Beschreibung:
This article investigates women's returns to schooling by exploiting Burundi's free primary education policy (FPE) of 2005 as a natural experiment. Credibly exogenous variation in education is identified through a fuzzy regression discontinuity design (RDD). Our results show that while educational attainment was positively influenced by Burundi's FPE for women situated at all wealth levels, the relevant downstream effects of schooling - measured by fertility, literacy and employment - reveal heterogeneous treatment effects by wealth. Poorer women profit in terms of higher literacy, employment as well as reduced fertility through policy induced education, while there are almost no effects of additional education for non-poor women. Our findings help in evaluating the generalisability of the nexus between women's education and fertility as well as associated factors.