It is widely believed that improving women's (and men's) education can increase age at first marriage, reduce fertility, and improve maternal and child health. Taking advantage of two unique randomized experiments, this proposed project will study the impact of education on early fertility and child health in Sub-Saharan Africa, and provide evidence on the channels through which these effects are obtained. This project is proposed as a renewal to grant R01 HD039922, Women Empowerment &Child Health in Developing Countries. In previous work, supported by the current grant, we have shown, in two settings (upper primary education in Kenya and secondary education in Ghana), that reducing the cost of education leads to significant increases in school participation. In Kenya, in 2003, free uniforms were distributed to grade 6 students in 163 schools, randomly selected out of 328, leading to a significant increase in school participation and a significant decrease in teenage pregnancy. In Ghana, in 2008, we identified 2,064 students who qualified for secondary school (grade 10 and onwards) but who did not enroll due to lack of funds. 682 of these students were randomly selected to receive a 4-year scholarship for secondary school. 81% of those who received the scholarship offer enrolled in secondary school, while the enrollment rate among control students was only 16%. In this project, we propose to exploit these unique experiments, which have generated exogenous variation in educational attainment in large samples, to examine the impact of education on early fertility and child health, and the causal pathways that explain these effects. In both settings, we are collecting detailed data, several years after the initial intervention, on students who were affected by it. In Kenya, the data collection will be completed by the end of 2010 with funding from other sources. In Ghana, we plan to follow the study cohort yearly, and to conduct detailed interviews in 2012, 2015 (and 2018, pending additional funding from other sources). In both cases, the data sets collected are very rich, including data on attitudes, cognitive and non-cognitive skills, preferences, etc., and will be made publicly available. In both settings, we will (a) compare the incidence of early fertility and the health of children (both measured and reported) between treatment and control women, to measure the impact of education (primary education in Kenya, secondary education in Ghana) on early fertility and child health;and (b) shed light on the channels through which access to education of men and women affects early fertility choices and child health.
The Millennium Development Goals, adopted by the UN, call for universal primary education for boys and girls, equality in education for boys and girls, and reduction in maternal and child mortality. In order to achieve these goals, we need reliable evidence on which policies can effectively help students, girls in particular, to stay in school. We also need evidence on the role of education in ensuring healthy reproductive behavior. The knowledge gained in this proposal will contribute to this objective by enriching our understanding of the determinants and the impact of women's (and men's) education in developing countries.