Empirical evidence suggests that when transfer payments are given to women rather than to their husbands, expenditures on children increase (Duflo 2003, Qian 2008, Bobonis 2009, Duflo 2012). The conventional interpretation of this fact is that women and men have different preferences, in the sense that women attach more weight to children’s welfare (Doepke, M. and Tertilt, M., 2011). This belief has led to a trend in development policy to channel more resources towards women and, more generally, to promote improvements in the empowerment of women as an effective spur to economic development.
Microcredit schemes are a widely used development policy tool that aims to improve the economic and living conditions of its clients. Most of the microcredit interventions have been directed exclusively at women in part because of the presumed link between female empowerment and the general well-being of the household, especially the children. However, there is limited empirical support of a positive link between women’s access to credit and either health and educational outcomes of women and children, or women’s empowerment. Recent randomized control trial studies find little or no effect of microcredit on investments in education or health, or on women’s decision-making power (Banerjee et. al 2016). In contrast, Pitt et al (2003) and Pitt, Khandker and Cartwright (2006), in examining non-random access to microcredit in Bangladesh in the 1990s, do find that credit provided to women helps to increase women’s empowerment, reduces malnutrition in children, and has large positive effects on relative changes in children’s health. They find credit provided to men has no significant effect. However, these studies use structural estimation techniques to identify effects. Given the non-experimental nature of these studies, questions have been raised about the robustness of the findings (Roodman and Morduch 2009; Duvendack and Palmer-Jones 2012; Roodman and Morduch 2014). An important question is whether the difference in findings is due to the short run nature of recent experimental field studies compared with the Bangladeshi study, the specifics of Bangladesh versus other locations, or the difference in identification strategies (Roodman and Morduch 2014).
In this project, we propose to address this question by providing a long run analysis of a randomized control trial in Bangladesh. We propose collecting follow-up data on a clustered randomized control trial of a credit program targeted towards farm microenterprises in Bangladesh, successfully implemented by BRAC in 2012. This would allow us to measure the long run impact of women’s access to microcredit on health and educational investment, and health-related outcomes of women and children. We will also be able to examine the effect on three different aspects of women’s empowerment.
The study’s broader impacts are to increase understanding, strengthen collaborations between academics and institutions, and improve teaching. Most importantly, the research team will partner with BRAC, the largest NGO in the world. This partnership will facilitate the integration of research results into policy guidelines and microenterprise development programs which have the potential to improve lives worldwide. More generally, upon completion of the study, the data will be made available to the public.