Essays in Gender and Development

Abstract

This dissertation consists of three essays at the intersection of gender and economics in developing countries. In chapter 1, I study the economic implications of a particular cultural practice: cousin, or consanguineous, marriage. One sixth of all marriages in Egypt are between first cousins, but there are important differences in the characteristics of individuals who select into such relationships relative to those who marry non-relatives. To measure the causal impact of the practice on socioeconomic outcomes abstracting from selection, I instrument for the probability of marrying a cousin using exogenous variation in family structure, and use weak instrument robust methods to estimate parameters and evaluate statistical significance. I find that individuals who marry a cousin because of exogenous attributes of their natal family structure are further in age from their spouse, predominantly driven by older men marrying cousins. I also find that women married to cousins receive higher levels of marital transfers that give them bargaining power within their marriages, likely as compensation for their spouse's attributes. This contrasts to patterns for those who select into cousin marriage; those individuals are younger at the time of marriage, match with partners closer to their own ages, and have no differences in the level of marital transfers exchanged. The contrast between OLS and IV results suggests that selection into cousin marriage may be motivated by anticipation of not matching on the wider marriage market, credit constraints, or the desire to consolidate property within the extended family. In chapter 2, I present baseline statistics from an experiment which examines the impact of random job offers on women's experiences of intimate partner violence in Bangladesh. This paper build on a larger study which aims to increase women's labor force participation and use of mobile money services. I collect supplementary data on women's experiences of intimate partner violence, men and women's agreement with conservative social norms, and second order beliefs regarding their community's sanction of intimate partner violence. I validate survey measures of intimate partner violence with a list randomization elicitation. I also present results from two incentivized decisionmaking activities conducted at baseline. I specify the outcomes I plan to test once endline data is available, as well as the econometric specifications I will use. Finally, I present power calculations using baseline data to determine the smallest effect sizes I can detect. Finally, in chapter 3, I study the impact of an exogenous negative shock to labor demand for female migrants within Bangladesh. I use a difference in differences strategy and compare outcomes between districts that have a history of sending migrants with those that do not, before and after the shock. I find that migrants respond to the initial shock and return to their households rather than remain unemployed in Dhaka, and that at least some of these women marry. I see no decrease in the level of investment in children's human capital, which suggests households do not revise their perceptions regarding the returns to education, and have access to other tools to smooth consumption. Finally, I see no changes in the daily agricultural wage rate for women in the years after the shock. I lack data on several important margins of adjustment which would allow us to discern the mechanisms behind the effects.