944,416 1,172,926 138,990 59,163 342,013 87,359 53,944 229,267 ES-3 4,944,416 1,172,926 138,990 59,163 84,489 40,763 252,157 67,316 ES-4 4,944,416 896,944 150,572 103,535 82,773 21,142 93,616 58,190 ES-5 5,783,286 125,136 21,486 217,512 58,190 Note: Rate of real growth of per capita GDP: 2% per year. MOZAMBIQUE: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2001 80.4 12.3 n/a n/a 3.1 0.9 n/a 1.1 82 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 14.0 1,189 51 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 10.6 899 47 ES-3 108.3 64.6 11.4 9.5 7.2 2.4 42.3 3.1 264 25 ES-4 108.3 49.4 12.4 16.6 7.1 1.2 15.7 2.7 228 20 ES-5 100.0 7.1 1.2 35.9 2.7 228 20 Note: Rate of real growth of per capita GDP: 2% per year. MOZAMBIQUE: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2001 3.9 9.4 23.8 2020 "Generous" 3.4/3.5b 5.2 6.6 "Spartan" 4.7 5.7 Level of per-student spending in % of per capita GDPa 2001 9 32 30 110 138 57 33 489 559 742 563 2020 "Generous" 15/16b 42 90 50 65 140 100 166 265 319 219 "Spartan" 21 90 50 34 140 100 166 265 319 219 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 155 156 MOZAMBIQUE: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 933 487 443 399 69 70 65 61 "Spartan" 772 333 289 245 63 64 58 53 ES-2 "Generous" 808 371 327 283 61 66 60 55 "Spartan" 687 255 211 167 56 59 53 47 ES-3 "Generous" 701 267 223 179 58 60 54 48 "Spartan" 600 170 126 82 53 54 46 39 ES-4 "Generous" 611 182 138 94 52 54 47 40 "Spartan" 532 98 106 62 18 49 48 39 31 ES-5 "Generous" 511 27 ­17 ­61 52 45 37 28 "Spartan" 501 153 18 ­26 ­70 52 44 36 27 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 157 NIGER NIGER: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2002 889,323 121,283 n/a n/a 21,414 n/a n/a 6,585 2020 ES-1 4,135,090 1,867,659 157,997 0 570,187 97,902 62,396 226,186 ES-2 4,135,090 1,411,120 189,597 55,841 430,808 73,970 47,143 170,896 ES-3 4,135,090 1,411,120 189,597 55,841 102,395 33,209 222,223 56,601 ES-4 4,135,090 1,079,092 205,397 97,722 101,701 17,462 82,334 49,604 ES-5 5,480,137 137,813 18,033 193,691 49,604 Note: Rate of real growth of per capita GDP: 2% per year. NIGER: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2002 38.7 9.7 n/a n/a 2.6 n/a n/a 0.7 49 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 13.1 1,001 50 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 9.9 757 47 ES-3 108.3 64.6 11.4 9.5 7.0 2.3 42.6 3.3 251 25 ES-4 108.3 49.4 12.4 16.6 6.9 1.2 15.8 2.9 220 20 ES-5 100.0 6.9 1.2 36.0 2.9 220 20 Note: Rate of real growth of per capita GDP: 2% per year. 158 NIGER: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2002 5.5 8.5 10.2 2020 "Generous" 3.6/3.7b 5.4 7.0 "Spartan" 5.0 6.0 Level of per-student spending in % of per capita GDPa 2002 20 49 820 110 157 820 33 424 484 643 488 2020 "Generous" 17/20b 50 90 50 75 140 100 212 339 409 280 "Spartan" 25 90 50 39 140 100 212 339 409 280 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. NIGER: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 875 479 448 417 94 78 74 71 "Spartan" 690 302 271 240 85 72 67 63 ES-2 "Generous" 752 367 336 305 83 74 70 66 "Spartan" 612 233 202 171 76 68 63 58 ES-3 "Generous" 657 276 245 214 78 70 66 61 "Spartan" 540 165 134 103 73 64 58 52 ES-4 "Generous" 575 200 169 139 72 66 61 55 "Spartan" 484 107 114 83 52 67 60 53 47 ES-5 "Generous" 485 52 21 ­9 76 60 53 47 "Spartan" 475 161 43 12 ­19 76 59 52 46 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 159 160 · Developing Post-Primary Education in Sub-Saharan Africa NIGERIA NIGERIA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 22,566,255 3,530,518 n/a n/a 2,717,890 50,350 n/a 1,494,080 2020 ES-1 29,032,374 10,664,552 1,184,950 0 4,588,183 530,240 489,089 2,879,106 ES-2 29,032,374 8,057,661 1,421,940 408,270 3,466,627 400,626 369,534 2,175,325 ES-3 29,032,374 8,057,661 1,421,940 408,270 1,438,815 314,082 1,465,991 774,759 ES-4 29,032,374 6,161,741 1,540,435 714,473 1,261,847 145,827 557,582 599,530 ES-5 39,272,275 1,261,847 145,827 1,384,580 599,530 Note: Rate of real growth of per capita GDP: 2% per year. NIGERIA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 106.3 37.2 n/a n/a 31.3 0.9 n/a 15.1 1,162 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 20.1 1,638 44 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 15.2 1,237 40 ES-3 108.3 64.6 11.4 9.5 12.2 4.0 35.9 5.4 441 25 ES-4 108.3 49.4 12.4 16.6 10.7 1.8 13.6 4.2 341 20 ES-5 100.0 10.7 1.8 33.9 4.2 341 20 Note: Rate of real growth of per capita GDP: 2% per year. NIGERIA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 4.9 7.2 7.2 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2005 14 29 33 110 39 33 33 83 94 125 95 2020 "Generous" 13/15b 36 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 161 162 NIGERIA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 14,884 8,287 7,431 6,575 636 64 58 52 "Spartan" 12,347 5,810 4,954 4,098 577 56 49 42 ES-2 "Generous" 12,789 6,273 5,417 4,561 556 58 51 44 "Spartan" 10,873 4,402 3,546 2,690 511 50 42 34 ES-3 "Generous" 10,713 4,223 3,367 2,511 529 49 41 33 "Spartan" 9,123 2,673 1,817 960 490 41 31 22 ES-4 "Generous" 9,231 2,787 1,931 1,075 483 41 32 23 "Spartan" 7,991 539 1,576 720 ­136 454 32 21 11 ES-5 "Generous" 8,021 1,062 206 ­650 599 32 22 11 "Spartan" 7,822 939 862 6 ­850 599 31 20 9 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 163 RWANDA RWANDA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2003 1,649,656 165,729 n/a n/a 96,755 9,141 n/a 21,696 2020 ES-1 2,078,165 738,745 82,083 0 308,242 52,926 33,514 102,883 ES-2 2,078,165 558,163 98,499 28,804 232,894 39,988 25,321 77,734 ES-3 2,078,165 558,163 98,499 28,804 56,468 18,314 118,850 31,013 ES-4 2,078,165 426,830 106,708 50,408 56,059 9,625 44,005 27,166 ES-5 2,782,335 56,059 9,625 100,673 27,166 Note: Rate of real growth of per capita GDP: 2% per year. RWANDA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2003 112.6 23.5 n/a n/a 14.2 1.3 n/a 2.8 246 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 10.9 833 48 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 8.2 629 44 ES-3 108.3 64.6 11.4 9.5 7.1 2.3 42.4 3.3 251 25 ES-4 108.3 49.4 12.4 16.6 7.1 1.2 15.7 2.9 220 20 ES-5 100.0 7.1 1.2 36.0 2.9 220 20 Note: Rate of real growth of per capita GDP: 2% per year. 164 RWANDA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2003 3.9 5.9 6.4 2020 "Generous" 3.0/3.2b 5.4 7.0 "Spartan" 5.0 6.0 Level of per-student spending in % of per capita GDPa 2003 8 51 12 110 63 14 33 588 671 891 676 2020 "Generous" 16/19b 47 90 50 71 140 100 185 296 357 245 "Spartan" 23 90 50 37 140 100 185 296 357 245 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. RWANDA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 380 193 176 159 34 72 67 63 "Spartan" 304 121 104 87 31 65 59 53 ES-2 "Generous" 330 148 131 114 29 67 62 57 "Spartan" 273 93 76 59 27 60 54 48 ES-3 "Generous" 291 112 95 78 27 63 57 51 "Spartan" 246 69 52 35 24 56 49 42 ES-4 "Generous" 255 79 62 45 24 58 51 44 "Spartan" 220 44 45 28 11 22 51 43 35 ES-5 "Generous" 228 22 5 ­12 29 53 45 38 "Spartan" 224 68 19 2 ­15 29 52 44 37 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 165 166 · Developing Post-Primary Education in Sub-Saharan Africa SENEGAL SENEGAL: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2003 1,271,815 257,993 n/a n/a 72,371 10,866 n/a 55,100 2020 ES-1 2,477,987 1,225,214 102,931 0 392,700 67,427 41,837 367,750 ES-2 2,477,987 925,718 123,517 35,273 296,707 50,945 31,610 277,855 ES-3 2,477,987 925,718 123,517 35,273 115,249 37,378 128,944 62,022 ES-4 2,477,987 707,902 133,810 61,728 103,457 17,764 48,595 49,127 ES-5 3,370,858 139,222 18,096 121,373 49,127 Note: Rate of real growth of per capita GDP: 2% per year. SENEGAL: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2003 68.7 23.0 n/a n/a 9.4 1.4 n/a 6.1 484 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 29.3 2,302 50 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 22.1 1,740 47 ES-3 108.3 64.6 11.4 9.5 11.4 3.7 36.9 4.9 388 25 ES-4 108.3 49.4 12.4 16.6 10.2 1.8 13.9 3.9 308 20 ES-5 100.0 10.2 1.8 34.1 3.9 308 20 Note: Rate of real growth of per capita GDP: 2% per year. SENEGAL: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2003 4.6 6.2 7.1 2020 "Generous" 3.0/3.2b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2003 15 22 67 110 36 370 33 208 238 316 240 2020 "Generous" 13/15b 37 90 50 55 140 100 151 242 291 200 "Spartan" 18 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 167 168 SENEGAL: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario public financing Annual recurrent financing Per-student education gap in gap in post-primary education Annual spending at spending primary or under 3 scenarios for domestic capital Total financing gap as share of post-primary (recurrent basic resource mobilization financing total public education spending (%) Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 1,347 831 763 694 53 68 63 58 "Spartan" 1,109 598 530 461 47 61 55 49 ES-2 "Generous" 1,135 626 558 490 45 62 56 50 "Spartan" 955 450 382 314 41 55 48 40 ES-3 "Generous" 893 386 318 250 43 52 44 36 "Spartan" 738 236 167 99 39 41 32 23 ES-4 "Generous" 759 257 189 121 38 43 34 25 "Spartan" 638 31 140 72 3 35 32 22 11 ES-5 "Generous" 624 90 22 ­47 43 31 20 9 "Spartan" 604 59 71 2 ­66 43 28 17 6 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 169 SIERRA LEONE SIERRA LEONE: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2004 1,191,556 140,071 n/a n/a 38,707 n/a n/a 13,693 2020 ES-1 1,301,934 458,427 50,936 0 191,114 32,814 20,788 68,284 ES-2 1,301,934 346,367 61,124 17,867 144,397 24,793 15,706 51,593 ES-3 1,301,934 346,367 61,124 17,867 34,477 11,182 73,963 18,943 ES-4 1,301,934 264,869 66,217 31,268 34,467 5,918 27,353 16,710 ES-5 1,737,957 34,467 5,918 62,503 16,710 Note: Rate of real growth of per capita GDP: 2% per year. SIERRA LEONE: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2004 149.8 39.7 n/a n/a 12.0 n/a n/a 3.7 258 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 11.9 882 49 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 9.0 666 45 ES-3 108.3 64.6 11.4 9.5 7.0 2.3 42.6 3.3 245 25 ES-4 108.3 49.4 12.4 16.6 7.0 1.2 15.8 2.9 216 20 ES-5 100.0 7.0 1.2 36.0 2.9 216 20 Note: Rate of real growth of per capita GDP: 2% per year. 170 SIERRA LEONE: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2004 4.2 5.9 5.9 2020 "Generous" 3.6/3.8b 5.6 7.2 "Spartan" 5.1 6.2 Level of per-student spending in % of per capita GDPa 2004 10 29 73 110 30 73 33 253 289 383 291 2020 "Generous" 18/21b 54 90 50 85 140 100 255 408 493 338 "Spartan" 26 90 50 43 140 100 255 408 493 338 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. SIERRA LEONE: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 237 129 120 110 20 75 71 68 "Spartan" 189 83 74 65 18 69 64 59 ES-2 "Generous" 203 98 89 80 17 71 67 62 "Spartan" 167 64 55 45 15 65 60 54 ES-3 "Generous" 172 69 60 51 15 66 61 55 "Spartan" 144 42 33 24 14 60 53 47 ES-4 "Generous" 151 50 41 31 14 61 55 49 "Spartan" 129 29 29 20 10 12 55 48 40 ES-5 "Generous" 138 15 6 ­3 19 58 51 44 "Spartan" 135 45 13 4 ­6 19 57 50 43 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 171 172 · Developing Post-Primary Education in Sub-Saharan Africa SUDAN SUDAN: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2003 3,852,981 480,759 n/a n/a 366,841 2020 ES-1 9,011,209 1,622,911 278,656 169,485 1,399,342 ES-2 9,011,209 1,622,911 278,656 169,485 1,399,342 ES-3 9,011,209 317,617 103,011 828,856 167,554 ES-4 9,011,209 290,734 49,919 427,856 135,328 ES-5 9,329,777 192,960 49,437 348,735 135,328 Note: Rate of real growth of per capita GDP: 2% per year. SUDAN: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2003 55.3 20.7 n/a n/a 13.3 1,032 n/a 2020 ES-1 108.3 53.9 9.3 16.6 37.5 2,944 55 ES-2 108.3 53.9 9.3 16.6 37.5 2,944 55 ES-3 108.3 10.6 3.4 81.3 4.5 352 25 ES-4 108.3 9.7 1.7 42.0 3.6 285 20 ES-5 100.0 9.7 1.7 34.5 3.6 285 20 Note: Rate of real growth of per capita GDP: 2% per year. SUDAN: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2003 2.2 n/a 3.4 2020 "Generous" 3.0/3.1b n/a 5.9 "Spartan" n/a 5.1 Level of per-student spending in % of per capita GDPa 2003 8 n/a 33 110 18 33 33 82 93 124 94 2020 "Generous" 12/13b n/a 90 50 55 140 100 151 242 291 200 "Spartan" n/a 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 173 174 SUDAN: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario public financing Annual recurrent financing Per-student education gap in gap in post-primary education Annual spending at spending primary or under 3 scenarios for domestic capital Total financing gap as share of post-primary (recurrent basic resource mobilization financing total public education spending (%) Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,982 1,753 1,573 1,394 97 62 56 50 "Spartan" 2,781 1,554 1,374 1,194 94 59 53 46 ES-2 "Generous" 2,982 1,753 1,573 1,394 97 62 56 50 "Spartan" 2,781 1,554 1,374 1,194 94 59 53 46 ES-3 "Generous" 1,618 407 227 48 78 30 19 7 "Spartan" 1,579 368 188 9 78 28 17 5 ES-4 "Generous" 1,332 121 ­58 ­238 78 15 1 ­12 "Spartan" 1,297 ­5 86 ­94 ­274 78 12 ­2 ­15 ES-5 "Generous" 1,340 ­67 ­246 ­426 85 15 2 ­12 "Spartan" 1,317 184 ­90 ­270 ­450 85 14 0 ­14 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 175 TANZANIA TANZANIA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2002 7,082,355 329,280 n/a n/a 24,890 n/a n/a 31,049 2020 ES-1 8,936,792 3,838,086 215,187 0 844,435 144,990 132,421 1,521,120 ES-2 8,936,792 2,899,887 258,224 108,945 638,018 109,548 100,051 1,149,291 ES-3 8,936,792 2,899,887 258,224 108,945 175,815 57,021 455,662 139,270 ES-4 8,936,792 2,217,561 279,743 190,654 169,246 29,060 169,310 118,294 ES-5 10,514,471 342,366 29,725 400,906 118,294 Note: Rate of real growth of per capita GDP: 2% per year. TANZANIA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2002 100.4 9.2 n/a n/a 1.5 n/a n/a 1.0 83 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 36.7 3,088 61 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 27.7 2,333 57 ES-3 108.3 64.6 11.4 9.5 8.1 2.6 41.2 3.4 283 25 ES-4 108.3 49.4 12.4 16.6 7.8 1.3 15.3 2.9 240 20 ES-5 100.0 7.8 1.3 35.5 2.9 240 20 Note: Rate of real growth of per capita GDP: 2% per year. 176 TANZANIA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2002 3.8 5.2 5.2 2020 "Generous" 3.1/3.2b 4.8 6.2 "Spartan" 4.4 5.3 Level of per-student spending in % of per capita GDPa 2002 12 44 140 110 44 160 33 394 450 597 453 2020 "Generous" 14/16b 41 90 50 61 140 100 171 274 330 226 "Spartan" 20 90 50 32 140 100 171 274 330 226 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. TANZANIA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario public financing Annual recurrent financing Per-student education gap in gap in post-primary education Annual spending at spending primary or under 3 scenarios for domestic capital Total financing gap as share of post-primary (recurrent basic resource mobilization financing total public education spending (%) Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,907 1,929 1,823 1,717 126 77 73 70 "Spartan" 2,455 1,495 1,388 1,282 109 73 68 64 ES-2 "Generous" 2,418 1,462 1,356 1,250 104 72 68 63 "Spartan" 2,077 1,133 1,027 921 91 68 63 57 ES-3 "Generous" 1,680 730 624 518 98 60 54 47 "Spartan" 1,381 443 337 231 86 51 44 36 ES-4 "Generous" 1,444 508 402 296 84 53 46 39 "Spartan" 1,212 180 285 179 73 75 45 36 27 ES-5 "Generous" 1,123 74 ­32 ­138 75 40 31 21 "Spartan" 1,092 301 44 ­63 ­169 75 38 29 19 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 177 178 · Developing Post-Primary Education in Sub-Saharan Africa TOGO TOGO: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 774,154 291,804 n/a n/a 50,650 13,171 n/a 28,236 2020 ES-1 1,401,969 688,727 57,876 0 220,044 37,782 23,498 149,004 ES-2 1,401,969 520,371 69,451 19,874 166,255 28,546 17,754 112,580 ES-3 1,401,969 520,371 69,451 19,874 44,521 14,439 81,438 24,016 ES-4 1,401,969 397,931 75,239 34,779 43,153 7,409 30,232 20,539 ES-5 1,903,347 58,106 7,558 71,205 20,539 Note: Rate of real growth of per capita GDP: 2% per year. TOGO: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 79.7 51.1 n/a n/a 13.0 3.4 n/a 6.1 472 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 21.4 1,707 54 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 16.1 1,289 51 ES-3 108.3 64.6 11.4 9.5 7.9 2.5 41.5 3.4 275 25 ES-4 108.3 49.4 12.4 16.6 7.6 1.3 15.4 2.9 235 20 ES-5 100.0 7.6 1.3 35.6 2.9 235 20 Note: Rate of real growth of per capita GDP: 2% per year. TOGO: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 6.2 8.7 9.0 2020 "Generous" 3.2/3.4b 4.9 6.3 "Spartan" 4.5 5.5 Level of per-student spending in % of per capita GDPa 2005 21 19 28 110 32 120 33 132 198 238 168 2020 "Generous" 14/16b 40 90 50 62 140 100 166 265 319 219 "Spartan" 20 90 50 33 140 100 166 265 319 219 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 179 180 TOGO: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 405 238 221 203 30 73 69 64 "Spartan" 323 160 143 126 26 66 61 55 ES-2 "Generous" 343 182 164 147 25 68 63 58 "Spartan" 282 123 106 88 22 61 55 49 ES-3 "Generous" 284 124 107 90 23 61 55 49 "Spartan" 233 76 58 41 20 53 45 38 ES-4 "Generous" 244 87 70 53 20 55 48 41 "Spartan" 204 27 49 32 15 18 46 38 29 ES-5 "Generous" 202 25 8 ­10 24 46 37 28 "Spartan" 197 43 20 3 ­14 24 44 35 27 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 181 UGANDA UGANDA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2002 7,390,881 719,900 16,000 16,000 79,774 n/a n/a 78,500 2020 ES-1 11,411,202 4,043,596 234,883 0 800,237 70,253 131,984 1,122,121 ES-2 11,411,202 3,055,162 281,860 125,582 604,623 53,080 99,721 847,825 ES-3 11,411,202 3,055,162 281,860 125,582 161,674 26,810 457,588 134,697 ES-4 11,411,202 2,336,300 305,348 219,768 158,004 13,871 169,481 116,153 ES-5 13,005,873 331,086 15,199 430,250 116,153 Note: Rate of real growth of per capita GDP: 2% per year. UGANDA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2002 129.8 27.4 1.2 2.2 6.8 n/a n/a 3.7 282 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 31.0 2,219 59 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 23.4 1,677 55 ES-3 108.3 64.6 11.4 9.5 7.8 2.5 41.5 3.7 266 25 ES-4 108.3 49.4 12.4 16.6 7.7 1.3 15.4 3.2 230 20 ES-5 100.0 7.7 1.3 35.6 3.2 230 20 Note: Rate of real growth of per capita GDP: 2% per year. 182 UGANDA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2002 3.2 7.4 7.4 2020 "Generous" 3.3/3.4b 5.1 6.6 "Spartan" 4.7 5.7 Level of per-student spending in % of per capita GDPa 2002 8 145 108 110 145 108 33 139 159 211 160 2020 "Generous" 15/16b 42 90 50 62 140 100 155 247 298 204 "Spartan" 21 90 50 33 140 100 155 247 298 204 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. UGANDA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 2,304 1,227 1,137 1,046 195 75 71 67 "Spartan" 1,902 842 752 662 178 70 65 60 ES-2 "Generous" 1,990 936 845 755 173 71 67 62 "Spartan" 1,686 645 555 465 160 66 61 55 ES-3 "Generous" 1,673 625 535 445 167 66 60 55 "Spartan" 1,404 368 277 187 154 59 53 46 ES-4 "Generous" 1,480 446 355 265 152 61 55 49 "Spartan" 1,271 310 246 156 66 143 55 48 41 ES-5 "Generous" 1,189 47 ­43 ­133 141 52 44 37 "Spartan" 1,164 429 22 ­68 ­158 141 51 43 35 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 183 184 · Developing Post-Primary Education in Sub-Saharan Africa ZAMBIA ZAMBIA: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2005 2,194,384 281,771 n/a n/a 181,622 6,932 n/a 24,553 2020 ES-1 3,081,232 634,219 104,804 0 413,783 71,047 44,021 59,071 ES-2 3,081,232 479,188 125,765 36,193 312,636 53,680 33,261 44,631 ES-3 3,081,232 479,188 125,765 36,193 104,042 33,743 143,466 55,913 ES-4 3,081,232 366,438 136,245 63,338 94,316 16,194 54,036 44,723 ES-5 3,585,991 94,316 16,194 128,471 44,723 Note: Rate of real growth of per capita GDP: 2% per year. ZAMBIA: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2005 96.0 48.0 n/a n/a 22.2 0.8 n/a 2.6 214 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 4.5 390 26 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 3.4 295 20 ES-3 108.3 64.6 11.4 9.5 9.8 3.2 39.0 4.3 370 25 ES-4 108.3 49.4 12.4 16.6 8.9 1.5 14.7 3.4 296 20 ES-5 100.0 8.9 1.5 34.9 3.4 296 20 Note: Rate of real growth of per capita GDP: 2% per year. ZAMBIA: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2005 3.1 3.7 4.7 2020 "Generous" 3.0/3.1b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2005 8 17 170 110 25 170 33 216 247 328 249 2020 "Generous" 14/15b 40 90 50 61 140 100 191 306 369 253 "Spartan" 20 90 50 32 140 100 191 306 369 253 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. 185 186 ZAMBIA: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 935 438 385 331 42 64 58 53 "Spartan" 774 281 228 174 38 56 50 43 ES-2 "Generous" 838 346 293 239 36 60 53 47 "Spartan" 716 227 174 121 34 53 45 38 ES-3 "Generous" 820 330 277 224 34 59 52 46 "Spartan" 729 242 189 135 32 54 46 39 ES-4 "Generous" 719 232 178 125 31 53 46 38 "Spartan" 647 118 162 109 56 30 48 40 31 ES-5 "Generous" 622 60 7 ­46 40 46 37 29 "Spartan" 609 185 47 ­7 ­60 40 45 36 27 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. Simulation Results by Country · 187 ZIMBABWE ZIMBABWE: Enrollments Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary School-to-work School-to-work transition transition Primary General TVET program General TVET program Tertiary 2003 2,265,360 712,646 10,000 10,000 38,023 n/a 1,045 76,000 2020 ES-1 2,536,623 1,088,474 60,715 0 245,091 63,113 37,876 538,663 ES-2 2,536,623 822,402 72,857 30,643 185,180 47,685 28,617 406,990 ES-3 2,536,623 822,402 72,857 30,643 74,730 36,349 114,913 58,337 ES-4 2,536,623 628,896 78,929 53,626 66,323 17,079 43,490 45,683 ES-5 2,980,598 132,957 17,138 108,368 45,683 Note: Rate of real growth of per capita GDP: 2% per year. ZIMBABWE: GER and Other Parameters Under Different Policy Scenarios, Base Year and 2020 Lower secondary Upper secondary GER (%) GER (%) Higher education GER (%) School- School- and other indicators Primary to-work to-work GER transition transition GER Students/ Unemployment (%) General TVET program General TVET program (%) 100,000 (simulated) (%) 2003 92.6 51.3 1.4 2.8 5.7 n/a 0.3 6.2 588 n/a 2020 ES-1 108.3 85.5 9.5 0.0 38.8 6.7 12.0 42.6 3,808 55 ES-2 108.3 64.6 11.4 9.5 29.3 5.0 9.0 32.2 2,877 52 ES-3 108.3 64.6 11.4 9.5 11.8 3.8 36.3 4.6 412 25 ES-4 108.3 49.4 12.4 16.6 10.5 1.8 13.7 3.6 323 20 ES-5 100.0 10.5 1.8 34.0 3.6 323 20 Note: Rate of real growth of per capita GDP: 2% per year. 188 ZIMBABWE: Teacher Salaries and Per-Student Spending by Level of Education, Base Year and 2020 Lower secondary Upper secondary Higher education School-to- School-to- work work Humanities transition transition & social Primary General TVET program General TVET program sciences Sciences Professional All Teacher salaries in multiples of per capita GDPa 2003 4.1 4.1 4.1 2020 "Generous" 3.0/3.1b 4.6 5.9 "Spartan" 4.2 5.1 Level of per-student spending in % of per capita GDPa 2003 13 23 140 110 23 160 33 171 195 259 197 2020 "Generous" 13/14b 37 90 50 55 140 100 151 242 291 200 "Spartan" 19 90 50 29 140 100 151 242 291 200 Note: Rate of real growth of per capita GDP: 2% per year. a. After adjustment for wealth, geographic dimension, and/or stimulation of demand. b. The first number pertains to the 6-year primary cycle (or whatever the duration of the primary cycle) and the second number to the 9-year primary cycle in ES-5. ZIMBABWE: Estimated Annual Public Spending on Education and Financing Gap in 2020 2005 US$ million unless otherwise indicated Aggregate Annual annual recurrent Policy scenario Annual recurrent financing public financing gap in post-primary education Per-student education gap in Annual under 3 scenarios for domestic Total financing gap as share of spending at spending primary or capital resource mobilization total public education spending (%) post-primary (recurrent basic financing Coverage levels and capital) education Scenario 1 Scenario 2 Scenario 3 gap Scenario 1 Scenario 2 Scenario 3 ES-1 "Generous" 1,642 1,160 1,087 1,014 19 72 68 63 "Spartan" 1,430 952 880 807 14 68 63 58 ES-2 "Generous" 1,344 868 796 723 12 66 60 55 "Spartan" 1,184 712 639 566 9 61 55 49 ES-3 "Generous" 831 357 284 212 11 45 36 27 "Spartan" 688 217 144 71 8 33 23 12 ES-4 "Generous" 699 228 155 83 8 34 24 13 "Spartan" 590 3 119 47 ­26 8 22 10 ­3 ES-5 "Generous" 561 30 ­43 ­115 17 18 5 ­8 "Spartan" 541 54 10 ­62 ­135 17 15 2 ­12 Notes: Rate of real growth of per capita GDP: 2% per year. Enrollment scenarios: The five scenarios for coverage or enrollment are described on pages 16­18. Resource mobilization scenarios: Scenarios 1­3 assume a budget allocation for education of 20%, 23%, and 26%, respectively, see also pages 43­44. Shaded cells: Dependency on external financing 35% or less. 189 References AFD (Agence Française de Développement). 2009. "Priorities for Education Finance and External Aid to Support Africa's Development in the Coming Years." PowerPoint presentation by Michel Jacquier, deputy CEO of AFD, held at the Conference of African Ministers of Finance and Education in Tunis, July 15­17. Brossard, M., and B. Foko. 2007. Couts et Financement de l'Enseignement Supérieur en Afrique Francophone. Washington, DC: World Bank and Pôle de Dakar. Bruns, B., A. Mingat, and R. Rakotomalala. 2003. Achieving Universal Primary Edu- cation by 2015: A Chance for Every Child. Washington, DC: World Bank. Cohen, J., and D. Bloom. 2006. Universal Basic and Secondary Education. Cam- bridge, MA: American Academy of Arts and Sciences. Fasih, T. 2008. Linking Education Policy to Labor Market Outcomes. Washington, DC: World Bank. Fields, G. 2007. "Labor Market Policy in Developing Countries: A Selective Review of the Literature and Needs for the Future." Policy Research Working Paper 4362, World Bank, Washington, DC. Hanushek, E., and L. Wößmann. 2007. "The Role of Education Quality in Economic Growth." Policy Research Working Paper 4122, World Bank, Washington, DC. Hoppers, W., and S. Obeegadoo. 2008. "Beyond Primary Education: Challenges and Approaches to Expanding Learning Opportunities in Africa. Pulling the Pieces . . . Together." General Synthesis Report Presented at the ADEA 2008 Biennale on Education in Africa. Maputo, Mozambique, May 5­9. IMF (International Monetary Fund). 2009. World Economic Outlook April 2009: Crisis and Recovery. World Economic and Financial Surveys. Washington, DC: IMF. Kingdon, G., and M. Soderbom. 2008. "Education, Skills, and Labor Market Out- comes: Evidence from Ghana." Education Working Paper Series 12, World Bank, Washington, DC. Lam, D. 2007. "The Demography of Youth in Developing Countries and Its Eco- nomic Implications." Background paper for World Development Report 2007: Development and the Next Generation. Washington, DC: World Bank. Lewin, K. 2008. "Strategies for Sustainable Financing of Secondary Education in Sub-Saharan Africa." Prepared under the Secondary Education in Africa 191 192 · References Initiative. Working Paper 136, Africa Human Development Series, World Bank, Washington, DC. Mingat, A. 2004. La Rémunération/Le Statut des Enseignants dans la Perspective de l'Atteinte des Objectifs du Millénaire dans les Pays d'Afrique Subsaharienne Francophone en 2015. Washington, DC: World Bank. Mingat, A., and F. Ndem. 2008. L'Enjeu de la Scolarisation en Milieu Rural et les Défis de Développement de la Couverture Scolaire au Niveau du Premier Cycle Sec- ondaire. Paris: Agence Française de Développement. UN (United Nations). 2009. The Millennium Development Goals Report 2009. New York: UN. Walther, R. 2005. Financing Vocational Training: A Europe-Africa Comparison. Paris: Agence Française de Développement. ------. 2006. "Vocational Training in the Informal Sector." Working Paper 30, Agence Française de Développement, Paris. Walther, R., and Filipiak, E. 2008. Towards a Renewal of Apprenticeship in Africa. Paris: Agence Française de Développement. World Bank. 2008. Accelerating Catch-Up: Tertiary Education for Growth in Sub- Saharan Africa, Washington, DC: World Bank. ------. Forthcoming. Education in Sub-Saharan Africa: A Comparative Analysis. Washington, DC: World Bank. Index Figures, notes, and tables are indicated by f, n, and t following page numbers. A per-student spending, 92t Africa Action Plan (World Bank), 3 teacher salaries, 92t African Development Bank, 4 primary education, 76t Asia's vs. SSA's secondary and tertiary growth in population (used in education levels, 9, 15n2 simulations), 78t, 91t Association for the Development of per-student spending, 92t Education in Africa school leavers and completion (ADEA), 3, 4 rate, 68t teacher salaries, 92t B school-to-work transition programs, benchmarks 91­92t for cost drivers in secondary spending and financing gap, 59, 93t education, 29, 30­34 technical and vocational for expenditures excluding teacher education and training salaries, 38­39 (TVET), 91­92t for government resources for tertiary education, 76t, 91t education, 45­46 per-student spending, 92t for marginal costs to reach rural upper secondary education, 74t, 91t populations, 39­40 per-student spending, 92t for per-student spending, 35 teacher salaries, 92t for public-private division of Burkina Faso educational responsibility, lower secondary education, 72t xx­xxi, 46­49 coverage as of 2005, 10t for teacher salaries, 37­38, 52 enrollment growth in, 12t, 13, Benin 70t, 94t lower secondary education, 72t logistic difficulties to reach 2020 enrollment growth in, 70t, 91t coverage goals, 14t logistic difficulties to reach 2020 per-student spending, 28t, 95t coverage goals, 14t teacher salaries, 28t, 95t 193 194 · Index Burkina Faso (Continued) C primary education, 76t Cameroon growth in population (used in lower secondary education, 72t simulations), 78t, 94t coverage as of 2005, 10t per-student spending, 95t enrollment growth in, 12t, school leavers and completion 70t, 100t rate, 10t, 68t logistic difficulties to reach 2020 projected for 2020, 11t, 12 coverage goals, 14t teacher salaries, 95t per-student spending, 28t, 101t school-to-work transition programs, teacher salaries, 28t, 101t 94­95t primary education, 76t spending and financing gap, 59, 96t growth in population (used in technical and vocational education simulations), 78t, 100t and training (TVET), 94­95t per-student spending, 101t coverage as of 2005, 10t school leavers and completion tertiary education, 76t, 94t rate, 10t, 68t coverage as of 2005, 10t projected for 2020, 11t, 12 per-student spending, 95t teacher salaries, 101t upper secondary education, 74t, 94t school-to-work transition programs, coverage as of 2005, 10t 100­101t per-student spending, 28t, 95t spending and financing gap, teacher salaries, 28t, 95t 58, 102t Burundi technical and vocational education lower secondary education, 72t and training (TVET), enrollment growth in, xvii, 13, 100­101t 70t, 97t coverage as of 2005, 10t logistic difficulties to reach 2020 tertiary education, 76t, 100t coverage goals, 14t coverage as of 2005, 10t per-student spending, 28t, 98t per-student spending, 101t teacher salaries, 28t, 98t upper secondary education, primary education, 76t 74t, 100t growth in population (used in coverage as of 2005, 10t simulations), 78t, 97t per-student spending, 28t, 101t per-student spending, 98t teacher salaries, 28t, 101t school leavers and completion capital costs in simulation model, rate, 68t xx, 36, 53 teacher salaries, 98t Central African Republic school-to-work transition programs, lower secondary education, 72t 97­98t enrollment growth in, 13, spending and financing gap, 59, 99t 70t, 103t technical and vocational education logistic difficulties to reach 2020 and training (TVET), 97­98t coverage goals, 14t tertiary education, 76t, 97t per-student spending, 104t per-student spending, 98t teacher salaries, 104t upper secondary education, 74t, 97t primary education, 76t per-student spending, 28t, 98t growth in population (used in teacher salaries, 28t, 98t simulations), 78t, 103t Index · 195 per-student spending, 104t competing claims on external aid, school leavers and completion xxiii, 4, 57 rate, 68t Conference for Ministers of Finance teacher salaries, 104t and of Education (2009), 4 school-to-work transition programs, Congo, Democratic Republic of 103­4t lower secondary education, 72t spending and financing gap, enrollment growth in, 70t, 109t 59, 105t logistic difficulties to reach 2020 technical and vocational coverage goals, 14t education and training per-student spending, 110t (TVET), 103­4t teacher salaries, 110t tertiary education, 76t, 103t primary education, 76t per-student spending, 104t growth in population upper secondary education, (used in simulations), 74t, 103t 78t, 109t per-student spending, 104t per-student spending, 110t teacher salaries, 104t school leavers and completion Chad rate, 68t lower secondary education, 72t teacher salaries, 110t enrollment growth in, 13, school-to-work transition programs, 70t, 106t 109­10t logistic difficulties to reach 2020 spending and financing gap, coverage goals, 14t 59, 111t per-student spending, 107t technical and vocational education teacher salaries, 107t and training (TVET), primary education, 76t 109­10t growth in population (used in tertiary education, 76t, 109t simulations), 78t, 106t per-student spending, 110t per-student spending, 107t upper secondary education, school leavers and completion 74t, 109t rate, 68t per-student spending, 110t teacher salaries, 107t teacher salaries, 110t school-to-work transition programs, Congo, Republic of 106­7t lower secondary education, 72t spending and financing gap, enrollment growth in, 13, 59, 108t 70t, 112t technical and vocational logistic difficulties to reach 2020 education and training coverage goals, 14t (TVET), 106­7t per-student spending, 113t tertiary education, 76t, 106t teacher salaries, 113t per-student spending, 107t primary education, 76t upper secondary education, 74t, 106t growth in population (used in per-student spending, 107t simulations), 78t, 112t teacher salaries, 107t per-student spending, 113t combining primary with lower school leavers and completion secondary education into rate, 68t basic cycle, xix, 52, 62 teacher salaries, 113t 196 · Index Congo, Republic of (Continued) Côte d'Ivoire school-to-work transition programs, lower secondary education, 72t 112­13t enrollment growth in, 70t, 115t spending and financing gap, 58, 114t logistic difficulties to reach 2020 technical and vocational education coverage goals, 14t and training (TVET), per-student spending, 116t 112­13t teacher salaries, 116t tertiary education, 76t, 112t primary education, 76t per-student spending, 113t growth in population (used in upper secondary education, 74t, 112t simulations), 78t, 115t per-student spending, 113t per-student spending, 116t teacher salaries, 113t school leavers and completion construction costs. See capital costs in rate, 68t simulation model teacher salaries, 116t costs, xix­xx, 25­43. See also school-to-work transition programs, benchmarks 115­16t adjusting benchmarks for spending and financing gap, expenditures excluding 59, 117t teacher salaries, 38­39 technical and vocational education adjusting marginal costs to reach and training (TVET), rural populations, 39­40 115­16t aggregate costs, simulations of, tertiary education, 76t, 115t 53­58 per-student spending, 116t capital costs, xx, 36, 52 upper secondary education, government commitment of 74t, 115t spending for, 62 per-student spending, 116t higher education, 35­36 teacher salaries, 116t non-salary curriculum diversification, 62 lower secondary education, 72­73t D primary education, 76­77t Dakar Forum (2000 World Education in simulation model, 33 Forum), 3 upper secondary education, Democratic Republic of Congo. See 74­75t Congo, Democratic per-student. See per-student spending Republic of repetition rate and, 33­34 diversity among SSA countries, xxiii, secondary education 7, 9, 10t, 11 benchmarks for cost drivers in, financing gap, 58­60 30­34, 37­38 per-student spending, 35 comparative perspectives on cost private education, 46 drivers in, 27­30 data comparison, 29, 29t E expenditures excluding teacher EdStats, 8 salaries, 33 educational development, policy teacher salaries. See teacher salaries context for, 3­5 technical and vocational education sustainable options, identification and training, 34­35, 36t of, xviii, 60, 62 Index · 197 Education for All (EFA) Ethiopia Fast Track Initiative (EFA-FTI), xx, lower secondary education, 72t 3, 30, 37, 41n4, enrollment growth in, 70t, 121t 45, 47, 52 logistic difficulties to reach 2020 goals for 2015, 3 coverage goals, 14t elementary education. See primary per-student spending, 28t, 122t education teacher salaries, 28t, 122t employment opportunities, 17 primary education, 76t adapting enrollment at growth in population (used in upper levels to match, simulations), 78t, 121t 21, 52, 56 per-student spending, 122t modern job sector and, 20 school leavers and completion enrollment growth, xix, 9­15 rate, 68t in primary education, 11­12, 11t, teacher salaries, 122t 78­79t school-to-work transition programs, projected post-primary education in 121­22t 2020, 21­23, 86­87t spending and financing gap, in secondary education, 12­15, 12t 59, 123t lower secondary education, xvii, technical and vocational education 70­71t and training (TVET), Eritrea 121­22t lower secondary education, 72t tertiary education, 76t, 121t enrollment growth in, 70t, 118t per-student spending, 122t logistic difficulties to reach 2020 upper secondary education, coverage goals, 14t 74t, 121t per-student spending, 119t per-student spending, 28t, 122t teacher salaries, 119t teacher salaries, 28t, 122t primary education, 76t expenditures. See costs growth in population (used in external aid needed. See financial gap simulations), 78t, 118t per-student spending, 119t F school leavers and completion financial crisis and economic rate, 68t downturn, effect of, 4 teacher salaries, 119t financial gap, xxi­xxii, 45­50. See also school-to-work transition programs, resources; specific countries 118­19t benchmarks affecting volume of spending and financing government resources for gap, 59, 120t education, 45­46 technical and vocational education benchmarks for public-private and training (TVET), division of educational 118­19t responsibility, xx­xxi, 46­49 tertiary education, 76t, 118t flexibility of framework and, 37 per-student spending, 119t higher education, 49 upper secondary education, increase with more affordable 74t, 118t education, 63 per-student spending, 119t primary and lower secondary teacher salaries, 119t education, 47 198 · Index financial gap, (Continued) expenses (non-salary), 33 school-to-work transition organization of service delivery, programs, 48 31­32, 31t, 52, 62 simulation results for, 53­58, 54t repetition rate, 33­34 technical and vocational education teacher salaries, 32­33 and training (TVET), 48­49 Ghana upper secondary education, 47­48 lower secondary education, 72t flexible parameters used for enrollment growth in, 13, simulations, xxi, 37­41 70t, 127t foreign direct investment, 4 logistic difficulties to reach 2020 FTI. See Education for All (EFA) coverage goals, 14t per-student spending, 128t G teacher salaries, 128t The Gambia primary education, 76t lower secondary education, 72t growth in population (used in enrollment growth in, 70t, 124t simulations), 78t, 127t logistic difficulties to reach 2020 per-student spending, 128t coverage goals, 14t school leavers and completion per-student spending, 125t rate, 68t teacher salaries, 125t teacher salaries, 128t primary education, 76t school-to-work transition programs, growth in population (used in 127­28t simulations), 78t, 124t spending and financing gap, 58, 129t per-student spending, 125t technical and vocational education school leavers and completion and training (TVET), 127­28t rate, 68t tertiary education, 76t, 127t teacher salaries, 125t per-student spending, 128t school-to-work transition programs, upper secondary education, 124­25t 74t, 127t spending and financing gap, per-student spending, 128t 58, 126t teacher salaries, 128t technical and vocational education girls' education, 21, 46 and training (TVET), Guinea 124­25t lower secondary education, 72t tertiary education, 76t, 124t enrollment growth in, xvii, 13, per-student spending, 125t 70t, 130t upper secondary education, logistic difficulties to reach 2020 74t, 124t coverage goals, 14t per-student spending, 125t per-student spending, 28t, 131t teacher salaries, 125t teacher salaries, 28t, 131t gender differences. See girls' education primary education, 76t generous vs. spartan scenario, growth in population (used in simulation model simulations), 78t, 130t aggregate costs and funding gaps per-student spending, 131t in 2020, xxii­xxiii, 54t, school leavers and completion 55­57 rate, 68t capital costs, 36 teacher salaries, 131t Index · 199 school-to-work transition programs, financial gap, 4, 49 130­31t humanities vs. natural sciences vs. spending and financing gap, professional studies, 36, 36t 58, 132t modern sector jobs and, 20 technical and vocational education per-student spending, 26, 26t, 27, and training (TVET), 76­77t 130­31t as benchmark for, 38 tertiary education, 76t, 130t privately financed, xxi, 49 per-student spending, 131t holistic approach, 1, 4, 51 upper secondary education, 74t, 130t I per-student spending, 28t, 131t International Monetary Fund (IMF), 8 teacher salaries, 28t, 131t Guinea-Bissau K enrollment growth in lower Kenya secondary education, 70t lower secondary education, 72t lower secondary education, 72t enrollment growth in, 70t, 136t enrollment growth in, 13, 133t per-student spending, 137t logistic difficulties to reach 2020 teacher salaries, 137t coverage goals, 14t primary education, 77t per-student spending, 134t growth in population (used in teacher salaries, 134t simulations), 78t, 136t primary education, 76t per-student spending, 137t growth in population (used in school leavers and completion simulations), 78t, 133t rate, 68t per-student spending, 134t teacher salaries, 137t school leavers and completion school-to-work transition programs, rate, 68t 136­37t teacher salaries, 134t spending and financing gap, school-to-work transition programs, 58, 138t 133­34t technical and vocational education spending and financing gap, 59, 135t and training (TVET), technical and vocational education 136­37t and training (TVET), tertiary education, 77t, 136t 133­34t per-student spending, 137t tertiary education, 76t, 133t upper secondary education, per-student spending, 134t 75t, 136t upper secondary education, per-student spending, 137t 74t, 133t teacher salaries, 137t per-student spending, 134t teacher salaries, 134t L Lesotho H lower secondary education, 73t higher education enrollment growth in, xvii, 13, access to, 49 71t, 139t cost assumptions for, 35­36 logistic difficulties to reach 2020 coverage levels, 9, 10t coverage goals, 14t 200 · Index Lesotho (Continued) technical and vocational education per-student spending, 140t and training (TVET), teacher salaries, 140t 142­43t primary education, 77t tertiary education, 77t, 142t growth in population (used in per-student spending, 143t simulations), 78t, 139t upper secondary education, per-student spending, 140t 75t, 142t school leavers and completion per-student spending, 143t rate, 69t teacher salaries, 143t teacher salaries, 140t Malawi private education, 46 lower secondary education, 73t school-to-work transition programs, enrollment growth in, 71t, 145t 139­40t logistic difficulties to reach 2020 spending and financing gap, coverage goals, 14t 58, 141t per-student spending, 28t, 146t technical and vocational education teacher salaries, 28t, 146t and training (TVET), primary education, 77t 139­40t growth in population tertiary education, 77t, 139t (used in simulations), per-student spending, 140t 79t, 145t upper secondary education, 75t, 139t per-student spending, 146t per-student spending, 140t school leavers and completion teacher salaries, 140t rate, 69t lower secondary education. teacher salaries, 146t See secondary education school-to-work transition programs, low-income countries, defined, 2n1 145­46t spending and financing gap, M 59, 147t Madagascar technical and vocational education lower secondary education, 73t and training (TVET), enrollment growth in, 13, 145­46t 71t, 142t tertiary education, 77t, 145t logistic difficulties to reach 2020 per-student spending, 146t coverage goals, 14t upper secondary education, per-student spending, 143t 75t, 145t teacher salaries, 143t per-student spending, 28t, 146t primary education, 77t teacher salaries, 28t, 146t growth in population (used in Mali simulations), 78t, 142t lower secondary education, 73t per-student spending, 143t coverage as of 2005, 10t school leavers and completion enrollment growth in, xvii, 12t, rate, 69t 13, 71t, 148t teacher salaries, 143t logistic difficulties to reach school-to-work transition programs, 2020 coverage 142­43t goals, 14t spending and financing gap, per-student spending, 149t 59, 144t teacher salaries, 149t Index · 201 primary education, 77t upper secondary education, 75t, 151t growth in population (used in per-student spending, 152t simulations), 79t, 148t teacher salaries, 152t per-student spending, 149t methodology and data, 7­8. See also school leavers and completion simulation model rate, 10t, 69t Millennium Development Goals projected for 2020, 11t (MDGs), 4, 9, 37 teacher salaries, 149t model. See simulation model school-to-work transition programs, modern sector jobs, 20, 47 148­49t Mozambique spending and financing gap, lower secondary education, 73t 59, 150t enrollment growth in, xvii, 13, technical and vocational education 71t, 154t and training (TVET), logistic difficulties to reach 2020 148­49t coverage goals, 14t coverage as of 2005, 10t per-student spending, 155t tertiary education, 77t, 148t teacher salaries, 154t coverage as of 2005, 10t primary education, 77t per-student spending, 149t growth in population (used in upper secondary education, 75t, 148t simulations), 79t, 154t coverage as of 2005, 10t per-student spending, 155t per-student spending, 149t school leavers and completion teacher salaries, 149t rate, 69t Mauritania teacher salaries, 155t lower secondary education, 73t school-to-work transition programs, enrollment growth in, 13, 154­55t 71t, 151t spending and financing gap, 59, 156t logistic difficulties to reach 2020 technical and vocational education coverage goals, 14t and training (TVET), per-student spending, 152t 154­55t teacher salaries, 151t tertiary education, 77t, 154t primary education, 77t per-student spending, 155t growth in population (used in upper secondary education, simulations), 151t 75t, 154t per-student spending, 152t per-student spending, 155t school leavers and completion teacher salaries, 155t rate, 69t teacher salaries, 152t N school-to-work transition programs, Niger 151­52t lower secondary education, 73t spending and financing gap, coverage as of 2005, 10t 58, 153t enrollment growth in, xvii, 12t, technical and vocational education 13, 71t, 157t and training (TVET), logistic difficulties to reach 2020 151­52t coverage goals, 14t tertiary education, 77t, 151t per-student spending, 158t per-student spending, 152t teacher salaries, 157t 202 · Index Niger (Continued) technical and vocational education primary education, 77t and training (TVET), growth in population (used in 160­61t simulations), 79t, 157t coverage as of 2005, 10t per-student spending, 158t tertiary education, 77t, 160t school leavers and completion coverage as of 2005, 10t, 11 rate, 10t, 11, 69t per-student spending, 161t projected for 2020, 11t, 12 upper secondary education, 75t, 160t teacher salaries, 158t coverage as of 2005, 10t school-to-work transition programs, per-student spending, 161t 157­58t teacher salaries, 161t spending and financing gap, 59, 159t P technical and vocational education per-student spending. See also and training (TVET), specific countries 157­58t current patterns by level of coverage as of 2005, 10t education, 25­27 tertiary education, 77t, 157t compared to other regions, coverage as of 2005, 10t, 11 25­27, 26t per-student spending, 158t future development of secondary upper secondary education, education and, 28­29 75t, 157t lower secondary education, 72­73t, coverage as of 2005, 10t 83t, 88­89t per-student spending, 158t primary education, 26, 26t, teacher salaries, 158t 76­77t, 83t Nigeria secondary education, 26, 26t lower secondary education, 73t tertiary education, 26, 26t, 27 coverage as of 2005, 10t SSA compared with other enrollment growth in, 12t, countries, 35 71t, 160t upper secondary education, logistic difficulties to reach 2020 74­75t, 83t coverage goals, 14t Pôle de Dakar, 7 per-student spending, 161t population growth rates used in teacher salaries, 160t simulations, 78­79t primary education, 77t post-primary education, 17­23. growth in population See also higher education; (used in simulations), secondary education; 79t, 160t technical and vocational per-student spending, 161t education and training school leavers and completion (TVET) rate, 10t, 11, 69t challenge of developing, 4, 63 projected for 2020, 11t, 12 student flow management teacher salaries, 161t options for, xxii­xxiii, 20­21 school-to-work transition programs, private vs. public financing, 48 160­61t projected 2020 enrollment, spending and financing gap, 21­23 58, 162t simulation assumptions in, 18­20 Index · 203 primary education R average teacher salary, 42n8, 76­77t repetition rate, in simulation model, benchmark of, 38 33­34 combining with lower secondary Republic of Congo. See Congo, into basic cycle, xix, 52, 62 Republic of completion rate, xvii, 1, 3, 9, resources. See also financial gap 10t, 68­69t benchmarks affecting, 45­46 construction costs, 36 mobilization module, 42n14 coverage for simulation model, 82t pace of growth, xxiii, 57­58, 63 enrollment growth in, 11­12, relationship between country 11t, 78­79t income level and expenditures (non-salary), 76­77t teacher salaries, financial gap, 47 29­30, 65­66 MDG goal for 2015, 9 rural populations, 43nn19­20 module in simulation model, 40 lower secondary education, 41 per-student spending, 76­77t, 83t marginal costs to reach, 39­40 privately financed, 47, 83t primary education, 40 students per teacher, 76­77t retaining out-of-school universal, xvii, xix, 3, 9, 39, 46, 52 population, 40 private education Rwanda benchmarks for public-private lower secondary education, 73t division of educational coverage as of 2005, 10t responsibility, xx­xxi, 46­49 enrollment growth in, 12t, 13, share of enrollment by education 71t, 163t level, 83t logistic difficulties to reach 2020 coverage goals, 14t Q per-student spending, 164t quality and reach of educational teacher salaries, 163t services, 25­43 primary education, 77t benchmarks for cost drivers in growth in population (used in secondary education, simulations), 79t, 163t 30­34 per-student spending, 164t organization of service delivery, school leavers and completion 31­32, 31t rate, 10t, 11, 69t teacher salaries and, 32­33, projected for 2020, 11t 37­38 teacher salaries, 164t comparative perspectives on private education, 46 secondary education cost school-to-work transition programs, drivers, 27­30, 28t 163­64t current patterns of per-student spending and financing gap, spending by level of 59, 165t education, 25­27, 26t technical and vocational education indicative framework, flexibility of, and training (TVET), 163­64t xxi, 37­41 coverage as of 2005, 10t relationship between student tertiary education, 77t, 163t learning and resources, 41n1 coverage as of 2005, 10t rural populations, 40 per-student spending, 164t 204 · Index Rwanda (Continued) neglect of, 47 upper secondary education, projected enrollment in 2020, 75t, 163t 21­23 coverage as of 2005, 10t upper secondary education per-student spending, 164t average teacher salary, 28t, teacher salaries, 164t 29­30, 74­75t construction costs, 36 S coverage as of 2005, 10t salaries. See teacher salaries expenditures (non-salary), school-to-work transition programs. 74­75t See also specific countries financial gap, 47­48 financial gap, 48 per-student spending, 28t, projected enrollment in 2020, 74­75t, 83t, 88­89t 86­87t privately financed, 47­48, 83t simulation assumptions for student projected enrollment in 2020, flow management with, xix, 21­23, 86­87t 18, 52 simulation model coverage, secondary education 31­32, 31t, 82t benchmarks for teacher salaries, 38 students per class, 27, 28t, cost drivers 74­75t benchmarks for, 30­34 students per teacher, 28t, 74­75t comparative perspectives on, teachers per class, 28t, 74­75t 27­30, 28t Senegal coverage level, 9 lower secondary education, 73t lower secondary education enrollment growth in, 71t, 166t average teacher salary, 28t, logistic difficulties to reach 2020 29­30, 42n8, 72­73t coverage goals, 14t construction costs, 36 per-student spending, 167t coverage as of 2005, 10t teacher salaries, 166t enrollment growth, xvii, 12­15, primary education, 77t 12t, 70­71t growth in population (used in expenses (non-salary), 72­73t simulations), 79t, 166t financial gap, 47 per-student spending, 167t logistic difficulties for countries school leavers and completion to reach 2020 coverage rate, 69t goals, 14­15, 14t teacher salaries, 167t module in simulation model, 41 school-to-work transition programs, per-student spending, 72­73t, 166­67t 83t, 88­89t spending and financing gap, 59, 168t privately financed, 47, 83t technical and vocational education projected enrollment in 2020, and training (TVET), 12­15, 12t, 21­23, 86­87t 166­67t simulation model coverage, tertiary education, 77t, 166t 31­32, 31t, 82t per-student spending, 167t students per class, 27, 28t, 72­73t upper secondary education, 75t, 166t students per teacher, 28t, 72­73t per-student spending, 167t teachers per class, 28t, 72­73t teacher salaries, 167t Index · 205 service delivery arrangements, 31­32, secondary education 31t, 52 non-salary expenses, 33 Sierra Leone organization of service delivery, lower secondary education, 73t 31­32, 31t, 52 enrollment growth in, 71t, 169t repetition rate, 33­34 logistic difficulties to reach 2020 teacher salaries, 32­33 coverage goals, 14t technical and vocational education per-student spending, 170t and training, cost teacher salaries, 169t assumptions for, 34­35, 52 primary education, 77t simulation results, xxi­xxii, 51­60 growth in population aggregate costs and funding gaps in (used in simulations), 2020, 53­58, 54t 79t, 169t diversity across 33 SSA countries, per-student spending, 170t 58­60 school leavers and completion enrollment in 2020 by education rate, 69t level, 22t teacher salaries, 170t key assumptions, summary private education, 46 of, 52­53 school-to-work transition programs, social returns on investment in 169­70t education, 20­21, 47 spending and financing gap, 59, 171t structural duality of SSA economies, 20 technical and vocational education student flow management, and training (TVET), post-primary 169­70t options for, 20­21 tertiary education, 77t, 169t proactive management of, 57 per-student spending, 170t simulation assumptions in, 18­20 upper secondary education, students per class 75t, 169t lower secondary education, 27, per-student spending, 170t 28t, 72­73t teacher salaries, 170t organization of service delivery simulation model, xviii­xix, 7­8. See in simulation model, also simulation results 31­32, 31t capital costs, xx, 36, 53 upper secondary education, 27, flexible parameters used, xxi, 37­41 28t, 74­75t indicative framework, flexibility of, students per teacher xxi, 37­41 lower secondary education, 28t, key assumptions, summary 72­73t of, 52­53 organization of service delivery in lower secondary module, 41 simulation model, 31­32, 31t post-primary student flow primary education, 76­77t management, xxii­xxiii, upper secondary education, 28t, 18­23, 52 74­75t primary education module, 40 Sudan public-private division, xx­xxi, 46­48 lower secondary education, 73t purpose of, 61 enrollment growth in, 71t, 172t school-to-work transition per-student spending, 173t programs, 48, 52 teacher salaries, 172t 206 · Index Sudan (Continued) tertiary education, 77t, 175t primary education, 77t coverage as of 2005, 10t growth in population (used in per-student spending, 176t simulations), 79t, 172t upper secondary education, per-student spending, 173t 75t, 175t school leavers and completion coverage as of 2005, 10t rate, 69t per-student spending, 176t teacher salaries, 173t teacher salaries, 176t school-to-work transition programs, tax revenues, 42n14 172­73t teacher placement issues, 40 spending and financing gap, teacher salaries. See also specific 59, 174t countries technical and vocational education benchmarks for, 37­38, 52 and training (TVET), organization of service delivery 172­73t in simulation model, tertiary education, 77t, 172t 31­32, 31t per-student spending, 173t primary education, 42n8, 76­77t upper secondary education, 75t, 172t relationship with country income per-student spending, 173t level, 29­30, 65­66 teacher salaries, 173t secondary education sustainable options, xviii, 60, 62 comparative data, 28t, 29­30, 30t T lower secondary education, 42n8 Tanzania simulation model, 32­33 lower secondary education, 73t technical and vocational education coverage as of 2005, 10t and training (TVET). See also enrollment growth in, xvii, 12t, specific countries 13, 71t, 175t costs assumptions for, 34­35, 36t logistic difficulties to reach 2020 coverage as of 2005, 10t coverage goals, 14t financial gap, 48­49 per-student spending, 176t per-student spending as benchmark teacher salaries, 175t for, 38 primary education, 77t projected enrollment in 2020, growth in population (used in 86­87t simulations), 79t, 175t simulation assumptions for per-student spending, 176t student flow management school leavers and completion with, 52 rate, 10t, 69t tertiary education. See higher projected for 2020, 11t education teacher salaries, 176t Togo school-to-work transition programs, lower secondary education, 73t 175­76t enrollment growth in, xvii, 13, spending and financing gap, 59, 177t 71t, 178t technical and vocational education logistic difficulties to reach 2020 and training (TVET), coverage goals, 14t 175­76t per-student spending, 28t, 179t coverage as of 2005, 10t teacher salaries, 28t, 178t Index · 207 primary education, 77t spending and financing gap, growth in population (used in 59, 183t simulations), 79t, 178t technical and vocational education per-student spending, 179t and training (TVET), school leavers and completion 181­82t rate, 69t tertiary education, 77t, 181t teacher salaries, 179t per-student spending, 182t school-to-work transition programs, upper secondary education, 75t, 181t 178­79t per-student spending, 28t, 182t spending and financing gap, teacher salaries, 28t, 182t 59, 180t United Nations and EFA goals, 3 technical and vocational education United Nations Educational, and training (TVET), Scientific, and Cultural 178­79t Organization (UNESCO) tertiary education, 77t, 178t Institute of Statistics, 8, 41n2 per-student spending, 179t Regional Office for Education upper secondary education, (BREDA), 7 75t, 178t World Conference on Higher per-student spending, Education, 3­4 28t, 179t universal primary education. teacher salaries, 28t, 179t See primary education training for occupation/profession. upper secondary education. See technical and vocational See secondary education education and training urban residents as percentage of (TVET) school-age population, 39, TVET. See technical and vocational 43nn19­20 education and training W U World Bank Uganda Africa Action Plan, 3 lower secondary education, 73t as source for simulation data, 7­8 enrollment growth in, xvii, 13, World Conference on Higher 71t, 181t Education (UNESCO), 4 logistic difficulties to reach 2020 World Education Forum (Dakar coverage goals, 14t Forum 2000), 3 per-student spending, 28t, 182t teacher salaries, 28t, 181t Z primary education, 77t Zambia growth in population (used in lower secondary education, 73t simulations), 79t, 181t coverage as of 2005, 10t per-student spending, 182t enrollment growth in, 12t, school leavers and completion 71t, 184t rate, 69t logistic difficulties to teacher salaries, 182t reach 2020 coverage private education, 46 goals, 14t school-to-work transition programs, per-student spending, 185t 181­82t teacher salaries, 184t 208 · Index Zambia (Continued) Zimbabwe primary education, 77t lower secondary education, 73t growth in population enrollment growth in, xvii, 13, (used in simulations), 71t, 187t 79t, 184t logistic difficulties to reach 2020 per-student spending, 185t coverage goals, 14t school leavers and per-student spending, 188t completion rate, 10t, teacher salaries, 187t 11, 69t primary education, 77t projected for 2020, 11t, 12 growth in population (used in teacher salaries, 185t simulations), 79t, 187t private education, 46 per-student spending, 188t school-to-work transition programs, school leavers and completion 184­85t rate, 69t spending and financing gap, teacher salaries, 188t 59, 186t school-to-work transition programs, technical and vocational education 187­88t and training (TVET), spending and financing gap, 184­85t 59, 189t coverage as of 2005, 10t technical and vocational education tertiary education, 77t, 184t and training (TVET), coverage as of 2005, 10t 187­88t per-student spending, 185t tertiary education, 77t, 187t upper secondary education, per-student spending, 188t 75t, 184t upper secondary education, coverage as of 2005, 10t 75t, 187t per-student spending, 185t per-student spending, 188t teacher salaries, 185t teacher salaries, 188t ECO-AUDIT Environmental Benefits Statement The World Bank is committed to preserving Saved: endangered forests and natural resources. · 8 trees The Office of the Publisher has chosen · 3 million BTUs of to print Developing Post-Primary Education total energy in Sub-Saharan Africa on recycled paper · 775 lbs of CO2 with 30 percent post-consumer waste, in equivalent of accordance with the recommended standards greenhouse gases for paper usage set by the Green Press Initia- · 3,732 gallons of tive, a nonprofit program supporting publish- waste water ers in using fiber that is not sourced from · 227 lbs of endangered forests. For more information, solid waste visit www.greenpressinitiative.org. C ountries in Sub-Saharan Africa have made substantial progress in universalizing primary school completion. Many young Africans are completing primary schooling, and many more will do so in the coming years. The pressure--already strong--to expand secondary and tertiary education is expected to intensify. Finding a sustainable path for such expansion is a challenge for all countries in the region. Given the diversity across African countries, Developing Post-Primary Education in Sub-Saharan Africa: Assessing the Financial Sustainability of Alternative Pathways offers no generic policy fix. Rather, it seeks to provide policy makers and their development partners with an analytical tool to inform discussion and debate about alternative options in light of country circumstances. It presents simulation scenarios that serve an illustrative function to draw attention to the implications of such options as raising the share of education in the national budget, reforming the service delivery arrangements to manage costs, diversifying the student flow beyond lower secondary education, and enlarging the role of private funding, particularly in post-primary education. The study captures the nature of the policy choices by presenting alternative packages of policies and using them to clarify the affordability of what the authors characterize as spartan and generous choices. One of the study's most valuable contributions is the flexibi- lity of the simulation model, which can be used to adapt the package of policies to national contexts. Developing Post-Primary Education in Sub-Saharan Africa: Assessing the Financial Sustainability of Alternative Pathways, which contains detailed annexes with results for 33 low-income countries, will be of interest to national education policy makers and development partners, as well as education researchers and education consultants. ISBN 978-0-8213-8183-0 THE WORLD BANK SKU 18183