94741 DIREC TIONS IN DE VELOPMENT Human Development Out-of-School Youth in ­Sub-Saharan Africa A Policy Perspective Keiko Inoue, Emanuela di Gropello, Yesim Sayin Taylor, and James Gresham Out-of-School Youth in Sub-Saharan Africa Direc tions in De velopment Human Development Out-of-School Youth in Sub-Saharan Africa A Policy Perspective Keiko Inoue, Emanuela di Gropello, Yesim Sayin Taylor, and James Gresham © 2015 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 18 17 16 15 This work is a product of the staff of The World Bank with external contributions. The findings, interpreta- tions, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, 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 judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Inoue, Keiko, Emanuela di Gropello, Yesim Sayin Taylor, and James Gresham. 2015. Out-of-School Youth in Sub-Saharan Africa: A Policy Perspective. Directions in Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-0505-9. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content ­ contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights ­ of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@ worldbank.org. ISBN (paper): 978-1-4648-0505-9 ISBN (electronic): 978-1-4648-0506-6 DOI: 10.1596/978-1-4648-0505-9 Cover photo: Youth at a fruit and vegetable stand. © Arne Hoel / World Bank. Further permission required for reuse. Cover design: Debra Naylor, Naylor Design, Inc. Library of Congress Cataloging-in-Publication Data has been requested. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Contents Acknowledgments xi About the Authors xiii Abbreviations xv Overview 1 Introduction 1 Six Key Factors of the Out-of-School Youth Problem 2 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? 11 Entry Point 1: Retention of At-Risk Youth in School 12 Entry Point 2: Remediation through Formal or Alternative Education 14 Entry Point 3: Integration with the Labor Market 16 Notes 18 References 18 Chapter 1 Policy Challenges Associated with Out-of-School Youth in Sub-Saharan Africa 19 Introduction 19 Notes 22 References 23 Chapter 2 Why Do Sub-Saharan African Youth Drop Out of School? 25 Introduction 25 Magnitude of the Out-of-School Youth Population in Sub-Saharan Africa 26 Gender Disparities 33 Urban/Rural Distinction 38 Household Characteristics and Out-of-School Youth 40 School Characteristics 50 Notes 54 References 55 Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   v   vi Contents Chapter 3 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth 61 Introduction 61 Education Expenditure as a Percentage of GDP 62 Share of Secondary Education Expenditure in Total Public Education Expenditure 63 Gross Enrollment and Gross Enrollment Growth 64 Investment in School Infrastructure 66 Population Growth 67 Share of Wage and Salaried Workers and Vulnerable Employment 69 Notes 71 Chapter 4 Paths Back to Formal or Informal Education or to the Labor Force 73 Introduction 73 Entry Point 1: Retention of At-Risk Youth in School 76 Entry Point 2: Remediation through Formal or Alternative Education 83 Entry Point 3: Integration with the Labor Market 89 Notes 95 References 95 Chapter 5 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? 101 Introduction 101 Notes 109 Reference 109 Appendix A Summary of Programs Reviewed 111 Appendix B Data Sources 123 Appendix C Country Typologies 125 Appendix D Distributions of Out-of-School Youth for Select Sub-Saharan African Countries 129 Appendix E Sequential Logistic Regression Model 133 Appendix F Summary Results of Feda and Sakellariou’s (2013) Findings 137 Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Contents vii Figures O.1 Out-of-School Youth, by Country and Country Typologies 3 O.2 Gender and School Attendance 5 O.3 Geographic Location and School Attendance 6 O.4 Parental Education and School Attendance 7 O.5 Adult Employment and School Attendance 9 O.6 Household Income and School Attendance 10 O.7 Share of Youth Who Cite Low School Quality as the Main Reason for Dropping Out 11 O.8 Interaction between Key Factors and Entry Points 12 1.1 Out-of-School Youth in Sub-Saharan Africa: Estimated Number and Share, by Age Cohort 20 2.1 School Attendance Status of Youth Aged 12–24 Years, by Country 27 2.2 Share of Out-of-School Youth Aged 12–24 Years, by Country Typologies 28 2.3 Share of Out-of-School Youth, by Age Cohort and Country Typologies 29 2.4 School Transitions of 15- to 24-Year-Old Youth 31 2.5 Over-Age Enrollment and Dropout Rates 32 2.6 Gender and Schooling Outcomes for 12- to 14-Year-Old Youth 34 2.7 Gender and Schooling Outcomes for 15- to 24-Year-Old Youth 35 2.8 Gender Disparities: Percentage Difference in Share of Out-of-School Girls and Boys, by Country Typologies 36 2.9 Marginal Impact of Gender on School/Work Outcomes for Females Aged 15–24 Years, Compared with Males, by Country Typologies 37 2.10 Geographic Location and Schooling Outcomes for 12- to 14-Year-Old Youth 39 2.11 Odds Ratio at Various Transitions for Urban Youth Aged 15–24 Years, by Country Typologies 40 2.12 Impact of Head of Household’s Education Level on 12- to 14-Year-Old Youths’ Marginal Probability of Attending School, by Country Typologies 42 2.13 Odds Ratio at Various Transitions for Youth Aged 15–24 Years 43 2.14 Impact of Household Income and Income-Generating Capacity on 12- to 14-Year-Old Youths’ Marginal Probability of Attending School 44 2.15 Impact of Household Income and Income-Generating Capacity on 15- to 24-Year-Old Youths’ Odds Ratio at Various Transitions 45 2.16 Cost and Household Spending on Education 47 2.17 Impact of Various Household Economic Characteristics on the Marginal Probability of 15- to 24-Year-Old Youths’ Work/School Outcomes 49 Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 viii Contents 2.18 Distance from Primary Schools for Out-of-School Youth 51 2.19 Youth Who Cite Low School Quality as the Main Reason for Dropping Out 52 2.20 Repeaters in the Lower and Upper Secondary Cycles 53 3.1 Out-of-School Youth, Difference from Regional Mean, by Education Expenditure as a Share of GDP 62 3.2 Out-of-School Youth, Difference from Regional Mean, by Secondary Education Expenditure as a Share of Total Public Education Expenditure 64 3.3 Out-of-School Youth, Difference from Regional Mean, by Gross Enrollment Rate 65 3.4 Out-of-School Youth, Difference from Regional Mean, by Gross Enrollment Rate Growth 66 3.5 Out-of-School Youth, Difference from Regional Mean, by Share of Schools with Toilets 67 3.6 Out-of-School Youth, Difference from Regional Mean, by Population Growth Rate 68 3.7 Out-of-School Youth, Difference from Regional Mean, by Share of Wage and Salaried Workers 70 3.8 Out-of-School Youth, Difference from Regional Mean, by Share of Vulnerable Employment 71 4.1 Policy Interventions for Out-of-School Youth 74 5.1 Interaction between Key Factors and Entry Points 104 E.1 Educational Transitions and Associated Probabilities 134 Tables 2.1 Transition Years and Dropout Rates, by Grade 30 3.1 Education Expenditure as a Share of GDP and ­Out-of-School Youth 62 3.2 Secondary Education Expenditure as a Share of Total Public Education Expenditure and Out-of-School Youth 63 3.3 Gross Enrollment Rates and Out-of-School Youth 65 3.4 Gross Enrollment Rate Growth and Out-of-School Youth 65 3.5 Share of Schools with Toilets and Out-of-School Youth 67 3.6 Population Growth Rate and Out-of-School Youth 68 3.7 Share of Wage and Salaried Workers and Out-of-School Youth 69 3.8 Vulnerable Employment and Out-of-School Youth 70 4.1 Countries Grouped by Compulsory Education, by Level 77 4.2 Newer Cash Transfer Programs in Africa with Some Education Support Component 79 A.1 Summary of Programs Reviewed 112 B.1 Survey Year and Survey Source for Data on Sub-Saharan African Countries 123 C.1 Income Levels of Sub-Saharan African Countries 125 Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Contents ix C.2 Colonial Heritage of Sub-Saharan African Countries 126 C.3 Conflict History of Sub-Saharan African Countries 126 C.4 Income Level, Colonial Heritage, and Conflict History of Sub-Saharan African Countries 127 D.1 Distributions of Out-of-School Youth for Select Sub-Saharan African Countries 130 F.1 Regional Classification of 20 Countries Studied 137 F.2 Impact of Household Characteristics on Schooling Decisions of 12- to 14-Year-Old Youth 137 F.3 Determinants of School/Work Outcomes for 15- to 24-Year-Old Youth 138 F.4 Determinants of Schooling Outcomes at Transition Points for 15- to 24-Year-Old Youth 141 Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Acknowledgments The preparation of this study was led by Keiko Inoue and Emanuela di Gropello, under the overall guidance of Peter Materu and Sajita Bashir. The study was written by Keiko Inoue, Emanuela di Gropello, Yesim Sayin Taylor, and James ­ Gresham, with significant contributions from Chelsea Coffin, Kebede Feda, and Chris Sakellariou. Himdat Bayusuf, Juan Manuel Moreno, and Owen Ozier also provided technical contributions in the report’s early stages. This policy report aims to address two priority issues that have emerged for African policy makers: (a) the growing pressure to provide universal secondary education and the resultant trade-off between expansion, quality, and relevance; and (b) the shorter-term imperative of out-of-school youth. In both cases, it is fundamental to better understand the drivers and constraints to transitions between each level of education, as well as the demand for secondary education and skills in the current and developing labor market. This report and its ­ underpinning diagnostic work aim to deepen the analytical base and inform operational programs, policy dialogue, and development projects designed to tackle the out-of-school youth challenge. Cristian Aedo, Ernesto Cuadra, and Michel Welmond peer-reviewed the report. The team also thanks reviewers at earlier stages including Peter Darvas, Deon Filmer, Kirsten Majgaard, Ana Ruth Menezes, and Reehana Raza. Sukhdeep Brar, Jaap Bregman, Helen Craig, Linda English, and Margo Hoftijzer provided guidance at critical points in the report’s preparation. This report also benefited from the excellent editorial work of Amy Gautam. The study was made possible by financial support provided under the Norwegian Post-Primary Education Fund and the Multi-Donor Education and Skills Fund, both of which are administered by the World Bank. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   xi   About the Authors Keiko Inoue is a senior education specialist at the World Bank who has worked in more than 15 countries spanning East Asia, Africa, Latin America, and Europe and Central Asia. In recent years, she has focused on the promotion of skills development and positive youth development in low-income, middle-income, and conflict-affected countries. Her areas of research include analysis of learning outcomes as measured by the Organisation for Economic Co-operation and Development Programme for International Student Assessment in Albania and Kazakhstan; public finance review of the education sector in Albania and Liberia; and education country status reports for Liberia and Rwanda. This report was conceptualized and developed during her tenure as the co–team leader for the Secondary Education in Africa program. Emanuela di Gropello is a program leader in the Mali, Niger, and Chad Country Management Unit. She was previously a lead economist in the East Asia and Latin American Education Sector Units, where she worked and published exten- sively on governance, financing, and skills issues. She holds a doctorate in eco- nomics from the University of Oxford. Yesim Sayin Taylor is the director of fiscal and legislative affairs at the District of Columbia’s Office of the Chief Financial Officer. She leads the team that is responsible for assessing the impact of proposed legislation on the District of Columbia’s budget. She also testifies on and assesses proposals that affect the tax revenues of the District of Columbia. Yesim writes and publishes on public finance and tax policy, both for the District of Columbia and on issues that largely affect the developing world. Recently, she has investigated what makes a resident leave the District of Columbia—a location with a large transitional population, high living expenses, and relatively high income taxes. Her work showed that tax rates have little explanatory power in people’s decision to leave the city. In international development, she has written public expenditure reviews in education and health, performed project feasibility studies, and has most recently written on the human development challenges in Mali after the recent political crisis. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   xiii   xiv About the Authors James Gresham has consulted for the World Bank since 2008. He has worked with the World Bank’s education teams on country and regional programs in Africa, the Middle East and North Africa, Europe and Central Asia, and the World Bank Institute. He has experience with investment projects, technical assistance, and analytical work related to primary and secondary education and youth skills development. He is currently a doctoral student in the education policy program at The George Washington University. He holds a master’s degree in international affairs from the School of International Service at American University in Washington, DC. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Abbreviations ALOZ Adult Literacy Organization of Zimbabwe ASAMA Accelerated Compressed Learning for Malagasy Adolescents (Madagascar) BEUPA Basic Education for Urban Poverty Areas (Uganda) COBET Complementary Basic Education in Tanzania COPE Care of the Poor (Nigeria) DFID U.K. Department for International Development DHS Demographic and Health Survey GDP gross domestic product LRA Lord’s Resistance Army (Uganda) LSMS Living Standards Measurement Survey MICS Multiple Indicator Cluster Survey NGO nongovernmental organization SSA Sub-Saharan Africa STEP Skills Training and Entrepreneurship Program (Ghana) UNDP United Nations Development Programme UNESCO United Nations Educational, Scientific, and Cultural Organization USAID United States Agency for International Development Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   xv   Overview Introduction The economic and social prospects are daunting for the 89 million out-of-school youth who comprise nearly half of all youth in Sub-Saharan Africa. Within the next decade when this cohort becomes the core of the labor market, an esti- mated 40 million more youth will drop out, and will face an uncertain future without work and life skills. Their lack of work and life skills will impair these youth’s ability to get good jobs in desirable occupations, resulting in low and unstable incomes while exposing them to potentially long periods of ­ unemployment. The adverse effects of staying out of school will also be felt by the next generation, since these youth’s poor economic outcomes will hurt their ability to provide favorable opportunities for their own children. Societies at large will feel the impact: economic growth will be constrained, limiting the revenue-raising capacity of governments, while the need for public expenditures to support these youth, who will be more likely to rely on government health care, public welfare, or housing assistance, will expand. They will have shorter lives than their educated peers, will be more likely to become teen parents and to commit crimes, and will be less likely to raise healthy children, engage in civic activity, or vote or volunteer in their communities. With a growing cohort of out-of-school youth, the opportunities to address the compounding policy challenges are fleeting. While East Asian countries turned their youth bulges into an engine for growth, the same phenomenon spells a potentially explosive economic and social disaster for Sub-Saharan Africa. Efforts to address out-of-school youth issues must be cross-sectoral and driven by leadership at the highest levels. Yet the reality is that out-of- school youth are often “policy orphans,” positioned in a no man’s land with little data to develop an evidence-based advocacy framework, low implemen- tation capacity, lack of interest in long-term sustainability of programs, ­ insufficient funds, and no coordination across the different governmental entities—ministries of labor, education, and human services, among others— that carry partial responsibility for these youth. The international focus from development agencies, including the World Bank, is also fragmented. The continent is mired with youth programs that were launched with much fan- fare and then either fizzled when the desired outcomes were not immediately achieved or were abruptly terminated when funds (often external) ran out. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   1   2 Overview Thus, it is critical to elevate the issue to the highest political office with authority to allocate sufficient funds and human resources to design, imple- ment, evaluate, and then sustain interventions. To support the design of better policies that target youth, this report explores the overarching factors that contribute to the out-of-school youth problem, ­ specifically focusing on the 12- to 24-year-old cohort, which includes those most likely to drop out during the secondary cycle.1 There are strong (and not surprising) links between a country’s socioeconomic and demographic characteristics and the magnitude of its out-of-school youth population. These factors are explored in detail in this report. In addition, the incidence of out-of-school youth is lower in countries that spend a larger share of their gross domestic product (GDP) on education and devote a larger share of their public education resources to secondary education. Youth, especially younger youth, benefit from schools with adequate ­ facilities. Countries with high population growth rates also experience a higher incidence of out-of-school youth. Finally, the presence of strong formal labor markets and the availability of stable jobs entice more youth (or ­ their parents) to choose school over work. When a larger share of the labor force holds wage and salaried jobs, youth tend to attend and stay in school— a reminder that labor and education policies and the general business climate in a country are deeply connected. This report draws from three background papers as well as from an exten- sive review of the literature on programs and policies for out-of-school youth in Sub-Saharan Africa (see appendix A) as well as in other regions. The back- ground studies include an overview of basic indicators related to out-of- school youth, a diagnostic analysis of the magnitude and nature of the out-of-school youth problem in the region, and an econometric model of schooling from lower to higher education cycles. The first two studies utilize data from household surveys and labor market surveys conducted between 2006 and 2011 for 31 countries across Africa. The econometric study uses household survey data from 20 countries in the region (see appendixes E and F). Although the background studies did not have data for all countries in the region, data were available for a majority of countries that represent the region well. Six Key Factors of the Out-of-School Youth Problem This report teases out the factors that explain youth’s decisions to attend and stay in school and their school/work choices. It finds that policy makers must consider six key factors that characterize out-of-school youth: (a) most out- of-school youth drop out before secondary school; (b) early marriage is a key detriment to female youth’s education; (c) residing in a rural area for all youth increase the likelihood of being out of school; (d) parental education level and (e) the number of working adults are important household factors; Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 3 and (f) lack of school access and low educational quality are binding supply- side constraints. 1. A growing contingency of young people either never enroll or drop out secondary cycle. In countries where the incidence of out- before reaching the ­ of-school youth is high, a larger share of youth has never set foot in a school. More than half of youth between the ages of 12 and 24 years are out of school, and one in five has never been enrolled. The out-of-school problem is particularly widespread in low-income countries, Francophone countries (which tend to be low-income), and fragile or conflict-affected countries (see figure O.1 and appendix C). Figure O.1 Out-of-School Youth, by Country and Country Typologies a. School attendance status of youth aged 12–24 years, by country 100 80 60 Percent 40 20 0 er . oz am . m da Ni nda Lib ria Sw Ken a az ya Ca o, R ia ín e o, Rw ipe bi ia Gh ue s a Zi io o Gu we Be ea M an nin Ga dag nia rk M e Bu oon Se The o m ep M Z Rep ba ia a, r r a F li m ie a di d’ ad d eo i te Ch l Co Nam nd i Le an bi a ge Pr n r an L w Cô ega Et oth as in a er ng ib am b m p oi m asc in De an ge q To S M un b a za é rra la c ila Ni a Iv . Ug n h r T Bu ng Co o Sã Out of school, completed secondary Out of school, dropout Out of school, never attended In school Note: Figures do not add to 100 percent in some cases because of missing data. figure continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 4 Overview Figure O.1  Out-of-School Youth, by Country and Country Typologies (continued) b. Share of out-of-school youth aged 12–24 years, by country typologies Low income Lower middle income Upper middle income Anglophone Francophone Lusophone Conflict history No conflict history 0 10 20 30 40 50 60 Percent Out of school, never attended Out of school, dropout Sources: Estimates based on various household surveys. 2. Early marriage has detrimental effects on the educational outcomes of young women. Girls are already more likely to be excluded from primary education, and even when they do enroll in and complete primary school, they are much less likely to attend secondary school, especially in countries where school availability is very limited. Family attitudes and expectations influence girls’ schooling many years before they get married. Comparisons of outcomes for older youth (aged 15–24 years) show that married girls face very poor odds all along: 1:5 for finishing primary school and 3:5 for enrolling at the secondary level compared with boys of similar age (see figure O.2). Girls are more likely to be engaged in home care or work and less likely to be in school (with or without working at the same time), and again, the effects are much stronger for married girls. The impact of mar- riage is greatest in countries with stronger schooling o ­ pportunities. For example, among Anglophone countries, where school ­ systems are more flexible, married girls between the ages of 15 and 24 years are 30 ­ percent more likely to be engaged in work only and 36 percent less likely to go to school without working compared with boys of similar age. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 5 Figure O.2  Gender and School Attendance a. Probability of attending school for 12- to b. Odds ratios for various transitions, females 14-year-old girls compared with boys aged 15–24 years, compared with males 0 3.5 3.0 –2 2.5 –4 Odds ratio 2.0 Percent –6 1.5 1.0 –8 0.5 –10 0 0 Girls 2.0 Complete Enroll in Stay at Complete primary secondary secondary secondary Anglophone Lower income before age Lower middle income Lusophone 24 years No conflict history Francophone Female: never married Sub-Saharan Africa Conflict history Female: has married c. Marginal impact of gender on school/work outcomes for females aged 15–24 years, Anglophone countries 30 20 10 0 Percent –10 –20 –30 –40 Inactive Work Work and School only (or home only school care) Female: never married Female: has married Source: Feda and Sakellariou 2013. Note: Horizontal dashed line signifies that when the odds ratio is 1, probability for the groups being compared is the same. 3. Rural youth are more likely to have never attended or to have dropped out of school compared with urban youth (see figure O.3). Across the region, seven out of 10 rural youth have never attended school. Among the youngest cohort of 12- to 14-year-olds, urban youth are eight percent more likely to be in school than are rural youth, and this effect is greatest among countries with supply constraints (such as Francophone and conflict-affected countries). Urban youth Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 6 Overview complete primary school and enroll at the secondary level more often than rural youth, but once they enroll at the secondary level, the divergences between rural and urban youth diminish: urban youth are only slightly more likely to stay in school and complete secondary education by the age of 24 years. Urban youth also tend to be exclusively in school rather than mixing work and schooling and less likely to exclusively work or help in the house compared with rural youth. Figure O.3  Geographic Location and School Attendance a. Probability of attending school for b. Odds ratios for various transitions, 12- to 14-year-old urban youth urban youth aged 15–24 years compared with rural youth compared with rural youth 14 1.8 12 1.6 1.4 10 1.2 Odds ratio Percent 8 1.0 6 0.8 0.6 4 0.4 2 0.2 0 0 Urban Enroll in Complete Enroll in Complete primary primary secondary secondary Francophone Sub-Saharan Africa Urban Conflict history Anglophone Lower middle income Lower income No conflict history Lusophone c. Marginal impact of location on school/work outcomes for 15- to 24-year-old youth 15 10 5 Percent 0 –5 –10 –15 Inactive Work Work and School only (or home only school care) Urban Sub-Saharan Africa Source: Feda and Sakellariou 2013. Note: Horizontal dashed line signifies that when the odds ratio is 1, probability for the groups being compared is the same. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 7 4. Parents’ education level is the most important determinant of schooling choices. The analysis supports a story of “transmission” in which schooling and life outcomes are correlated across generations. When parents complete secondary education or achieve higher degrees, they are much more likely to keep their children in secondary school and push them to finish this cycle. Younger youth (aged 12–14 years) from households in which the head of household completed secondary education are 20 percent more likely to be in school compared with youth from households in which the head of household has little or no education (see figure O.4). This effect is strongest in Francophone countries (more than 25 percent improvement in the probability of attending school). Older youth in households in which the head of household has com- pleted at least secondary education are twice as likely to stay in and complete school before the age of 24 years in comparison with those whose parents have no education (see figure O.4). Figure O.4 Parental Education and School Attendance a. Impact of parental education on education b. Odds ratios for various transitions, decisions of 12- to 14-year-olds 15- to 24-year-old youth, compared with head of household with incomplete primary education 30 6.0 25 5.0 20 4.0 Odds ratio Percent 15 3.0 10 2.0 5 1.0 0 0 Primary Secondary Tertiary Complete Enroll in Stay at Complete primary secondary secondary secondary Francophone Anglophone before age Sub-Saharan Africa Lusophone 24 years Head of household primary Head of household secondary Head of household tertiary Source: Feda and Sakellariou 2013. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 8 Overview 5. After controlling for parental education levels, household earning capacity (measured as the number of working adults)—not household income level— ­ has the greatest impact on both schooling choices and on school/work decisions. For younger youth, having even one working adult in the house- ­ hold increases the probability of attending school by 14 percent, and the impact of having two working adults is 21 percent (see figure O.5). Earning capacity matters the most in countries with relatively good schooling ­ outcomes—such as lower-middle-income countries and Anglophone countries (this is in contrast to income effects, which are greatest in supply- ­ constrained ­countries). Similarly, in a household with working adults, youth aged 15–24 years are much more likely to complete primary school, enroll at the secondary level, and stay in school: one working adult produces an odds ratio of 2:1 for staying the course at the secondary level. A household’s earning capacity also determines youth’s priorities: in a household with working adults, youth tend to focus on school rather than on work. Having one or two working adults (compared with none) decreases youth’s proba- bility of working instead of going to school, by 15–18 percent, and increases the probability of solely attending school without work obligations by 14–15 percent. This latter effect can be as high as 30 percent in the southern countries of Sub-Saharan Africa. This finding once again underscores the strong links between labor markets and education. Not only can families with working adults afford to send their children to school, but integration with labor markets could act as a constant reminder of the importance of skills development (figure O.5). This is not to say that household income level does not matter; it is still important, but more for some groups than for others. For 12- to 14-year- olds, the impacts of income are greatest for middle-income households and most ­ pronounced for countries where education outcomes are poor, such as Francophone and low-income countries (see figure O.6). For the older cohort of 15- to 24-year-olds, the effects of income on the decision to enroll in secondary school are substantial, but the importance of income subsides in the subsequent decision to stay and complete secondary school before the age of 24 years. In this last transition in the secondary cycle, no discern- ible differences exist for youth who come from the poorest households and youth from households with incomes in the second and third quintiles (figure O.6). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 9 Figure O.5  Adult Employment and School Attendance a. Number of working adults employed and probability b. Odds ratios at various transitions of attending school for 12- to 14-year-olds for 15- to 24-year-old youth (baseline: households with no working adults) (baseline: households with no working adults) 30 2.5 25 2.0 20 Odds ratio 1.5 Percent 15 1.0 10 5 0.5 0 0 One employed Two employed Three or more Complete Enroll in Stay at Complete household household employed primary secondary secondary secondary member 25+ member 25+ household before age years of age years of age member 25+ 24 years years of age One employed household member 25+ years of age Anglophone Francophone Two employed household member Lower middle income Lusophone 25+ years of age Lower income Conflict history Three or more employed household No conflict history member 25+ years of age c. Income-generating capacity (baseline: households with no working adults) 20 15 10 Percent 5 0 –5 –10 –15 –20 Inactive Work only Work and school School only (or home care) One employed household member 25+ years of age Two employed household member 25+ years of age Three or more employed household member 25+ years of age Source: Feda and Sakellariou 2013. Note: Horizontal dashed line signifies that when the odds ratio is 1, probability for the groups being compared is the same. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 10 Overview Figure O.6  Household Income and School Attendance a. Odds ratios at various b. Household income level and schooling transitions for youth aged 15–24 years outcomes for 12- to 14-year-olds (baseline: poorest quintile) (baseline: poorest quintile) 2.0 16 1.8 14 1.6 12 1.4 10 Odds ratio 1.2 Percent 8 1.0 6 0.8 0.6 4 0.4 2 0.2 0 0 –2 Complete Enroll in Stay at Complete Q2 Q3 Q4 Q5 primary secondary secondary secondary before age 24 years Lusophone Francophone HH income Q5 Lower income Lower middle HH income Q4 No conflict history income HH income Q3 Anglophone Conflict history HH income Q2 Source: Feda and Sakellariou 2013. Note: HH = household; Q = quintile. Horizontal dashed line signifies that when the odds ratio is 1, probability for the groups being compared is the same. 6. Lack of schooling and poor-quality schools amplify the impact of demand- side constraints on schooling choices. Access is a key problem for poor and rural communities, where barriers such as distance and entrance require- ments or exit exams are often binding. In rural areas, where secondary schools are far away, parents do not send children or girls to school because they fear for their safety. It is also common for parents to take children out of primary school with the expectation that they won’t have access to sec- ondary schooling. Households also factor quality into their schooling deci- sions, especially as youth get older (see figure O.7). Low school quality and lack of motivation at school act as push factors by hindering student achievement and progress. Administrative features of schools, such as a lack of clear and rigorous school goals, amplify such push factors: youth who drop out perceive teachers to be less interested in them and view school discipline as ineffective and inequitably applied. Similarly, youth are more likely to drop out when teachers are regularly absent or when schools are unresponsive to local needs and preferences (such as religious beliefs), are unaccountable to parents or students, or lack basic infrastructure (such as drinking water and bathrooms) (figure O.7). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 11 Figure O.7 Share of Youth Who Cite Low School Quality as the Main Reason for Dropping Out 30 25 20 Percent 15 10 5 0 ia e a da da a ea o ria r i ad ge aw ny bi Th as op n an an ge m aF Ch Ke Ni i al a, Gu hi Rw Ug Za Ni M bi in Et m rk Ga Bu Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: Gresham 2013. What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? Programs for out-of-school youth must address three policy areas to have an impact: retention, remediation, and integration. Retaining at-risk youth in the education system requires programs that address a mix of demand- and supply- side obstacles. Improving the access to and quality of education systems, while providing targeted scholarships and financial aid packages or cash transfers, are key ingredients for effective retention programs. Remediation requires identify- ing the youth most likely to successfully complete second chance programs and then designing programs that enhance their cognitive, technical, and life skills. Integration programs should prepare youth for Sub-Saharan Africa’s largely informal labor markets and thus should be targeted, decentralized, and coordi- nated across layers of implementing levels. No simple policy solutions exist for the dropout problem and the retention, remediation, and integration challenges. On the demand side, the characteristics that seem to matter most for schooling choice (i.e., early marriage or rural resi- dence, low education levels of parents, and parents’ inability to find work) are indicative of other social problems and cannot be addressed by policies designed in isolation to support out-of-school youth. On the supply side, providing sec- ondary education is costly, and alternative education and workforce training programs face tough competition from other policy initiatives for government and donor funding. Targeted and multisectoral programs that address the most pertinent demand- and supply-side constraints in a given country are likely to be the most effective. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 12 Overview Figure O.8 Interaction between Key Factors and Entry Points Entry points 1. 2. 3. Retain Remediate Integrate at-risk through with the youth in formal or labor 1. Drop out before secondary school school alternative market education 2. Early marriage for girls 3. Rural youth Key factors 4. Parental education/attitudes Policy world 5. Family income/earning capacity 6. Lack of access/poor quality of schools Policy discussions on out-of-school youth are framed by the six key factors that characterize out-of-school youth along with the three entry points of reten- tion, remediation, and integration (see figure O.8). This framework allows for making recommendations on which policy path may be preferable for a given subgroup, across different typologies of countries that generally have similar supply-side problems. It also provides a means for discussing short-term policies that are necessary to prepare current youth for work as well as longer-term poli- cies that focus on broader enrollments and lower dropout rates. These recom- mendations are not meant to be prescriptive, but should be read as summary findings of the regional analysis loosely mapped to various interventions reviewed in this report. Entry Point 1: Retention of At-Risk Youth in School For youth in school, the most immediate policy intervention is preventing at-risk youth from dropping out in the first place. Given that most youth drop out before they start secondary school, retention efforts must begin before youth enter the secondary cycle. Early intervention is likely to be cost-effective, since at the secondary level the factors that drive students to drop out are even stronger because academic gaps are bigger, income constraints are amplified (older youth’s labor is worth more), family demands on youth (such as work and mar- riage) are greater, and supply is scarcer while the costs of providing secondary schooling are higher than those of primary schooling. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 13 The main challenge for developing retention policies is determining the right mix. Policies that address one underlying cause of exclusion could exacerbate others. For example, compulsory or free education at the secondary level indeed helped increase initial enrollments in the region, but also resulted in declining quality and deteriorating academic standards, which have acted as push factors. Scholarship and financial aid programs should be paired with education of families and young people on the potential benefits of completing secondary education. To retain students who are interested in learning but cannot afford it, cash incentives or subsidies should be the main policy tools. The simplest way of ­ providing financial stability to households is cash transfers, with either soft or hard conditions. Experience with such cash transfer programs shows that they make an immediate and important difference in schooling outcomes. Unconditional transfers are more effective for middle-income families, espe- cially in countries where availability of schooling is very limited. Expanding unconditional cash transfers to relatively better-off families may be politically difficult (since these programs compete with other social welfare and education initiatives), but general subsidies to education in the form of reduced fees may make a big difference.2 Rather than limiting these benefits to the poorest house- holds, where many factors other than income may affect schooling decisions,3 governments should experiment with expanding these benefits to households all the way up to middle-income families. Furthermore, paying families for pri- vate school tuition might be the most expedient and cost-effective way of expanding secondary education in Sub-Saharan Africa, where private education is growing rapidly. To improve retention, governments must also focus on low-cost interven- tions to improve quality (for example, through remedial support) and infra- structure (through low-cost improvements) as an immediate step. Households report that quality of schooling plays an important role in the schooling deci- sions of older youth, but for younger youth and during earlier transitions, increasing teacher attentiveness and providing remedial support may reduce dropout rates. Teacher mentorship motivates youth to stay in school, while teachers with low academic expectations and teachers who dismiss low-­ achieving students push students out. Community-based teacher training and programs that increase parental involvement in school management, such as those in Mali and Uganda, can help increase teacher awareness. This outcome is seen over and over again in simple interventions on the continent; for exam- ple, the Girls’ Education Project in Eritrea reports improved grade promotion as a result of tutoring. In the longer run, early intervention, especially through early child develop- ment programs, may be the most cost-effective preventive measure to reduce the out-of-school youth population. Children who begin schooling later than they should are more likely to drop out, but, of course, over-age enrollment itself is a manifestation of other demand-side constraints such as poverty and low interest in education. Early child development programs are shown to reduce later Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 14 Overview incidences of dropping out, mainly by assuring that children enroll in school when they are supposed to and that parents are motivated to choose schooling over work or marriage (for girls). Entry Point 2: Remediation through Formal or Alternative Education For youth already out of school, the most likely path to complete their education is alternative education systems such as equivalency programs. Successful alterna- tive education programs are those with multiple entry and exit points and close associations with formal education. The two most important factors in imple- menting nationwide alternative education programs are the coordination between national government and subnational entities (regional authorities, communities, local governments, or other stakeholders) and the availability of sustainable funding. Small-scale alternative education schemes that target out-of-school youth ­ are especially successful when they mix cognitive and technical skills development with life skills training as well as mentoring. But survival concerns such as hunger or homelessness make it hard for marginalized youth to take advantage of skills- training programs. To overcome these barriers, life-skills programs must include components such as provision of daycare for youth with children, stipends to cover transportation and basic costs, and health care to ensure youth are physically and mentally fit. This can only be done with cross-sectoral support across govern- ment entities. Finally, many remediation programs disappear either because they run out of money or cannot compel their target youth to attend or complete their offerings. Rigorous evaluations are necessary, first to collect more information on the profile of youth who are most likely to complete remediation programs, and second, to provide ways to make these programs more cost-effective over time. For youth who were excluded from education before the secondary level, second chance programs can bridge the gap. Across the region, an estimated 33 percent of youth never make it to the secondary level; the shares are particu- larly high among Francophone countries (57 percent) and fragile countries or countries with a history of conflict (48 percent). Second-chance programs can achieve rapid results for out-of-school youth and encourage them to continue with formal education at the secondary level. Youth who never attended school may be easily persuaded to do so if they come across the right program. Madagascar, Uganda, and Zimbabwe have implemented successful second chance programs with compressed curricula. These second chance programs are desperately needed in low-income countries, Francophone countries, and coun- tries with conflict histories, where exclusion rates are very high. In the longer term, eliminating the secondary school enrollment bottleneck is a necessary condition for reducing the exclusion problem. Even for younger children who complete primary education, secondary enrollment can remain a challenge due to the lack of high-quality schools. Countries that have instituted compulsory education at the secondary level have achieved significant enroll- ment increases, but most times at the expense of quality. As an intermediary step, expanding the primary cycle by a few years might curb the deterioration in Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 15 quality while lowering dropout rates. This will allow countries to take advantage of the physical infrastructure and teaching force already in place at the primary level, especially in rural areas. It would also push the natural exit point a few years into the future. Whether improving retention or providing incentives to return to education, all programs must involve and educate parents, especially those of younger youth. Parental education and parents’ attitudes toward education are the most important determinants of school and work outcomes for youth, especially for earlier transitions in and out of the primary cycle. When parents themselves have poor economic opportunities, they tend to discount the importance of education. The effect of parental education on secondary education is strongest in countries where schooling outcomes are better, such as lower-middle-income and Anglophone countries, where parents are more likely to have completed second- ary education. Mali, Guinea, and Uganda have implemented smaller-scale parent education programs that also require participation in school management. These programs, which take advantage of the culture of mutual support in such coun- tries, report higher levels of parental involvement in school management; increases in school attendance for youth and in civic participation for adults; and decreases in dropout rates, reports of domestic violence, and early marriage. Parental involvement in school management can also lead to small but significant improvements in school infrastructure and facilities. For example, in Guinea, the Community Participation in Education for Equity and Quality Program enlisted parents’ help in simple investments such as adding latrines, with positive impacts on girls’ enrollment levels. While parental attitudes matter in earlier transitions, youth’s attitudes toward education play a larger role later, especially when it comes to completing second- ary education. Many youth drop out of secondary education because they are expected to support themselves and their families. In addition, many young people in the region engage in risky sexual behavior, which often results in early parent- hood. Finally, youth may not fully understand the potential benefits from educa- tion because they do not see many good examples, perhaps because they live in rural areas, slums, or conflict-affected regions. Mentoring programs that go beyond academic or work-skills training, and emphasize non-cognitive skills have success- fully helped such at-risk youth. For example, South Africa’s USIKO Program recruits male volunteers specifically to mentor young men, to encourage school attendance and discourage risky behavior such as gang activities. Since 2000, more than 600 at-risk youth have successfully completed the mentorship program, and of this group, more than 90 percent obtained high school qualifications. Related to attitudes toward education is the differential treatment of boys and girls when it comes to schooling. Programs focused on girls, especially girls who are likely to marry, must provide economic incentives for them to stay in school. Across the region, girls are much less likely to enroll in school compared with boys, and married girls have the worst outcomes—they most likely never enrolled in primary school or dropped out. But once enrolled at the secondary level, girls, especially married girls, do significantly better: they are three times more likely Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 16 Overview than boys to graduate from secondary school. This suggests that investing in getting girls to enroll and complete primary education has a high payoff. For ­ younger girls, targeted subsidies, accessible only if parents or the girls themselves change their attitudes toward education, may be the best intervention tool. For older girls, however, staying in school may not be possible or viable, and the appropriate alternative education and workforce development programs must provide strong linkages to work outcomes, satisfying the general expectation that these young women must provide for their families. Alternative education programs for older youth must recognize that work is a part of youth’s lives in the region. Children in Sub-Saharan Africa start working at very young ages, either at home or elsewhere, sometimes for a salary and some- times not. Youth from poorer households are more likely to work and less likely to attend school full-time, and these effects are strongest among low-income countries. For working youth, alternative education programs that recognize that youth are expected to work and provide for themselves and their families have proved to be successful. Such programs include Sierra Leone’s training courses for ex-combatant and other war-affected youth, the Republic of South Sudan’s alter- native education courses for out-of-school youth in conflict areas, and Kenya’s experiment with vouchers for technical and vocational training. Work constraints are particularly binding for programs that target rural youth. In every country in the region, rural youth are more likely to be out of school compared with urban youth, and more likely to have never attended school. This is particularly true in low-income and Francophone countries such as Mali, Burkina Faso, and Senegal. Geographic location has the highest impact when youth enroll at and complete the primary level, but makes little difference at later transitions. Rural youth are also more likely to work and less likely to attend school full time. While lack of availability of formal schooling may limit rural youth’s schooling outcomes, rural youth would still benefit from programs that recognize their work obligations. Some examples of these programs can be found in Ghana and Ethiopia. Entry Point 3: Integration with the Labor Market Youth who are not likely to go back to school require practical training and expe- rience to increase their employability. Given that a large share of the economy remains informal in Sub-Saharan Africa, informal apprenticeships are the key mechanism for out-of-school youth to learn skills and find employment, but they are difficult to scale up. Seventeen African countries earmark funds for national training, but the bulk of these funds go to train those who cannot find jobs when they complete their formal education. Large-scale national programs that intend to train unemployed and undereducated youth are not very successful. Wage subsidies and large public works projects can support large groups of youth but rarely produce long-lasting employability. A handful of programs follow the examples set by the Latin American Jovenes programs, which are targeted, decentralized, and coordinated across the state, civilian entities, and the Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Overview 17 private sector. Just like alternative education schemes, workforce development programs are most effective when they are comprehensive, thus meeting a myriad of youth needs such as provision of a range of skills development, ­ business training, and microfinance programs. Workforce development programs must consider that most youth will be self-employed or work for a small, informal enterprise. Many youth in the region ­ drop out of school because they are unlikely to improve their employability by completing middle or high school. Because formal work opportunities are rare, formal education carries less intrinsic value and parents often do not send their children to school, as in Zambia, Uganda, Kenya, and rural Ghana, because they think education will not lead to a job. In the region, programs that recognize the difficulty in finding wage and salaried work in a formal establishment have dem- onstrated some success, but the biggest constraint has been the lack of a legal framework defining how financial institutions can lend to youth. Even when they have ideas worth investing in, youth who want to start their own business are left to their own devices, relying on their friends or parents, or resorting to stealing or prostitution. In the region, countries that do better in terms of keeping their youth in school also try to push multiple paths to alternative degrees or jobs for out-of- school youth.4 Providing multiple pathways for youth is important, but diversi- fied programs complicate resource allocation and coordination decisions harder within the country and across the donor base. Needless to say, the challenge is much bigger for low-income countries: without a strong formal education system with which to form close ties, developing a nonformal alternative education program is more challenging. While resource problems are hard to solve, a review of existing programs in Sub-Saharan Africa points to two overarching changes in managing existing resources. First, countries must invest in developing the capacity of and coordina- tion among national governments, subnational governments, and communities. The success of alternative education and workforce development programs increases when training areas and venues are not limited; when the relations between national and subnational governments are strong; when the division of labor is clear; and when the program design is transparent. Uganda and Kenya are experimenting with programs similar to Latin America’s Jovenes, with encouraging results. Second, countries must leverage donor funds through better coordination. This is indeed another capacity problem since in the absence of a ­ national framework or vision, donor-funded programs and interventions remain fragmented. Nongovernmental organizations (NGOs) and nonprofit institutions run a myriad of workforce development programs that provide a mixture of training, on-the-job experience, life skills counseling, and mentoring. These pro- grams are generally successful in integrating out-of-school youth into labor mar- kets, but their effectiveness is limited by their size and funding source; some of these programs cannot expand beyond a certain point before they do not have the resources, others lose resources when donors move on to other things. Because many of the countries in the region share similar challenges with Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 18 Overview out-of-school youth, regional goals that transcend borders can help focus donors and create longer-term commitments. Most important, the act of dropping out of school itself signals a complex set of problems, and for this reason, out-of-school youth do not fit neatly into a single policy area. Sometimes programs for out-of-school youth fall under education policy, and other times under workforce development, social develop- ­ ment, and welfare, whereas the appropriate policy mix will have elements of all. This fragmented policy space is further complicated by the challenge presented by the age group. Out-of-school youth (aged 12–24 years), unlike younger children, fend for themselves or their families, and their behavior is not always ­ easy to change or modify. This leaves many existing programs disjointed and inef- fective, and serves as a constant reminder of the need for coordination and focus at the highest levels of the policy world. Notes 1. The focus is on youth at the secondary level because with low enrollment, high repeti- tion, and high dropout rates, secondary education presents the biggest bottleneck toward completing a formal degree. Given the magnitude of the out-of-school youth population, rapid expansion and greater availability of formal secondary education are unlikely to help youth in this cohort, who instead must take alternative paths to improve their employability. 2. Secondary school fees account for a much larger share of household expenditures— up to 60 percent in Mauritania and 70 percent in Rwanda, and for older youth, more than half of households rank cost the primary reason for not attending or for dropping out of school. Elimination of school fees in Cameroon, Kenya, Lesotho, Malawi, Uganda, Tanzania, and Zambia has typically led to significant increases in enrollment. 3. Unconditional cash transfers may not be as effective in the poorest households if parental attitudes reflect constraints not related to income but to expectations for youth (such as marriage). Similarly, the usefulness of funding schools, public or pri- vate, would be limited for those families who live far away from schools. 4. The best examples from the region in this regard are Uganda, Kenya, and Ghana. References Feda, Kebede, and Chris Sakellariou. 2013. “Out of School, School-Work Outcomes and Education Transitions of Youth in Sub-Saharan Africa—A Diagnostic.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Gresham, James. 2013. “Out-of-School Youth in Africa—Diagnostic Note.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Cha p t e r 1 Policy Challenges Associated with Out-of-School Youth in Sub-Saharan Africa Introduction An estimated 89 million young people aged 12–24 years—nearly half of all youth in Sub-Saharan Africa—are out of school (figure 1.1).1 If current dropout trends continue, a conservatively estimated 40 million more of their peers will drop out of school over the next decade,2 just as these young people become the core of the labor force. Out-of-school youth are likely to face a future without the ben- efits of proper credentials, and in most cases, work and life skills. Their lack of skills will impair these youth’s ability to get good jobs in desirable occupations, leaving them with low and unstable incomes, long periods of unemployment, and a low quality of life (Sum et al. 2007). The adverse effects of staying out of school will also be felt by the next generation, since these youth’s poor economic out- comes will hurt their ability to provide favorable educational opportunities for their own children. A meager labor market potential for such a large share of youth will have significant economic and social costs. Less will be produced and the economy will settle on a lower path of growth (Heckman and Masterov 2007). A weaker economy will constrain the revenue-raising capacity of governments, while the need for public expenditures to support these youth will be higher (Thornberry, Moore, and Christenson 1985). If they follow the documented path of school dropouts elsewhere, these youth will be more likely to commit crimes (Raphael 2004), to rely on government health care (Muennig 2005), or to rely on public welfare or housing assistance (Waldfogel, Garfinkel, and Kelly 2005). They will have shorter lives than high school graduates (Muennig 2005), will be more likely to become teen parents and less likely to raise healthier children (Haveman, Wolfe, and Spaulding 1991). Their children will be far less likely to graduate from high school compared with children of parents with high school degrees (Wolfe and Haveman 2002). Furthermore, youth without viable future Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   19   20 Policy Challenges Associated with Out-of-School Youth in Sub-Saharan Africa Figure 1.1 Out-of-School Youth in Sub-Saharan Africa: Estimated Number and Share, by Age Cohort a. Estimated number 100 90 19 million Out-of-school youth (in millions) 80 15- to 24-year-olds dropped out of secondary 70 60 50 55 million 15- to 24-year-olds 40 dropped out before 30 secondary 20 10 15 million 12- to 14-year-olds out of school 0 b. Share, by age cohort 100 80 Share in the cohort population (%) 60 40 20 0 Ages 12 to 14 years Ages 15 to 18 years Ages 19 to 24 years Completed secondary Out of school In school Source: Estimates from Feda and Sakellariou (2013), based on various household surveys. options in the labor market upon completion of their education may be more likely to find themselves working at very young ages (Kahraman 2011). Finally, these youth will be less likely to be engaged in civic activity, vote, or volunteer in their communities (Junn 2005)—activities that lead to better social outcomes (Levin 2005). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Policy Challenges Associated with Out-of-School Youth in Sub-Saharan Africa 21 This report provides a comprehensive review of the state of out-of-school youth in Sub-Saharan Africa, the determinants of school and work outcomes for youth, and the policy initiatives in place to support these youth. The goals of the report are to explore the overarching factors that contribute to the out-of-school youth phenomenon and to make recommendations for improving existing poli- cies and programs. The report focuses on the 12- to 24-year-old cohort, covering youth who are most likely to drop out through the secondary cycle. This focus is chosen because with low enrollment, high repetition, and high dropout rates, secondary education presents the biggest bottleneck toward completing a formal degree. Given the magnitude of the out-of-school youth population, youth in this cohort are unlikely to benefit from rapid expansion and greater availability of formal secondary education, and must instead take alternative paths to improve their employability. In addition, policies in place to strengthen second- ary enrollment or support out-of-school youth who drop out of secondary school are particularly weak across the region, either because of resource constraints or poor program design. And the challenges policy makers face in designing effec- tive programs for out-of-school youth are further complicated by Sub-Saharan Africa’s growing youth population.3 Even if countries were to expand resources available for secondary education, the fiscal burden of supporting the education of youth will continue to increase, and without major reforms, just adding more funds might be sufficient to prevent possible further deterioration in the levels or shares of out-of-school youth. The report covers three main areas, as follows. First, the report reviews the state of out-of-school youth in Sub-Saharan Africa, looking for overarching trends. Household characteristics, traits of youth, social norms and expectations, and characteristics of the school system all matter in understanding why youth drop out and then stay out of school. The diagnostic analysis uses household sur- vey data from 31 countries in the region as well as other regional data on national and regional socioeconomic characteristics. It shows that youth who drop out of school are indeed qualitatively different, both in their demographic characteris- tics and their access to schooling, from those who stay in school (Gresham 2013). These youth tend to come from poorer households with fewer working or edu- cated adults, live in rural areas, and are more likely to be female. In the region, a significant number of youth have never attended school, especially in many Francophone, low-income, and fragile or conflict-affected countries. Among those who do continue with secondary education, dropout rates are especially high between the ages of 15 and 18 years, when youth tend to seek work. That said, many youth, especially urban youth, leave school but remain jobless. Research shows that out-of-school youth consider their prospects in the labor market when deciding whether and when they attempt to go back to school or training. The decision path youth take as they progress from primary to lower second- ary to upper secondary school is also explored, specifically focusing on household characteristics that explain these decisions. In examining this decision path, this report relies on Feda and Sakellariou’s (2013) analysis, which utilizes household survey data from 20 countries to explore why youth drop out of school. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 22 Policy Challenges Associated with Out-of-School Youth in Sub-Saharan Africa Second, the report looks at the relationship between countries’ socioeco- nomic and demographic characteristics and the magnitude of their out-of-school youth population. The incidence of out-of-school youth is lower in countries that spend a larger share of their gross domestic product (GDP) on education and devote a larger share of their public education resources to secondary educa- tion. Youth, especially younger youth, benefit from schools with adequate school facilities. Countries with high population growth rates also experience a higher incidence of out-of-school youth. Finally, formal labor markets entice more youth (or their parents) to choose school over work. When a larger share of the labor force holds wage and salaried jobs, youth tend to attend and stay in school. Third, the report reviews the policies and programs in place for out-of-school youth. The review shows that programs in Sub-Saharan Africa are organized around three entry points: retention of youth in school, remediation through formal or alternative education programs, and integration into the labor market. No simple policy solutions exist to the dropout problem and the reenrollment challenge. Even when the overarching reasons for dropping out are well known, designing programs to retain students in school or marshal them into alternative paths is notoriously difficult. Comprehensive and multisectoral programs seem to have the most effect, but any program must be designed with a rigorous evalu- ation in mind so as to gain from lessons learned as the program is implemented and adapted. The rest of the report is organized as follows: Chapter 2 provides an overview of the state of out-of-school youth and related characteristics across the region, including the magnitude of the out-of-school youth population and the over- arching characteristics of the household, school system, and macro variables cor- related with youth’s decisions to stay out of school. This chapter also explores the determinants of the school/work and school/dropout decisions across various transitions (primary to lower secondary to upper secondary), combining the review of literature of the determinants of schooling outcomes with a diagnos- tic study of the region, and an econometric analysis of household data from 20 countries in the region. Chapter 3 provides interregional comparisons of schooling outcomes for youth across various economic and demographic dimen- sions. Chapter 4 reviews the policies and programs in place in Sub-Saharan Africa, linking intervention tools to the underlying causes of dropping out, such as household constraints, youth behavior, and supply-side problems. Chapter 5 concludes with policy recommendations for the region in general and for sub- groups of countries that share similar characteristics. The policy discussion is framed by the determinants of schooling choices and the entry points for at-risk and out-of-school youth, focusing on both short- and long-term options. Notes 1. This regional estimate is projected based on analysis of household survey data from 20 countries in the region, including (survey years in parentheses): Burkina Faso (2008/09), Cameroon (2007), Chad (2011), Côte d’Ivoire (2008), Ethiopia (2004/05), Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Policy Challenges Associated with Out-of-School Youth in Sub-Saharan Africa 23 The Gambia (2009), Ghana (2005/06), Guinea (2012), Kenya (2004/05), Malawi (2010/11), Mozambique (2008/09), Niger (2007), Nigeria (2010), Rwanda (2005/06), São Tomé and Príncipe (2010), Senegal (2005), Sierra Leone (2003), Tanzania (2010), Uganda (2010), and Zambia (2010). 2. This is an extrapolation based on the estimated dropout rates among youth under the age of 14 in the 20 countries for which data are available. 3. According to the UNESCO Institute for Statistics, the secondary school–age population grew by 9 percent between 2006 and 2010 in Sub-Saharan Africa; during that period, the same population declined by 1 percent worldwide. While the population is very slowly aging in the region, the African youth population is still projected to grow by 1.6 percent in 2024, or seven times the rate in the rest of the world. References Feda, Kebede, and Chris Sakellariou. 2013. “Out of School, School-Work Outcomes and Education Transitions of Youth in Sub-Saharan Africa—A Diagnostic.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Gresham, James. 2013. “Out-of-School Youth in Africa—Diagnostic Note.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Haveman, Robert, Barbara Wolfe, and James Spaulding. 1991. “Childhood Events and Circumstances Influencing High School Completion.” Demography 28 (1): 133–57. Heckman, James J., and Dimitriy V. Masterov. 2007. “The Productivity Argument for Investing in Young Children.” Paper presented as the T. W. Schultz Award Lecture at the Allied Social Sciences Association Annual Meeting, Chicago, January 5–7. Junn, Jane. 2005. “The Political Costs of Unequal Education.” Paper prepared for the Symposium on Social Costs of Inadequate Education, Teachers College, Columbia University, New York, October 24–25. Kahraman, Berna. 2011. “Youth Employment and Unemployment in Developing Countries: Macro Challenges with Micro Perspectives.” Graduate Doctoral Dissertations, Paper 36, University of Massachusetts, Boston. Levin, Henry M. 2005. “The Social Costs of Inadequate Education.” Paper prepared for the Symposium on Social Costs of Inadequate Education, Teachers College, Columbia University, New York, October 24–25. Muennig, Peter. 2005. “The Economic Value of Health Gains Associated with Education Interventions.” Paper prepared for the Symposium on Social Costs of Inadequate Education, Teachers College, Columbia University, New York, October 24–25. Raphael, Steven. 2004. “The Socioeconomic Status of Black Males: The Increasing Importance of Incarceration.” Working Paper, Goldman School of Public Policy, University of California, Berkeley. Sum, Andrew, Ishwar Khatiwada, Joseph McLaughlin, and Paulo Tobar. 2007. “An Assessment of the Labor Market, Income, Health, Social, Civic and Fiscal Consequences of Dropping Out of High School: Findings for Massachusetts Adults in the 21st Century.” Center for Labor Market Studies Publications Paper 2, Northeastern University, Boston. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 24 Policy Challenges Associated with Out-of-School Youth in Sub-Saharan Africa Thornberry, Terence P., Melanie Moore, and R. L. Christenson. 1985. “The Effect of Dropping Out of High School on Subsequent Criminal Behavior.” Criminology 23 (1): 3–18. Waldfogel, Jane, Irwin Garfinkel, and Brendan Kelly. 2005. “Public Assistance Programs: How Much Could Be Saved with Improved Education?” Paper prepared for the Symposium on Social Costs of Inadequate Education, Teachers College, Columbia University, New York, October 24–25. Wolfe, Barbara L., and Robert H. Haveman. 2002. “Education in the 21st Century: Meeting the Challenges of a Changing World.” Paper prepared for Conference Series 47, Federal Reserve Bank of Boston, Boston. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Cha p t e r 2 Why Do Sub-Saharan African Youth Drop Out of School? Introduction Chapter 2 explores the magnitude and nature of the out-of-school youth population and the reasons why youth drop out of school. The analyses show the ­ following: • The out-of-school youth problem is widespread in Sub-Saharan Africa, with more than half of all youth between the ages of 12 and 24 years out of school. • The out-of-school problem is particularly large in low-income countries, Francophone countries, and fragile or conflict-affected countries. • Most youth drop out before the beginning of the secondary cycle. In countries where the incidence of out-of-school youth is high, a larger share of youth has never attended school. • Young girls, rural youth, and older youth are more likely to be out of school. Poverty amplifies these effects. Gender and urban-rural disparities and the gap in school attendance between the poorest and richest households increase with age. • Early marriage has a detrimental effect on the educational outcomes of young women, but married young women, if they can make it to secondary school, are more likely to stay in school and finish, compared with their male counterparts. • Parental attitudes toward education and households’ income-earning potential are the two most important determinants of education outcomes: youth from households where the head of the household is more educated are less likely to work and more likely to attend school. Similarly, the more working adults in a household, the more likely youth are to focus on school only and to continue to upper secondary education. • When one controls for the education level of the head of household, the explan- atory power of household income decreases, but remains significant. Incremental Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   25   26 Why Do Sub-Saharan African Youth Drop Out of School? increases in household income have a much greater impact on school/work outcomes in low-income countries and those with a history of conflict. • School/work outcomes for youth are very sensitive to the household’s s ­ ector of employment and the income-generating capacity of the household measured as the number of working adults). Youth from households engaged (­ in agriculture are less likely to focus just on school and more likely to work only, or to juggle work and school. Youth from households with working adults are more likely to be focused on school only. This chapter draws from two background studies: a diagnostic analysis of the magnitude and nature of the out-of-school youth problem in the region (Gresham 2013) and an econometric model of schooling outcomes of youth aged 12–14 years, the school/work decisions of youth aged 15–24 years, and the transition of youth from lower to higher education cycles (Feda and Sakellariou 2013). Out-of-school youth are defined as those between the ages of 12 and 24 years who are definitively out of school—youth who have either never attended school or who attended for a period of time but dropped out before completing secondary school.1 The diagnostic study builds on data from Demographic and Health Surveys for 19 countries and other household or labor market surveys for 12 additional countries, all con- ducted between 2006 and 2011 (see appendix B for a list of the survey data). The econometric study of the determinants of out-of-school youth uses household survey data from 20 countries in the region (see appendix D for the basic results). The study focuses on three sets of outcomes and their rela- tionship to household characteristics: schooling choices of youth between the ages of 12 and 14 years (a simple logistic regression model), school/work outcomes of those between the ages of 15 and 24 years (a multivariate logis- tic regression model), and the transitions to higher education levels for those between the ages of 15 and 24 years (a sequential logistic regression model, explained in appendix E). Magnitude of the Out-of-School Youth Population in Sub-Saharan Africa Approximately half of the youth population in Sub-Saharan Africa is out of school, though students’ experiences with school systems vary greatly across countries. Countries with a high incidence of out-of-school youth tend to have a disproportionately large share of youth who never attend school. For example, in Mali, where 68 percent of youth are out of school, only one in seven out-of-school youth is a dropout, while the other six never attended school. In Burkina Faso, Chad, and Sierra Leone, this figure is one in five. On the other hand, in the Republic of Congo, Uganda, and Zimbabwe, most out- of-school youth have attended school at some point in their lives (figure 2.1). More out-of-school youth live in low-income countries, but the problem is widespread across the continent. Upper-middle-income countries have the Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 27 Figure 2.1 School Attendance Status of Youth Aged 12–24 Years, by Country 100 80 60 Percent 40 20 0 er . oz am . m da Ni nda Lib ria Sw Ken a az ya Ca o, R ia Bu oon ín e o, Rw ipe bi ia Gh ue s a Zi io o Gu we B ea M n in Ga dag nia Se The rk M e o m ep M Z Rep ba ia a, r r a F li m ie a di d’ ad Co Nam nd d eo i te Ch l i Le an bi a ge Pr n r an L w Cô ega Et oth as in a er ng ib am b m p Ta en oi m asc in De an ge q To S M un b a za é rra la c ila Ni a Iv . Ug n h r Bu ng Co o Sã Out of school, completed secondary Out of school, dropout Out of school, never attended In school Source: Gresham 2013. Note: Figures do not add to 100 percent in some cases because of missing data. lowest rate of out-of-school youth (33 percent of youth), with most youth dropping out after some exposure to schooling. Among low-income ­ countries, more than 45 percent of youth are out of school, and nearly half of these chil- dren have never attended school. Francophone countries have the worst track record, with more than half of youth out of school, and nearly one-third never enrolled. Among the 31 countries analyzed, Anglophone countries perform the best. Fragile or conflict-affected countries have a slightly higher incidence of out-of-school youth, but compared with countries with no ­ conflict history, a larger share of these youth have never attended school (figure 2.2). As they get older and the opportunity cost of their time increases, youth are more likely to drop out. While primary education is associated with little loss of household income (and this can be due to poor measurement of the value of household production or family work), and tertiary education is strongly linked with a wide wage differential, at the secondary level, the loss Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 28 Why Do Sub-Saharan African Youth Drop Out of School? Figure 2.2 Share of Out-of-School Youth Aged 12–24 Years, by Country Typologies Low income Lower middle income Upper middle income Anglophone Francophone Lusophone Conflict history No conflict history 0 10 20 30 40 50 60 Percent Out of school, never attended Out of school, dropout Source: Gresham 2013. of income from nonwage earnings (for example, by working on a family farm) appears to be large enough to counter the positive income effects of education (Appleton 2001). For example, an analysis of household survey data from 1990 and 2000 from Uganda shows that the net income effect of secondary schooling is minimal: although secondary schooling makes it more likely that households will receive wage earnings, this is broadly offset by corresponding reductions in the probability of receiving income from self-employment (agri- cultural and nonagricultural) (Appleton 2001). Both dropout rates and labor participation increase with age. The diagnostic work shows that while only 20 percent of children aged 12–14 years drop out across the region, this rate increases to 40 percent among 15- to 18-year-olds and to almost 70 percent among 19- to 24-year-olds. This pattern holds true for all countries in the diagnostic study, but is more pronounced in low-income coun- tries and countries of Francophone heritage (figure 2.3). Labor force participa- tion also increases with age: participation is 55 percent among children aged 12–14 years, 61 percent among those aged 15–18 years, and 75 percent among those aged 19–24 years. For most countries in the region, dropout rates increase in the last year of each cycle. For example, in Burkina Faso, Ghana, Mali, and Namibia, the dropout rates in the last year of the primary cycle are almost twice the average dropout rate earlier in the primary cycle. In Uganda and Kenya, almost twice as many students drop out in the last year of the lower secondary cycle compared with ­ earlier years. In Ghana, the dropout rate increases almost fivefold, from an average of 17.5 percent in the earlier years of the lower secondary cycle to ­ 86 percent the last year of lower secondary. This suggests that these transition years act like bottlenecks, sometimes because of exit or entrance examinations Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 29 Figure 2.3 Share of Out-of-School Youth, by Age Cohort and Country Typologies a. Income b. Colonial heritage 80 80 70 70 60 60 50 50 Percent Percent 40 40 30 30 20 20 10 10 0 0 Low Lower-middle Upper-middle Anglophone Francophone Lusophone c. Conflict history 80 70 60 50 Percent 40 30 20 10 0 No conflict history Conflict history/fragile Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: Gresham 2013. (table 2.1). In Kenya, for example, the high scores needed for acceptance appear to prevent children from attending (Ozier 2010). The analysis of household data from 20 countries in the region shows that most youth aged 15–24 years never enroll or drop out before the beginning of the secondary cycle (Feda and Sakellariou 2013). Across the region, an esti- mated 33 percent of youth never make it to the secondary level; the shares Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 30 Table 2.1 Transition Years and Dropout Rates, by Grade Percentage dropping out Country Gr1 Gr2 Gr3 Gr4 Gr5 Gr6 Gr7 Gr8 Gr9 Gr10 Gr11 Gr12 Madagascar 20 31 38 45 44 31 31 36 33 20 19 68 Kenya 1 4 4 11 13 18 27 62 17 29 18 67 Ghana 8 10 14 15 17 24 17 18 86 14 19 92 Liberia 13 13 20 21 19 21 16 16 14 11 9 82 Mali 9 12 17 17 26 19 13 15 21 6 16 16 Nigeria 3 3 5 5 7 46 6 10 19 9 9 76 São Tomé and Príncipe 21 13 15 54 32 31 26 26 17 13 31 43 Sierra Leone 4 6 8 9 11 15 10 16 23 8 15 55 Benin 10 15 19 20 21 15 9 12 14 15 8 10 Burkina Faso 17 22 33 34 34 52 30 33 32 34 15 14 Cameroon 8 12 15 18 28 46 51 31 34 34 22 12 Congo, Dem. Rep. 16 23 27 28 32 39 29 35 35 33 28 70 Guinea 4 9 11 14 16 15 8 10 9 10 6 10 Niger 11 18 20 25 45 32 36 24 25 30 15 9 Senegal 11 21 24 27 47 30 14 16 23 20 14 13 Malawi 7 13 17 22 28 31 39 43 33 43 23 71 Zambia 6 12 14 19 23 25 48 24 54 18 15 91 Namibia 8 10 12 13 17 19 28 29 46 63 41 83 Swaziland 7 11 13 19 20 29 44 33 39 44 28 85 Uganda 6 12 19 22 32 42 65 34 42 38 66 18 Source: Gresham 2013. Note: Shaded cells are transition points between cycles. Why Do Sub-Saharan African Youth Drop Out of School? 31 Figure 2.4 School Transitions of 15- to 24-Year-Old Youth 100 90 80 70 Percentage of youth 60 50 40 30 20 10 0 n le e e e e ne ne y or m m m on ea gi ho ho co co co ist fra m ph op op in in in th d y/ co te gl s w le e lic or Lu an dl gh Lo d An st nf id id Fr hi ei co m rm -w ct No er pe fli p w Po n Up Lo Co Secondary grad or in In primary or primary higher education grad, no secondary Still in secondary Never enrolled or Dropped secondary dropped primary Source: Feda and Sakellariou 2013. are particularly high among Francophone countries (57 percent) and fragile countries or countries with a conflict history (48 percent). Anglophone countries ­ perform the best in age-appropriate enrollment. Across this group of countries, 40 percent of youth enroll at the secondary level or complete secondary (figure 2.4). Kenya, Nigeria, and Zambia do particularly well, with approxi- ­ mately half of youth aged 15–24 years either in secondary school or graduated. Cameroon is the best-performing country among the Francophone group, with 35 percent of this cohort enrolled at the secondary and 22 percent enrolled at the primary level. Over-age enrollment is a very important predictor of dropout patterns. Across Sub-Saharan Africa, delaying enrollment by one year increases the prob- ability of dropping out by 9.2 percent, but the estimates vary greatly across countries. For example, in Uganda, each year of over-age enrollment increases the probability of dropping out by 2.2 percent, whereas this rate is more than 20 percent in Burundi and Swaziland (figure 2.5). Of course, over-age enroll- ment is not the cause for dropping out, but is an indicator of underlying issues Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 32 Why Do Sub-Saharan African Youth Drop Out of School? Figure 2.5 Over-Age Enrollment and Dropout Rates a. Age/enrollment profile 25 23 21 19 Age (years) 17 15 13 11 9 7 5 1 2 3 4 5 6 7 8 9 10 11 12 Grade Liberia Sub-Saharan Africa Expected age by grade Burkina Faso Ghana Togo Zimbabwe b. Change in the probability of dropping out with each additional year of late enrollment Swaziland Burundi Sierra Leone Rwanda Gabon Tanzania Congo, Rep. Lesotho Gambia, The Burkina Faso Ethiopia Mali Kenya Togo Niger Ghana Benin Mauritania Malawi Senegal Guinea Namibia São Tomé and Príncipe Congo, Dem. Rep. Comoros South Sudan Sub-Saharan Africa average Mozambique Côte d’Ivoire Sudan Zambia Chad Nigeria Cameroon Madagascar Zimbabwe Liberia South Africa Uganda 0 5 10 15 20 25 30 Percent Source: Household survey data for various years. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 33 that could be related to demand- or supply-side factors. On the demand side, children might start school late because their parents do not feel particularly compelled to get their children to school. On the supply side, late enrollment could signal lack or difficulty of access. It may also signal that quality is a con- cern, as when repetition rates are high, over-age enrollment in later grades is a consequence. Gender Disparities Girls and young women are more likely to be out of school compared with boys and young men across the entire region, but the prospects are poor for both genders. Lesotho and Namibia are the only countries in the region where gender disparities appear to favor girls (measured as the difference between the share of females and males out of school). In both countries, only a handful of youth are out of school among the youngest cohort. Girls are less likely to attend school and more likely to drop out, but the disparities change as youth get older. Even in countries where gender disparities decrease with age, such as in Niger, it is not because the outcomes for girls improve with age, but rather because the outcomes for boys worsen. That is, the outcomes are dismal for both genders: although boys tend to remain in school a bit longer than girls on average, many youth will eventually drop out of school. Among 12- to 14-year-olds, 22 percent of girls and 20 percent of boys are out of school. Within this youngest cohort, dropout rates for boys and girls are similar, at about 6 percent, but more girls (16 percent) are likely to never have attended school compared with boys. On average in the region, being a girl reduces the marginal probability of attending school by 4 percent, but this effect is much larger in Francophone countries (8.3 percent decrease in marginal probability) and countries with conflict history (9.2 percent decrease) ­ (figure 2.6). Anglophone and lower-middle-income countries see a smaller ­ disadvantage associated with gender compared with the rest of the region; in these countries, being a female results in a 3 percent reduction in the prob- ability of attending school. Young women aged 15–18 years and 19–24 years are at a larger disadvan- tage relative to young men of the same age cohorts, but the effects are reversed for those women who actually make it to secondary school. Among 15- to 18-year-olds, one in five young women has never attended school, and a quar- ter dropped out after enrolling; these are relatively larger shares compared with young men in this cohort, among whom 14 percent never attended school and 21 percent dropped out. Among 19- to 24-year-olds, 29 percent of young women have never attended school compared with 18 percent of young men, and 47 percent of the women dropped out after enrolling compared with 43 percent of the men (figure 2.7, panel a). The gender effect is amplified with early marriage, especially in earlier decisions of enrolling, dropping out, or staying with schooling (Feda and ­ Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 34 Why Do Sub-Saharan African Youth Drop Out of School? Figure 2.6  Gender and Schooling Outcomes for 12- to 14-Year-Old Youth a. Share of out-of-school youth, by gender b. Probability of attendance for girls compared with boys 18 0 16 14 –2 12 –4 Percent Percent 10 8 –6 6 4 –8 2 0 –10 Girls Boys Girls Boys Girls Never in school Dropout Anglophone Lower income Lower middle income Lusophone No conflict history Francophone Sub-Saharan Africa Conflict history Sources: Feda and Sakellariou 2013; Gresham 2013. Sakellariou 2013). Unmarried girls in the region have similar chances of attend- ing and completing primary school as young men—the odds ratio unmarried girls face for this transition is 0.95 (an odds ratio of 1 implies equal probability with young men). However, among married girls, the odds ratio is 0.22—that is, the odds of enrolling in and completing primary school favor males five to one. Once they complete primary education, unmarried girls face similar chances of both enrolling and remaining in secondary school. Married females face lower odds once again, 3:5 of enrolling and nearly 1:10 of not dropping out and continuing with secondary education. The effect of marriage is reversed once a young woman overcomes the odds of enrollment and not dropping out: unmarried young women face an odds ratio of 1.23 of completing the second- ary cycle by the age of 24 compared with young men, while married women are three times more likely to complete secondary education by the same age (figure 2.7, panel b). The effects of poverty and early marriage are well documented in the literature. A longitudinal study of youth based in KwaZulu-Natal Province, ­ South Africa, found that while both girls and boys in poorer households are more likely to not continue into secondary education, the effect of poverty on females is much stronger (Hallman and Grant 2004). A considerable propor- tion of the school delays experienced by African women between the ages of 16 and 24 is due to pregnancy: 5 percent of 16- to 17-year-olds, 20 percent of 18- to 19-year-olds, 25 percent of 20- to 22-year-olds, and 28 percent of Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 35 Figure 2.7  Gender and Schooling Outcomes for 15- to 24-Year-Old Youth a. Out-of-school youth, by gender 50 45 40 35 30 Percent 25 20 15 10 5 0 Girls Boys Girls Boys Never in school Dropout Ages 15–18 years Ages 19–24 years b. Schooling outcome, by gender 3.5 3.0 2.5 Odds ratio 2.0 1.5 1.0 0.5 0 Complete Enroll in Stay at Complete primary secondary secondary secondary before age 24 years Female: never married Female: has married Sources: Feda and Sakellariou 2013; Gresham 2013. 23- to 24-year-olds become pregnant. And pregnancy reduces the probability of completing secondary education by half. Similarly, poverty increases the rural disadvantage. Cardoso and Verner’s (2007) study of adolescents in large urban areas in Brazil concluded that parenthood at an early age and extreme poverty are the key contributors to dropping out of school. Gender differentials are greatest among low-income countries, Francophone countries, and fragile and conflict-affected countries (figure 2.8). The only middle-income country in this sample of 20 countries is Namibia, which is also ­ the only country where, for all age groups, a smaller share of girls are out of Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 36 Why Do Sub-Saharan African Youth Drop Out of School? Figure 2.8  Gender Disparities: Percentage Difference in Share of Out-of-School Girls and Boys, by Country Typologies 25 20 15 10 Percent 5 0 –5 –10 e e e ne e ne y m m m on or ho ho co co co st ph op op hi in in in co ct s gl w e e Lu an dl dl fli Lo An id id n Fr Co m rm er pe w Up Lo Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: Gresham 2013. school compared with boys. In every other country in the region for which data are available, girls face a disadvantage. Whether married or not, girls are more likely to be engaged in home care or work and less likely to be in school (with or without working at the same time). For unmarried girls, the probability of being in school is not that different from boys (the differences are statistically significant, but not greater than 3 percent in any group of countries), and in countries where work opportunities are rare, unmarried girls are more likely to be inactive than working compared with boys. The impact of marriage is greatest among countries that otherwise per- form ­better than the regional average. Among Anglophone countries, for example, married girls between the ages of 15 and 24 years are 16 percent more likely to be inactive, 16 percent less likely to combine school with work, 30 percent more likely to be engaged in work only, and 36 percent less likely to be engaged in schooling only, all compared with boys of similar age. In Lusophone countries, marriage does not necessarily reduce the probability of attending school, but it appears to force more young women to combine work and school compared with the more desirable outcome of attending school only (figure 2.9). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 37 Figure 2.9 Marginal Impact of Gender on School/Work Outcomes for Females Aged 15–24 Years, Compared with Males, by Country Typologies a. Anglophone b. Francophone 4 20 3 15 10 2 5 1 Percent 0 Percent 0 –5 –1 –10 –2 –15 –3 –20 –4 –25 Inactive (or Work only Work and School Inactive (or Work only Work and School home care) school only home care) school only c. Lusophone d. Lower income 4 25 20 3 15 2 10 5 1 Percent Percent 0 0 –5 –10 –1 –15 –20 –2 –25 –3 –30 Inactive (or Work only Work and School Inactive (or Work only Work and School home care) school only home care) school only Female: never married Female: has married figure continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 38 Why Do Sub-Saharan African Youth Drop Out of School? Figure 2.9  Marginal Impact of Gender on School/Work Outcomes for Females Aged 15–24 Years, Compared with Males, by Country Typologies (continued) e. Lower middle income f. Conflict history 4 30 3 25 20 2 15 1 10 Percent Percent 0 05 –1 0 –2 –05 –3 –10 –4 –15 Inactive (or Work only Work and School Inactive (or Work only Work and School home care) school only home care) school only g. No conflict history 3 2 1 0 Percent –1 –2 –3 –4 Inactive (or Work only Work and School home care) school only Female: never married Female: has married Source: Feda and Sakellariou 2013. Urban/Rural Distinction In every country in the region, rural youth are more likely to be out of school and more likely to have never attended school compared with urban youth. Across the region, 25 percent of youth aged 12–14 years in rural areas are out of school, compared with 14 percent in urban areas. In this age group, among rural out-​ of-school youth, seven out of 10 have never attended school and only three out of 10 have dropped out after some exposure to schooling (figure 2.10, panel a). Exclusion is a rural phenomenon: for this cohort, those countries with the Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 39 Figure 2.10  Geographic Location and Schooling Outcomes for 12- to 14-Year-Old Youth a. Share of youth out of school, by location b. Marginal impact of location on attendance 20 14 18 12 16 14 10 12 Percent Percent 8 10 8 6 6 4 4 2 2 0 0 Dropout Never in Dropout Never in Urban school school Francophone Sub-Saharan Africa Conflict history Anglophone Rural Urban Lower middle income Lower income No conflict history Lusophone Sources: Feda and Sakellariou 2013; Gresham 2013. highest share of rural out-of-school youth also have the highest share of youth never attending school. In Mali, Burkina Faso, and Senegal, for example, only one in 10 out-of-school youth in rural areas has ever attended school. Among this youngest cohort, urban out-of-school youth are more likely to have attended school, and in countries such as the Republic of Congo, Uganda, and Zimbabwe, almost all out-of-school youth who live in urban areas have attended some school before dropping out. Feda and Sakellariou estimate that across the region, urban youth are 8 ­percent more likely to be in school compared with rural youth. The effect is greatest among Francophone countries (13 percent) and countries with conflict histories (10 percent) (figure 2.10, panel b).2 Urban-rural disparities increase with age. Among 15- to 18-year-olds, 45 ­ percent of rural youth are out of school compared with 31 percent of urban youth. These shares are 77 percent and 57 percent, respectively, for those between the ages of 19 and 24 years. Among the older cohort, a larger share of out-of-school youth, whether urban or rural, have some exposure to schooling, indicating that most youth enroll in school relatively late. For example, half of rural out-of-school youth aged 15–18 years were in school before dropping out, and seven out of 10 urban out-of-school youth enrolled in some schooling before dropping out. For 19- to 24-year-olds, these rates are 60 percent and 76 percent, respectively. Geographic location has the highest impact during earlier transitions. In the region, urban youth aged 15–24 years are 1.5 times more likely to complete primary school, and 1.6 times more likely to enroll in secondary school. Odds decline slightly during later transitions, with urban youth facing the favorable odds ratio of 1.18 of staying in school at the secondary level, and 1.14 of completing secondary educa- ­ tion by the age of 24 years. There is, however, great cross-country variation. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 40 Why Do Sub-Saharan African Youth Drop Out of School? Figure 2.11 Odds Ratio at Various Transitions for Urban Youth Aged 15–24 Years, by Country Typologies a. Colonial history b. Income level 2.5 1.8 1.6 2.0 1.4 1.2 Odds ratio Odds ratio 1.5 1.0 1.0 0.8 0.6 0.5 0.4 0.2 0 0 Complete Enroll in Stay at Complete Complete Enroll in Stay at Complete primary secondary secondary secondary primary secondary secondary secondary before age before age 24 years 24 years Lusophone Anglophone Francophone Lower income Lower middle income c. Conflict history 2.0 1.8 1.6 1.4 Odds ratio 1.2 1.0 0.8 0.6 0.4 0.2 0 Complete Enroll in Stay at Complete primary secondary secondary secondary before age 24 years Conflict history No conflict history Source: Feda and Sakellariou 2013. Among Francophone countries, urban youth are almost twice as likely to complete primary education and enroll in secondary school. Once enrolled, however, geo- graphic location provides no advantages for secondary students—urban and rural youth are equally likely to stay in school, but urban youth face unfavorable odds of 0.6 of completing their secondary education (figure 2.11). Household Characteristics and Out-of-School Youth Parents’ Characteristics Parents’ attitudes toward education play an important role in schooling out- comes. One important indicator of parental attitudes is their own education level (Oreopoulos, Page, and Stevens 2006). For example, Huebler (2011) found that among the 20 African countries he analyzed, children whose parents c ­ ompleted secondary schooling were 20 percent more likely to be in school. The diagnostic study and econometric analyses corroborate this finding. Youth of all ages are Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 41 more likely to attend school when their parents have completed primary educa- tion (compared with those with parents who have not attended school or com- pleted primary); when the head of household has a secondary or higher degree, the probability of youth dropping out of school is reduced significantly. Parental education appears to be the most important determinant of schooling choices for 12- to 14-year-olds. Youth from households where the head of house- hold completed primary education are 15 percent more likely to be in school compared with youth from households in which the head of household has little or no education. The gap in the probability of attending school goes up to 20 percent when the head of household completed upper secondary, suggesting that adults who themselves invested in education are more likely to invest in their children’s education. Other studies have found that children of parents who are generally more involved in schooling and other aspects of their children’s lives are less likely to be out of school (Rumberger 2001), and parental education could be a good proxy for general parental involvement. The analyses show that this effect is strongest in Francophone countries, with more than 25 percent improvement in the probability of attending school (figure 2.12). Parents may also choose to keep their children out of school if they perceive that education does not increase earnings or the ability to find a job. Sub-Saharan Africa has relatively high returns to education at any level of schooling (Schultz 2003; Psacharopoulos and Patrinos 2004), but these return differentials do not appear to be sufficient to entice more youth to stay in school.3 This may be because returns are uneven across different groups (Fasih et al. 2012), or perhaps because even those who finish school may find themselves jobless (Carneiro, Hansen, and Heckman 2003). Household survey data support these points: youth with secondary or tertiary education do not always achieve better employment outcomes. For example, in Burundi, Cameroon, Côte d’Ivoire, Kenya, Madagascar, and Nigeria, more educated youth have higher rates of unemployment than youth with lower educational attainments (Garcia and Fares 2008). Part of the problem is that formal work opportunities are rare.4 Boyle, Brock, and Mace (2002) report that in Kenya, Uganda, and Zambia, par- ents often do not send their children to school because they think education will not lead to a job. In Ghana, measured private returns to secondary education increased rapidly in the 1990s (Sackey 2008), but not for those who lacked access to nonagricultural wage and salaried jobs—that is, largely rural youth. The majority (82.5 percent) of the employed workforce are self-employed or work in agriculture, and returns to education for this segment of the population are very low (Kingdon and Söderbom 2007). In rural Ghana, parents find education to be irrelevant to their children’s lives, since most expect their children to become farmworkers (Pryor and Ampiah 2003). Thus, high paying jobs in the formal ­ sector are not particularly relevant to many people’s lives. Low perceived returns also affect youth’s decisions to attend school. Youth are strongly influenced in their schooling decisions by the perceived quality of their prospective school, and if their perceptions of education outcomes are changed, for example, by learning more about the benefits of completing their education, Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 42 Why Do Sub-Saharan African Youth Drop Out of School? Figure 2.12 Impact of Head of Household’s Education Level on 12- to 14-Year-Old Youths’ Marginal Probability of Attending School, by Country Typologies a. Sub-Saharan Africa and colonial heritage b. Income level 30 25 25 20 20 15 Percent Percent 15 10 10 5 5 0 0 Primary Secondary Tertiary Primary Secondary Tertiary Francophone Anglophone Lower income Lower middle income Sub-Saharan Africa Lusophone c. Conflict history 25 20 15 Percent 10 5 0 Primary Secondary Tertiary Conflict history No conflict history Source: Feda and Sakellariou 2013. youth may change their behavior. Using survey data for 8th-grade boys in the Dominican Republic, Jensen (2010) showed that perceived returns to secondary education are extremely low, and a simple intervention such as just telling stu- dents how much more they could earn if they finished school keeps them in school for three to four more months over the next four years. Head-of-household characteristics matter in earlier transitions, but once youth enroll in secondary school, parental influence on education subsides. Youth whose parents have completed primary education face much better odds of completing primary education and enrolling at the secondary level (3:1 and almost 2:1, respectively), but once they enroll, they are not more likely to stay in school than their peers whose parents have not completed the primary cycle. When parents complete secondary education or achieve higher degrees Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 43 Figure 2.13 Odds Ratio at Various Transitions for Youth Aged 15–24 Years a. Household head—incomplete primary education b. Household head—male and not in agriculture 6.0 1.8 1.6 5.0 1.4 4.0 1.2 Odds ratio Odds ratio 1.0 3.0 0.8 2.0 0.6 0.4 1.0 0.2 0 0 Complete Enroll in Stay at Complete Complete Enroll in Stay at Complete primary secondary secondary secondary primary secondary secondary secondary before age before age 24 years 24 years HH head primary HH head tertiary HH head female HH head in agriculture HH head secondary Source: Feda and Sakellariou 2013. Note: HH = household. themselves, they are much more likely to keep their children in secondary school and push them to finish this cycle (the odds are 2:1, and much higher earlier— the odds of enrolling in primary are 5:1, favoring youth whose parents have completed secondary, compared with those whose parents have no education (figure 2.13, panel a). The effects of parental education are strongest in countries ­ where schooling outcomes are better, such as lower-middle-income and Anglophone countries, and weakest in countries with conflict history. Finally, when the head of household works in the agriculture sector, youth are much less likely to complete primary education and enroll at the secondary level (odds are 0.8:1). Once enrolled at the secondary level, the influence of the agri- culture sector diminishes. Female heads-of-household have higher odds of their children completing primary education and enrolling in secondary school (1.5:1 and 1.6:1, respectively). Once in secondary school, students perform similarly regardless of the gender of the household head, except in low-income countries, where youth from households headed by a female are much more likely to com- plete secondary education once they enroll (figure 2.13, panel b). Household Income and School Attendance Perhaps one of the most important determinants of schooling outcomes is poverty. Regardless of the methodology used, almost every study on educational attain- ment and enrollment outcomes finds that children from poorer households are less likely to complete formal education.5 The diagnostic analysis shows great disparities between the education outcomes for youth from the poorest and richest households in the region. Roughly one in five children aged 12–14 is out ­ of school in the region, but 12-year-olds from the richest households in the region are three times more likely to be in school compared with similarly aged children Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 44 Why Do Sub-Saharan African Youth Drop Out of School? from the poorest households.6 When combined with the education level of the head of household, household income effects are reduced but remain significant. percent For 12- to 14-year-olds, the probability of attending school increases by 4–5 ­ for households in each subsequent income quintile, compared with youth from the poorest households. The effects are maximized for households in the middle- income group across different country groups, and are most pronounced in ­ countries where education outcomes are poor, such as Francophone and low- income countries. Lusophone countries are the only group for which the mar- ginal impact on the probability of attending school increases with income levels (figure 2.14, panel a). ­ While parental education outcomes dwarf the income effects for the 12- to 14-year-old group, the effect of households’ earning capacity, measured as the number of working adults, remains large. Among this group, youth are 14 percent more likely to attend school if there is at least one working adult in their house- hold compared with youth in households with no working adults. The impact of having two working adults is 21 percent. As opposed to household income level, the marginal effects of having working adults are greatest in countries with rela- tively good outcomes—such as lower-middle-income and Anglophone countries (figure 2.14, panel b). The effects of income are lower when controlling for parental education, but are still substantial for 15- to 24-year-olds. Household income increases the odds of better education outcomes at all transitions, but the impact is strongest at the Figure 2.14 Impact of Household Income and Income-Generating Capacity on 12- to 14-Year-Old Youths’ Marginal Probability of Attending School a. Impact of household income b. Impact of household members’ employment 16 30 14 25 12 10 20 Percent Percent 8 15 6 4 10 2 5 0 –2 0 Q2 Q3 Q4 Q5 One employed Two employed Three or more HH member HH members employed HH Lusophone Francophone aged 25+ years aged 25+ years members Lower income Lower middle income aged 25+ years No conflict history Conflict history Anglophone Francophone Anglophone Lower middle income Lusophone Lower income Conflict history No conflict history Source: Feda and Sakellariou 2013. Note: HH = household; Q = quintile. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 45 second transition—the odds of enrolling at the secondary level are 1.8:1 for youth from the highest-earning households, but this effect subsides with upper transitions, going down to 1.4:1 for completing secondary education. At this transition, there is no discernible difference for youth who come from the poor- est households and youth from households with incomes in the second and third quintiles (figure 2.15, panel a). The income effects are particularly important in Lusophone countries, where youth from the highest-income households are more than twice as likely to enroll in and complete secondary education. Similarly, in a household with working adults, youth aged 15–24 are much more likely to complete primary school, enroll at the secondary level, and stay in school: even one working adult increases the odds of staying the course at the secondary level to 2:1. Interestingly, the results are reversed for completing sec- ondary education before the age of 24 (figure 2.15, panel b). Household expenditure surveys suggest that for the poorest households, out- of-pocket costs of attending school are often prohibitively high. A recent study of 15 African countries found that households spend, on average, 4.2 percent of their budget on education, accounting for about 1.7 percent of gross domestic product (GDP) in this group of countries—this is half the share of public expen- diture in education (Foko, Tiyab, and Husson 2012).7 Generally, poorer house- holds spend a smaller share of their income on education—2.6 percent across the same countries. This may appear to be low, but if one looks at d ­ iscretionary spending (excluding food and shelter), the share of education expenditure could Figure 2.15 Impact of Household Income and Income-Generating Capacity on 15- to 24-Year-Old Youths’ Odds Ratio at Various Transitions a. Impact of household income b. Impact of household members’ employment 2.0 2.5 1.8 1.6 2.0 Odds ratio 1.4 Odds ratio 1.2 1.5 1.0 0.8 1.0 0.6 0.4 0.5 0.2 0 0 y y y y ye ary y y ye ary ar ar ar ar ar ar s nd s nd im 24 d im nd nd 24 d ar ar e on e on co pr pr co co co ag ec ag sec se se se se e e re e s et et at l in at n e pl pl fo let ll i fo let ay ay m m ol ro be mp be mp St Co Co St r re En En Co Co HH income Q5 HH income Q3 One employed HH member aged 25+ years HH income Q4 HH income Q2 Two employed HH members aged 25+ years Three or more employed HH members aged 25+ years Source: Feda and Sakellariou 2013. Note: HH = household; Q = quintile. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 46 Why Do Sub-Saharan African Youth Drop Out of School? be as high as one-third of all expenditures (Boyle, Brock, and Mace 2002, 71–72).8 At the secondary level, more household expenditures go to cover school fees compared with primary education. In Côte d’Ivoire, for example, secondary school fees account for approximately 38 percent of household expenditures, as opposed to 27 percent at the primary level.9 In Mauritania, secondary school fees go as high as more than 60 percent of household expendi- tures, and in Rwanda, close to 70 percent. This is a binding constraint. The diag- nostic analysis of household survey data also finds similar results ( ­figure 2.16). Across the region, household expenditure on education constitutes about percent of total expenditures. One-third of out-of-school 12- to 14-year-olds 5 ­ come from households that identify cost as the primary reason for not attending or for dropping out of school, but the numbers are much higher for low-income countries such as Burkina Faso and Ethiopia. In Uganda, 25 percent of the poor- est households cite lack of money as the main reason why children don’t attend secondary school in the first place, and more than half of the poorest households take children out of secondary school, at least tem­ porarily, because they cannot afford the fees. In Zambia, comparable figures are 36 percent and 62 percent at the secondary level, respectively. Elimination of school fees at the primary level in Cameroon, Kenya, Lesotho, Malawi, Tanzania, Uganda, and Zambia has typically led to significant increases in enrollment, and similar policies may help ­ students enroll in and complete secondary schooling. Work and School Choice Work is a part of life for many young children and youth in Sub-Saharan Africa, whether they are in school or not. Children start working at very young ages— based on data from 25 countries in the region, on average, 38 percent of children aged 7–14 are economically active, working at home or elsewhere, sometimes for a salary and sometimes not, and 15 percent work only and do not attend school. Among 15- to 24-year-olds, an estimated 60 million youth work and 28.3 million youth both work and attend school. In Sudan and the Republic of South Sudan, roughly one in four working children attends school at the same time; in Burkina Faso, Mali, and Niger, this figure is three in 10. On the other side of the spectrum, in Malawi and Uganda, roughly nine in 10 economically active children juggle work and school at the same time. Children typically work to help with household chores or to augment household income, and sometimes at their p­ arents’ suggestion because working is seen as a positive attribute (Moyi 2011). Just like education outcomes, child labor is negatively associated with household income—for example, in Sub-Saharan Africa, children from the richest households are 16 percent less likely to work compared with children from the poorest quintile (Huebler 2011).10 Shocks to household income often force youth out of school. Youth may drop out temporarily from school to help family or because the family lacks funds to pay school fees as they deal with fluctuations in their income. The United Nations reported that in Niger, 47,000 children left school to help their families deal with drought in 2012 (Huyghe and Mebrahtu 2012).11 Ferreira and Schady (2008) Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 47 Figure 2.16 Cost and Household Spending on Education a. Share of out-of-school youth citing cost as the primary reason for dropping out, by country 60 50 40 Percent 30 20 10 0 ia o e a a da da a ria r i ad aw ny ne bi ge Th as op an an ge m Ch aF Ke i Ni al a, Gu hi Rw Ug Ni Za M bi in Et m rk Ga Bu Ages 12–14 years Ages 15–18 years Ages 19–24 years b. Household spending on education, by country 12 10 Percentage of household 8 expenditures 6 4 2 0 bi pia m so Ga Eth oire Gu na Ke a M M ya Si n pe nz e Ug nia Za da Ni e ín a an w a d’ ad Gh e a Ni er L l bi i n e ra a am law Ta eon Pr d é R eri qu bi Th oo n Ca a Fa er eg in g a an ci d an te Ch a m io m Iv g a, er oz a in S e rk Cô Bu m To o Sã Sources: Feda and Sakellariou 2013; Gresham 2013. showed that among the poorest, economic shocks and the consequent loss of parental income are more likely to result in students leaving school even when the adverse economic environment also reduces youth’s income potential. The authors showed that economic shocks have led to declines in enrollment in low-income countries, but increases in enrollment in high-income countries. Furthermore, girls are more likely to be hit by economic shocks, especially girls in countries where female enrollment levels are already low (Gubert and Robilliard 2006). Other events that reduce family income, such as the death of a parent, can push children out of school. For example, in South Africa, parental death has been shown to affect educational outcomes (Case and Ardington 2004). Similarly, in Ethiopia, Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 48 Why Do Sub-Saharan African Youth Drop Out of School? a child who loses his or her mother between the ages of 8 and 12 is less likely to attend school or to learn to read or write (Himaz 2009).12 In general, if more adults are working in a household, the impact of these shocks can be absorbed easier, without affecting youth’s schooling outcomes. School/work outcomes for youth are very sensitive to household sector of employment, income level, and income-generating capacity. The marginal proba- bility of going to school decreases by 6 percent for youth from households engaged in agriculture, whereas the marginal probability of just working ­ (without attending school) increases by 13 percent for this group (figure 2.17, panel a). The effect of household sector is strongest in lower-income countries. The impact of household income level, even when controlling for the education level of the head of the household, is still significant. Compared with the poorest households, youth from the richest households are 4 percent more likely to only attend school and percent less likely to only be engaged in work (figure 2.17, panel b). Incremental 7 ­ increases in household income have a much greater impact on the school/work outcomes in low-income countries and those with a conflict history. Youth in households with higher earning potential, measured as the number of working adults, are more likely to only attend school than to only work. When a household has one or two working adults (compared with none), youth are 15–18 percent less likely to just work, and 14–15 percent more likely to just attend school (figure 2.17, panel c) The effects of working adults on youth outcomes are particularly pronounced in Southern Africa,13 where the presence of two working adults increases the full-time school outcome for youth by 30 percent. Attitudes and habits of youth, shaped by their environment and expectations, also play a large role in schooling outcomes. An emerging body of research emphasizes the importance of noncognitive abilities in explaining gaps in school- ing and life outcomes. This research, summarized by Heckman (2000), Carneiro and Heckman (2003), and Cunha et al. (2006), suggests that soft skills—­ behavior and personality traits, goals, motivations, and preferences—strongly influence schooling decisions and a wide variety of risky decisions among youth (Rumberger and Lamb 2003; Heckman, Stixrud, and Urzua 2006). Youth who would appear to benefit financially from schooling do not pursue it because they lack the skills to study, listen, or stay on task; have low expectations of them- selves; and are easily demoralized when they fail academically—that is, they face high “psychic costs” of attending school (Heckman and Kautz 2012).14 Thus, low aspirations, especially when compounded with lack of motivation and academic support and poor understanding about the personal returns to education, may play a larger role in dropping out than previously envisioned (Thomas, Webber, and Walton 2002). For example, a study of 16- to 25-year-olds high school dropouts revealed that seven out of 10 dropouts were not motivated or inspired ­ to work hard, and they dropped out even when they thought that they would have graduated if they applied themselves (Bridgeland, Dilulio, and Morison 2006). In addition to academic struggles, limited ways to make up for failed course credits and boredom appear to be important reasons why students drop out (Berliner et al. 2008). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 49 Figure 2.17 Impact of Various Household Economic Characteristics on the Marginal Probability of 15- to 24-Year-Old Youths’ Work/School Outcomes a. Household sector 15 10 5 Percent 0 –5 –10 –15 Inactive Work only Work and school School only (or home care) HH head in agriculture b. Household income 6 4 2 Percent 0 –2 –4 –6 –8 Inactive Work only Work and school School only (or home care) HH income Q2 HH income Q4 HH income Q3 HH income Q5 c. Income-generating capacity 20 15 10 Percent 5 0 –5 –10 –15 –20 Inactive Work only Work and school School only (or home care) One employed HH member aged 25+ years Three or more employed Two employed HH members aged 25+ years HH members aged 25+ years Source: Feda and Sakellariou 2013. Note: HH = household; Q = quintile. Social abilities, learned or inherited, are also important in youth’s schooling outcomes for reasons other than poverty. The literature on the importance of soft skills suggests that interventions that remove or relax income constraints or improve cognitive skills may not always be successful if children do not learn the requisite social abilities at early ages. Furthermore, traits that are Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 50 Why Do Sub-Saharan African Youth Drop Out of School? passed genetically or otherwise from parents to children, such as low self- esteem, will lead to poor schooling and life outcomes across generations. Policy interventions or corrective actions that address lack of social skills might be an uncomfortable policy choice for school officials. First, “values” that matter for schooling and life outcomes are typically seen as things that should be tackled in the family domain. For example, would families be comfortable with schools teaching youth the importance of aggressiveness or self-promotion? Furthermore, it is almost impossible to define the right mix of such traits (Bowles, Gintis, and Osborne 2001) or directly relate them to employment outcomes. For example, a randomized experiment conducted by the World Bank showed that soft skills development had a longer-lasting effect than wage subsidies, but not necessarily on immediate job outcomes. In addition, the effects were hard to measure. The wage subsidies led to extremely large short- term gains in employment that almost completely disappeared by the time the subsidies expired. The soft skills training had no short-term impact on employ- ment, but did help to improve beneficiaries’ life outlook and reduce depres- sion, suggesting that training may have benefits outside of the labor market (Groh et al. 2012). Finally, peers influence youth’s intentions about school. When their peers are out of school, youth are more likely to drop out (Audas and Willms 2001; Thomas, Webber, and Walton 2002) and to engage in risky behavior such as drug and alcohol use (Ellickson et al. 1998; Roebucka, French, and Dennis 2004). Secondary school dropouts from poor urban areas are more likely to have other dropouts as friends (Chugh 2011). In fact, youth who drop out have very differ- ent peers from those who stay in school: youth at risk for dropping out have fewer friends, and among those, at least one is a dropout, has already graduated, or is working (Ellenbogen and Chamberland 1997). School Characteristics Among the key reasons why children do not attend school is that they do not have access or easy access to a school. When there is not enough capacity, access becomes a problem for poor and rural communities, for which barriers such as distance and entrance requirements or exit exams become binding. In rural areas, where secondary schools are far away, parents do not send younger children or girls to school because they fear for their safety (Boyle, Brock, and Mace 2002; Nekatibeb 2003; Verspoor and Bregman 2008). No data are available at the sec- ondary level, but household surveys suggest that about half of households live 30 minutes or more away from a primary school (figure 2.18). In rural Africa, it is also common for parents to take children out of primary school with the expectation that they won’t have access to secondary schooling. Evidence from Chad and Senegal suggests that what may appear to be parents’ indifference to schooling may actually represent concerns about not having schools in close proximity—when the distance to school is greater than 1 kilometer, enrollment rates are reduced to negligible levels (Lehman 2003). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 51 Figure 2.18  Distance from Primary Schools for Out-of-School Youth 100 90 80 70 60 Percent 50 40 30 20 10 0 n e ia ad i e a da e ia da l ria aw ga an Th on ag oo op an Ch an an ge ne al Gh er Le er a, nz hi Rw Ug Ni M Se bi Av m Et Ta ra m Ca er Ga Si 60 minutes or more 30–59 minutes Less than 30 minutes Source: Feda and Sakellariou 2013. Education quality is also an important factor for attendance.15 Retention increases with school quality. Using data collected in a longitudinal survey of pri- mary school students in the Arab Republic of Egypt, Hanushek, Lavy, and Hitomi (2006) show that students of all achievement and ability levels are more likely to stay in school if the school quality is high. When school quality is low, both parents and youth reduce the importance they attach to education. Low school quality and lack of motivation at school act as push factors by hindering student achieve- ment and progress (Hardre and Reeve 2003). Organizational features of schools such as lack of clear and rigorous school goals appear to amplify such push factors: youth who dropped out perceive teachers to be less interested in them, and view school discipline as ineffective and inequitably applied (Audas and Willms 2001). Youth are more likely to drop out when teachers are regularly absent, when schools are unresponsive to local needs and preferences (such as religious beliefs) or unaccountable to parents or students, or when basic infrastructure (such as drinking water and bathrooms) is lacking (Hunt 2008). Teacher mentorship moti- vates youth to stay in school (Thomas, Webber, and Walton 2002), and teachers with low academic expectations or who treat low-achieving students in a negative way push students out: secondary school dropouts in the slums of Delhi report that teachers often punish low-performing students through verbal and physical Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 52 Why Do Sub-Saharan African Youth Drop Out of School? abuse, or by giving such students cleaning tasks in school (Chugh 2011). In some instances, teacher attitudes toward students are why children stay out of school in the first place. In the Democratic Republic of Congo, students report widespread sexual abuse perpetrated by teachers, and the fear of or actual sexual abuse some- times leads to girls dropping out (Seymour 2011). Household data from 12 Sub-Saharan African countries suggest that quality is an important concern across all age groups. Only 5 percent of all youth aged 12–14 cite quality as a reason for dropping out across these countries (unweighted average), but among this group, only a quarter are out of school. Among 15- to 18-year-olds, the share of students who drop out because of poor quality increases to 10 percent, but the total share of youth out of school is also much higher, at about 45 percent. Similarly, 15 percent of youth indi- cate that low quality was the reason they dropped out, but among this group, three-quarters are out of school. Overall, quality concerns play a larger role as youth become older. The share of youth who report quality as the main reason for dropping out of school increases with age in all countries in the sample (figure 2.19). In Niger, where enrollment declines to 20 ­ percent among the older cohort, more than half of out-of-school youth in this group (up from one-fifth among 12- to 14-year-olds) report quality as the main reason for dropping out. When learning standards are low and academic support is lacking, students who are unsuccessful drop out. Those youth who struggle at the secondary level because of poor understanding at the elementary level or lack of aca- demic or remedial support are more likely to drop out. Repetition or detention at the same grade level can cause embarrassment and lead to dropout. One study showed that one in five children who drop out of secondary school in the slums of New Delhi cites failure at school as the most significant reason for Figure 2.19  Youth Who Cite Low School Quality as the Main Reason for Dropping Out 30 25 20 Percent 15 10 5 0 a e a da da a a o ria r i ad ge aw pi ny bi ne Th as an an ge io m aF Ch Ke Ni i al a, Gu h Rw Ug Za Ni M bi in Et m rk Ga Bu Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: Gresham 2013. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 53 dropping out. Repetition rates are particularly high at the secondary level in Sub-Saharan Africa. In Angola, Burundi, Cape Verde, and Namibia, almost a ­ quarter of the students repeat their first year in the secondary cycle. The percent- age of youth who repeat declines over the second and third year of secondary school, and jumps again in the fourth year—the year for transition into upper secondary schools in most Sub-Saharan African countries. The percentage of repeaters among the upper secondary cycle is highest in Chad, Benin, Côte d’Ivoire, Mali, and Togo (figure 2.20). Figure 2.20 Repeaters in the Lower and Upper Secondary Cycles a. Repeaters in lower secondary 30 25 20 Percent 15 10 5 0 Le n Gh he Eth eria ur a ine biq o ia, a d’I a E ia Na ola Ga Uga nia a Za pia nd Ve . Co N bia Ma azil ea to Ma one Se oire Prí rde Bu Bur nin Cô l Gu wi a-B ue G u s au Lib ius Be o so An blic Sie me car Ca gas d an C ali na di Mo Le gal é a pe Rep R e had om Ca go, er To e mb nd M a an te ine ine rra roo Gu zam ot h g mb d a an ip Sw ritr ria la rki un T Fa n ig M a it g io mi ne is nc pu v nz s Ta ric ua Af Eq r al oT nt Sã Ce b. Repeaters in upper secondary 40 35 30 25 Percent 20 15 10 5 0 Ch s o d ru n d’ nin go n zi a rk uri e a a a hi ia ne . amNig e re a ua T Lib ia G u z a m uin a o, ea Za ibia -B ue ng in e Gh pia i n b ea Ve la aF s M M s au Le lic é B e r o er fri Sw Erit ar Toali Se Rep as i Bu Ma cipi te Be d Ca An gal Si Repland in tiu ca a re ag aw to an eri Ga nd i Ug an Pr nd rd an u o Co Gu on M i al G z a n Et mb b pe g o oi a a c ra b ea iq M m o is m er u Iv a d al ín Na C r Cô o m lA Eq To ra o nt Sã Ce Source: UNESCO Institute for Statistics. Note: Data are for 2011 for all countries except: Angola and Central African Republic (2009); Ethiopia, Guinea-Bissau, Lesotho, and Namibia (2010); and Tanzania (2012). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 54 Why Do Sub-Saharan African Youth Drop Out of School? Notes 1. Such youth are not enrolled in the age-appropriate level of education or at any other level of education, most typically the primary level. The analysis excludes youth who have successfully completed secondary school—a small minority of the total youth population in most countries—and youth who are enrolled at any other level of edu- cation, such as a 15-year-old enrolled in primary school. 2. Armed conflict or emergency situations make it harder for children to attend or remain in school: children are forced to migrate, going to school becomes dangerous for them, or sometimes they are enlisted as soldiers (Hunt 2008). According to the United Nations, 20 countries in Sub-Saharan Africa experienced armed conflict between 1999 and 2008, with devastating implications for the education of children and youth. Mozambique’s civil war, which lasted 15 years, reduced the expansion in education in such a way that at the end, average schooling was five years less than what it would have been had preconflict trends continued. Rwanda’s deep but rela- tively short violent episode (four years) cost the country 1.2 years of schooling (UNESCO 2011, 136). The conflict that started in 2002 in Côte d’Ivoire and lasted over three years is estimated to have reduced average years of education by a range of 0.2–0.9 years (Dabalen and Saumik 2012). Furthermore, the nature of conflict is also changing, with more direct attacks on civilians and civilian infrastructure including schools and schoolchildren (UNESCO 2011, 143). 3. There is some debate on whether returns to education have been constant or if they have declined with economic stagnation in Africa. In addition, linkages between quality and outcomes, which are not even for all segments of the society, may lead to ­ overestimation of actual returns (Hanushek, Lavy, and Hitomi 2006). 4. In Sub-Saharan Africa, 40 percent of households rely on household enterprises (non- farm enterprises operated by a single individual or with the help of family members) as an income source, and although the fast-growing economies generated new private nonfarm wage jobs at high rates, household enterprises generated the most new jobs outside agriculture (Fox and Sohnesen 2012). 5. See Hunt (2008) for a general review, as well as Brown and Park (2002) and Dachi and Garrett (2003). See Edmonds (2004) for the impact of low-income subsidies on labor and schooling decisions for children in South Africa. 6. The gap in school attendance across the poorest and the richest households widens with age. Among the older cohort of 15- to 18-year-olds, when not controlling for parental education levels, youth from the poorest households are almost twice as likely to be out of school compared with youth from the richest households. For the 19- to 24-year-old cohort, half of youth are out of school for the richest families, a large number, but not nearly as high as that of the poorest households, where 84 ­percent of youth are out of school. 7. The study includes Benin, Burkina Faso, Cameroon, Chad, the Republic of Congo, Côte d’Ivoire, Gabon, Madagascar, Malawi, Mali, Mauritania, Niger, Rwanda, Sierra Leone, and Tanzania. 8. The authors estimate that household expenditure on education is as high as 33 percent of discretionary spending in Uganda and Zambia, and 15.2 percent and 10.6 percent of all spending, respectively. 9. This calculation is based on data from 2002. 10. Using Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) from 35 countries (one was a survey of Palestinians in the Syrian Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 55 Arab Republic), Huebler (2011) found that children from the richest quintile are 19 percent less likely to work and 20 percent more likely to be in school. For the 20 African countries in the sample, children from the richest quintile households are 22 percent more likely to be in school. 11. Migration for noneconomic reasons can also affect schooling outcomes (Beegle and Poulin 2012). And intentions to leave the area in the future can induce youth, especially boys, to stay in school. 12. Maternal death appears to have a much larger impact on education than paternal death, regardless of income levels of families. There is some evidence that paternal death effects are mostly spurious, and significantly correlated with unobserved socio- economic variables (Chen, Chen, and Liu 2009). 13. For this analysis, “Southern Africa” includes data from Malawi, Mozambique, and Zambia. 14. Such soft skills are also strongly correlated with life outcomes. For example, measures of conscientiousness—the tendency to be organized, responsible, and hardworking— predict educational attainment, health, and labor market outcomes as strongly as, and sometimes much better than, measures of cognitive ability. This finding would explain, for example, why GED earners in the United States, although they have demonstrated having the same cognitive skill set as high school graduates, earn much lower wages (Cunha et al. 2006). Another study finds that fatalism is associated with lower salaries, and aggressiveness may increase earnings for men, and for those in high- status jobs (Bowles, Gintis, and Osborne 2001). 15. There is contradictory evidence on the linkages between education quality, cogni- tive skills, and economic outcomes. See, for example, Breton (2011) and Lee and Newhouse (2011), who find that attainment is a much stronger indicator for school- ing and employment outcomes, and cognitive abilities and education quality are more likely to impact job quality than employment, enrollment, unemployment, or labor force participation. Hanushek (2003) argues that overall resource policies have not led to discernible improvements in student performance. References Appleton, Simon. 2001. “Education, Incomes and Poverty in Uganda in the 1990s.” Working Paper 01/22, University of Nottingham, Centre for Research in Economic Development and International Trade, Nottingham, UK. Audas, Richard, and J. Douglas Willms. 2001. “Engagement and Dropping Out of School: A Life-Course Perspective.” Human Resources Department Canada Applied Research Branch Working Paper Series W-01-1-10E, HRDC Publications Centre, Québec. Beegle, Kathleen, and Michelle Poulin. 2012. “Migration and the Transition to Adulthood in Contemporary Malawi.” Policy Research Working Paper WPS6200, World Bank, Washington, DC. Berliner, BethAnn, Vanessa X. Barrat, Anthony B. Fong, and Paul B. Shirk. 2008. Reenrollment of High School Dropouts in a Large, Urban School. Issues & Answers Report, REL 2008 No. 056. Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory West. Bowles, Samuel, Herbert Gintis, and Melissa Osborne. 2001. “Incentive-Enhancing Preferences: Personality, Behavior and Earnings.” American Economic Review 91 (2): 155–58. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 56 Why Do Sub-Saharan African Youth Drop Out of School? Boyle, Siobhan, Andy Brock, and John Mace. 2002. Reaching the Poor: The “Costs” of Sending Children to School. Synthesis Report. London: Department for International Development. Breton, Theodore R. 2011. “The Quality versus the Quantity of Schooling: What Drives Economic Growth?” Economics of Education Review 30 (4): 765–73. Bridgeland, John M. Jr., John J. Dilulio, and Karen Burke Morison. 2006. The Silent Epidemic: Perspectives of High School Dropouts. A Report by Civic Enterprises in Association with Peter D. Hart Research Associates for the Bill & Melinda Gates Foundation. Civic Enterprises, Washington, DC. Brown, Philip H., and Albert Park. 2002. “Education and Poverty in Rural China.” Economics of Education Review 21 (6): 523–41. Cardoso, Ana Rute, and Dorte Verner. 2007. “School Dropout and Push-Out Factors in Brazil: The Role of Early Parenthood, Child Labor, and Poverty.” Policy Research Working Paper 4178, World Bank, Washington, DC. Carneiro, Pedro, Karsten T. Hansen, and James J. Heckman. 2003. “Estimating Distributions of Treatment Effects with an Application to the Returns to Schooling and Measurement of the Effects of Uncertainty on College Choice.” IZA Discussion Paper 767, Institute for the Study of Labor, Bonn, Germany. Carneiro, Pedro, and James J. Heckman. 2003. “Human Capital Policy.” In Inequality in America: What Role for Human Capital Policies? edited by James J. Heckman, Alan B. Krueger, and Benjamin Freidman. MIT Press: Cambridge, MA. Case, Anne, and Cally Ardington. 2004. “The Impact of Parental Death on School Enrolment and Achievement: Longitudinal Evidence from South Africa.” CSSR Working Paper 97, Centre for Social Science Research, University of Cape Town, South Africa. Chen, Stacy Y., Yen-Chien Chen, and Jin-Tan Liu. 2009. “The Impact of Unexpected Maternal Death on Education—First Evidence from Three National Administrative Data Links.” American Economic Review 99 (2): 149–53. Chugh, Sunita. 2011. “Dropout in Secondary Education: A Study of Children Living in Slums of Delhi.” NUEPA Occasional Paper 37, National University of Educational Planning and Administration, Delhi. Cunha, Flavio, James J. Heckman, Lance J. Lochner, and Dimitriy V. Masterov. 2006. “Interpreting the Evidence on Life Cycle Skill Formation.” In Handbook of the Economics of Education, edited by Eric A. Hanushek and Finis Welch. Elsevier B.V. Dabalen, Andrew L., and Paul Saumik. 2012. “Estimating the Causal Effects of Conflict on Education in Côte d’Ivoire.” Policy Research Working Paper 6077, World Bank, Washington, DC. Dachi, Hillary A., and R. M. Garrett. 2003. “Child Labour and Its Impact on Children’s Access to and Participation in Primary Education: A Case Study from Tanzania.” U.K. Department for International Development Education Research Paper 48, Department for International Development, London. Edmonds, Eric V. 2004. “Does Illiquidity Alter Child Labor and Schooling Decisions? Evidence from Household Responses to Anticipated Cash Transfers in South Africa.” NBER Working Paper 10265, National Bureau of Economic Research, Cambridge, MA. Ellenbogen, Stephen, and Claire Chamberland. 1997. “The Peer Relations of Dropouts: A Comparative Study of At-Risk and Not At-Risk Youths.” Journal of Adolescence 20 (4): 355–67. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 57 Ellickson, Phyllis, Khanh Bui, Robert Bell, and Kimberly A. McGuigan. 1998. “Does Early Drug Use Increase the Risk of Dropping Out of High School?” Journal of Drug Issues 28 (2): 357–80. Fasih, Tazeen, Geeta Kingdon, Harry Anthony Patrinos, Chris Sakellariou, and Mans Soderbom. 2012. “Heterogeneous Returns to Education in the Labor Market.” Research Working Paper 6170, World Bank, Washington, DC. Feda, Kebede, and Chris Sakellariou. 2013. “Out of School, School-Work Outcomes and Education Transitions of Youth in Sub-Saharan Africa—A Diagnostic.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Ferreira, Francisco H. G., and Norbert Schady. 2008. “Aggregate Economic Shocks, Child Schooling, and Child Health.” World Bank Research Observer, World Bank, Washington, DC. Foko, Borel, Beifith Kouak Tiyab, and Guillaume Husson. 2012. “Household Education Spending: An Analytical and Comparative Perspective for 15 African Countries.” Working Paper Pôle D’analyse Sectorielle en Éducation of UNESCO-BREDA, Pôle de Dakar, Dakar, Senegal. Fox, Louise, and Thomas Pave Sohnesen. 2012. “Household Enterprises in Sub-Saharan Africa: Why They Matter for Growth, Jobs, and Livelihoods.” Policy Research Working Paper WPS6184, World Bank, Washington, DC. Garcia, Marito, and Jean Fares. 2008. “The Effect of Education on Income and Employment.” In Youth in Africa’s Labor Market, edited by Marito Garcia and Jean Fares. Washington, DC: World Bank. Gresham, James. 2013. “Out-of-School Youth in Africa—Diagnostic Note.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Groh, Matthew, Nandini Krishnan, David McKenzie, and Tara Vishwanath. 2012. “Soft Skills or Hard Cash? The Impact of Training and Wage Subsidy Programs on Female Youth Employment in Jordan.” Policy Research Working Paper 6141, World Bank, Washington, DC. Gubert, Flore, and Anne-Sophie Robilliard. 2006. “Risk and Schooling Decisions in Rural Madagascar: A Panel Data Analysis.” DIAL Working Paper DT/2006-08, Institut de Recherche pour le Développement (IRD), Paris. Hallman, K., and M. Grant. 2004. Poverty, Educational Attainment, and Livelihoods: How Well Do Young People Fare in KwaZulu Natal, South Africa? Horizons Research, Population Council, Washington, DC. Hanushek, Eric A. 2003. “The Failure of Input-Based Schooling Policies.” The Economic Journal 113: F64–98. Hanushek, Eric A., Victor Lavy, and Kohtaro Hitomi. 2006. “Do Students Care about School Quality? Determinants of Dropout Behavior in Developing Countries.” NBER Working Paper 12737, National Bureau of Economic Research, Cambridge, MA. Hardre, Patricia L., and Johnmarshal Reeve. 2003. “A Motivational Model of Rural Students’ Intentions to Persist in, versus Drop Out of, High School.” Journal of Educational Psychology 95 (2): 347–56. Heckman, James J. 2000. “Policies to Foster Human Capital.” Research in Economics 54 (1): 3–56. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 58 Why Do Sub-Saharan African Youth Drop Out of School? Heckman, James J., and Tim D. Kautz. 2012. “Hard Evidence on Soft Skills.” NBER Working Paper 18121, National Bureau of Economic Research, Cambridge, MA. Heckman, James J., Jora Stixrud, and Sergio Urzua. 2006. “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior.” Journal of Labor Economics 24 (3): 411–82. Himaz, Rozana. 2009. “The Impact of Parental Death on Child Outcomes: Evidence from Ethiopia.” Young Lives Policy Brief 7, University of Oxford, with support from the Department for International Development and Irish Aid. Huebler, Friedrich. 2011. “Child Labour and School Attendance: Evidence from MICS and DHS Surveys.” Paper presented at the seminar on Child Labour, Education and Youth Employment, Understanding Children’s Work Project, Universidad Carlos III de Madrid, September 11–12. Hunt, Frances. 2008. Dropping Out from School: A Cross Country Review of Literature. Transitions and Equity Research Monograph 16, Consortium for Research on Educational Access Falmer, UK. Huyghe, Laura, and Shushan Mebrahtu. 2012. “In Niger, Children Are Forced to Drop Out of School to Support Their Families.” At a Glance: Niger, UNICEF. Jensen, Robert. 2010. “The (Perceived) Returns to Education and the Demand for Schooling.” Quarterly Journal of Economics 125 (2): 515–48. Kingdon, Geeta, and Måns Söderbom. 2007. “Education, Skills, and Labor Market Outcomes: Evidence from Ghana.” Centre for Education and International Development, University of Cambridge, Cambridge, UK. Lee, Jean N., and David Newhouse. 2011. “Cognitive Skills and Youth Market Outcomes.” Background paper for the World Development Report 2013, World Bank, Washington, DC. Lehman, Douglas. 2003. “Bringing the School to the Children: Shortening the Path to EFA.” Education Notes 26884, World Bank, Washington, DC. Moyi, Peter. 2011. “Child Labor and School Attendance in Kenya.” Educational Research and Reviews 6(1): 26–35. Nekatibeb, Teshome. 2003. “Low Participation of Female Students in Primary Education: A Case Study of Dropouts from the Amhara and Oromia Regional States in Ethiopia.” UNESCO International Institute for Capacity Building in Africa. unesdoc.unesco.org​ /images/0013/001322/13224leo.pdf. Oreopoulos, Philip, Marianne E. Page, and Ann Huff Stevens. 2006. “The Intergenerational Effects of Compulsory Schooling.” Journal of Labor Economics 24 (4): 729–60. Ozier, Owen. 2010. “The Impact of Secondary Schooling in Kenya: A Regression Discontinuity Analysis.” Manuscript, Department of Economics, University of California at Berkeley. Pryor, John, and Joseph Ghartey Ampiah. 2003. Understandings of Education in an African Village: The Impact of Information and Communication Technologies. Report on U.K. Department for International Development Research Project Ed2000-88, Department for International Development, London. Psacharopoulos, George, and Harry Anthony Patrinos. 2004. “Returns to Investment in Education: A Further Update.” Education Economics 12 (2): 112–34. Roebucka, M. Christopher, Michael T. French, and Michael L. Dennis. 2004. “Adolescent Marijuana Use and School Attendance.” Economics of Education Review 23 (2): 133–41. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Why Do Sub-Saharan African Youth Drop Out of School? 59 Rumberger, Russell W. 2001. “Who Drops Out of School and Why?” Paper prepared for the National Research Council, Committee on Educational Excellence and Testing Equity Workshop, School Completion in Standards-Based Reform: Facts and Strategies, National Research Council, Washington, DC, July 17–18. Rumberger, Russell W., and Stephen P. Lamb. 2003. “The Early Employment and Further Education Experiences of High School Dropouts: A Comparative Study of the United States and Australia.” Economics of Education Review 22 (4): 353–66. Sackey, Harry A. 2008. “Private Returns to Education in Ghana: Implications for Investments in Schooling and Migration.” AERC Research Paper 174, African Economic Research Consortium, Nairobi, Kenya. Schultz, T. Paul. 2003. “Evidence of Returns to Schooling in Africa from Household Surveys: Monitoring and Restructuring the Market for Education.” Yale University Economic Growth Center Discussion Paper 875, New Haven, CT. Seymour, Claudia. 2011. Gaps in Accessing Formal and Non-Formal Education Opportunities for Youth in DRC. Final report, Save the Children. Thomas, Wayne, Don J. Webber, and Fiona Walton. 2002. “School Leaving Intentions at the Age of Sixteen: Evidence from a Multicultural City Environment.” Economic Issues 7 (1): 1–14. UNESCO (United Nations Educational, Scientific, and Cultural Organization). 2011. The Hidden Crises: Armed Conflict and Education. Translated by EFA Global Monitoring Report. Paris: UNESCO. Verspoor, Adriaan, and Jacob Bregman. 2008. At the Crossroads: Challenges for Secondary Education in Africa. Synthesis Report. Washington, DC: Africa Human Development Department, World Bank. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Cha p t e r 3 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth Introduction Chapter 3 provides a regional comparison of the incidence of out-of-school youth across various demographic and economic variables. Macroeconomic, socioeconomic, and education data (from the UNESCO Institute for Statistics) are used for an i­ntra-regional analysis of the incidence of out-of-school youth, comparing the performance of groups of countries with shared characteristics average. Specifically, for each variable considered (such as edu- to the regional ­ cation expenditures as a share of gross domestic product [GDP] or public expenditure on secondary e ­ ducation), countries are grouped in four quartiles (lowest, lower, higher, highest), and the schooling o ­ utcomes for each quartile are compared with the regional performance. For example, in the region, countries spend, on average, 4.8 percent of their GDP on education, but among the lowest quartile, the share is 2.8 percent, and among the highest, 7.3 percent. When countries are grouped in this manner, the group that spends the highest share of its GDP on education has the lowest incidence of out-of-school youth for all three age cohorts: among the highest-spending ­ countries, the shares of out-of-school youth are 15–20 percent lower than the regional average. The analysis shows that youth are less likely to be out of school in countries that dedicate a larger share of their GDP to education and a larger share of their public funds to secondary education, and invest more in school infrastructure. The incidence of out-of-school youth is lower in countries with high gross enroll- ment rates, but also in countries that are achieving the fastest growth in gross enrollment rates. Youth are more likely to be in school if the population is stable, and when a larger share of labor is hired in the formal sector. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   61   62 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth Education Expenditure as a Percentage of GDP Across the 25 countries for which data are available,1 countries commit on ­ average 4.8 percent of their GDP to education. Among the lowest quartile, coun- tries devote approximately 2.8 percent of their GDP to education. In this group, the share of out-of-school youth is about the same as the regional average in each age cohort. However, among those that devote the highest share of their GDP to education in the region, the share of out-of-school youth is significantly below the regional average for each cohort. For example, only 8 percent of youth aged 12–14 are out of school within this quartile compared with 23 percent in the region (table 3.1 and figure 3.1). Table 3.1 Education Expenditure as a Share of GDP (by Quartile) and ­Out-of-School Youth Education expenditure Share of out-of-school youth (%) as a share of GDP (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 2.8 23 44 72 Lower 4.1 33 51 74 Average 4.8 23 42 71 Higher 5.3 25 45 76 Highest 7.3 8 27 60 Source: UNESCO Institute for Statistics and World Bank EdStats. Figure 3.1 Out-of-School Youth, Difference from Regional Mean, by Education Expenditure as a Share of GDP (in Quartiles) 15 10 Difference from regional mean (%) 5 0 –5 –10 –15 –20 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth 63 Share of Secondary Education Expenditure in Total Public Education Expenditure Among the countries that spend a large share of their public education funds on secondary education, the incidence of out-of-school youth is lowest. In the region, countries2 spend approximately 28 percent of their public education resources on secondary education, but this figure is 41 percent for the top quar- tile, which includes the Cameroon, the Republic of Congo, Democratic Republic of Congo, Ghana, Mali, Rwanda, and Swaziland (incidentally, Mali, the country with one of the highest incidences of out-of-school youth, is the outlier in this group). In this group, the share of out-of-school youth is 5 percentage points below the regional average for the 12- to 14-year-old and 19- to 24-year-old cohorts, and 6 p­ ercentage points below the regional average for the 15- to 18-year-old cohort. On the other hand, across the lowest quartile (Burkina Faso, The Gambia, Ethiopia, Kenya, Madagascar, Niger, and Tanzania), the share of secondary education expenditure is only 20 percent of total public education expenditure, and this is reflected in the magnitude of the out-of- school youth problem, which is 4–8 percent higher than the regional average for the three cohorts (table 3.2 and figure 3.2). Table 3.2 Secondary Education Expenditure as a Share of Total Public Education Expenditure (by Quartile) and Out-of-School Youth Secondary education expenditure as share Share of out-of-school youth (%) of total public education expenditure (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 19.7 27 50 75 Lower 25.4 21 38 68 Average 28.7 23 42 71 Higher 29.5 24 43 74 Highest 41.0 18 36 66 Source: UNESCO Institute for Statistics and World Bank EdStats. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 64 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth Figure 3.2 Out-of-School Youth, Difference from Regional Mean, by Secondary Education Expenditure as a Share of Total Public Education Expenditure (in Quartiles) 10 8 Difference from regional mean (%) 6 4 2 0 –2 –4 –6 –8 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Gross Enrollment and Gross Enrollment Growth While it is not surprising that the incidence of out-of-school youth is lowest in countries with the highest gross enrollment rates, the relationship between the share of out-of-school youth and growth in gross enrollment rates is sur- prising. First, for the 26 countries for which both out-of-school youth and gross enrollment rate data are available, the average gross enrollment rate is 39.1 percent, and among those with the highest gross enrollment rates (The Gambia, Ghana, Kenya, Namibia, São Tomé and Príncipe, and Swaziland), the share of out-of-school youth is up to 10 percent points below the regional averages for each cohort (table 3.3 and figure 3.3). For the 24 countries with gross enrollment rate data available roughly for the period between 2000 and 2011, those with the lowest gross enrollment rate growth have the highest incidence of out-of-school youth for all cohorts in the region, suggesting that these countries have not improved secondary school expansion at all in the last decade or so. Among the second-lowest quartile of gross enrollment growth rate, on the other hand, the out-­ of-school youth incidence is lowest across the region for the 15–18 and 19–24 age cohorts, suggesting that coun- tries that have hit a certain enrollment level may no longer try to improve expansion in secondary education, even when their performance is below ideal levels (table 3.4 and figure 3.4). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth 65 Table 3.3  Gross Enrollment Rates (by Quartile) and Out-of-School Youth Share of out-of-school youth (%) Gross enrollment rate (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 23.9 26 47 74 Lower 34.9 23 45 76 Average 39.1 21 41 70 Higher 41.3 25 40 67 Highest 59.0 11 31 62 Source: UNESCO Institute for Statistics and World Bank EdStats. Figure 3.3 Out-of-School Youth, Difference from Regional Mean, by Gross Enrollment Rate (in Quartiles) 8 6 4 Difference from regional mean (%) 2 0 –2 –4 –6 –8 –10 –12 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Table 3.4  Gross Enrollment Rate Growth (by Quartile) and Out-of-School Youth Share of out-of-school youth (%) Gross enrollment rate growth (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 0.8 27 47 73 Lower 1.9 19 36 65 Average 2.1 21 41 70 Higher 2.2 17 40 71 Highest 3.4 21 40 70 Source: UNESCO Institute for Statistics and World Bank EdStats. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 66 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth Figure 3.4 Out-of-School Youth, Difference from Regional Mean, by Gross Enrollment Rate Growth (in Quartiles) 8 6 Difference from regional mean (%) 4 2 0 –2 –4 –6 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Investment in School Infrastructure Countries with adequate investment in school infrastructure appear to have a lower incidence of out-of-school youth across all three age cohorts. Data on schools with bathroom facilities are available for 20 countries3 and show a clear negative correlation between the incidence of out-of-school youth and availabil- ity of facilities in the school, a reasonably good proxy for capital investments. Among the lowest quartile, only 35 percent of schools have toilets, and across this group the share of out-of-school youth is greater than the regional average, especially among the youngest cohort of 12- to 14-year-olds. On the other hand, among the highest quartile, where bathroom facilities are ubiquitous, the inci- dence of out-of-school youth is 14 percentage points below the regional average for the youngest cohort, 11 percentage points lower for the 15- to 18-year-old cohort, and 2 percent lower for the 19- to 24-year-old cohort, suggesting that parents of younger children are especially sensitive to lack of adequate facilities (table 3.5 and figure 3.5). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth 67 Table 3.5 Share of Schools with Toilets (by Quartile) and Out-of-School Youth Share of out-of-school youth (%) Share of schools with toilets (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 35 29 45 72 Lower 57 27 47 73 Average 64 23 43 71 Higher 72 24 40 68 Highest 94 9 32 69 Source: UNESCO Institute for Statistics and World Bank EdStats. Figure 3.5 Out-of-School Youth, Difference from Regional Mean, by Share of Schools with Toilets (in Quartiles) 10 5 Difference from regional mean (%) 0 –5 –10 –15 –20 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Population Growth Population growth, especially that of youth, is one of the biggest challenges in addressing the out-of-school youth problem, as evident in cross-country data. Across the 31 countries in the region for which out-of-school youth data are available, the average population growth rate is 2.53 percent, but cross-country variations are large (with a population growth rate of 1.03 percent in Lesotho and 4.16 percent in Zambia). Among the countries with the lowest population Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 68 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth growth rate, the share of out-of-school youth is 7 percentage points below the regional mean among 12- to 14-year-olds, and 4 percent below the regional mean among 15- to 18-year-olds. On the other hand, those countries with the highest population growth rate face the largest magnitude of out-of-school youth: in this group, the youngest show the strongest correlation, with the inci- dence of out-of-school youth 6 percentage points above the regional average (table 3.6 and figure 3.6). Table 3.6 Population Growth Rate (by Quartile) and Out-of-School Youth Share of out-of-school youth (%) Population growth (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 1.71 14 36 69 Lower 2.37 21 38 66 Average 2.53 21 40 69 Higher 2.76 23 42 71 Highest 3.30 27 44 71 Source: UNESCO Institute for Statistics and World Bank EdStats. Figure 3.6 Out-of-School Youth, Difference from Regional Mean, by Population Growth Rate (in Quartiles) 8 6 Difference from regional mean (%) 4 2 0 –2 –4 –6 –8 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth 69 Share of Wage and Salaried Workers and Vulnerable Employment A strong formal labor market reduces the incidence of out-of-school youth. In general in Sub-Saharan Africa, formal labor markets are not very robust, but even small increases in the share of wage and salaried workers predict a smaller share of youth out of school. Data are available for 11 countries, and among those, the share of wage and salaried workers is 14 percent (table 3.7). Among those with the lowest share of formal employment (Burkina Faso, Ethiopia, and Niger), the share of out-of-school youth is 16 percentage points above the regional average for the 12- to 14-year-old cohort, 19 percentage points above for the 15- to 18-year-old cohort, and 13 percentage points above for the 19- to 24-year-old cohort. Among those with the most robust ­ formal markets (Uganda, Ghana, and the Republic of Congo), on the other hand, the share of out-of-school youth is below the regional average for all cohorts: by 16, 15, and 11 percentage points, in order of age cohort, suggesting that younger youth benefit the most from having parents with formal employment (figure 3.7). Similarly, when a larger share of the population holds jobs characterized as vulnerable (that is, self-employed and contributing family workers), the inci- dence of out-of-school youth increases. Across the 11 countries in the region for which both vulnerable employment and out-of-school youth data are avail- able, on average, 83 percent of those employed are classified as “vulnerable.” Among the group with the lowest incidence of vulnerable employment, the share of out-of-school youth aged 19–24 years is also lowest—9 percent below the regional average for 12- to 14-year-olds, 14 percentage points below the average for 15- to 18-year-olds, and 12 percentage points below for the 19- to 24-year-old cohort. For the top two quartiles for vulnerable employment, these findings are reversed (table 3.8 and figure 3.8). The data suggest that older youth benefit the most when the economy is less reliant on household production. Table 3.7 Share of Wage and Salaried Workers (by Quartile) and Out-of-School Youth Share of out-of-school youth (%) Share of wage and salaried workers (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 7 41 62 82 Lower 12 30 55 80 Average 14 25 43 69 Higher 17 19 28 57 Highest 20 9 28 58 Source: UNESCO Institute for Statistics and World Bank EdStats. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 70 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth Figure 3.7 Out-of-School Youth, Difference from Regional Mean, by Share of Wage and Salaried Workers (in Quartiles) 25 20 Difference from regional mean (%) 15 10 5 0 –5 –10 –15 –20 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Table 3.8 Vulnerable Employment (by Quartiles) and Out-of-School Youth Share of out-of-school youth (%) Vulnerable employment (%) Ages 12–14 years Ages 15–18 years Ages 19–24 years Lowest 78 16 30 57 Lower 81 11 25 59 Average 83 25 43 69 Higher 86 43 68 86 Highest 89 28 49 76 Source: UNESCO Institute for Statistics and World Bank EdStats. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Shared Characteristics of Countries with a High Incidence of Out-of-School Youth 71 Figure 3.8 Out-of-School Youth, Difference from Regional Mean, by Share of Vulnerable Employment (in Quartiles) 30 25 20 Difference from regional mean (%) 15 10 5 0 –5 –10 –15 –20 –25 Lowest Lower Higher Highest Ages 12–14 years Ages 15–18 years Ages 19–24 years Source: UNESCO Institute for Statistics and World Bank EdStats. Notes 1. These countries are, in ascending order of the share of GDP spent on education: Zimbabwe, the Democratic Republic of Congo, Madagascar, Chad, Guinea, Uganda, Cameroon, Sierra Leone, The Gambia, Burkina Faso, Niger, Côte d’Ivoire, Ethiopia, Rwanda, Mali, Benin, Malawi, Senegal, Burundi, Tanzania, the Republic of Congo, Kenya, Ghana, Swaziland, and Namibia. 2. Data are available for the same 25 countries. 3. These data are available at the primary level only for Niger, Chad, Ethiopia, Côte d’Ivoire, Cameroon, the Republic of Congo, Mali, Ghana, Madagascar, Benin, Burkina Faso, Guinea, Namibia, the Democratic Republic of Congo, Swaziland, Malawi, Burundi, Rwanda, Tanzania, and Uganda (in ascending order of the share of schools with toilets). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Cha p t e r 4 Paths Back to Formal or Informal Education or to the Labor Force Introduction The transition from school to work is difficult for youth in Sub-Saharan Africa, and programs designed for out-of-school youth must overcome multiple odds. Poor employment outcomes are not just a manifestation of poor education outcomes—demand for labor has been declining, and this decline affects youth ­ disproportionately. The large youth cohort, combined with internal migration, increases the supply of labor in large, urban areas and further reduces the odds of youth finding gainful employment in formal markets. In most countries, formal wage and salaried positions are rare, but even rarer for youth: in Uganda, Burkina Faso, and Burundi, fewer than one in 20 youth has a wage or salaried position— this figure is fewer than one in 10 for Cameroon, Madagascar, and Zambia, and fewer than one in five for The Gambia, Kenya, Malawi, and Mozambique. These odds are further worsened by poor health and pandemics ­ (specifically HIV/AIDS) and conflict and war (Garcia and Fares 2008). The ­ difficulty of finding a job is also evident from the number of years it takes to transition from school to employ- ment. According to one estimate, the median number of years it takes to find a job after leaving school in Sub-Saharan Africa is five (Guarcello et al. 2008). In fact, many youth who complete their formal education also face unemployment. Thus, programs that target better education and employment outcomes for out-of-school youth must tackle multiple problems. To be most effective, broad- based programs should focus on increasing opportunities for employment for everyone—not just a certain set of youth—which can only be achieved through better economic policies that promote growth. That said, there is still a significant role for programs that improve skills, human c ­ apital, and life outcomes. Countries in the region have tried both demand- and supply-side interven- tions but most programs are crippled with resource constraints or fail to achieve long-lasting results. A review of these programs is still useful, because they help delineate various entry points for at-risk and out-of-school youth, and these dif- ferent entry points help frame both long- and short-term policy interventions Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   73   74 Paths Back to Formal or Informal Education or to the Labor Force suitable under various conditions.1 A review of existing programs in the region shows that interventions can be grouped, albeit loosely, by the pathway each program proposes for out-of-school youth (Lerman 2005). Three such key entry points are: (a) retention of at-risk youth in school; (b) remediation programs that focus on bringing out-of-school youth back to formal schooling or alternative education programs; and (c) integration of youth into the labor market through workforce development (figure 4.1). For those youth who have enrolled in school, the most immediate policy intervention is to retain them in school. Identifying and helping at-risk youth is a common theme, not just in Sub-Saharan Africa but across the world, but this group is notoriously hard to remedy, especially when both demand- and supply- side factors lead to the decision to drop out. Youth could be at risk of dropping out because of academic reasons, the family or youth’s own preferences, or income constraints. At the secondary level, compared with primary education, these factors are even stronger because academic gaps are bigger, income con- straints are amplified (because youth’s labor is worth more), and family demands on youth (such as work and marriage) are greater. Furthermore, policies that address one of these underlying causes could exacerbate other causes. For exam- ple, implementation of compulsory education at the secondary level has indeed Figure 4.1 Policy Interventions for Out-of-School Youth School access and quality Household characteristics Youth characteristics • Lack of schools • Poverty • Cognitive abilities Reasons for dropping out • Poor academic quality • Misinformation • Noncognitive skills • Misinformation • Poor support • Attitudes toward education • Risk taking At-risk youth Early intervention 1. Retention in Improve More schools school quality Compulsory education Change the cost of Cognitive Soft skills education Change Academic support, perception skills development motivation • Eliminate school fees development of education Out-of-school • Reduce out-of pocket Key entry points youth costs Target group 2. Remediation Organizational reforms • Incentives to stay in through formal school or alternative education Alternative diploma programs Second chance programs Workforce development/skills training 3. Integration with the labor Private on-the-job training market Life skills training Direct expenditures for Subsidies or conditional job training Direct education cash transfers for Subsidies for jobs for youth expenditures households Tax expenditures supporting youth employment Intervention tools Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 75 helped increase initial enrollments, but it has also resulted in declining quality and shortages and deteriorating academic standards, which have acted as push factors. Scholarships and financial aid programs, while effective in removing financial barriers, help only a handful of students who come from households where education is valued greatly. For the rest, large-scale cash transfer programs have produced some positive results in enrollment, but these programs rarely cover the targeted population in a comprehensive way because they compete for resources with other government programs. For youth already out of school, the most likely path to complete their educa- tion is alternative education systems such as equivalency programs. Successful alternative education programs are those with multiple entry and exit points and close associations with formal education. But because lack of adequate formal education is a significant problem in Sub-Saharan Africa, designing effective alternative education programs in the region is especially hard. The two biggest constraints in implementing nationwide alternative education programs are the coordination between national government and subnational entities (regional authorities, communities, local governments, or other stakeholders) and avail- ability of funding. A myriad of small-scale alternative education schemes target out-of-school youth; these are especially successful when they mix academics or cognitive skills with training in life skills as well as mentoring. But survival concerns appear to limit participation in such programs, especially for marginal- ­ ized youth. Finally, little information exists on how successful second chance programs are as a bridge to formal education. For those youth who are not likely to go back to school, the alternative path is practical training and experience to increase their employability. Given that a large share of the economy remains informal in Sub-Saharan Africa, informal apprenticeships are the key mechanism for out-of-school youth to learn skills and find employment. But experience in the region, such as in Ghana, shows that this mechanism is notoriously hard to formalize. There is no shortage of large-scale regional or national training and workforce development programs across the region with varying degrees of success in creating long-term employment for participating youth. Seventeen African countries earmark funds for national training, but the bulk of these funds goes to train those who cannot find jobs when they complete their formal educa- tion. Large-scale national programs that intend to train unemployed and undereducated youth are not very successful. Wage subsidies and large public works projects can support large groups of youth, but rarely produce long- lasting employability, perhaps due to the stigma attached to them. Reviews of earlier programs are dismal—public works programs have not led to per- manent increases in employment and have been found to be too costly com- pared with the benefits they generate. But some recent public works programs that combine subsidized jobs with skills development (such as the Jovenes programs in Latin America) have been found to perform better (Betcherman, ­ Olivas, and Dar 2004). Sub-Saharan Africa now has a handful of programs following the examples set by the Jovenes programs, which are targeted, Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 76 Paths Back to Formal or Informal Education or to the Labor Force decentralized, and coordinated across the state, civilian entities, and the private sector. Just like alternative education schemes, nongovernmental orga- ­ nizations (NGOs) and nonprofits run a myriad of workforce development ­ programs that provide a mixture of training, on-the-job experience, life skills counseling, and mentoring. These programs are sometimes successful in inte- grating out-of-school youth into labor markets, but their effectiveness is limited by their size and funding, especially those programs that rely on international donor support. Business training and microfinance programs also demonstrate success. While underlying causes for dropping out of school can be discussed in isolation, these indicators are strongly correlated, and most programs tackle multiple issues (for example, skills development, financial constraints, and poor health or risky behavior). Thus, many interventions differ not in “what” problem they are trying to solve, but in “how” they solve it. For example, pro- grams that aim at improving noncognitive skills can be school-based, targeting youth at risk of dropping out, and offer mentoring, or can focus on teaching students life skills, studying or listening skills, or optimism and confidence (such as the worldwide Aflatoun and Afloteen programs or the Big Brother/ Big Sister programs in the United States) combined with academic support. When offered as workforce readiness courses, such programs might focus on traits valued by employers. Similar programs that aim at reducing the direct or indirect cost of attending school—addressing income constraints—vary from generalized subsidies for schools to subsidies for businesses offering work opportunities for out-of-school youth, to targeted or conditional cash trans- fers. Programs focused on jobs can vary anywhere from massive public works programs to apprenticeships or on-the-job training, sometimes combined with life skills education. Entry Point 1: Retention of At-Risk Youth in School Little data exist on how frequently Sub-Saharan African students return to school after dropping out, or on how such students fare. Studies from developed countries find that students tend to return to school when they are faced with low employment prospects and when they are convinced that they are unlikely to find jobs without further credentialing or skills development. However, these studies suggest that retention is a key problem for this group when they reenroll, suggesting that the underlying cause of dropping out is more complicated than a poor decision about future earning potential. For example, a U.S. study found that students were more likely to return to school when motivated and offered immediate reenrollment and support and counseling by the school staff. However, a majority of the reenrollees dropped out again in their first year of high school, while some tried to reenroll multiple times and returned to school (and generally their original school) only for one year after they reenrolled. Furthermore, one-third failed to complete even one course, and only one in five graduated (Berliner et al. 2008). Another study of reenrollees aged 12 and older Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 77 found that dropouts of higher socioeconomic status and better cognitive skills (measured as higher achievement test scores) were more likely to attain some sort of degree—a GED or a diploma. Youth who dropped out later were more likely to obtain a diploma, and ethnicity and gender were not significant at all when controlling for other factors (Wayman 2001). One of the immediate responses to the retention problem is compulsory education. In fact, many countries in the region now require compulsory ­ lower secondary education (table 4.1). While compulsory education, which defines the number of years that children and youth must attend school, does not necessarily ensure immediate higher enrollment, it does signal a country’s com- mitment to provide access to education (Oreopoulos, Page, and Stevens 2006). It may also counter poor attitudes toward education and poor perceptions about the returns to schooling. A recent study of lower secondary enrollment around the world found that 80 percent of all adolescents of lower secondary school age live in countries where lower secondary is compulsory, but among this group, 15 percent are out of school (Bruneforth and Wallet 2010). Table 4.1 Countries Grouped by Compulsory Education, by Level Primary Lower secondary Upper secondary Angola Botswana Mauritius Benin Burkina Faso São Tomé and Príncipe Burundi Chad Seychelles Cameroon Comoros Cape Verde Congo, Dem. Rep. Central African Republic Congo, Rep. Equatorial Guinea Côte d’Ivoire Gambia, The Eritrea Guinea-Bissau Gabon Mozambique Ghana Niger Guinea Sierra Leone Kenya Swaziland Liberia Tanzania Madagascar Uganda Malawi Zambia Mali Zimbabwe Mauritania Namibia Nigeria Rwanda Senegal South Africa Sudan Togo Uganda Source: Coffin 2013. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 78 Paths Back to Formal or Informal Education or to the Labor Force However, the introduction of compulsory education puts other strains on the education system. The Ugandan experience with compulsory free secondary education is illustrative. Uganda implemented universal secondary education in 2007, the first country in Sub-Saharan Africa to do so. While attendance is con- ditional on achievement at the primary level, the government has invested in increasing the number of public schools. It also subsidizes private school fees for eligible students by approximately US$52 per student per year. In the program’s first year, 69 percent of primary school graduates continued on to secondary school as opposed to 50 percent prior to its implementation, and enrollment increased the most among girls from poor households (Asankha and Takashi 2011). However, there is general concern that compulsory secondary education has resulted in deteriorating academic standards, since promoting students has become a bigger concern (Hedger et al. 2010). Similarly, Kenya, which intro- duced universal secondary education in 2008, has been struggling with shortages of classrooms and teachers and delayed disbursement of government funds, as well as declining quality (Kavuma 2011). These examples suggest that while compulsory education might lead to rapid enrollment increases, the requirement may be too ambitious for countries where there are significant supply-side and funding problems. By relaxing income constraints, financial aid and scholarships can improve enrollment and retention. There are many scholarship programs in Sub-Saharan Africa, and with the expansion of universal primary education, a renewed focus on expanding scholarships for secondary school youth. One example is the Ambassadors’ Girls’ Scholarship Program, a U.S. Agency for International Development (USAID) initiative started in 2004 with a commit- ment to provide more than half a million scholarships to primary school–age girls in 13 countries in Africa.2 Scholarships are focused on supporting the retention of girls in school at the primary level, but in certain countries have been extended to support girls at the lower and upper secondary school levels.3 Scholarship items vary from country to country but mostly cover school fees and out-of-pocket expenses such as textbooks. In The Gambia, Guinea, and Senegal, more than half of the scholarship funds go toward the purchase of books. In Nigeria, approximately 12 percent goes toward the payment of school fees. The program is implemented through regional NGOs. By 2009, the pro- gram had distributed 144,134 scholarships (125,210 of which went to girls) in 1,639 schools through 43 NGOs, and had worked with close to 2,000 mentors. By this time, 25 percent of scholarship recipients (more than 35,000 students) were secondary school–aged youth who successfully completed primary educa- tion but lacked the funds to continue into secondary education. The program led the way for scholarships for secondary-level students in Benin and Mali, mostly leveraging private donations. Mali’s secondary-level program has sup- ported more than 660 girls to date. In Benin, 229 graduates of the Ambassadors’ Girls’ Scholarship Program, received scholarships, and the program aims at sup- porting approximately 300 girls per year. Mentoring girls to build soft skills such as confidence is a key part of the Beninese program. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 79 In Northern Uganda, a small financial aid program (Acholi Education Initiative Scholarships) supports out-of-pocket expenses and school fees for displaced children. The program targets adolescents who had been abducted and con- scripted into the rebel ranks but returned home, orphans, and displaced children, especially young women with children of their own. The program also offers other interventions such as counseling to reduce risky behavior and dropout rates and to support the psychological well-being of recipients. The program is funded through private donations, and some recipients may find themselves out of funds when the donors terminate their relationship with the local NGO. In addition, financial constraints keep the program small and limited to formal secondary education, although there is great demand among recipients to use funds toward vocational education and tertiary education (Acholi Education Initiative 2013). On the other end of the financial support spectrum are cash transfers, mainly designed to alleviate household poverty. Sub-Saharan Africa has many cash trans- fer programs that target very poor households and vulnerable children and youth (table 4.2). Cash transfer programs are relatively easy to administer and could lead to significant enrollment increases among youth,4 but these programs must compete with other social protection programs, including education subsidies, for funds from the government and international donors. The largest cash transfer program in the region with an education component is South Africa’s national child support grant, which reaches more than 9 million children. Among the newer and more current programs is Kenya’s Program for Orphans and Vulnerable Children, started in 2004 and expanded to more than 100,000 house- holds in almost all districts of the country by 2010. The program offers K Sh 1,500 per month per household, regardless of the number of eligible c ­ hildren or youth up to age 18. While the transfer is not conditional, the funds are expected to cover health care–related costs such as immunization and school expenditures such as fees and textbook costs. The program also has educational requirements for parents such as lectures on health and nutrition. However, it competes with other broad-based interventions such as free education or school improvement: when fully implemented, the program could require up to 1.7 percent of ­ government expenditures and half a percent of GDP (Ikiara 2009). Table 4.2 Newer Cash Transfer Programs in Africa with Some Education Support Component Date started and Country Program expansion Targeted households Kenya Program for Orphans and 2004–08 30,000–50,000 Vulnerable Children 2009–15 300,000 Zambia Social Cash Transfer Scheme 2007–08 (pilot) 9,600 households in 15 districts 2009–12 72 districts Malawi Social Cash Transfer Scheme 2007–08 25,000 2009–12 260,000 Uganda Cash Transfer Pilot Program 2007–10 9,000 Ethiopia Productive Safety Nets Program 2005 8 million Source: Adato and Hoddinott 2008. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 80 Paths Back to Formal or Informal Education or to the Labor Force The Zambian Social Cash Transfer Scheme, which targets households with no substantive income and more than three dependents, provides cash transfers of approximately US$10 per family, and an additional US$2.50 per child per month, paid out on a bimonthly basis. Impact evaluations show that the transfers have improved health and nutrition outcomes, livelihood opportunities, and par- ticipants’ outlook. Improvements in education outcomes were also substantial: half of youth who were not in school at the beginning of the program were enrolled during final evaluation (Schüring, Rompel, and Stanfield 2009). If expanded to all extremely poor households in the country, the program could cost US$46 million per year, approximately 0.4 percent of Zambian GDP, 1.4 percent of the government budget, or 4 percent of the annual foreign aid flow (Schubert 2005). Malawi’s Social Cash Transfer Scheme, which provides direct monthly payments to poor households (approximately US$4 per adult) with additional education-related bonuses (US$1.50 per elementary school stu- dent and US$3 per secondary school student), currently supports approximately 25,000 ultrapoor and labor-constrained households. The evaluation of the initial pilot program conducted in the Mchinji district showed increased attendance (children from households that received the transfers were 6 percent less likely to miss school) and some positive impact on enrollment (Miller 2009). If imple- mented to cover the estimated 250,000 poor households in the country, the program could cost approximately US$42 million per year, or about one-third of the government’s total expenditures on social protection programs and emer- gency aid (Schubert 2007). Liberia started a similar program in 2010, with households receiving US$10–US$25 per month depending on their size, and an additional US$2 (primary) to US$4 (secondary) per child as long as the children attend school. Approximately 2,000 households participate in the program, which has an annual cost of US$12 million. Ghana’s Livelihood Empowerment Against Poverty program, a large-scale cash transfer program implemented in 2008, has a component for orphans and vulnerable children up to age 14 that is intended to be conditional on school enrollment, attendance, and health out- comes, but the conditions are not always fully enforced (Jones, Ahadzie, and Doh 2009, 59). Conditional cash transfers provide incentives for families to keep their children in school and help improve life outcomes by discouraging risk-taking ­ behavior.5 The results from a conditional cash transfer program implemented in Malawi’s Zomba District (a district with high population density, high drop- out rates, low educational attainment, and high HIV/AIDS infection rates) showed important improvements. The program offered unmarried young women aged 13–22 and their families a combination of random cash awards (which went to the guardians and youth) and payment of school fees, which are the students’ responsibility at the secondary level in Malawi. The amount was about US$10 per month—approximately 15 percent of the household consumption budget, of which US$1–US$4 were cash awards for girls. The intervention was designed as a randomized trial, and the initial evaluation of the program showed that school enrollment among dropouts increased Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 81 significantly: 61.4 percent of dropouts among the treatment group returned to school compared with 17 percent among the control group. However, the effect of conditional transfers on dropout rates among those who were already enrolled at the beginning of the study was similar, suggesting that given their administrative cost advantages, unconditional cash transfers may be as good a policy tool for encouraging youth to stay in school (de Brauw and Hoddinott 2008; Garcia and Moore 2012). At the same time, the Malawi evaluation showed that conditional cash transfers have helped reduce risky behavior such as sexual activity among youth (up to 40 percent) and reduced the incidence of early marriage and teenage pregnancy. Among those who were out of school at the beginning of the program, the probability of getting married—the main alternative to attending school—and becoming pregnant declined by more than 40 percent and 30 percent, respectively. Other Sub-Saharan African countries are implementing cash transfer pro- grams to improve schooling outcomes. Burkina Faso’s recent pilot program in the Nahouri and Sanmatenga region is designed to compare the impact of condi- tional cash transfers to unconditional transfers on health and education o ­ utcomes. A group of participating households in randomly selected villages received cash transfers conditional on health visits, school enrollment, and attendance (Garcia and Moore 2012). A recent evaluation of the program suggested that as in the Malawi program, conditional and unconditional transfers produced comparable enrollment improvements for children already in school or those most likely to be enrolled, such as boys, older children, and high- ability children. But condi- tional transfers appear to be significantly more effective than unconditional transfers in improving the enrollment of children who are initially less likely to go to school, such as girls, younger children, and lower- ability children. The evaluation found that conditional cash transfers (annual amounts of approxi- mately US$18 for younger children and US$35 for older children) led to statisti- percent for cally significant increases in enrollment of 20.3 percent for girls, 37.3 ­ younger children, and 36.2 percent for low- ability children relative to mean enrollment in those subgroups (Akresh, de Walque, and Kazianga 2013). Nigeria’s In Care of the Poor (COPE) Program, implemented in 2008, provides cash transfers to extremely poor and vulnerable households on the condition that adult members attend training sessions, keep their children in school, and utilize health services. Mali’s Bourse Maman cash transfer program offered US$10 per month to poor households on the condition that children attend school 80 ­percent of the school year, and reportedly helped triple enrollment in one of the two participating regions between 2002 and 2007 (UNICEF 2009). The program also revealed that conditional cash transfers are not easy to implement since their administration requires effective targeting and enforcing of the condi- tions (Pereznieto 2009). Remedial support and mentoring can be a cost-effective means of helping at-risk youth so they stay in or return to school. One example from the region is ­ the Girls’ Education Project run by the National Union of Eritrean Youth and Students in Eritrea. The program offers tutoring to girls transitioning to Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 82 Paths Back to Formal or Informal Education or to the Labor Force secondary education and to those in secondary school who are academically weak. Tutorial classes offer support in both English and mathematics and cur- rently cover about 1,440 academically poor girls. Thus far, no systematic evalua- tion or monitoring of the project has been conducted, but girls involved in the project have been able to catch up in basic subjects in school, and in general, they have been promoted from grade to grade (World Bank 2008, 25–26). South Africa’s USIKO Program specifically targets at-risk youth. One of its programs provides school-based remedial support and mentoring that focuses both on academic and noncognitive skills. The program recruits male volunteers specifi- cally to mentor young men, to encourage school attendance and discourage risky behavior such as gang activities (World Bank 2008, 35). A similar program for girls is run with female volunteers (van Wyk and Naidoo 2009). Since 2000, more than 600 at-risk youth have successfully completed the mentorship pro- gram; of this group, more than 90 percent obtained high school qualifications. The program, which started as a small project at the psychology department of Stellenbosch University, has evolved into an independent NGO run by local community members (Naidoo and van Wyk 2008). Another important factor in retention is involving parents in the management of schools. Parental involvement not only improves student success, but also leads to more responsive schools. Across the region, many projects aim at providing literary skills to parents so that they can become a part of their children’s learning process. For example, between 2003 and 2008 Mali implemented, with funding from USAID, the Support for the Quality and Equity of Education Program—a community-based program with four objectives: improve teaching skills, improve curriculum, increase adult literacy, and increase parent participation in school management. The last two components aim at increasing parental involvement in children’s education. Literacy and motivational trainings for parents were held at community centers by voluntary facilitators with close links to the community. The facilitators and trainers did not receive any financial incentives, which helped lower program costs. The project served approximately 700 schools in its fifth (and last) year, the 2007–08 school year. By this time, the literacy program had served 17,637 basic literacy learners (of whom 6,524 were women) and 6,831 postliteracy learners (of whom 2,260 were women). Parents and commu- nity members had participated in 756 school management committees or parent associations. The implementers of the program noted an improvement in the quality of schools managed by the literacy program participants as well as in increased involvement of parents in their children’s education (UNESCO 2009). Similar programs have been implemented across Sub-Saharan Africa. For example, in Guinea the Community Participation in Education for Equity and Quality Program has led to increases in girls’ enrollment through increasing parental involvement as well as simple, low-cost, but important interventions such as adding latrines (Midling et al. 2006, 23). Uganda’s Family Basic Education Program, implemented by a local NGO and funded by international donors, started as a small pilot program in a single district (Bugiri district of Eastern Uganda) in 2000–01, and has since expanded to more than 600 villages in eight Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 83 districts in war-affected northern Uganda. Program evaluations report increased attendance (girls’ attendance increased by 67 days per year on average) and civic participation and decreased dropout rates, reports of domestic violence, and early marriage. In addition, parental involvement in education school management increased (Nyamugasira, Dorothy-Angura, and Robinson 2005). Entry Point 2: Remediation through Formal or Alternative Education Alternative education schemes in Sub-Saharan Africa typically consist of either accelerated learning programs that offer equivalency diplomas or vocational and technical education. Vocational and technical education is generally seen as the most direct link to the job market, but the evidence on the usefulness of voca- tional and technical alternative education for out-of-school youth is mixed, even in developed countries. For example a longitudinal study of out-of-school youth aged 14 and older from Australia found that about 37 percent of youth who dropped out pursued some kind of vocational education. The study found that among the youth followed, dropouts from parochial schools were more likely to participate, while poorer youth who lived in high- unemployment areas were less likely to participate. Most enrollees took trade-related courses, and students’ aca- demic achievement prior to dropping out, income levels, socioeconomic back- ground, and native language (in this case, English) were highly correlated with students’ ability to complete the vocational education (Ball and Lamb 2001). Second chance equivalency programs allow out-of-school youth to obtain credentials while continuing with their work or family obligations. One such example is the Complementary Basic Education in Tanzania (COBET) program. Tanzania experienced significant reductions in school enrollment and attendance starting in the 1980s, mostly in response to deteriorating school quality. One estimate put the number of school-age children who were out of school at 3 ­million, or about 40 percent (Helgesson 2001). The Ministry of Education and Culture implemented the COBET program in 1999 as a means of providing education to out-of-school children, with a particular emphasis on girls. The classes are conducted in COBET centers, and students do not incur any out-of-pocket expenses. No uniforms are required, and school days are only three and a half hours, allowing time for work and household chores. The program started with 20 centers in the two districts with the highest dropout rates (Masasi and Kisarawe) and had expanded to 50 centers in five districts by 2005. The program serves youth aged 11–18 years (in two separate cohorts) and provides numeracy and literacy skills, as well as vocational training for the older cohort, condensed into three years. Communities in collaboration with NGOs, religious groups, and the private sector are involved in implementing, monitoring, and evaluating the program. In 2005, a total of 617,131 youth enrolled in the COBET program, approximately 36 percent of the targeted enrollment for the same year (Macpherson 2007). While no evaluation of the program has been conducted, an earlier review of the pilot phase found that the COBET program did not meet its stated goals of expanding enrollment among girls (Helgesson 2001). Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 84 Paths Back to Formal or Informal Education or to the Labor Force Equivalency programs with flexible entry and exit and close links to the ­ formal education system have demonstrated greater success. Uganda’s Basic Education for Urban Poverty Areas (BEUPA) Program was one of the more suc- cessful equivalency programs offered in Sub-Saharan Africa. The program tar- geted youth aged 9–18 and offered a condensed curriculum of three years compared with the five-year basic education cycle; similar to the COBET pro- gram, it was delivered over a shorter school day (Thompson 2001; Lamichhane, Prasad, and Wagle 2008). While data from more recent years are not available, a 2002 review of the program found that of the more than 3,000 students served through 54 centers in Kampala, 55 percent were girls, more than a quarter trans- ferred to formal schools, and only about 10 percent dropped out (Ilon and Kyeyune 2002). The same evaluation found that the success of the BEUPA pro- gram was the result of a combination of factors: integration with a formal cur- riculum, provision of a combination of academic and life skills as well as career guidance, and strong community involvement in centers. Madagascar has two complementary programs, Planet of Alphas, and Accelerated Compressed Learning for Malagasy Adolescents (ASAMA), that col- lectively serve out-of-school youth aged 10–18 years to obtain their basic educa- tion credentials. ASAMA is an equivalency program offered to youth aged 12–14 who have had no access to formal school or who dropped out of grade 1 or grade 2 before completing the elementary level. The program covers the five-year basic education curriculum in 10 months, over three semesters. Participants can choose to complete the program and sit for an equivalency test for a diploma or return to school if they can demonstrate grade-level knowledge in a placement test. The program started as a pilot in Fianarantsoa, a city in south central Madagascar, and half the initial graduates were able to successfully pass equivalency tests and obtain their primary certificate. Initial cost evaluations showed that the per stu- dent cost of the program for 2003 was approximately US$100. The Planet of Alphas is a six-week intensive literacy and numeracy program intended for ado- lescents who cannot read or write. Despite the short program design, initial evalu- ations showed that more than two-thirds of the students passed the final test, approximately one in five enrolled in formal primary schools, and about one in 20 enrolled in the ASAMA Program (Joseph and Harison 2007). Calibrating second chance programs to meet the needs of target audiences can be tricky. In Zimbabwe, children typically drop out of school because they cannot afford school fees, and oftentimes cannot provide a birth certificate necessary to enroll or receive support (Kajura 2012). A recent program imple- ­ mented by the Children First project and the Open Society Initiative for Southern Africa was initially designed in response to a finding that more than half of Zimbabwean children failed to transition into form 1 (at the end of 7th grade). The remedial material was designed as a year-long curriculum to fast- track 12- to 15-year-old out-of-school youth so they could sit for the grade 7 exam. The materials were distributed to learning centers run by a national NGO (the Adult Literacy Organization of Zimbabwe, ALOZ). Program partners report that increasing numbers of youth, including youth who have never Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 85 attended school, are showing up at the sites. The majority of young people at the sites do not have basic literacy skills and will need more than one year to acquire the necessary skills (Bantwata Initiative 2012). Large-scale alternative education programs tend to target both youth and adults, and involve collaboration between national entities and localities. However, funding remains a key constraint. Sierra Leone’s Ministry of Education recently implemented the Support to Strengthen the Capacity of the Community Education Centres for Literacy and Vocational Skills for Women and Girls Program with financial support from UNESCO and in collaboration with local NGOs. The program targets out-of-school children and functionally illiterate youths and adults, particularly those from deprived social backgrounds, such as war victims, returnees, orphans, refugees, and internally displaced persons. The 10-month program offers a combination of basic literacy and skills, vocational education, civic education, and life skills that focus on financial responsibility, gender relations, and health. While the government provides supervision and technical assistance, the courses are offered by community-based learning centers with the help of volunteers who receive training and a small stipend from the ­ government. In its pilot phase, program participants (about 2,500 per year) gen- erally completed the coursework. However, its expansion is limited by funding, slow progress in rehabilitation of community centers, and lack of facilitators, especially females (and perhaps because of the erratic stipends).6 Ghana’s National Functional Literacy Program was successful in recruiting volunteer facilitators through effective use of rewards such as bicycles or sewing machines. This ­ program, like the one in Sierra Leone, was designed and overseen by the central government, but implemented through local organizations. It mixed functional literacy education with occupational skills, life skills, and health educa- tion. The program served more than 2 million adults (including its earlier form, initiated in 1992), and volunteers kept costs low. However, when international support ended, funding became a key constraint. In addition, while volunteers performed well teaching basic literacy skills, they were not always skilled in other areas, especially occupational training (Aryeetey and Kwak 2006). In 2005, Liberia’s Ministry of Education implemented the Advancing the Youth’s Alternative Education Project to provide alternative education to out-of-school youth aged 13–35 in livelihood training. The program supported community- based training in reading, writing, and math as well as work readiness, health, and life skills. It also supported youth networks, and has committed to supporting public-private partnerships with Liberian and international businesses so youth can find practical, work-based learning. The project was implemented through local NGOs, enrolling more than 70,000 learners in its peak year, but was phased out in 2011. An evaluation of the program suggested that it reached out to the most vulnerable groups in the country, and achieved consistently high attendance rates (Manda 2011). Local training centers can help generate funding for trainees. One such exam- ple is Ethiopia’s Community Based Nonformal Livelihood Skills Training for Youths and Adults (EXPRO) Program, which combines literacy training with Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 86 Paths Back to Formal or Informal Education or to the Labor Force a soft skills and entrepreneurial skills program. The program began in 2000, targeting adults and out-of-school children who have never completed their for- ­ mal education, especially those who live in rural areas of extreme poverty. The program, a collaboration between the Ministry of Education, regional education bureaus, and local technical and vocational education commissions, trains approximately 2,000 people annually. The participants undergo intensive train- ing in a trade (examples include tailoring, woodworking, tire repair, food prepara- tion, or animal husbandry) at local community centers or vocational education centers for an average of three months; sometimes the training includes a literacy component. The program was designed with the goal of linking training with access to credit to improve job outcomes, but this component of the project was not very successful (Sandhaas 2005). A 2004 review of the program showed that three-quarters of the participants in training modules were females. Similar to other large-scale programs, coordination between central and district- level representatives was weak, and the expansion of the program was limited by staff ­ shortages. Funding for the centers, which is provided by the government, was limited, which led to shortages in training materials and equipment. Some train- ing centers generate their own funding by selling milk or things like furniture that can former trainees made to be sold to schools or offices. Others have used for- mer trainees in construction to build additions to the center, such as offices, which could be used to generate additional income. No data exist to evaluate the impact of the program on the employment of trainees, but trainees have reported improvement in their qualifications as well as in their motivation to engage in income-generating activities. Recent innovative programs, some of them small-scale, mix vocational training with life skills programs in recognition of the links between employment pros- pects, poverty, and risky behavior by youth. In Uganda, a small randomized intervention that mixes vocational training with HIV/AIDS prevention skills shows that youth who received vocational training by local artisans in an appren- ticeship model in conjunction with prevention skills engaged in fewer delinquent behaviors and experienced greater improvements than the control group, which received delayed vocational training. For example, youth who received vocational training earlier were more likely to be employed and reported signifi- cantly greater increases in quality of life and social support. Youth were moni- tored at recruitment, 4 months, and 24 months, by which time 74 percent of the participants were employed. By the last assessment, both groups had received vocational training and both groups demonstrated improvements such as sustained employment and reduced risky behavior such as fewer sexual partners ­ and increased abstinence and condom use, and decreased alcohol, marijuana, and hard drug use (Rotheram-Borus et al. 2011). Approximately half the partici- pants had never been employed at the time they were recruited; by the end of the pilot, 86 percent were employed. Implemented by an NGO (Uganda Youth Development Link), the program was privately funded. A similar ­ program in Malawi (the Apprenticeship Training Program and Entrepreneurial Support for Vulnerable Youth) was set up to test whether youth who receive more Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 87 information on the impacts of their risky behavior and youth who improve their employability through training reduce engagement in risky behavior (Cho 2010), but the impact evaluation of the program has not yet been completed. In Sierra Leone, a midterm evaluation of a nonformal training program (part of the Employment Promotion Program implemented in 2006) that offers com- bined literacy and skills development courses with counseling for the youth demonstrated reduction of conflicts and violence among trainees, both at training facilities and in the community (German Technical Cooperation 2010). South Africa’s Urban Conflict Management Project (active through 2010) was built on a voluntary conflict resolution program, called the Community Peace Workers, and volunteers were offered cognitive and soft skills training as well as appren- ticeships. The program reduced crime (although hard numbers are not available), and almost all youth who volunteered as peace workers received vocational training, which eventually led to jobs, the majority of them in the security sector (German Technical Cooperation 2009). There is some evidence that vocational training outcomes improve when youth are presented with a menu of options. Kenya’s Technical and Vocational Vouchers Program was implemented in 2008 with the recruitment of approxi- mately 2,160 out-of-school youth (ranging in age from roughly 18–30). Only a quarter of participating youth had completed secondary education, and 27 ­percent of participants had dropped out before eighth grade. The program was designed to measure the effectiveness of training vouchers. Approximately half of the participants in the program were offered a voucher for vocational training, while the other half served as the control group. Of the voucher ­ winners, a ran- dom half were awarded a voucher restricted to public institutions, while the other half received an unrestricted voucher that could be used in either private or public institutions. The vouchers were approximately US$460 in value, suffi- cient to cover the cost of public or private vocational training. The project also included information sessions offered to half of all the participants (randomly selected from both the treatment and control groups) on the earnings gains from vocational education. The evaluation showed that offering young adults vouchers that cover program costs encouraged them to enroll, and that those who can use the voucher for a private training program are more likely to sign up and stay in school. Seventy-four percent of participants who received vouchers enrolled in some type of vocational training, compared with less than 4 percent of those in the control group. Participation rates were higher among those who received unrestricted vouchers (79 percent versus 64 percent), and dropout rates among those who received restricted vouchers were higher (the gap was approximately 16 percent). One interesting finding of the evaluation was that participants who had not completed secondary education were less likely to drop out compared to those who began the program with a secondary degree, suggesting that the less educated youth attributed a greater value to vocational training (Hamory Hicks et al. 2011). There are not sufficient data to demonstrate the success of the pro- gram in improving employment among participants, but qualitative assessments suggest that most graduates were able to find at least part-time jobs. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 88 Paths Back to Formal or Informal Education or to the Labor Force The effectiveness of alternative learning may be limited by survival concerns of participants. The Ugandan Non-Formal Education and Livelihood Skills for Marginalized Street and Slum Youth Training Program, implemented by an NGO between 2005 and 2009, specifically targeted out-of-school and socioeco- nomically vulnerable youth. UNESCO provided the funds. The training program focused on practical skills in specific trades including hairdressing, tailoring, motor mechanics, carpentry, electronics, welding, and cookery. It also offered life skills training with a focus on HIV/AIDS, reproductive health, nutrition, child rearing, and drug and alcohol abuse. The program recruited facilitators who were responsible for community mobilization, coordination, program monitoring, and recruiting of local skilled practitioners in the vocational trades, and who also acted as teachers and mentors for local youth. Facilitators also identified youth who were likely to benefit the most from the program. Most training was infor- mal and involved teaching of a trade by the local practitioner. The health educa- tion component of the program had some formal instruction such as seminars, group discussions, and lectures and involved easy-to-read learning materials. Participating youth were assessed by facilitators, social workers, and master arti- sans in a given trade. The program served 184 youth between 2004 and 2006, the majority of whom were female (152). Most of the participants (170) reported that they left school because of their inability to pay school fees. While data on completion rates do not exist, the program evaluation document reported that youth who dropped out noted that the program did not address their immediate economic problems and needs. In addition the informal nature of the program sometimes resulted in poor matches between the artisan and the apprenticing youth, conflicts about payments, and irregular work experience. Artisans involved in the training were not necessarily equipped to teach youth with little formal education. Commonly reported problems included slow learn- ing and destructive, disruptive, and unprofessional behavior such as begging for money from the customers among the participating youth (UNESCO 2006). But those who completed the program managed to find gainful employment, and generally reported improvements in their self-perception. Alternative education is particularly needed in conflict-affected areas, but large-scale, coordinated interventions are rare. A 2007 U.K. Department for International Development report found that alternative education is largely left to the domain of NGOs and community organizations in postconflict areas such as Northern Uganda, Somaliland, Eastern Cape, South Africa, Namibia, Sierra Leone, and the Democratic Republic of Congo (Dennis and Fentiman 2007). In Northern Uganda, where the conflict between the Government of Uganda and the rebel Lord’s Resistance Army (LRA) displaced more than 2 million people, interventions have targeted vulnerable groups including orphans of the war and HIV/AIDS, former abductees, and children and young people in camps. Education is highly decentralized in Uganda, and the local governments do not have the necessary means to support a large-scale intervention. The local devel- opment agency has put temporary school structures in camps and support teach- ing staff, but NGOs have also stepped in to provide alternative education. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 89 For example, St. Monica Girls’ Tailoring School, affiliated with a religious entity, has been running a vocational skills training program since 1982 for girls who were formerly abducted by the rebel forces. The program offers a combination of vocational skills and counseling services, similar to the scholarship program which includes a counseling component (Bell 2008). However, compared with the need, the numbers remain small: the tailoring school educated on average 50 students per year since 1982 (UNICEF 2008). In Somaliland, a local NGO, the Horn of Africa Voluntary Youth Committee, provides vocational training for youth intending to return to school. In place since 1998, the program was initially designed to provide training for poor returnees, but over time shifted its focus to out-of-school youth. Given the high demand, the program now tests incoming attendees and charges tuition. To date, it has trained approximately 3,000 youth. One of the larger programs for out-of-school youth in postconflict areas is Sierra Leone’s now-discontinued Youth Reintegration Training and Education for Peace Program. This program was active between 2001 and 2004 and targeted ex-combatants and other war-affected youth aged 15–34 years. The program offered nonformal education through community centers, and the curriculum included both academic education and life skills and reintegration training. The program reached 40,000 youth and adults within a year of implementation. A qualitative evaluation through a survey of program participants found that the program reached most of its goals at relatively low costs (US$150 per partici- pant) (Hansen, Wolf, and Sommers 2002). Participants reported improvement in personal skills such as conflict and stress management, problem solving, and management of their lives. They also reported improvements in their literacy and numeracy skills and in civic participation (Martin and Wingate 2001). The Republic of South Sudan’s Accelerated Learning Program, initially implemented by the Sudanese government, targets out-of-school youth, demobilized child soldiers, and young women, aged 12–18, who missed out on formal education. The program compresses the eight-year basic education curriculum into four years and incorporates additional life skills and health training. The program is designed so that participants can move between formal education and the accel- erated learning program. The program uses school buildings (Nicholson 2006). It has reached more than 120,000 learners in 260 centers run by NGOs and 815 run by the Ministry of Education. Girls’ participation rate in the program is higher than the participation rates of girls in formal education. Entry Point 3: Integration with the Labor Market Countries in Sub-Saharan Africa have a multitude of national workforce devel- opment programs that vary from large-scale training schemes to job exchanges and subsidized employment. These programs generally target the unemployed, and may not always serve the most vulnerable out-of-school youth because some academic qualification is usually required to participate in training and employ- ment schemes. Programs that specifically target out-of-school and disadvantaged youth tend to focus on vocational and life skills components (for example, Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 90 Paths Back to Formal or Informal Education or to the Labor Force empowerment programs in Kenya and Nigeria; the training programs offered to youth who work in the informal sector in Zimbabwe; and the combined appren- ticeship and training program in Côte d’Ivoire), and sometimes combine ele- ments of entrepreneurial training and support, such as training on writing proposals, business plans, financial and feasibility analysis, and legal analysis (for example, there are multiple entrepreneurial support programs in Burkina Faso, Tanzania, and Uganda), or programs that provide seed money or access to micro- finance. Most small-scale programs that specifically support out-of-school youth are run by NGOs, churches, or for-profit entities and are funded by NGOs, international donors, or multilateral organizations (Rother 2006). A review of the most common types of workforce development programs shows that these pro- grams are trying to address issues beyond the relatively plainer goal of returning youth to some type of education. First, the structure and the content of workforce development programs reflect the informal and undeveloped nature of labor markets. In Sub-Saharan Africa, informal apprenticeships are the key mechanism for out-of-school youth to learn skills and find employment. Formal apprenticeships are a part of techni- cal and vocational education systems in many Sub-Saharan African countries,7 but informal apprenticeships offered by businesses and tradespeople play an important role in promoting employability of out-of-school youth. This is partly because formal vocational training programs are not able to provide the relevant labor market training, and partly because nonformal and informal training appears to be a better fit for the largely informal labor market needs. In countries such as The Gambia, Ghana, Madagascar, Malawi, Mali, Senegal, Tanzania, and Zambia, an estimated 50–90 percent of young people participate in informal apprenticeships (Biavaschi et al. 2012) that tend to be purely practical, with no formal contracts or compensation schemes. Recruitment is through social net- works, usually favoring youth with some basic education. Because they are closely linked to the needs of the informal labor market, the apprenticeships generally lead to either self-employment or jobs with the business that provided the train- ing in the first place (Aggarwal, Hofmann, and Phiri 2010; Nübler, Hofmann, and Greiner 2010). One study from Ghana show that apprenticeships increase earn- ings of youth with little formal training by up to 50 percent (Monk, Sandefur, and Teal 2008). Governments’ attempts to formalize these informal apprenticeship systems have not always been successful (for example, in Ghana and Guinea). Seventeen Sub-Saharan African countries have national training funds ear- marked for preemployment or in-service training, as well as training for unem- ployed and disadvantaged groups such as out-of-school youth.8 As opposed to training provided by ministries or the formal education sector, training funds generally support third parties (training institutions, enterprises, or intermediary entities such as microfinance groups or the informal sector) to train the target group. The funds are usually set as autonomous entities governed by representa- tives from the government, employers, and in some cases, employees. The monies can be used to support training services or to provide wage subsidies for trainees, tax benefits for participating enterprises, or direct subsidies to trainees. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 91 Contracting for training is usually done through competitive bidding, which reduces training costs. The experience with training funds across Sub-Saharan Africa is generally positive—for example, funds in Benin, Burkina Faso, Côte d’Ivoire, Ghana, Madagascar, Mali, and Mauritania, all financed by international aid, have trained substantive numbers of participants, albeit with mixed employ- ment outcomes (Gyampo 2012), but the projects typically face sustainability issues when donor monies run out. One notable exception is South Africa, where a fifth of the payroll levy imposed on formal sector firms is transferred to the National Skills Fund, which provides training for disadvantaged groups and minorities. Approximately one-fourth of the National Skills Fund project expen- ditures are used to support youth training (Johanson 2009). Large-scale training and work programs run by national governments are not always successful. For example, training offered in the context of sector reforms, especially agricultural reforms, such as those in the Central African Republic (World Bank 1999) and Burundi (World Bank 1997), have not made any signifi- cant impact on the labor market outcomes of participating youth. Similarly, large public works programs for youth, such as the United Nations Development Programme (UNDP)-supported National Youth Environment Corps projects in The Gambia and Lesotho (which are now discontinued), South Africa’s Extended Public Works Program, and Sierra Leone’s Cash for Work programs, can be successful in employing young people en masse (Lal et al. 2010; Lieuw- Kie-Song and Philip 2010), and, when effectively targeted, can help with youth integration, but their effect on the long-term employability of youth is not strong, perhaps because these projects involve low-status jobs (such as construction of roads or municipal waste collection) with little future earning potential (Andrews and Kryeziu 2012). Nigeria’s Open Access Apprenticeship Program, despite reaching large numbers of youth (Haftendorn and Salzano 2004), suffered from high dropout rates (about 42 percent). The Côte d’Ivoire’s Appui à l’Introduction de la Formation Professionnelle par Alterance Program, which used a combina- tion of on-the-job and in-class training and job placement services for out-of- school youth, hit a roadblock in 2002 when economic and political conditions deteriorated. The program ended despite some evidence of success (albeit small scale) (Brewer 2004). In Ghana, preemployment training offered to secondary school graduates or dropouts, mostly in rural areas, suffered from low quality and high dropout rates among the trainees and coordination problems between the national and regional governments (Palmer 2007, 217). The program, offered by the Integrated Community Centres for Employable Skills Agency, was initially designed as a community-based alternative to formal education, where local artisans were expected to train youth and provide work opportunities. However, the training centers were never able to garner the necessary financial support from the communities (Palmer 2007). Again in Ghana, a short-term skills devel- opment training program called the Skills Training and Entrepreneurship Program ran into similar problems. Training costs were covered by the govern- ment, and the program offered training, apprenticeship, and job placement ser- vices and microfinance support to a combined 27,500 trainees through 2006. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 92 Paths Back to Formal or Informal Education or to the Labor Force However, despite high attendance, the Skills Training and Entrepreneurship Program was riddled with coordination problems between regions and the cen- tral government. The training areas and curricula, which were centrally dictated, did not meet regional needs, and funding flows from the central government to the regions experienced delays. Graduates of the training program oftentimes had a hard time accessing microcredit. The employment outcomes of the Skills Training and Entrepreneurship Program are unknown, and the program was eventually absorbed into the National Youth Employment Program. Second, workforce development programs try to ease the transition from school to work. Schools can be disconnected from the needs of the labor market, and it is not uncommon for workers with higher degrees to have a harder time finding work than those with less education. As a result, many programs combine some sort of training with some on-the-job experience, but these can run into problems because of lack of capacity or lack of coordination between the govern- ment entity running the program and the targeted businesses. Transitioning for students who dropped out of formal education could be particularly difficult if there is a stigma attached to being out of school or to graduating from a work- force development program. Experience from Latin America shows that large-scale, informal apprentice- ships and on-the-job training programs can be successful when they are targeted, decentralized, and coordinated across the state, civilian entities, and the private sector. Impact evaluations of the Jovenes programs have been very positive com- pared with similar large-scale workforce development programs elsewhere. The first Joven program was implemented in Chile in 1990, and then expanded to Argentina, Colombia, Peru, and Uruguay. Like many similar programs in Sub- Saharan Africa, the Jovenes programs offer a combination of training in basic academic skills and a specific trade, as well as job search assistance. Participants get work experience in a firm engaged in the trade, but firms are not expected to pay youth or to offer employment at the end of the training. The job training and practical training period is short—about six months combined. Participants get financial support for out-of-pocket expenses such as travel to and from work. In their review of the program, Betcherman, Olivas, and Dar (2004) outlined the following common characteristics of the Jovenes programs across countries: The programs are provided in collaboration between the state, which designs, super- vises, and provides financial support, and state agencies or civil entities, which implement local programs. The private sector provides the training and work experience. The programs target the intended youth and not broader groups by carefully calibrating the training areas, conditions for work experience, subsidy levels, educational qualifications, and geographic locations. While the central government supervises the programs, the implementation is local, and private firms that offer training or work experience components must go through a bid- ding process. Finally, the training programs offer noncognitive skills development such as good personal relations and self-esteem. The Jovenes programs are not suitable for all dropouts; they have been found to benefit youth with some edu- cation at the secondary level. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 93 Programs similar to Jovenes are now being implemented in Sub-Saharan Africa. Kenya’s Youth Empowerment Project includes a private sector internship and training component that offers youth aged 15–29 years work experience and skills training through internships and relevant training in the formal and informal sectors. To be eligible, youth must have at least eight years of schooling and have been out of school for at least a year. The program has placed a cap on the percent of youth with tertiary education (40 percent). The program plans to train approx- imately 10,000 youth over four years. The evaluation of the program will focus on a combination of employment and life outcomes such as changes in earnings and engagement in risky behavior. To date, 1,095 youth have attended life skills training (425 of them female), with an attendance rate of 80 percent, and 916 internships have been created, in both the informal and formal sectors. A majority of these interns have received additional business skills training (KEPSA 2011). Another successful program in the region is Uganda’s Youth Opportunities Program, implemented as part of a US$100,000 World Bank–funded project (the Northern Uganda Social Action Fund). The Youth Opportunities Program t ­argets unemployed or underemployed youth aged 15–35 years living in conditions of poverty. The program is a highly decentralized, community- and district-driven system of youth vocational training, offering up to US$10,000 in grants to youth groups that submit proposals in which they identify a vocational skill of interest and a vocational training institute. If their proposals are approved, youth groups receive a cash transfer deposited in a community bank account. These funds are used to enroll in the vocational training institute, purchase training materials, and equip graduates with the tools and start-up costs for practicing the trade after graduation. District technical offices under the Northern Uganda Social Action Fund manage the grants (Blattman, Fiala, and Martinez 2009). The program also includes components addressing conflict resolution. The program was evaluated through a randomized trial that measured the impact of the grants on employment and training outcomes. The initial evaluation showed that groups that received community grants: were almost four times more likely to participate in vocational training and twice as likely to be engaged in skilled work; improved their profits by 50 percent compared with the control group; and increased their savings by 20 percent. They were more likely to engage in civic activities and less likely to engage in aggressive activities ( ­ especially men) (Blattman, Fiala, and Martinez 2011). Training and apprenticeship programs offered by NGOs and other third p ­ arties tend to be small and limited in duration, especially when funded by international aid. For example, the Alliance for Youth Employability was a f ­ive-year program funded by USAID, Nokia, and Lions Clubs that provided training and employ- ment services to out-of-school and at-risk youth aged 14–29 years. The program trained approximately 900 youth across Rwanda, South Africa, and Tanzania in trades such as hospitality, carpentry, driving, tailoring, salon, media, and studio arts and also provided business start-up assistance. More than half of trained youth in South Africa were placed in employment (IYF 2009). The program ended after its five-year implementation period. The Liberian government recently implemented the Economic Empowerment of Adolescent Girls and Young Women Program, Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 94 Paths Back to Formal or Informal Education or to the Labor Force a training and job placement program for young women aged 16–27 years, with the help of World Bank financing and funds from the Nike Foundation. The pro- gram offers six months of vocational or business development training, combined with six months of support for job placement or microcredit for those interested in starting their own business, mixed with life skills training, mentorship, and financial training, as well as other support such as child care and transportation for participants. The stated goal of the program was to train and create employment for 2,500 girls and young women, and initial reports for the first two implementa- tion years showed that the project was at least reaching its training goals (World Bank 2010). The World Bank is funding a similar project in Rwanda that aims to reach 2,700 adolescent girls with similar services (World Bank 2011). Third, workforce development programs reflect the difficulty of finding financing for youth who want to start their own business. Again, because of the informal nature of labor markets, many youth go down the path of self-­ employment. Thus, programs that combine business training with access to finance to help out-of-school youth start their own businesses are widespread across Sub-Saharan Africa. Access to microfinance, combined with business train- ing and services, can make a significant impact on the lives of youth. However, youth’s access to microfinance is usually limited because of legal restrictions as well as the general perception among financing entities that youth are a risky group. A USAID study counted eight different microfinance entities operating in Uganda with some services for youth (Uganda has one of the most developed microfinance sector in Sub-Saharan Africa), but also noted that less than 10 ­percent of youth aged 15–24 years who start a business use these microfinance entities. The biggest sources of funding for young men are wages ( ­ approximately 20 percent) and small savings and credits associations ( ­approximately 14 per- cent), followed by parents and theft. For girls, parents (about 14 percent) and savings and credit associations (about 17 percent) provide the easiest access to capital, followed by funding from own savings, boyfriends, and even prostitution (USAID 2006). Microfinance programs generally aim to overcome legal and financial barriers for youth. For example, a quasi-experimental evaluation of Kenya’s Tap and Reposition Youth, a microfinance program that combined train- ing (financial, life skills, and health issues were covered) and mentorship with business grants to groups of young women, so long as the group met a certain savings threshold, showed that program participants earned higher wages (20 percent compared with the control group), increased their savings (doubled their savings compared with baseline, and their savings amounts were 50 percent higher than the control group), and were more likely to use a bank to keep their savings as opposed to keeping it at home (Erulkar and Chong 2005). The study also found some evidence that girls who participated in the program were more likely to refuse sex or to use a condom. The evaluation also showed that dropout rates were high, especially among younger girls, and especially when access to credit was delayed because of implementation problems. Other exam- ples across the region include: Street Kids International, an NGO that provides entrepreneurial support to marginalized children, with projects in Ethiopia Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 95 (about 2,500 youth were reached in its first five years) and Sierra Leone (plans to reach approximately 3,500 youth in three years); and Ghana’s Venture Capital Trust Fund, established in 2004, which provides low-cost credit to enterprises and business start-ups through tax-exempt intermediary institutions.9 Notes 1. By all means, this is not the only way of framing the out-of-school youth problem at the secondary level. For example, Lewin (2007) offers the concept of “zones of e ­ xclusion” that are defined by the degree of access students have to schooling. This approach is useful in classifying countries in different groups based on the degree of exclusion or presence of different types of exclusion zones in each country. This kind of approach is also useful for judging the resource needs across different types of countries. 2. Benin, Burkina Faso, Cape Verde, The Gambia, Ghana, Guinea, Liberia, Mali, Mauritania, Nigeria, Senegal, Sierra Leone, and Togo. 3. For details, see http://agsp.worlded.org/. The program is part of the larger Africa Learning Initiative, which aims at providing scholarships for children and youth and preservice and in-service training to teachers, distributing textbooks and other learning materials, and increasing the role of parents in their children’s education. 4. Depending on the policy goal, the results could vary significantly. For example, one study showed that in Malawi, targeting households with children yields an increase in enrollment of 5 percentage points among children aged 6–17 years, while targeting households with orphans yields an increase of 4.2 points (Handa and Stewart 2008). 5. For a review of conditional cash transfers, see Fiszbein and Schady (2009). One of the first rigorously reviewed examples of such a program was of the Oportunidades ­(formerly called Progresa) program in Mexico, which replaced a broad food subsidy with targeted conditional cash transfers. The evaluations showed that the biggest schooling improvements were achieved among secondary school students, increasing enrollment by 10 percent and 20 percent for boys and girls, respectively (Schultz 2000) and reducing incidences of work (Skoufias and Parker 2001). 6. For details, see http://www.unesco.org/uil/litbase/?menu=9&targetgroup=6​ &programme=102. 7. For example, see Bennell et al. (1999) for a review of apprenticeship systems in Tanzania and Zimbabwe, and see Walher (2008) for a review of systems in Benin, Mali, Senegal, and Togo. 8. These include Benin, Burkina Faso, Botswana, the Democratic Republic of Congo, Côte d’Ivoire, The Gambia, Kenya, Malawi, Mali, Mauritius, Namibia, Nigeria, Senegal, South Africa, Tanzania, Togo, and Zimbabwe. 9. For details, see http://venturecapitalghana.com.gh. References Acholi Education Initiative. 2013. “Special Scholarship for War Affected Youths” (accessed January 12, 2013), http://www.aei.org.ug/index.php?option=com_content&view=​ article&id=69&Itemid=116. Adato, Michelle, and John Hoddinott. 2008. “Lessons from Cash Transfers in Africa and Elsewhere: Impacts on Vulnerability, Human Capital Development and Food Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 96 Paths Back to Formal or Informal Education or to the Labor Force Security.” Presentation for “Investing in Social Protection in Africa: Regional Inter Governmental Experts Meeting,” Cairo, May 13–14. Aggarwal, Ashwani, Christine Hofmann, and Alexander Phiri. 2010. A Study on Informal Apprenticeship in Malawi. Geneva: International Labour Organization. Akresh, Richard, Damien de Walque, and Harounan Kazianga. 2013. “Cash Transfers and Child Schooling: Evidence from a Randomized Evaluation of the Role of Conditionality.” Impact Evaluation Series 82, World Bank, Washington, DC. Andrews, Colin, and Adea Kryeziu. 2012. “Public Works and the Jobs Agenda: Pathways for Social Cohesion?” Background paper for the World Development Report 2013, World Bank, Washington, DC. Aryeetey, Ernest, and Evelyn Kwak. 2006. “National Functional Literacy Programme, Ghana.” U.K. Department for International Development Inter-Regional Inequality Facility Policy Brief 9, Department for International Development, London. Asankha, Pallegedara, and Yamano Takashi. 2011. “Impacts of Universal Secondary Education Policy on Secondary School Enrollments in Uganda.” Unpublished manu- script, National Graduate Institute for Policy Studies (GRIPS), Tokyo. Ball, Katrina, and Stephen Lamb. 2001. Participation and Achievement in VET of Non- Completers of School. Research Report 20, Longitudinal Surveys of Australian Youth (LSAY), Australian Council for Educational Research, Camberwell, Victoria. Bantwata Initiative. 2012. Zimbabwe (accessed December 2, 2012). http://bantwana.org​ /­where-we-work/zimbabwe. Bell, Kate. 2008. “Case Study: St. Monica’s Girls’ Tailoring Center.” Paper by the Center for Global Engagement, Northwestern University, Evanston, IL. Bennell, Paul, Shane Bendera, Godfrey Kanyenze, Emrode Kimambo, Sixtus Kiwia, Tichafa Mbiriyakura, Faustin Mukyanuzi, N. Munetsi, Jo Muzulu, Willy Parsalaw, and John Temu. 1999. “Vocational Education and Training in Tanzania and Zimbabwe in the Context of Economic Reform.” Education Research Paper 28, Department for International Development, London. Berliner, Beth Ann, Vanessa X. Barrat, Anthony B. Fong, and Paul B. Shirk. 2008. Reenrollment of High School Dropouts in a Large, Urban School. Issues & Answers Report, REL 2008 056, U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory West, Washington, DC. Betcherman, Gordon, Karina Olivas, and Amit Dar. 2004. “Impacts of Active Labor Market Programs: New Evidence from Evaluations with Particular Attention to Developing and Transition Countries.” Social Protection Discussion Paper Series 402, World Bank, Washington, DC. Biavaschi, Costanza, Werner Eichhorst, Corrado Giulietti, Michael J. Kendzia, Alexander Muravyev, Janneke Pieters, Nuría Rodríguez-Planas, Ricarda Schmidl, and Klaus F. Zimmermann. 2012. “Youth Unemployment and Vocational Training.” Background paper for the World Development Report 2013, World Bank, Washington, DC. Blattman, Christopher, Nathan Fiala, and Sebastian Martinez. 2009. “Northern Uganda Social Action Fund Youth Opportunities Program.” Impact Evaluation Baseline Report, World Bank, Washington, DC. ———. 2011. “Can Employment Programs Reduce Poverty and Social Instability? Experimental Evidence from a Ugandan Aid Program.” World Bank, Washington, DC. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 97 Brewer, Laura. 2004. “Youth at Risk: The Role of Skills Development in Facilitating the Transition to Work.” ILO Skills Working Paper 19, International Labour Organization, Geneva. Bruneforth, Michael, and Peter Wallet. 2010. “Out-of-School Adolescents.” UNESCO Institute for Statistics, Montreal, Canada. Cho, Yoonyoung. 2010. “Impact Evaluation—TVST OVAY.” Presentation for the SIEF Working Camp, April 15. Coffin, Chelsea. 2013. “Secondary Education in Africa: Diagnostic Note.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA). de Brauw, Alan, and John Hoddinott. 2008. “Must Conditional Cash Transfer Programs Be Conditioned to Be Effective?” IFPRI Discussion Paper 00757, International Food Policy Research Institute, Washington, DC. Dennis, Carolyne, and Alicia Fentiman. 2007. “Alternative Basic Education in African Countries Emerging from Conflict: Issues of Policy, Co-ordination and Access.” U.K. Department for International Development Researching the Issues 67, Department for International Development, London. Erulkar, Annabel S., and Erica Chong. 2005. “Evaluation of a Savings and Micro-Credit Program for Vulnerable Young Women in Nairobi.” Population Council, Nairobi. Fiszbein, Ariel, and Norbert Schady. 2009. Conditional Cash Transfers: Reducing Present and Future Poverty. World Bank Policy Report. Washington, DC: World Bank. Garcia, Marito, and Jean Fares. 2008. “Working in Bad Jobs or Not Working at All.” In Youth in Africa’s Labor Market, edited by Marito Garcia and Jean Fares. Washington, DC: World Bank. Garcia, Marito, and Charity M. T. Moore. 2012. The Cash Dividend: The Rise of Cash Transfer Programs in Sub-Saharan Africa. Directions in Development. Washington, DC: World Bank. German Technical Cooperation. 2009. “Urban Conflict Management Peace and Development Project.” Program Brochure, Pretoria, South Africa. ———. 2010. “Non-Formal Education and Training.” Project Report. Guarcello, Lorenzo, Marco Manacorda, Furio Rosati, Jean Fares, Scott Lyon, and Cristina Valdivia. 2008. “School-to-Work Transitions: Regional Overview.” In Youth in Africa’s Labor Market, edited by Marito Garcia and Jean Fares. Washington, DC: World Bank. Gyampo, Ransford. 2012. “Youth Participation in Youth Programmes: The Case of Ghana’s National Youth Employment Programme.” Journal of Pan African Studies 5 (5): 13–28. Haftendorn, Klaus, and Carmela Salzano. 2004. “Facilitating Youth Entrepreneurship, Part II—A Directory of Awareness and Promotion Programmes in Formal and Non- Formal Education.” ILO SEED Working Paper 59, International Labour Organization, Geneva. Hamory Hicks, Joan, Michael Kremer, Isaac Mbiti, and Edward Miguel. 2011. “Vocational Education Voucher Delivery and Labor Market Returns: A Randomized Evaluation among Kenyan Youth.” Report for Spanish Impact Evaluation Fund (SIEF) Phase II, World Bank, Washington, DC. Handa, Sudhanshu, and Scott Stewart. 2008. “The Orphan Targeting Dilemma in Eastern and Southern Africa.” In Poverty in Focus: Cash Transfers—Lessons from Africa and Latin America, edited by Degol Hailu and Fábio Veras Soares. Brasilia: International Poverty Centre. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 98 Paths Back to Formal or Informal Education or to the Labor Force Hansen, Art Nenon, Joy Wolf, and Marc Sommers. 2002. Final Evaluation of the Office of Transition Initiatives’ Program in Sierra Leone. Final Report, USAID, Washington, DC. Hedger, Edward, Tim Williamson, Tom Muzoora, and Justina Stroh. 2010. Sector Budget Support in Practice—Case Study: Education Sector in Uganda. Final Draft Report, Overseas Development Institute, London. Helgesson, Linda. 2001. (E)Quality: Girls’ and Boys’ Basic Education in Masasi and Kisarawe Districts. Ministry of Education and Culture and UNICEF Tanzania, Dar es Salaam. Ikiara, Gerrishon K. 2009. Political Economy of Cash Transfers in Kenya. Report prepared for the Overseas Development Institute, London. Ilon, Lynn, and Robinah Kyeyune. 2002. Cost Evaluation for Complementary Basic Education Programs in Uganda. Kampala: USAID. IYF (International Youth Foundation). 2009. “IYF Youth Employability Programs.” Presentation at the “Youth-Employability-Opportunity” conference, Nairobi, Kenya, April 16–17. Johanson, Richard. 2009. “A Review of National Training Funds.” SP Discussion Paper 0922, Social Protection and Labor Unit, World Bank, Washington, DC. Jones, Nicola, William Ahadzie, and Daniel Doh. 2009. “Social Protection and Children: Opportunities and Challenges in Ghana.” Overseas Development Institute and UNICEF, London. Joseph, Jean-Baptiste, and Rakotozafy Harison. 2007. “Higher Education Research to Generate Alternative Educational Models in View of the Inadequacies of the Formal Educational System: Cases Observed in Madagascar.” Paper presented at the “Contribution of Higher Education to National Education Systems: Current Challenges for Africa” conference, Accra, Ghana, March 22. Kajura, Susan. 2012. “Bridging the Gap: Non-Formal Education as a Means to Re-engage Youth.” Paper presented at the annual meeting of the 56th Annual Conference of the Comparative and International Education Society, San Juan, Puerto Rico, April 22. Kavuma, Richard M. 2011. “Free Universal Secondary Education in Uganda Has Yielded Mixed Results.” Poverty Matters Blog, The Guardian, October 25. KEPSA (Kenya Private Sector Alliance). 2011. “KYEP Project Achievements.” Nairobi. Lal, Radhika, Steve Miller, Maikel Lieuw-Kie-Song, and Daniel Kostzer. 2010. “Public Works and Employment Programs: Towards a Long-Term Development Approach.” Working Paper, International Policy Centre for Inclusive Growth (IPC-IG), Brasilia. Lamichhane, Shreeram Prasad, and Mana Prasad Wagle. 2008. “Consolidating National Policies on Equivalency towards Accomplishing Universal Primary Education (UPE).” UNESCO Kathmandu Series of Monographs and Working Papers 16, UNESCO, Kathmandu, Nepal. Lerman, Robert I. 2005. “Helping Out-of-School Youth Attain Labor Market Success: What We Know and How to Learn More.” Unpublished report, American University and the Urban Institute, Washington, DC. Lewin, Keith. 2007. “Improving Access, Equity and Transitions in Education: Creating a Research Agenda.” Research Monograph 1, Consortium for Research on Educational Access, Brighton, UK. Lieuw-Kie-Song, Maikel, and Kate Philip. 2010. Mitigating a Jobs Crisis: Innovations in Public Employment Programmes. ILO Employment Report 6, International Labour Organization, Geneva. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Paths Back to Formal or Informal Education or to the Labor Force 99 Macpherson, Ian. 2007. “Tanzania Non-Formal Education.” Country Profile commissioned for the EFA Global Monitoring Report 2008, Education for All by 2015: Will We Make It? UNESCO, Geneva. Manda, Stella. 2011. Evaluation of the Accelerated Learning Programme (ALP) in Liberia. UNICEF. Martin, Angela, and Patrick Wingate. 2001. “Impact Evaluation: Youth Reintegration Training and Education for Peace (YRTEP) Program.” Prepared for the U.S. Agency for International Development Office of Transition Initiatives, U.S. Agency for International Development, Washington, DC. Midling, Michael, Louise Filion, Emmanuel M. David-Gnahoui, Mbarou Gassama-Mbaye, Amadou Tidjane Diallo, and Abdoul Karim Diallo. 2006. Program Evaluation for USAID/Guinea Basic Education Program Portfolio. Final Report under Contract EDH- I-00-03-00002-00, U.S. Agency for International Development, Washington, DC. Miller, Candace. 2009. “Economic Impact Report of the Mchinji Social Cash Transfer Pilot.” Draft Report prepared for the Government of Malawi and USAID, Center for Global Health and Development (CGHD), Boston University. Monk, Courtney, Justin Sandefur, and Francis Teal. 2008. “Does Doing an Apprenticeship Pay Off? Evidence from Ghana.” CSAE Working Paper Series 2008–08, Centre for the Study of African Economies, University of Oxford, Oxford, UK. Naidoo, Tony, and Sherine van Wyk. 2008. “Community Projects: Stellenbosch USIKO Youth Project.” Stellenbosch University, South Africa. Nicholson, Sue. 2006. “Accelerated Learning in Post-Conflict Settings.” Discussion Paper, Inter-Agency Network for Education in Emergencies (INEE), New York. Nübler, Irmgard, Christine Hofmann, and Clemens Greiner. 2010. “Understanding Informal Apprenticeship: Findings from Empirical Research in Tanzania.” ILO Employment Sector Employment Working Paper 32, International Labour Organization, Geneva. Nyamugasira, Warren, Aanyu Dorothy-Angura, and Clinton Robinson. 2005. Report of the Final Evaluation of “Literacy and Continuing Education in Uganda 2000–2005” and “Family Basic Education Uganda” 2003–05 Programmes. Prepared for Education Action International and Literacy and Adult Education (LABE), Uganda. Oreopoulos, Philip, Marianne E. Page, and Ann Huff Stevens. 2006. “The Intergenerational Effects of Compulsory Schooling.” Journal of Labor Economics 24 (4): 729–60. Palmer, Robert. 2007. “Skills Development, the Enabling Environment and Informal Micro-Enterprise in Ghana.” Dissertation, University of Edinburgh. Pereznieto, Paola. 2009. “Social Protection to Tackle Child Poverty in Mali.” Project Briefing 25, Overseas Development Institute, London. Rother, Friederike. 2006. “Interventions to Support Young Workers in Sub Saharan Africa: Regional Report for the Youth Employment Inventory.” World Bank, Washington, DC. Rotheram-Borus, Mary Jane, Marguerita Lightfoot, Rogers Kasirye, and Katherine Desmond. 2011. “Vocational Training with HIV Prevention for Ugandan Youth.” AIDS and Behavior 16 (5): 1133–37. Sandhaas, Bernd. 2005. “Poverty Reduction and Capacity Building through Livelihood Skill Training at CSTCs and VTCs—The EXPRO in Ethiopia.” DVV International, Addis Ababa, Ethiopia. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 100 Paths Back to Formal or Informal Education or to the Labor Force Schubert, Bernd. 2005. “The Pilot Social Cash Transfer Scheme, Kaloma District— Zambia.” Chronic Poverty Research Centre (CPRC) Working Paper 52. Chronic Poverty Advisory Network, London. ———. 2007. Piloting the Scale Up of the Malawi Social Cash Transfer Scheme. Report on a Consultancy Financed by UNICEF Lilongwe, Malawi. Schultz, T. Paul. 2000. The Impact of Progresa on School Enrollments. Washington, DC: International Food Policy Research Institute. Schüring, Esther, Matthias Rompel, and Michelo Stanfield. 2009. “Experiences with Social Cash Transfers as a Tool to Mitigate the Impact of AIDS in Rural Families in Zambia.” GTZ, on behalf of the Federal Ministry for Economic Cooperation and Development, Kiel, Germany. Skoufias, Emmanuel, and Susan W. Parker. 2001. “Conditional Cash Transfers and Their Impact on Child Work and Schooling: Evidence from the Progresa Program in Mexico.” Discussion Paper 123, Food Consumption and Nutrition Division, International Food Policy Research Institute (IFPRI), Washington, DC. Thompson, Ekundayo J. D. 2001. “Successful Experiences in Non-Formal Education and Alternative Approaches to Basic Education in Africa.” Discussion paper presented at the 2001 Biennial Conference of the Association for the Development of Education in Africa (ADEA), Arusha, Tanzania. UNESCO (United Nations Educational, Scientific, and Cultural Organization). 2006. Non-Formal Education and Livelihood Skills for Marginalized Street and Slum Youth in Uganda. Uganda Youth Development Link, Project Report, Kampala. ———. 2009. Support for the Quality and Equity of Education. Paris: UNESCO. UNICEF (United Nations Children’s Fund). 2008. “Trauma to Triumph: Restoring Hope in Post-Conflict Communities.” UNICEF. ———. 2009. “Child Poverty: A Role for Cash Transfers in West and Central Africa?” Briefing paper, UNICEF. USAID (U.S. Agency for International Development). 2006. Microfinance, Youth and Conflict: Central Uganda Case Study. Washington, DC. van Wyk, Sherine, and Tony Naidoo. 2009. “The Jamestown/USIKO Girls Mentoring Programme.” Stellenbosch University Stellenbosch, South Africa. Walher, Richard. 2008. “Towards a Renewal of Apprenticeship in West Africa: Enhancing the Professional Integration of Young People.” Agence Française de Développement, Paris. Wayman, Jeffrey C. 2001. “Factors Influencing GED and Diploma Attainment of High School Dropouts.” Education Policy Analysis Archives 95 (14). World Bank. 1997. Burundi—Agricultural Services Sector Project, Volume 1, Implementation Completion Report. Report 16843, World Bank, Washington, DC. ———. 1999. Implementation Completion Report: Agricultural Services Development Project. Report No. 19556. World Bank, Washington, DC. ———. 2008. “Transitions in Secondary Education in Sub-Saharan Africa Equity and Efficiency Issues.” Working Paper 125, World Bank, Washington, DC. ———. 2010. “Empowering Girls and Young Women in Liberia: Adolescent Girls Initiative.” Project Brief, World Bank, Washington, DC. ———. 2011. Promoting Economic Empowerment of Adolescent Girls and Young Women, Project Appraisal Document. Report AB6460, World Bank, Washington, DC. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Cha p t e r 5 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? Introduction More than half of youth in Sub-Saharan Africa are out of school, and youth participation in formal employment markets is very low. Neither of these two ­ facts provides optimism for the future: with an increasing share of the youth population out of formal employment markets, countries in Sub-Saharan Africa should expect continued economic stagnation, sluggish public revenue growth, and high expenditure needs. In addition, given the informal nature of most Sub-Saharan African economies, labor markets continue to be weak and access ­ to certain types of jobs is restricted; these factors reduce the incentives to start or complete secondary education. The problem is worse in some countries com- pared with others. Ghana, Kenya, Liberia, Nigeria, Swaziland, and Uganda have relatively low proportions of out-of-school youth, and most out-of-school youth between the ages of 12 and 24 years have some experience with school (only 12 ­percent of youth among this group have never enrolled, which is half the regional average). On the other hand, more than three-quarters of youth aged 12–24 years are out of school in Burkina Faso, Chad, Côte d’Ivoire, Madagascar, Mali, Niger, Rwanda, and Zimbabwe (in Burkina Faso and Niger, the share is almost 90 percent). In Burkina Faso, Chad, Mali, and Niger, more than half of youth have never enrolled in any kind of schooling. Schooling decisions and school/work choices correlate strongly with house- hold characteristics, gender, geographic location, and lack of school access, but some factors are more important than others. This report presents six key find- ings about out-of-school youth: 1. Most youth drop out even before the beginning of the secondary cycle, and in countries where the incidence of out-of-school youth is high, a larger share of youth have never set foot in a school. The out-of-school problem is Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   101   102 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? particularly widespread in low-income countries, Francophone countries (which tend to be low-income), and fragile or conflict-affected countries. 2. Early marriage has detrimental effects on the educational outcomes of young women, which are already poor compared with those of young men. Family ­ attitudes and expectations influence girls’ schooling many years before they get married. Comparisons of the outcomes for older youth show that married girls all along face very poor odds: 1:5 for finishing primary school and 3:5 for enrolling at the secondary level compared with boys of similar age. 3. Rural youth are more likely to have never attended or to drop out of school compared with urban youth. Across the region, seven out of ten rural youth have never attended school. The differences are also wide for work/school choices: urban youth tend to be exclusively in school compared with mixing work and schooling, and are less likely to exclusively work or help in the house compared with rural youth. 4. Parental attitudes toward education constitute the most important determi- nant of schooling choices: the analysis supports a story of “transmission” in which schooling and life outcomes are correlated across generations. When parents complete secondary education or achieve higher degrees themselves, they are much more likely to keep their children in secondary school and push them to finish this cycle. 5. After controlling for parental attitudes toward education, it is households’ earning capacity (not income level) that matters most, both on schooling choices and on school/work decisions. Household income effects are still important, and are most pronounced in countries where education outcomes are poor, such as Francophone and low-income countries. In these countries, income level starts to matter once families reach middle-income status. 6. Lack of schooling and poor-quality schools amplify the impact of demand-side constraints on schooling choices. Access is especially a problem for poor and rural communities, where barriers such as distance and entrance requirements or exit exams are often binding. Households also factor quality into their schooling decisions, especially as youth get older. Low school quality and lack of motivation at school act as push factors by hindering student achievement and progress. Overall, youth from poorer households, with fewer working adults and with parents who are not educated or are less engaged in the schooling and other aspects of their children’s lives, are less likely to complete primary school or enroll at the secondary level. And parental influence on schooling outcomes subsides with age. The effect of parental education is weaker for the transition from the lower to upper secondary cycle, replaced by societal choices such as early marriage and parenthood and other factors such as HIV/AIDS status. Countries with a large share of out-of-school youth share similar socioeco- nomic and demographic characteristics. Low-income countries have higher shares of their 12- to 24-year-old youth out of school (70 percent); so do Francophone countries. Half of the top 10 countries with the smallest share of Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? 103 out-of-school youth are lower-middle-income (in the region, this ratio is 27 ­percent), and many are Anglophone. Countries with relatively lower popula- tion growth rates, that invest a larger share of their gross domestic product (GDP) in education, that have stronger formal labor markets, and that have a larger share of literate adults are less likely to have their youth out of school. Analysis of the household data also shows that some of the pull factors (related to household characteristics) can be generalized across countries of similar income levels (which are correlated, to some extent, with colonial history). For example, household income matters less in low-income countries, where the sup- ply of schooling is already tight, but more in Anglophone countries (which tend to be lower-middle-income and to have greater access to education). No simple policy solutions exist to the dropout problem and the reenrollment challenge. Even when the overarching reasons for dropping out are well known, designing programs to retain students in school or marshal them into alternative paths is notoriously difficult. On the demand side, the characteristics that seem to matter most for schooling choice (i.e., poverty, low education levels of parents, and parents’ inability to find work) are indicative of other social problems and cannot be addressed by policies designed in isolation to support out-of-school youth. Labor is worth more for youth at the secondary level, making the opportunity cost of schooling higher. Expectations for youth are not always consistent with contin- ued schooling: youth are expected to support their families, and if necessary, drop out of school temporarily or permanently to find work. Young women are expected to get married or to take on significant chores in the house. Finally, changing attitudes toward education of both household heads and youth is hard. ­ On the supply side, providing secondary education is expensive, and alterna- tive education and workforce training programs face tough competition from formal education and other policy initiatives such as welfare or social support programs for government and donor funding. This problem will be exacerbated in the future given the rapid growth of the youth population in Sub-Saharan Africa. Countries also have a hard time developing nonformal alternative educa- tion programs, since without close ties to a strong formal education system, alternative education is not very successful. Finally, resource mobilization for secondary education is much harder. For example, coordinating donors, as hap- pened with the push for universal primary education, is harder because it requires strong central government organizations that can coordinate donor funds available for multiple purposes. The six key factors about out-of-school youth revealed by this analysis of determinants of out-of-school youth and the three entry points around which current policy interventions in the region are organized should frame the policy discussion on out-of-school youth (figure 5.1). This framework allows for recom- mendations on what policy path may be preferable for a given subgroup, across different typologies of countries that generally have similar supply-side problems. It also provides a means for discussing short-term policies that might be neces- sary to prepare current youth for work and longer-term policies that would focus on broader enrollment and lower dropout rates. These recommendations are not Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 104 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? Figure 5.1 Interaction between Key Factors and Entry Points Entry points 1. 2. 3. Retain Remediate Integrate at-risk through with the youth in formal or labor 1. Drop out before secondary school school alternative market education 2. Early marriage for girls 3. Rural youth Key factors 4. Parental education/attitudes Policy world 5. Family income/earning capacity 6. Lack of access/poor quality of schools meant to be prescriptive; they should be read as summary findings of the regional analysis loosely mapped to various interventions reviewed in this report. countries To retain students who are interested in learning but cannot afford it, ­ could experiment with removing cost constraints by providing cash incentives or subsidies. Especially in low-income and Francophone countries, expanding these benefits to households all the way up to middle-income families could improve their effectiveness. The simplest way of providing financial stability to house- holds is cash transfers, with either soft or hard conditions. Experience with such cash transfer programs shows that they make an immediate and important dif- ference in schooling outcomes. The analysis suggests that, perhaps due to lack of access to schooling around the poorest households, income supplements do not make as much of a difference in education outcomes for the poorest families. Thus unconditional transfers are more effective not for the poorest, but for middle-income families, especially in countries where availability of schooling is very limited. Expanding unconditional cash transfers to relatively better-off fami- lies may be politically difficult (since these programs compete with other social welfare and education initiatives), but general subsidies to education, in the form of reduced fees, may make a big difference.1 Paying families so that they can afford private schools, for example, might be the most expedient and cost-effective way of expanding secondary education in Sub-Saharan Africa. Vouchers for private secondary education can be ­ particularly effective in countries where supply of public schools is limited. Some ­ experience from Sub-Saharan Africa shows that the recent increase in s ­econdary school expansion is largely supported by the expansion in private secondary education. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? 105 For example, in Guinea, where secondary enrollments increased by more than 400 percent between 1997 through 2012, there are more private schools today than public schools (compared with almost no private schools in 1997). In addi- tion, evidence shows that private schools in Guinea could be very cost-effective (World Bank 2014). Comparing household expenditures on tuition for private schools to per pupil costs of public schools reveals that unit costs are comparable, even when one includes other out​ -of-pocket costs of education such as uniforms, transportation, or tutoring. Thus private schools need not be more expensive if one could afford to pay the tuition, and a voucher program for private schools could allow poorer families to take advantage of private schools without increas- ing per pupil public expenditures necessary to increase school availability. Unconditional cash transfers may not be as effective in the poorest households if parental attitudes reflect constraints not related to income but to expectations for youth (such as marriage). Similarly, for families who live in rural or conflict- affected areas, the absence of schools, public or private, would limit the useful- ness of funding. For this group of students, low-cost interventions to improve quality, infrastructure, and remedial support can help alleviate quality and avail- ability problems. Households report that quality of schooling plays an important role in the schooling decisions for older youth, but for younger youth in the earlier part of their education, increasing teacher attentiveness and providing remedial support may reduce dropout rates. Teacher mentorship motivates youth to stay in school; teachers with low academic expectations and teachers who treat low-achieving students in a negative way push students out. Community-based teacher training and programs that increase parental involve- ment in school management, such as those in Mali and Uganda, can help increase teacher awareness. The Girls’ Education Project in Eritrea reports that tutorial classes have improved grade promotion among participating girls. For youth who were excluded from education at the primary level, second chance programs can bridge the gap. Across the region, an estimated 33 percent of youth never make it to the secondary level; the shares are particularly high among Francophone countries (57 percent) and fragile countries or countries with conflict history (48 percent). Experience in the region suggests that sec- ond chance programs can achieve rapid results for out-of-school youth, encour- aging them to continue with formal education at the secondary level. Youth who never attended school may be easily persuaded to do so if they come across the right program. Madagascar, Uganda, and Zimbabwe have imple- mented successful second chance programs with compressed curricula. These second chance programs are desperately needed in low-income countries, Francophone countries, and countries with conflict histories, where exclusion rates are very high. Eliminating the secondary school enrollment bottleneck is a necessary ­ condition for reducing the exclusion problem. Even for younger children who complete primary education, secondary enrollment can remain a challenge because of lack of high-quality schools. Countries that have instituted compulsory education at the secondary level have achieved significant enrollment increases, Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 106 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? but most times at the expense of quality. As an intermediary step, expanding the primary cycle by a few years might help with quality problems while lowering dropout rates. This will allow countries, especially in their rural areas, to take advantage of the physical infrastructure and teaching force at the primary level. It would also push the natural exit point a few years into the future, when attitudes about education might become more positive. ­ In the longer run, early intervention, especially through early child ­development programs, may be the most cost-effective way of retaining youth in schools. The analysis shows that over-age enrollment is strongly correlated to dropout rates, but of course, over-age enrollment itself is a manifestation of other demand-side constraints such as poverty or low interest in education. Early child development programs are shown to reduce later incidences of dropping out, mainly by assur- ing that children enroll in school when they are supposed to and that p ­ arents are motivated to choose schooling over work, or in the case of girls, marriage. Whether improving retention or providing incentives to return to education, all programs must involve and educate parents, especially for younger youth. Parental education and parents’ attitudes toward education are the most impor- tant determinants of school and work outcomes for youth, especially for earlier transitions in and out of the primary cycle. When parents themselves have poor economic opportunities, they tend to discount the importance of education. The effect of parental education on secondary education is strongest in coun- tries where schooling outcomes are better, such as lower-middle-income and Anglophone countries, where parents themselves are more likely to have completed secondary education. Guinea, Mali and Uganda have implemented ­ smaller-scale parent education programs that also require participation in school management. These programs report higher levels of parental involvement in school management; increases in school attendance for youth and in civic partici- pation for adults; and decreases in dropout rates, reports of domestic violence, and early marriage. Parental involvement in school management can also lead to small but significant improvements in school infrastructure and facilities. For example, in Guinea, the Community Participation in Education for Equity and Quality Program has employed parents’ help in simple investments such as add- ing latrines, with positive impacts on girls’ enrollment levels. While parental attitudes matter in earlier transitions, youth’s attitudes toward education play a larger role later, especially when it comes to completing second- ary education. Many youth drop out of secondary education because they are expected to support themselves and their families. In addition, many young people in the region engage in risky sexual behavior, which often results in early parenthood. Finally, youth may not fully understand the potential benefits from education because they don’t observe these outcomes in their immediate sur- roundings, perhaps because they live in rural areas, slums, or conflict-affected regions. Successful programs in the region target at-risk youth and mix non- cognitive skills training with vocational training. For example, South Africa’s USIKO Program recruits male volunteers specifically to mentor young men, to encourage school attendance and discourage risky behavior such as gang activities. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? 107 Since 2000, more than 600 at-risk youth have successfully completed the men- torship program, and of this group more than 90 percent obtained high school qualifications. Related to attitudes toward education is the differential treatment of boys and girls when it comes to schooling. Programs focused on girls, especially married girls, must provide economic incentives to stay in school and away from risky behavior. Across the region, girls are much less likely to enroll in school compared with boys, and girls to be married face the worst prospects at the beginning of their education—their odds of enrolling and completing primary school are very poor. But once enrolled at the secondary level, girls, especially married girls, do significantly better: they are three times more likely to graduate from secondary school. This suggests that investment in getting girls to enroll and complete pri- mary education has a high payoff. For younger girls, targeted subsidies, accessible only if parents or girls themselves change their attitudes toward education, may be the best intervention tool. For older girls, however, staying in school may not be possible or viable, and appropriate alternative programs and workforce devel- opment programs must provide strong linkages to work outcomes, satisfying the general expectation that these young women must provide for their families. Alternative education programs for older youth must recognize that work is a part of youth’s lives in the region. Children in the region start working at very young ages either at home or elsewhere, sometimes for a salary and sometimes not. Youth from poorer households are more likely to work and less likely to attend school full-time, and these effects are strongest among low-income countries. For working youth, alternative education programs that recognize that ­ youth are expected to work and provide for themselves and their families have proved to be successful. Such programs include Sierra Leone’s training courses for ex-combatant and other war-affected youth, the Republic of South Sudan’s alternative education courses for out-of-school youth in conflict areas, and Kenya’s experiment with vouchers for technical and vocational training. Work constraints are particularly binding for programs that target rural youth. In every country in the region, rural youth are more likely to be out of school compared with urban youth, and rural youth are more likely to never have attended school. This is particularly true for low-income and Francophone coun- tries such as Mali, Burkina Faso, and Senegal. Geographic location has the highest impact when youth enroll at and complete primary level, but makes little differ- ence at later transitions. Rural youth are also more likely to work and less likely to attend school full-time. While availability of formal schooling may limit rural youth’s schooling outcomes, rural youth would still benefit from programs that recognize their work obligations. Some examples of these programs can be found in Ethiopia and Ghana. Workforce development programs must consider that most youth will be self- employed or work for a small, informal enterprise. Many youth drop out of school in the region because they are unlikely to improve their employability when they complete middle or high school. Because formal work opportunities are rare, formal education carries less intrinsic value and parents often do not Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 108 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? send their children to school, as evidenced by studies from Ghana, Kenya, Uganda, and Zambia, because they think education will not lead to a job. In the region, programs that recognize the difficulty in finding wage and salaried work in a formal establishment have demonstrated some success, but the biggest con- straint has been the lack of a legal framework that defines how financial institu- tions can lend to youth. Youth who want to start their own business are left to their own devices even when they have ideas worth investing in, having to rely on wages, Rotating Savings and Credit Associations, their parents, and even theft or prostitution. Providing multiple pathways for youth is important, though resource con- straints limit the scale of alternative options. Secondary education is more expen- sive to provide than primary education, and in searching for resources from governments and donors, alternative education and workforce training programs face tougher competition from the public education sector and other policy ini- tiatives such as welfare or other social support programs. But coping with out-of- school youth requires more than expanding the formal education sector. In the region, countries that do better in terms of keeping their youth in school also try to push multiple paths to alternative degrees or jobs for out-of-school youth.2 The importance of providing multiple paths makes resource allocation and coor- dination issues much harder within countries and across the donor base. Needless to say, the challenge is much bigger for low-income countries: without close ties to a strong formal education system, developing a nonformal alternative educa- tion program is much harder. While resource problems are hard to solve, the review of the existing pro- grams in Sub-Saharan African countries points to two policies that can increase the effective use of resources and are immediately available to most countries in the region. First, it is imperative that countries invest in developing the capacity of and coordination among national governments, subnational governments, and communities. In general, large-scale programs, whether alternative education systems or workforce development programs, suffer from implementation and coordination problems. The success of these programs increases when training areas and venues are not limited, when the relations between national and subnational governments are strong, when the division of labor is clear, and when ­ the program design is transparent. Uganda and Kenya are experimenting with programs similar to the Jovenes programs of Latin America, with encouraging results. Second, additional leveraging of funds can be achieved through donor coordination. NGOs and nonprofits run a myriad of workforce development programs that provide a mixture of training, on-the-job experience, life skills counseling, and mentoring. These programs are generally successful in integrating out-of-school youth with labor markets, but their effectiveness is limited by their size and funding, especially for those programs that enjoy international donor support, which leads to scalability issues. With many options and paths, coordina- tion of donor funds is particularly difficult for this level. Countries in the region may consider seeking collective goals and focal points around which donors could be coordinated. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 What Can Policy Achieve for Out-of-School Youth in Sub-Saharan Africa? 109 Notes 1. Secondary school fees account for a much larger share of household expenditures— up to 60 percent in Mauritania and 70 percent in Rwanda, and for older youth, more than half the households rank cost as the primary reason for not attending or for drop- ping out of school. Elimination of school fees in Cameroon, Kenya, Lesotho, Malawi, Tanzania, Uganda, and Zambia has typically led to significant increases in enrollment. 2. The best examples from the region in this regard are Ghana, Kenya, and Uganda. Reference World Bank. 2014. “Republic of Guinea: Public Expenditure Review in Education.” World Bank, Washington, DC. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Appendix A Summary of Programs Reviewed Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   111   112 Table A.1 Summary of Programs Reviewed Type of intervention Program Country Funding source Implementing agency Target group Salient points Cash transfer Program for Kenya Government funds Education Ministry Orphans and Currently covers approximately 100,000 Orphans and and international vulnerable children households, or about one-third of the Vulnerable development eligible households. If expanded fully, Children grants can require up to 1.7 percent of government expenditures and 0.5 percent of GDP. Cash transfer Social Cash Zambia Government funds Education Ministry “Destitute” households Initial impact evaluations showed half of Transfer and international with no substantive youth who were not in school at the Scheme development income and more baseline level for the evaluations were grants than three enrolled during final evaluation. dependents If expanded to all extremely poor households in the country, the program could cost US$46 million per year, approximately 0.4 percent of Zambian GDP, 1.4 percent of the government budget, or 4 percent of annual foreign aid flow. Cash transfer with Social Cash Malawi Government funds Education Ministry Ultrapoor households Evaluations show reductions in missed an education Transfer and international with no income days. Currently covers only one-tenth bonus Scheme development of the eligible 250,000 households. If grants implemented fully, would cost US$42 million per year, or about one-third of total government expenditures on social protection programs and emergency aid. Scholarship Ambassadors’ Girls’ 13 countries in International Local NGOs Primary and secondary By 2009, the program distributed 144,134 Scholarship Africa development school girls scholarships, 125,210 of which went Program grants (USAID) to girls. Scholarship Northern Mali Girls’ Mali Private Local NGO Secondary-level To date, more than 660 girls have been Scholarship students supported through this program in Program Northern Mali. Scholarship Batonga Girls’ Benin Private Local NGO Secondary-level Approximately 300 girls per year, includes Education students soft skills training for recipients of Program scholarships. table continues next page Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Scholarships and Acholi Education Uganda Private donations Local NGO Youth affected by Financial assistance to the most interventions for Initiative conflict vulnerable among war-affected Acholi support and life Scholarships adolescents, with special preference skills for young women who have children of their own. Conditional and Burkina Faso Pilot Burkina Faso Public funds and National Council Rural households with To qualify for the conditional cash unconditional Cash Transfer international against AIDS and STI low income, and transfer portion of the program, cash transfers Scheme development and the University orphans and children aged 7–15 years enroll in grants (World of Ouagadougou vulnerable children school and attend at least 90 percent Bank) of the time (in addition to health check requirements). Conditional transfers were US$17.60 for children aged 7–10 years and US$35.20 for children aged 11–15 years. Conditional cash Bourse Maman Mali International Local NGOs Poor rural households The program was implemented in two transfer development with school-age regions, Mopti and Kayes, with grants (UNICEF) children persistently low school enrollment. Provides US$10/month transfer on condition that children attend school at least 80 percent of the school year. Qualitative evaluations suggest significant improvements in enrollment but also lots of administrative difficulties. Approximately 400 households participate in the program. Conditional cash Zomba Cash Malawi International NGO Adolescent girls aged The program pays school fees and up to transfer Transfer development 13–22 years US$10/month cash payments Program grants (World conditional on attendance. The Bank) and program improved attendance and foundation grants delayed pregnancy and marriage. table continues next page 113 114 Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Academic support Stellenbosch South Africa Local funds NGO At-risk adolescent boys A combination of life skills and academic and mentoring USIKO Youth and girls in skills program of approximately Project Jamestown, a small 18 months. Since 2000, approximately working-class 600 youth have participated; more community of than 90 percent received their high about 10,000 school diploma. Quality Support for the Mali International Local NGOs Teachers and parents Training support for teachers, quality improvements Quality and development of young children improvements at schools, literacy and parental Equity of grants (USAID) support for parents, development of education Education parent associations, federations, and school management committees to increase local involvement in schools. Quality Community Guinea International Local NGOs Teachers and parents The program expanded from a single improvements Participation in development of young children district to eight districts more than and parental Education for grants (USAID) 600 villages, but faces funding education Equity and challenges. Quality Accelerated learning Access to Zimbabwe International Local NGO Out-of-school youth Initially designed to help youth who drop and remedial Education for development aged 12–15 years out of school but eventually attracted support Out-of-School grants and local a larger group, including those who Youth resources had never been to school. Second chance Complementary Tanzania Local funds and Ministry of Education Out-of-school youth Three-year condensed program with equivalency Basic Education international and Culture, local aged 11–18 years shorter days and focus on numeracy, program in Tanzania development NGOs literacy, and vocational skills. No grants (UNESCO) overall evaluation exists, but evaluation of the pilot program found that it did not achieve its stated goal of providing education for girls. table continues next page Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Second chance Basic Education for Uganda Government funds, Kampala City Council Out-of-school urban BEUPA program had the highest transfer equivalency Urban Poverty international (KCC) Ministry of youth aged to universal primary education (UPE) program Areas (BEUPA) development Education and 9–18 years schools of any Government of Uganda grants (German Sports (MoES) complementary basic education government) program. In 2002, out of 3,440 participants in BEUPA, 26.4 percent transferred to UPE schools; 54.8 percent of the participants were girls. The dropout rate was 10.3 percent. A preliminary evaluation noted its successes in collaborating with artisans from the community, the attempts to organize apprenticeships for learners seeking practical experience, and the general advantage of this type of program over conventional literacy classes. However, it has been argued that the full impact on occupational trends cannot be attained without a more holistic approach in literacy education. Second chance Planet of Alphas, Madagascar International Government Out-of-school youth After 10 months of ASAMA training, equivalency and Accelerated development aged 10–18 years 52.6 percent of participants passed the program Compressed grants (UNDP) primary equivalency exam. Of those Learning for attending the Planet of Alphas Malagasy program, 66.7 percent of children Adolescents passed the final test, 37.5 percent at (ASAMA) the advanced level. After intensive training in reading, 18.2 percent of those leaving the program enrolled in a formal primary school and 5.3 percent registered for the ASAMA program. table continues next page 115 116 Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Vocational and life Community Sierra Leone International Ministry of Education Out-of-school and In its pilot phase, the 10-month program skills training Education development with support from functionally served 2,500 learners per year in eight Centres for grants (UNESCO) local NGOs and illiterate youth and districts, most of whom have managed Literacy and volunteers adults, particularly to complete all the modules, but its Vocational Skills those from deprived expansion is limited by funding social backgrounds, constraints. such as war victims, returnees, orphans, refugees, and internally displaced persons Vocational training Vocational Training Uganda Private Local NGO Youth aged This randomized trial combined combined with with HIV 13–23 years apprenticeship with a life skills HIV/AIDS Prevention for program and demonstrated education Ugandan Youth improvements such as sustained employment, reduced risky behavior (such as fewer sexual partners and increased abstinence and condom use), and decreased alcohol, marijuana, and hard drug use. Apprenticeship Non-Formal Uganda International Local NGO Out-of-school and The program served 184 youth between combined with Education and development vulnerable youth 2004 and 2006, the majority of whom health and life Livelihood Skills grants (UNESCO) from slums were female (152). Use of local artisans skills trainings for Marginalized kept training costs low, but lack of and mentoring Street and Slum training experience and pedagogical Youth knowledge hindered success. Those who completed the program managed to find gainful employment, and generally reported improvements in their self-perception. table continues next page Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Basic literacy and life Literacy and Ghana Government funds, Ministry of EducationOut-of-school and The program, combined with its earlier skills training Community international with support from functionally phase, served 2.5 million adults, and Development development local NGOs and illiterate youth and the dropout rate was approximately Program grants (World volunteers adults, particularly 7 percent. The biggest constraint Bank), local donors those from deprived was funding. social backgrounds and rural areas Vocational training St. Monica Girls’ Uganda Local funds Religious NGO Young girls in The program offers a combination of for at-risk youth Tailoring School postconflict regions vocational skills and counseling of Northern Uganda services. It has educated 1,400 students since 1982. Vocational training Horn of Africa Somaliland Local funds Local NGO Youth returnees in The program, in place since 1998, was Voluntary Youth postconflict areas initially designed to provide training Committee for poor returnees, but over time Vocational shifted its focus to out-of-school Training youth. Given the high demand, the Program program now tests incoming attendees and charges tuition. Alternative Youth Sierra Leone International International NGO Ex-combatant and The program expanded to more than education Reintegration development other war-affected 40,000 participants in a year. Training and grants (USAID) youth aged Qualitative evaluations report that the Education for 15–34 years participants found the program Peace improved their lives and academic skills. Alternative Alternative Republic of International Sudanese government, Out-of-school youth in The program has reached more than education Learning South Sudan development international and conflict areas, aged 120,000 learners in 260 centers run by Program (initially grants (Save the local NGOs 12–18 years NGOs and 815 run by Ministry of implemented Children UK) Education. Girls’ participation rates in by Sudanese the program are higher than their government) participation rates in formal education. table continues next page 117 118 Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Vocational Technical and Kenya International NGOs Youth aged The results of a randomized trial show education Vocational development 18–30 years that vouchers increase enrollment in Vouchers grants (World vocational training significantly, Program Bank, USAID) especially when they are not restricted to certain types of programs. The results also show that youth who did not complete secondary education attributed a greater value to the vouchers, and were less likely to drop out of the program. Apprenticeship Apprenticeship Malawi International National AIDS Youth aged The 1,900 youth participating in the training Training development Commission of 15–24 years, project will receive skills training and Program and grants (Global Malawi including many start-up kits with the aim of improving Entrepreneurial Fund) orphans and entrepreneurship and self- Support for out-of-school and employment among the most Vulnerable HIV-affected youth vulnerable, as well as life skills training Youth to reduce high-risk sexual behavior by providing marketable skills and follow-up support for entrepreneurship. Alternative Community-Based Ethiopia Local funds NGO Adults and out-of- Trains approximately 2,000 adults education Nonformal school youth, annually in trades, the majority of Livelihood Skills especially in rural whom are female. Participants report Training for areas improvement in their qualifications to Youths and compete in the local labor market and Adults have gained a strong motivation to actively strive for any sort of self-employment or income- generating activity, both individually and in groups. table continues next page Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Vocational training Employment Sierra Leone International Ministry of Labour and Rural youth aged and livelihood Promotion development Social Security with 15–25 years skills Program grants (German NGO (GTZ) Federal Ministry for Economic Cooperation and Development) Vocational and Urban Conflict South Africa International National Department Youth aged The program, started in Western Cape in livelihood skills Management development of Safety and 18–30 years 1997, combines voluntary work (in training with Project grants Security, with NGO conflict resolution) with training. volunteer (GTZ) Volunteers also participate in life skills activities training as well as in a bridge program for business English, mathematics, computer training, and career guidance. After a one-year service as community peace workers, participants can take up a professional training or an apprenticeship. The program led to permanent employment for most participants, with the added benefit that program participants committed fewer crimes. table continues next page 119 120 Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points On-the-job training Youth Kenya International Ministry of Youth and Youth aged 15–29 The project is expected to cost Empowerment development Sports with a private years with at least approximately US$17 million. It targets Project grants (World sector–led eight years of training approximately 10,000 youth Bank) organization education (there is a over four years. The evaluation of the participation cap for project will focus on a combination of those with tertiary employment and life outcomes such education) as changes in earnings and engagement in risky behavior. To date, 1,095 youth have attended life skills training (425 female) with an attendance rate of 80 percent, and 916 internships have been created, both in the informal and formal sectors. A majority of these interns have received additional business skills training. Training National Training 17 countries Generally donor Typically autonomous Funds that target Provides training, internships, and Funds funds, government fund disadvantaged practical learning opportunities. support, levies populations usually Training is generally provided by have a youth private entities, which may also component organize work opportunities. Levies support participating youth, enterprises, and tradesmen. Training and job Alliance for Youth Rwanda, South International International NGO with Disadvantaged and Youth received job training, career placement Employment Africa, and development local NGOs out-of-school youth counseling, direct job placement, and Tanzania grants (USAID) and aged 14–29 years business training. The total cost was private funds approximately US$2 million; 900 youth (Nokia, Lions participated over five years. Clubs) table continues next page Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Training and job Appui à Côte d’Ivoire Government and Ministry of Technical Out-of-school youth The program offered a combination of placement l’Introduction de international Education and aged 14–24 years on-the-job and in-class training in car la Formation donors Vocational Training mechanics and electronics, tailoring, Professionnelle building site supervisors, commercial par Alterance administrative agents, computer equipment technicians, and industrial maintenance technicians. The program suffered setbacks in 2002, when the coaching experts were called back in response to the political crisis. A tracer study conducted in 2006 showed completion and job placement rates were high—57 percent of car mechanics and 71 percent of commercial trainees were offered a job. Training and job Economic Liberia International Ministry of Gender and Women aged The program offers a combination of placement Empowerment development Development 16–27 years training and job placement services as of Adolescent grants and donor well as microfinance, life skills, and Girls and Young funds mentoring. The goal is to train 2,500 Women girls so they can get wage or salaried jobs or start their own business. In its first two years, the program trained approximately 2,500 girls. table continues next page 121 122 Table A.1  Summary of Programs Reviewed (continued) Type of intervention Program Country Funding source Implementing agency Target group Salient points Training and Youth Uganda International Central and regional Youth aged The program offered competitive grants microfinance Opportunities development government 15–35 years to youth (of up to US$10,000). Youth Program grants agencies groups had to apply for grants with a proposal that identified a desired skill and a vocational training institute. A randomized experiment showed that receivers of grants: were almost four times more likely to participate in vocational training and twice as likely to be engaged in skilled work; improved their profits by 50 percent compared with the control group; and increased their savings by 20 percent. They were more likely with engage in civic activities and less likely to engage in aggressive activities. Training and Tap and Reposition Kenya NGOs and nonprofits Nonprofit entity (K-Rep Out-of-school women Women were required to form groups of microfinance Youth Development aged 16–22 years 25 that registered as a formal Agency) organization, and to open a savings account. The group then received one week of training on basic business skills, life skills, and health, and were required to save a certain amount per week as collateral for a loan. After eight weeks of saving, the group was approved for a loan. Program participants had higher earnings and savings; however, dropout rates were high, especially among younger women. Note: GDP = gross domestic product; GTZ = German Agency for Technical Cooperation; NGO = nongovernmental organization; UNDP = United Nations Development Programme; UNESCO = United Nations Educational, Scientific, and Cultural Organization; UNICEF = United Nations Children’s Fund; USAID = U.S. Agency for International Development. Appendix B Data Sources Table B.1 Survey Year and Survey Source for Data on Sub-Saharan African Countries Country Survey year Source Benin 2006 DHS Burkina Faso 2010 LSMS Burundi 2010 DHS Cameroon 2011 DHS Chad 2010 LSMS Congo, Dem. Rep. 2007 DHS Congo, Rep. 2009 DHS Côte d’Ivoire 2008 LSMS Ethiopia 2011 DHS Gambia, The 2009 LSMS Ghana 2008 DHS Guinea 2010 LSMS Kenya 2008 DHS Lesotho 2009 DHS Liberia 2007 DHS Madagascar 2008 DHS Malawi 2010 DHS Mali 2006 DHS Mozambique 2009 LSMS Namibia 2006 DHS Niger 2006 LSMS Nigeria 2010 LSMS Rwanda 2010 DHS São Tomé and Príncipe 2010 LSMS Senegal 2011 DHS Sierra Leone 2008 DHS Swaziland 2006 DHS table continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   123   124 Data Sources Table B.1  Survey Year and Survey Source for Data on Sub-Saharan African Countries (continued) Country Survey year Source Tanzania 2010 LSMS Uganda 2010 LSMS Zambia 2010 LSMS Zimbabwe 2010 DHS Note: DHS = Demographic and Health Survey; LSMS = Living Standards Measurement Survey. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Appendix C Country Typologies Table C.1 Income Levels of Sub-Saharan African Countries Low income Lower middle income Upper middle income High income Benin Cameroon Angola Equatorial Guinea Burkina Faso Cape Verde Botswana Burundi Congo, Rep. Gabon Central African Republic Côte d’Ivoire Mauritius Chad Djibouti Namibia Comoros Ghana Seychelles Congo, Dem. Rep. Lesotho South Africa Eritrea Nigeria Ethiopia São Tomé and Príncipe Gambia, The Senegal Guinea South Sudan Guinea-Bissau Sudan Kenya Swaziland Liberia Zambia Madagascar Malawi Mali Mauritania Mozambique Niger Rwanda Sierra Leone Somalia Tanzania Togo Uganda Zimbabwe Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   125   126 Country Typologies Table C.2 Colonial Heritage of Sub-Saharan African Countries Anglophone Francophone Lusophone Botswana Benin Angola Eritrea Burkina Faso Cape Verde Ethiopia Burundi Guinea-Bissau Gambia, The Cameroon Mozambique Ghana Central African Republic São Tomé and Príncipe Kenya Chad Lesotho Comoros Liberia Congo, Dem. Rep. Malawi Congo, Rep. Mauritius Côte d’Ivoire Namibia Djibouti Nigeria Gabon Seychelles Guinea Sierra Leone Madagascar Somalia Mali South Africa Mauritania South Sudan Niger Sudan Rwanda Swaziland Senegal Tanzania Togo Uganda Zambia Zimbabwe Table C.3 Conflict History of Sub-Saharan African Countries Angola Liberia Burundi Mozambique Central African Republic Rwanda Chad Sierra Leone Comoros Somalia Congo, Dem. Rep. South Sudan Congo, Rep. Sudan Côte d’Ivoire Togo Eritrea Uganda Guinea Zimbabwe Guinea-Bissau Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Country Typologies 127 Table C.4 Income Level, Colonial Heritage, and Conflict History of Sub-Saharan African Countries Income level Colonial heritage Lower Upper Fragile/ Low middle middle High conflict Country income income income income Anglophone Francophone Lusophone affected Angola X X X Benin X X Botswana X X Burkina Faso X X Burundi X X X Cameroon X X Cape Verde X X Central African Republic X X X Chad X X X Comoros X X X Congo, Dem. Rep. X X X Congo, Rep. X X X Côte d’Ivoire X X X Djibouti X X Equatorial Guinea X Eritrea X X X Ethiopia X X Gabon X X Gambia, The X X Ghana X X Guinea X X X Guinea-Bissau X X X Kenya X X Lesotho X X Liberia X X X Madagascar X X Malawi X X Mali X X Mauritania X X Mauritius X X Mozambique X X X Namibia X X Niger X X Nigeria X X Rwanda X X X São Tomé and Príncipe X X Senegal X X Seychelles X X table continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 128 Country Typologies Table C.4  Income Level, Colonial Heritage, and Conflict History of Sub-Saharan African Countries (continued) Income level Colonial heritage Lower Upper Fragile/ Low middle middle High conflict Country income income income income Anglophone Francophone Lusophone affected Sierra Leone X X X Somalia X X X South Africa X X South Sudan X X X Sudan X X X Swaziland X X Tanzania X X Togo X X X Uganda X X X Zambia X X Zimbabwe X X X Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Appendix D Distributions of Out-of-School Youth for Select Sub-Saharan African Countries Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   129   130 Table D.1  Distributions of Out-of-School Youth for Select Sub-Saharan African Countries millions Ages 12–14 years School transitions (ages 15–24 years) School-work activity (ages 15–24 years) Never enrolled/ Still in primary/ Inactive (no dropped primary Dropped Still in Secondary Both work work or Country Out-of-school primary graduate secondary secondary graduate School only Work only and school school) Unemployed Burkina Faso 0.633 1.75 0.26 0.095 0.236 0.024 0.095 1.726 0.26 0.142 0.142 Cameroon 0.22 1.157 0.718 0.479 1.596 0.04 0.559 1.756 1.277 0.279 0.12 Chad 0.298 1.008 0.371 0.071 0.318 — 0.371 0.424 0.248 0.566 0.159 Côte d’Ivoire 0.476 2.404 0.515 0.215 0.987 0.172 1.159 1.674 0.086 1.331 0.043 Ethiopia 2.25 7.671 4.197 0.579 1.737 0.289 3.184 6.368 2.461 2.171 0.289 Gambia, The 0.046 0.175 0.046 0.05 0.076 0.034 0.107 0.156 0.004 0.095 0.019 Ghana 0.251 1.009 0.57 0.921 1.36 0.527 1.58 1.623 0.439 0.57 0.176 Guinea 0.349 1.071 0.268 0.156 0.692 0.045 0.781 0.803 0.067 0.424 0.156 Kenya 0.169 1.191 1.27 1.747 2.7 1.032 3.573 2.064 0.556 1.191 0.556 Malawi 0.099 0.618 0.451 0.404 0.76 0.143 0.594 0.903 0.546 0.309 0.024 Mozambique 0.284 1.597 0.662 0.506 1.013 0.117 0.312 2.064 1.052 0.312 0.156 Niger 0.485 1.697 0.358 0.119 0.191 0.024 0.191 1.314 0.239 0.263 0.382 Nigeria 1.79 4.45 3.62 0.83 12.52 6.4 12.8 4.45 4.17 5.29 1.11 Rwanda 0.087 1.094 0.759 0.022 0.313 0.045 0.29 1.273 0.558 0.067 0.045 São Tomé and Príncipe 0.455 1.915 0.186 0.106 0.426 0.027 0.532 1.33 0.027 0.612 0.16 Senegal 0.077 0.342 0.145 0.051 0.265 0.051 0.137 0.368 0.291 0.034 0.026 Sierra Leone 0.001 0.012 0.007 0.005 0.009 — 0.012 0.006 0.001 0.01 0.004 Tanzania 0.588 1.779 0.692 3.557 2.767 1.087 2.273 4.941 1.087 1.285 0.296 Uganda 0.19 0.974 2.209 0.779 2.274 0.26 0.585 2.014 3.443 0.325 0.13 Zambia 0.2 0.719 0.343 0.651 1.37 0.343 1.576 0.822 0.034 0.719 0.274 Total 8.9 32.6 17.6 11.3 31.6 10.7 30.7 36.1 16.8 16.0 4.3 SSA estimate 15.0 54.7 29.6 19.0 53.0 17.9 51.5 60.5 28.3 26.8 7.2 Source: Feda and Sakellariou 2013. Note: — = not available; SSA = Sub-Saharan Africa.. Distributions of Out-of-School Youth for Select Sub-Saharan African Countries 131 Reference Feda, Kebede, and Chris Sakellariou. 2013. “Out of School, School-Work Outcomes and Education Transitions of Youth in Sub-Saharan Africa—A Diagnostic.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Appendix E Sequential Logistic Regression Model Sequential Logit Model Feda and Sakellariou (2013) use a sequential logit model 1 to explore the ­education transition and dropout decisions of youth. The model consists of four transitions: first, the decision whether to continue/finish primary school (versus never enroll- ing or dropping out of primary school); second, given that the youth continued/ finished primary school, the decision of whether to get into secondary education or not; third, given that the youth enrolled in secondary education, whether the youth eventually dropped out or not; and fourth, given that the youth continued with secondary education, whether the youth completed secondary education by the age of 24 (very few may also have completed a higher level), versus still participating in secondary education. The purpose of the sequential logit model is to model the influence of explanatory variables on the probability of passing a set of transitions. This model is best suited to studying the relationship between families’ socioeco- nomic status and children’s educational attainment. In particular, the model explores the association between family socioeconomic status and the probabil- ity of passing from one educational level to the next, as opposed to focusing on the relationship between socioeconomic status and the highest achieved level of education. Knowing the influence of socioeconomic status at each of these tran- sitions can give a more complete picture of how inequalities in opportunities to attain education come about, describing inequalities in the process not in the outcomes. A schematic of the model is shown in figure E.1. In this model, one has to have passed all lower transitions (“at risk” of passing a transition) to make a deci- sion to continue or leave. Given the assumption that each decision is indepen- dent of the others, one can estimate the Inequality of Educational Opportunity “proper” model (see Buis 2010) by running a series of logit regressions for each transition on the appropriate subsample. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   133   134 Sequential Logistic Regression Model Figure E.1 Educational Transitions and Associated Probabilities 4 3 p4 2 Still in Completed p3 secondary or secondary 1 p2 in higher Secondary dropout education p1 Primary Never attended 1–p3 1–p4 Exit or dropped out 1–p2 of primary Exit Exit Did not complete 1–p1 secondary by age 24 years After assigning a value to each level of education (pseudo-years) (lk), one can also study the effect of the explanatory variables on the expected final outcome. The probability that person i passes transition k, pk, is given by: exp( a1 + l1 ∗ SESi + b1 ∗ xi ) p1i = 1 + exp( a1 + l1 ∗ SESi + b1 ∗ xi ) exp( a 2 + l 2 ∗ SESi + b2 ∗ xi ) p2i = if pass1 = 1 1 + exp( a 2 + l 2 ∗ SESi + b2 ∗ xi ) exp( a 3 + l 3 ∗ SESi + b 3 ∗ xi ) p3i = if pass2 = 1 1 + exp( a 3 + l 3 ∗ SESi + b 3 ∗ xi ) exp( a 4 + l 4 ∗ SESi + b 4 ∗ xi ) p4i = if pass3 = 1 1 + exp( a 4 + l 4 ∗ SESi + b 4 ∗ xi ) where the “Inequality of Opportunity” (such as father’s education, race/ethnicity, and so on) associated with transition k is l k, the constant for transition k is ak, and the effect of the control variable xk is bk. The two coefficients (l k and bk) have similar means but they show the effects of different variables. Coefficient l k shows the effect of the variable of interest— in this case, socioeconomic status, or SES. bk shows the effect of other variables in the model (for example, age, distance to schools) if they are not part of the vari- ables of interest. Thus, the two coefficients are meant to measure the association between the outcome variable and the variables of interest and control variables, both of which influence the outcome. The concept of marginal effects is useful when one wants to examine the effect of one variable by controlling the effects of the rest and it is more associated with the variable of interest (l k) than the control variables. For example, to see the effect of household head education on a child’s school attendance by controlling all other variables (for example, age, Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Sequential Logistic Regression Model 135 wealth, area, and so on), the coefficient of household head education can be interpreted as the marginal effect of household head education on a child’s school attendance. Where pass1 = 1, person i transited from p1 to p2. The counterfactual, pass1 = 0, means that person i did not transit from p1 to p2, indicating that the person dropped out of that level of school or left the system. Similarly, pass2 = 1 refers to the transition from p2 to p3, and pass3 = 1 refers to the transition from p3 to p4, with the corresponding counterfactuals and explanations as given for pass1. Note 1. Also known under a variety of other names, such as Sequential Response Model (Maddala 1983), Continuation Ratio logit (Agresti 2002), Model for Nested Dichotomies (Fox 1997), and Mare model (Mare 1981; Shavit and Blossfeld 1993). For an extended discussion, see also Buis (2010). References Agresti, Alan. 2002. Categorical Data Analysis. 2nd ed. Hoboken, NJ: Wiley-Interscience. Buis, Maarten. 2010. “Not All Transitions Are Equal: The Relationship between Inequality of Educational Opportunities and Inequalities of Educational Outcomes.” In Inequality of Educational Outcome and Inequality of Educational Opportunity in the Netherlands during the 20th Century, edited by M. Buis, PhD dissertation, Vrije University, Amsterdam. Feda, Kebede, and Chris Sakellariou. 2013. “Out of School, School-Work Outcomes and Education Transitions of Youth in Sub-Saharan Africa—A Diagnostic.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Fox, John. 1997. Applied Regression Analysis, Linear Models and Related Methods. Thousand Oaks, CA: Sage. Maddala, G. S. 1983. Limited Dependent and Qualitative Variables in Econometrics. Cambridge, UK: Cambridge University Press. Mare, Robert D. 1981. “Change and Stability in Education Stratification.” American Sociological Review 46 (1): 72–87. Shavit, Yossi, and Hans-Peter Blossfeld. 1993. Persistent Inequality: Changing Educational Attainment in Thirteen Countries. Boulder, CO: Westview Press. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Appendix F Summary Results of Feda and Sakellariou’s (2013) Findings Table F.1 Regional Classification of 20 Countries Studied East South Central West Ethiopia Malawi Cameroon Burkina Faso Kenya Mozambique Chad Côte d’Ivoire Rwanda Zambia Guinea Gambia, The Tanzania São Tomé and Príncipe Ghana Uganda Niger Nigeria Senegal Sierra Leone Table F.2 Impact of Household Characteristics on Schooling Decisions of 12- to 14-Year-Old Youth Sub-Saharan Africa Regional impact Youth characteristics Age Not significant Not significant Female 4 percent less likely to be in school 2–3 percent in East Strongest effect at 10 and South percent in Central Africa Urban vs. local Urban youth are 8 percent more Less important in More important in Central likely to be in school East and South and West Household composition Number of children 14 years of Not important Minor but significant positive effect in the South age and younger Number of children aged Not important When there are more children older than 14 years 15–24 years of age, youth are more likely to attend school in the South (2 percent marginal impact) Head of household characteristics Female 8 percent less likely to be in school More important in the West (10 percent marginal impact) Married Not significant table continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9   137   138 Summary Results of Feda and Sakellariou’s (2013) Findings Table F.2  Impact of Household Characteristics on Schooling Decisions of 12- to 14-Year-Old Youth (continued) Sub-Saharan Africa Regional impact Head of household characteristics (cont.) Primary completed 15 percent more likely to be in Less important in the South and more in the West school compared with HH with no education Upper secondary completed 21 percent more likely to be in Much lower effect in the South (8 percent) school compared with HH with no education Higher education 18 percent more likely to be in Much lower effect in the South (11 percent) school compared with HH with no education Works in agriculture 8 percent less likely to attend school Insignificant in the South Household income Household income quintile: 3–5 percent more likely to attend Income effect much stronger for higher-earning 2–5 school than the poorest households in the South and Central regions household One working adult older than 13 percent more likely to attend Significant everywhere, of similar magnitude, 25 years of age compared school except in the Central region, where there is no with none significant impact Two working adults older than 20 percent more likely to attend Significant everywhere, of similar magnitude, 25 years of age compared school except in the Central region, where there is no with none significant impact Three or more working adults 17 percent more likely to attend Significant everywhere, of similar magnitude, older than 25 years of age school except in the Central region, where there is no compared with none significant impact Country characteristics Colonial: Francophone 9 percent less likely to attend school compared with Anglophone Colonial: Lusophone 3 percent more likely to attend school compared with Anglophone Conflict history 2 percent more likely to attend school Source: Feda and Sakellariou 2013. Table F.3  Determinants of School/Work Outcomes for 15- to 24-Year-Old Youth Inactive (or home care) Work only Work and school School only Youth characteristics Age Older youth are less likely Older youth are more Older kids are less Older kids are less likely to be inactive, likely to work, likely to be in to be in school only, especially in the much stronger school or work, especially in Central and West effects in the especially in low-income and regions South (12 percent) low-income and conflict-affected and in lower- conflict-affected countries income and countries conflict-affected countries table continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Summary Results of Feda and Sakellariou’s (2013) Findings 139 Table F.3  Determinants of School/Work Outcomes for 15- to 24-Year-Old Youth (continued) Inactive (or home care) Work only Work and school School only Youth characteristics (cont.) Female: never 7 percent more likely to Significant positive Small negative Small negative impact married be inactive in impact in the impact in general in general but lower-income South only, and but stronger in stronger in countries significant conflict-affected conflict-affected negative impact in countries countries lower-income countries Female: has married 13 percent more likely to 26 percent more 11 percent less likely 28 percent less likely to be inactive, stronger likely to work only, to juggle school attend school only, in lower-income almost 50 percent and work, bigger almost 60 percent in countries in the South impact in the South low-income countries Urban vs. rural 7 percent more likely to 13 percent less likely 3 percent less likely 9 percent more likely to be inactive, strongest to work only, to combine work be in school only, in the South strongest in the and school, no especially strong in South and in effect in West Central Africa and conflict-affected Africa lower-income countries countries Household composition Number of children Minor impact Minor impact Minor impact Minor impact aged 14 years and younger Number of children Minor impact 2–3 percent less likely Minor impact Minor impact aged 15–24 years to work only, stronger in countries with a conflict history Head of household characteristics Female Small negative effect Small negative Small positive Small positive impact, impact, stronger in impact, stronger stronger in lower-­ the South and in in lower-middle- middle-income lower-middle- income countries countries income countries Married Small negative impact, Small negative Small positive Small positive impact stronger in lower- impact, stronger in impact middle-income lower-middle- countries income countries Primary completed Negative impact, Reduces the Positive impact, 11 percent higher than strongest in the South probability by highest in those whose parents (7 percent) 11 percent, but up Lusophone have no education, to 18 percent in countries at strongest effect in the West 9 percent the West and in lower-middle-income countries table continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 140 Summary Results of Feda and Sakellariou’s (2013) Findings Table F.3  Determinants of School/Work Outcomes for 15- to 24-Year-Old Youth (continued) Inactive (or home care) Work only Work and school School only Head of household characteristics (cont). Upper secondary Small positive impact 19 percent less likely Small negative 16 percent more likely completed to work only, up to impact to just be in school 30 percent in the compared with those West and in with household lower-middle- heads with no income countries education, strongest in the West, lower-middle-income countries Higher education Positive impact, but not 23 percent less likely Not significant 17 percent more likely significant in the East to work only, to just be in school and Central regions 41 percent in compared with those Lusophone household heads countries with no education, 26 percent in lower-middle-income countries Works in agriculture 10 percent less likely to 13 percent more Small positive Negative impact, be idle likely to work only impact especially in lower- income countries Household income Household income Small negative impact At the highest Small positive Up to 5 percent more quintile: 2–5 quintile, the impact likely to be in school probability of only, up to 9 percent working is 7 in conflict-affected percent less than countries those in the poorest households, stronger effects in low-income countries One working adult 6 percent more likely to Small negative 18 percent less likely 14 percent more likely older than 25 years be inactive impact to be just working to be in school only of age compared with none Two working adults 3 percent less likely to be 15 percent less likely Small positive 15 percent more likely older than 25 years inactive, strongest to work only, impact to be in school only, of age compared effect in Central stronger in the strongest in the with none region, with 13 South at 26 South at 29 percent percent marginal percent reduction probability Three or more 7 percent less likely to be 8 percent less likely Small positive 12 percent more likely working adults inactive to be working impact to be in school only, older than 25 years only, stronger in strongest in the of age compared the East and South South at 26 percent with none Source: Feda and Sakellariou 2013. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Summary Results of Feda and Sakellariou’s (2013) Findings 141 Table F.4  Determinants of Schooling Outcomes at Transition Points for 15- to 24-Year-Old Youth Enroll in primary Finish primary Enroll in secondary Complete secondary Youth characteristics Female: never Odds similar to males, Odds similar to males Odds similar to males in Odds in favor of females, married but much lower in Francophone and 1.2:1 conflict-affected Lusophone countries, countries (0.6:1 in slightly lower favor of males) otherwise Female: has The odds of enrolling in Lower odds for Married females face Once they enroll in married primary school are married females odds of 0.14:1 secondary, married very low, 0.2:1, (0.6:1), except in females are three times compared with the East, where more likely to complete males there are no this cycle, and almost significant four times in conflict- differences affected countries Urban vs. rural Higher odds of Higher odds of Slightly better odds for Slightly better odds, enrolling in and enrolling in and urban youth especially in the South, finishing primary finishing primary but lower odds in compared with rural compared with Francophone countries (1.5:1 and 1.6:1), rural (1.5:1 and almost twice in 1.6:1), almost twice Francophone, in Francophone, Lusophone, and Lusophone, and conflict-affected conflict-affected countries countries Head of household characteristics Female More likely to enroll in More likely to enroll in Similar odds Especially important in primary (1.5:1) and finish primary lower-income compared with (1.5:1) compared countries, where odds households headed with households are 1.5:1 by males headed by males, especially in the East Married Slightly higher odds of Slightly higher odds Slightly higher odds of Slightly higher odds of passing all of passing all passing all transitions, passing all transitions, transitions, at about transitions, at at about 1.1 to 1 at about 1.1 to 1 1.1 to 1 about 1.1 to 1 Primary Three times more likely Odds of finishing Almost similar odds Almost similar odds completed to enroll in primary primary are 1.8:1, compared with but lower in youth who live in countries with household where conflict history the head has no education Secondary Five times more likely Three times more Approximately twice as Approximately twice as completed to enroll in primary likely to finish likely to enroll in and likely to enroll in and school, six times primary, except in complete secondary complete secondary more likely in the countries with cycle, lower in cycle, lower in West and in conflict history, countries with conflict countries with conflict lower-middle- where the odds are history history income countries slightly less than 2:1 table continues next page Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 142 Summary Results of Feda and Sakellariou’s (2013) Findings Table F.4  Determinants of Schooling Outcomes at Transition Points for 15- to 24-Year-Old Youth (continued) Enroll in primary Finish primary Enroll in secondary Complete secondary Head of household characteristics (cont). Higher Four times more likely Three times more Approximately twice as Approximately twice as education to enroll in primary likely to finish likely to enroll in and likely to enroll in and school primary, except in complete secondary complete secondary countries with cycle, lower in cycle, lower in countries conflict history countries with conflict with conflict history history Works in Less likely to enroll in Less likely to enroll in Similar odds of two Similar odds of two agriculture and complete and complete enrolling in and enrolling in and primary education primary education completing completing secondary (0.8:1), lowest in (0.8:1), lowest in secondary cycle cycle Lusophone countries Lusophone countries Household income Household Better odds of enrolling Better odds of At the highest quintile, Smaller impact compared income in primary increasing finishing in the odds are 1.5:1, with earlier transitions, quintile: 2:5 with income quintile primary increasing especially important the odds are about the (odds increase from with income in Lusophone same at lower quintiles, 1.2 to 1.5), especially quintile (odds countries and 1.3:1 at the highest important in increase from 1.2 quintile Lusophone to 1.8, especially countries important in Lusophone and lower-middle- income countries One or more Odds around 1.3:1.6 Odds around 1.3:1.6 Twice as likely to enroll Odds are reversed employed in secondary, almost household three times more member likely in the South over 25 Source: Feda and Sakellariou 2013. Reference Feda, Kebede, and Chris Sakellariou. 2013. “Out of School, School-Work Outcomes and Education Transitions of Youth in Sub-Saharan Africa—A Diagnostic.” Background paper prepared for the World Bank program on Secondary Education in Africa (SEIA), World Bank, Washington, DC. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 Environmental Benefits Statement The World Bank Group is committed to reducing its environmental footprint. In support of this commitment, the Publishing and Knowledge Division lever- ages electronic publishing options and print-on-demand technology, which is located in regional hubs worldwide. Together, these initiatives enable print runs to be lowered and shipping distances decreased, resulting in reduced paper consumption, chemical use, greenhouse gas emissions, and waste. ­ The Publishing and Knowledge Division follows the recommended standards for paper use set by the Green Press Initiative. Whenever possible, books are printed on 50 percent to 100 percent postconsumer recycled paper, and at least 50 percent of the fiber in our book paper is either unbleached or bleached using Totally Chlorine Free (TCF), Processed Chlorine Free (PCF), or Enhanced Elemental Chlorine Free (EECF) processes. More information about the Bank’s environmental philosophy can be found at http://crinfo.worldbank.org/wbcrinfo/node/4. Out-of-School Youth in Sub-Saharan Africa  •  http://dx.doi.org/10.1596/978-1-4648-0505-9 The economic and social prospects are daunting for the 89 million out-of-school youth who comprise nearly half of all youth in Sub-Saharan Africa. Within the next decade, when this cohort becomes the core of the labor market, an estimated 40 million more youth will drop out and will face an uncertain future with limited work and life skills. Furthermore, out-of-school youth often are considered “policy orphans,” positioned between sectors with little data, low implementation capacity, lack of interest in long-term sustainability of programs, insufficient funds, and little coordination across the different government agencies. Out-of-School Youth in Sub-Saharan Africa: A Policy Perspective provides a diagnostic analysis that focuses on the 12- to 24-year-old cohort. The authors find that individual and household characteristics, social norms, and characteristics of the school system are important factors in understanding why youth drop out and remain out of school. Six key factors characterize out-of-school youth: (i) most out-of-school youth drop out before secondary school; (ii) early marriage for female youth and (iii) rural residence increase the likelihood of being out of school; (iv) parental education level; (v) the number of working adults are important household factors; and (vi) lack of school access and low educational quality are binding supply-side constraints. Policy discussions on out-of-school youth are also framed by three entry points for intervention: retention, remediation, and integration. Out-of-School Youth in Sub-Saharan Africa also reviews existing policies and programs for out-of-school youth across the continent. Ultimately, this report aims to inform public discussion, policy formulation, and development practitioners’ actions working with youth in Sub-Saharan Africa. Out-of-School Youth in Sub-Saharan Africa will be of significance to policymakers and international development experts focusing on education, youth development, and youth employment. While the report focuses on youth in Sub-Saharan Africa, the recommendations draw on global good practice and provide a guiding framework for addressing out-of-school youth issues in countries from other regions. ISBN 978-1-4648-0505-9 SKU 210505