January 2024 SCHOOLGIRLS NOT BRIDES: AFRICA GENDER SECONDARY EDUCATION AS A INNOVATION LAB SHIELD AGAINST CHILD MARRIAGE The Gender Innovation Lab (GIL) conducts impact evaluations of Authors: Hélène Giacobino, Elise Huillery, Bastien Michel and Mathilde Sage1 development interventions in Sub-Saharan Africa, seeking KEY MESSAGES to generate evidence on • Child marriage is recognized as a major burden in low-income countries, how to close gender gaps in with severe consequences on women’s life trajectories. It is particularly earnings, productivity, assets, pervasive in Niger, where middle school enrollment for girls is low, and dropouts and agency. The GIL team is currently working on over 80 are high. In the context of a multi-sectoral World Bank project, researchers impact evaluations in more than evaluated the impact of a three-year intervention eliminating the financial and 30 countries with the aim of logistical barriers for girls admitted to middle school. building an evidence base with • Offering adolescent girls from vulnerable households in rural Niger lessons for the region. scholarships and tutoring upon admission to middle school improved The impact objective of GIL is educational outcomes and wellbeing. Girls that received the program were increasing take-up of effective 53 percent less likely to have dropped out of school at the time of follow up. policies by governments, They also reported a higher degree of life satisfaction. Importantly, there is no development organizations, evidence that the positive effects on beneficiaries have been at the expense of and the private sector non-beneficiaries. to address the underlying • The program was effective in postponing girls’ marriage or engagement. causes of gender inequality Girls who received a scholarship program were 49 percent less likely to be in Africa, particularly in terms of women’s economic and married in the summer following their third year of middle school and 30 social empowerment. The Lab percent less likely to be engaged. aims to do this by producing • The intervention raised girls’ educational and professional aspirations for and delivering a new body themselves as well as parents’ aspirations for their daughter, plausibly of evidence and developing a due to changes in girls’ human capital and preferences. These results compelling narrative, geared suggest that the intervention’s effects are likely to last beyond the mere duration towards policymakers, on what of the scholarship. works and what does not work in promoting gender equality. 1 The authors would like to thank Amy Geist, Gaëlle Conille, and Raffaella Pizzamiglio for preparing this brief.. https://www.worldbank.org/en/programs/africa-gender-innovation-lab CONTEXT could find a host family living closer to their school. Child marriage remains widespread in West and Central In the middle schools where scholarship girls were Africa. In Niger, 76 percent of women aged 20-24 got enrolled, tutoring was also set up (4 weekly hours of married before age 18 and 28 percent before age 15. In tutoring lessons in French and Mathematics). It was addition, the fertility rate is the highest in the world with open to all and not mandatory for scholarship girls. 6.2 children per woman.2 • Mixed group: In 95 villages, a randomly selected half Child marriage has major consequences on women’s life of all eligible girls received the intervention while the trajectories, including reducing educational attainment other half did not receive it. and future participation in the labor market. Once married, • Control group: In 95 villages, no girls received the girls often drop out of school to devote their time to their intervention. new household responsibilities. Although the primary school enrollment rate in Niger has increased significantly The program spanned the first three years of middle over the past 20 years - from 27 percent in 2000 to 65 school. In the summer following the third year (a few percent in 2017 – the middle school enrollment rate months after the last scholarship payment) researchers remains low, especially for girls (21 percent).3 Generally surveyed the study participants and measured the speaking, school dropout is reinforced by the costs of impact of the program on educational and labor market education. Indeed, to enroll their daughters in middle outcomes, marriage, aspirations, and empowerment. school, parents must often find housing near the school, and face the direct costs of education and the opportunity cost of sending a child to school instead of having them FIGURE 1: IMPACT ON EDUCATION do housework or engage in economic activities. Control Treatment In this context, the Ministry of Population, the Ministry Percentage of those surveyed of Women’s Promotion and Child Protection, and the 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% Ministry of Secondary Education in Niger designed an intervention to reduce marriage among adolescent girls Dropouts by eliminating the direct costs of secondary education. This intervention is a component of the Sahel Women’s *** Empowerment and Demographic Dividend (SWEDD) project, a regional multi-sector World Bank project aiming to address constraints to family planning and women’s Enrolled in grade 6 empowerment in the Sahel. WHAT WE DID Enrolled in In the context of the SWEDD project in Niger, researchers grade 7 evaluated the impact of Toutes les Filles à L’Ecole (“all *** girls in school”) or TFE, an intervention that provided scholarships and tutoring to adolescent girls starting in middle school for a period of three years. Nearly 2,400 Enrolled in adolescent girls across 285 rural villages with no middle grade 8 schools were randomly assigned to three groups.4 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% • Treatment group: In 95 villages, every girl eligible for a scholarship received a scholarship (US $306 per Note: The symbols */**/*** denote statistical significance year) covering housing and feeding costs so that girls at the 10%, 5%, and 1% levels, respectively. 2 Institut National de la Statistique. ENAFEME Niger 2021. 3 Enrollment rates from https://data.worldbank.org/indicator/SE.SEC.ENRR.FE (last accessed in February 2022). 4 Villages included in the sample were located in five of the country’s seven regions: Dosso, Maradi, Tahoua, Tillabéry, and Zinder. Only two regions, Agadez and Diffa, were excluded from the study, the former because its fertility rate is lower than in the rest of the country and the latter for security reasons. WHAT WE FOUND so. Beneficiaries of the program were 53 percent more The intervention improved educational outcomes likely to want higher education. In terms of professional among adolescent girls. Girls in treatment communities aspirations, girls who received the scholarship were more were 53 percent less likely to have dropped out of likely to desire qualified and administrative occupations school at the time of follow up. Girls’ overall numeracy (e.g. employee, intellectual and scientific profession, scores increased by 0.18 standard deviations, but the administrative authority, manager or technician, etc.) program had no statistically significant effect on overall compared to a non-qualified and non-administrative literacy scores. These findings show that education occupation (e.g. farming, household work, self-employed costs constitute an important barrier to girls’ secondary in trade, craftswoman, etc.). The intervention similarly education, and that scholarships can be very effective at changed parents’ aspirations for their daughters: the reducing dropouts. parents of the adolescent girls in the treatment group were more likely to prefer that their daughters pursue higher The program also led to postponed marriage. Girls in education, have qualified and administrative occupations, treatment communities were 50 percent less likely to be and get married later, suggesting that parental preference married in the summer following their third year of middle for early marriage can respond positively to financial school and 30 percent less likely to be engaged. While it support for their daughter’s education. was too early to measure an impact on pregnancy (only 3 percent of girls in the control group had ever been These results (in particular the ones on girls’ and pregnant at endline), the ideal age for having a first child parents’ aspirations) suggest that the effects of increased by one year. the intervention may extend beyond its duration. Notably, after the program, girls had a preference for The program increased life satisfaction, along with later marriage for both men and women, and longer educational and professional aspirations. Notably, education for their children for both sons and daughters. the intervention induced an increase in reported life This indicates that similar programs can have a positive satisfaction of 25 percent of a standard deviation, which impact on the education of future generations. is a substantial welfare gain, suggesting that during adolescence, remaining in middle school is beneficial Overall, the TFE program has been found to be relatively for individual wellbeing, even for those who struggle in cost-effective in comparison to other similar initiatives, school and have to repeat a grade. with a total program cost of US$1,027 (2019 USD) per girl over the three years of implementation. While it falls Girls who received the scholarship intervention wished to behind some initiatives in relation to education and child attend and finish high school and felt more able to do marriage outcomes, the program increases education FIGURE 2: IMPACT ON MARRIAGE Control Treatment 16% Percentage of those surveyed 14% 12% 10% 8% 6% *** * 4% 2% 0% Married Engaged Note: The symbols */**/*** denote statistical significance at the 10%, 5%, and 1% levels, respectively. years by 0.04 for every US$100 and prevents 0.17 child marriages per US$1,000 invested, outperforming several similar programs.5 Moreover, there is no evidence suggesting that the program caused externalities (negative or positive) on non-beneficiary adolescent girls. Indeed, in the TFE intervention, the positive impact on treated girls did not come at the cost of lower education or earlier marriage for non-treated girls living in the same households or villages. POLICY IMPLICATIONS As suggested by the study’s results, alleviating financial and logistical barriers to access to secondary education for disadvantaged girls in rural areas can be effective at increasing enrollment in middle school, reducing early marriage, and improving their wellbeing. A longer-term follow-up would be needed to better assess the impact of the intervention on fertility. While the longer term impacts of the program need to be studied, the study’s results on parents’ and daughters’ educational, professional, and family- related aspirations suggest that the effects of similar interventions are likely to last beyond the mere duration of the scholarship and may even extend to future generations. In the context of Niger, characterized by strong gender inequalities and conservative gender norms, offering scholarships to girls starting middle school can have unambiguously large and positive effects at both the individual and the community levels. For more information on this study, see the Journal Article: HERE 5 For instance, in relation to education outcomes, the program is more cost-effective than the unconditional cash transfers intervention evaluated in Baird et al. (2011), but falls below the scholarships in Ghana evaluated by Duflo et al. (2022). Regarding child marriage outcomes, it surpasses the voucher program in Colombia evaluated by Angrist et al. (2002), as well as several other programs with no impact. However, it is less cost-effective than BRAC’s ELA in Uganda and Kenya’s Free School Uniform Program, which prevent 0.42, and 1.11 marriages per US$1,000 spent, respectively. FOR MORE INFORMATION, PLEASE CONTACT Estelle Koussoube Photo credit: Borgarello/World Bank, Brice Delagneau, John Hogg/World Bank, Nestle mkoussoube@worldbank.org Raffaella Pizzamiglio The authors gratefully acknowledge funding from the Sahel Women Empowerment and Demographic Dividend (SWEDD) rpizzamiglio@worldbank.org Project in Niger and J-PAL Post-Primary Initiative. This project would not have been possible without the support of the Ministry of Secondary Education, the Ministry of Women’s Promotion and Child Protection, and the Ministry of Public Health, Population and Social Affairs of Niger, the Project Coordination Unit of the SWEDD Project (in particular, Ali Africa Gender Innovation Lab Doungou Boubacar and Ibrahim Moctar Moustapha), the World Bank Health Nutrition and Population Global Practice, and the World Bank Africa Gender Innovation Lab. The authors are also grateful to NGOs ONEN and Hikima for implementing afrgenderlab@worldbank.org the intervention, and to the National Institute of Statistics of Niger for collecting the survey data. Finally, we are also grateful to Olivier Francès for excellent research assistance. 1818 H St NW Washington, DC 20433 USA This work has been funded in part by the Umbrella Facility for Gender Equality (UFGE), which is a multi-donor trust fund administered by the World Bank to advance gender equality and women’s empowerment through experimentation and www.worldbank.org/africa/gil knowledge creation to help governments and the private sector focus policy and programs on scalable solutions with sustainable outcomes. The UFGE is supported with generous contributions from Australia, Canada, Denmark, Finland, Germany, Iceland, Ireland, Latvia, the Netherlands, Norway, Spain, Sweden, Switzerland, United Kingdom, United States, the Bill and Melinda Gates Foundation, and the Wellspring Philanthropic Fund.