Empirical essays on environmental and health economics in developing countries

Health and environment have recently been active research areas in development economics. However, estimating impacts on development has been hampered by concerns that there may be confounding variables that bias the estimates. My dissertation evaluates a credible relationship between the two by identifying an empirical context in which the roles of confounding variables are mitigated.;The first essay quantifies the impacts of air pollution and related regulations on infant mortality in China. I exploit plausibly exogenous variations in air quality generated by environmental regulations since 1995. The results suggest that the regulations led to significant reductions in air pollution and infant mortality rate (IMR). I estimate that 25,400 fewer infants died per year than would have died in the absence of the regulations, corresponding to about a 21 percent decline in IMR. The instrumental variable estimates indicate that a one percent reduction in total suspended particulates results in a 0.95 percent reduction in IMR, whereas a one percent reduction in sulfur dioxide results in a 0.82 percent reduction in IMR. The estimated impact of a unit change in TSP is of similar magnitude to that found in the U.S., but the elasticity is substantially higher in China, highlighting the greater benefits associated with regulations when pollution is already quite high.;The second essay quantifies the returns of health infrastructure to child health status, as measured by weight-for-age z-scores. By exploiting plausibly exogenous changes in access to health services induced by the health policy in South Africa after the end of apartheid, I show that gaining access to health institutions improves nutritional status of boys but not of girls among newly born babies and children with low health status.;The third essay investigates whether improved access to health services leads to better educational achievements. The health policy I examine in the second chapter provides a rare opportunity to credibly evaluate a relationship. The results indicate that access to health services has little impact on educational outcomes, except that boys who gained substantial increases in health access at the time of enrollment are likely to start school earlier.