Access to Credit, Education, and Women’s Say in Bangladesh

A significant empirical literature on women’s say in the household focuses on the effects of microcredit, but there is little evidence on the relative roles of access to credit and education. Using household survey data from Bangladesh, this paper provides a comparative analysis of the effects of education and credit on women’s decision-making power in the household with regard to their own and children’s health, large purchases, and geographic mobility. In the absence of credible exclusion restrictions, the paper implements Oster’s (2019) bias-adjusted OLS estimator, which extends Altonji et al.’s (2005) approach in which “selection on observables” serves as a guide to “selection on unobservables.” In addition, it uses a doubly robust radius matching estimator due to Lechner et al. (2011). The evidence suggests a limited impact of credit on the decision-making power of women, and this is consistent with the recent evidence from RCT-based studies in other developing countries. In contrast, education is much more important for enhancing women’s say in a range of household decisions. There is no significant interaction effect between education and credit. Evidence from Gelbach’s (2016) decomposition suggests that outside employment is an important mediating mechanism, but household wealth and assortative marriage matching on education are not important. The impact of education on women’s decision-making remains strong even after controlling for these mediating factors, pointing to the importance of other mechanisms, such as self-confidence and better negotiation skills of educated women.