Gender-Specific Labor and Health Responses to Program

This dissertation explores gender-specific labor and health responses to various programs. Often evaluations of programs, interventions, or regulations examine their effects on the intended outcomes of the overall populations that they target, but do not fully explore sources of heterogeneity. Many differential impacts of such programs occur on the basis of gender. Even if outcomes are evaluated separately by gender, the set of outcomes itself may be limited to gender-neutral measures. The chapters in this dissertation examine gender-specific responses to programs in settings that include the U.S. health care system, the agricultural household in Zambia, and the U.S. university environment.;The first chapter examines the impact of health care access for women on birth outcomes. Studies in both economics and medicine find that poor mental health during pregnancy is associated with poor birth outcomes, but little is known about the ability of mental health care access and treatment to counteract these effects. I use a difference-in-differences strategy exploiting the staggered enactment of state mental health parity laws in twenty-five states from 1995-2002 to identify the impact of mental health care access on the probability of an adverse birth outcome. These state mental health parity laws are insurance mandates requiring coverage of mental health care be equivalent to physical health care. Using birth records, I find that, among the group of mothers most likely to have private insurance, introduction of a mental health parity law in a state decreased the probability of an adverse birth outcome. Furthermore, I find that the parity laws decreased the likelihood that a pregnant woman hospitalized for delivery would receive a mental illness diagnosis.;The second chapter examines the impact of soil conservation adoption on intra-household allocations in Zambia. This chapter is joint work with Conor Carney. Anecdotal evidence suggests that household adoption of soil conservation techniques may lead to greater female involvement in agricultural production, but little is known about the causal relationship. We test this relationship empirically, examining the impact of soil conservation adoption on female agricultural labor contributions within households in Zambia. Moreover, we explore the short-run effects of increased female agricultural productivity on female empowerment through measurement of changes in gender-specific resource allocations within the household. The extension of funding of conservation farming (CF) training sessions in 2007 in specific districts in Zambia provides variation in CF take-up. We use this variation to implement a difference-in-differences strategy on a number of datasets. Our results suggest that expansion of funding for CF training sessions increased take-up of CF, increased female labor hours, and shifted household expenditures towards goods associated more strongly with female preferences than male preferences. These results show the importance of understanding the impact of development programs, specifically promotion of agricultural technologies, on household gender dynamics.;The final chapter explores the impact of a gender-biased shared interest within networks on labor market outcomes. There are many explanations for the gender wage gap in the U.S., but differences in strength and number of ties within a worker's networks are not well-studied. These network differences may be due to the dissimilar ways in which men and women network. In this paper, I specifically explore a gender-biased shared interest within networks, football team performance, in order to better understand the impact of these networking differences. In the college setting, a college football team's stronger presence on campus, as measured by win-loss record, may lead to stronger male networks between students and possibly alums, which might improve male educational and labor market outcomes. It is unclear, however, whether the same effect would exist for female students, especially as the sport has a larger male following. In order to explore the impact of a change in the strength of a college football team on labor market outcomes by gender, I estimate the effects of fluctuations in football win-loss records on wages. Specifically, I examine the impact of college football success at NCAA Division I Football Bowl Subdivision (FBS) schools relative to a student's period of attendance on future earnings. Furthermore, I explore the potential mechanisms through which this effect could be occurring, including network development during and after college.