"Essays on Gender Inequality and Human Capital Formation"

Chapter 1 analyzes the effect of intergenerational coresidence on women's labor supply decisions. Women's labor force participation is stymied by childcare and housework duties, as well as long-held social norms that restrict their autonomy and mobility in developing countries. A coresiding mother-in-law may restrict women's labor force participation as the custodian of gender-specific social norms, but may also help by taking on housework responsibilities. Using a nationally representative panel dataset from India, my coauthor and I use the exogenous variation in the mother-in-law's death to empirically investigate which effect dominates. We show that a mother-in-law's death reduces her daughter-in-law's labor force participation by 10 percent in an individual fixed-effects model. A placebo test reveals no effect of a coresiding father-in-law's death on his daughter-in-law's labor force participation, which alleviates concerns about demographic changes as the drivers of our results. Also, women with four or more children drive the effects of the mother-in-law's death. We provide suggestive evidence to show that by sharing the burden of household production tasks, coresiding mothers-in-law free up their daughter-in-law's time, which allows them to participate in the labor market. Overall, our results suggest that long-established gender roles that limit women's role as homemakers and caregivers play a critical role in shaping women's labor supply decisions in India.

Chapter 2 examines the unintended effects of making in-utero (prenatal) sex detection illegal. Ultrasound technology gives parents control over fertility and enables them to influence their children's gender composition through prenatal sex detection. To address declining female-to-male ratios, the Indian government put a legal ban on prenatal sex detection. A successful ban can increase the probability of a female birth. However, in the absence of prenatal sex-detection techniques and the presence of strong son preferences, parents can respond by investing fewer resources in 'unwanted' girls they would have otherwise aborted. Using a difference-in-differences strategy, individual-level survey data, and the World Health Organization's z-score based measures for child health, I find that girls born after the ban are more likely to be stunted/malnourished compared to boys as a result of the
ban. Also, the probability of stunting is significantly increasing in girls' age after the ban, suggesting that parents respond by investing fewer resources in girls. While helpful, existing papers only focus on the effect of prenatal sex detection on sex ratios and child mortality. This paper adds to the literature by examining the impact of prenatal sex detection on female child health and gender discrimination.

Chapter 3 examines the effects of communal violence on women's marital outcomes. Using individual-level survey data from India and a difference-in-differences approach, the study shows that the Hindu-Muslim riots in Gujarat in the year 2002 led to a decrease in women's age of marriage and an increase in the probability of getting married before the age of 18. However, event-study and synthetic controls methods suggest that the effects were not immediate and are prominent two years after the riots. Women married after the riots also had fewer years of education, and poorer social and economic status.