We exploit a change in compulsory schooling laws in Turkey to estimate the causal effects of education on religiosity and women's empowerment. A new law implemented in 1998 resulted in individuals born after a specific date to be more likely to complete at least 8 years of schooling while those born earlier could drop out after 5 years. This allows the implementation of a Regression Discontinuity (RD) Design and the estimation of meaningful causal estimates of schooling. Using the 2008 Turkish Demographic Health Survey, we show that the reform resulted in a one-year increase in years of schooling among women on average. Over a period of ten years, this education increase resulted in women reporting lower levels of religiosity, greater decision rights over marriage and higher household consumption (of durables). We find that these effects work through different channels, depending on women's family background. For women whose mothers had no formal education, the reform resulted in them only finishing the compulsory schooling and having higher labor force participation. For women whose mothers had some formal education, the reform had persistent effects beyond compulsory schooling, and these women were subsequently married to more educated (and possibly wealthier) husbands but remained outside the labor force. We interpret these findings as evidence that education may empower women across a wide spectrum of a Muslim society, yet, depending on pre-reform constraints to participation, its effects may not be strong enough to fully overcome participation constraints (in education or the labor force).

Download the article here or read it on our SlideShare channel.