Essays in Development Economics /

The following dissertation evaluates methods to improve the delivery of education and health care in low income countries. In "The Education Gender Gap in Developing Countries : The Role of Female Teachers", joint with Karthik Muralidharan, we add to the limited empirical literature on whether female teachers improve girls' education outcomes in developing countries, a policy frequently advocated. Using a difference-in-difference estimate with fixed effects, we find that teachers are relatively more effective at teaching students of their own gender. However, female teachers are more effective overall, resulting in improvement of girls' test scores by .036 standard deviations per year and a lack of adverse effects for boys. In "The Distributional Consequences of Micro Health Insurance : Can a Pro-Poor Program Prove to be Regressive?", I estimate heterogeneous effects of heath care consumption by poverty status and the related redistribution of premiums. Understanding these effects can inform optimal design of MHI contracts to maximize benefits and reduce unintended adverse effects. I document that poorer households consume significantly less health care at baseline, suggesting MHI may unintentionally lead to poorer households subsidizing wealthier households. But strikingly, twenty months after the introduction of MHI, there is no significant relationship between health care consumption and income among enrolled households. Thus, even though ex-ante health care consumption suggests MHI will result in regressive premium redistribution, ex-post behavior suggests the poor will not subsidize wealthier households. The next two chapters discuss community based micro health insurance (MHI). MHI, insurance targeted at low income populations, has been an increasingly popular policy, though empirical evidence of its effectiveness has been limited. In "Evaluating Health-Seeking Behavior, Utilization of Care, and Health Risk : Evidence from a Community Based Insurance Model in India", I assess the extent to which MHI reduces vulnerability and increases access to health care. I exploit a staggered expansion of MHI in which the villages offered the contract were randomly selected. I fail to find support for increased health-seeking behavior, but find suggestive evidence of reduced health shocks. This suggests the potential of MHI to improve the poor's health and implies strengthened financial sustainability of MHI program