Pré-publication, Document de travail: In this paper, we examine the effect of an education policy on women's well- being through the analysis of the impact of a school construction program in Benin. We exploit a sharp increase in school constructions in the 1990s in this country, to assess the causal impact of a primary education program on primary school attendance, age at marriage and tolerance of intimate partner violence (IPV). Using a double dfference method, along with a regression kink design, we find that the program increased the probability to attend primary school in rural areas. The policy also increased age at marriage and decreased the probability to find wife beating tolerable. We show that, in this context, the benefits of girls' education have percolated down to women's well-being beyond the initial goal of the policy.