Three Essays on the Economics of Health in Developing Countries.

摘要

This dissertation consists of three chapters that address health issues in developing countries. The first two chapters study Ghana's social health insurance program, the National Health Insurance Scheme. Many developing countries have recently instituted social health insurance schemes (SHIs) to ease financial barriers to utilization of healthcare services and help mitigate the effects of adverse health shocks. Although these SHIs offer generous terms and benefits, enrollment remains low especially among the poorest households who are the intended primary beneficiaries. The first two chapters are based on randomized field interventions implemented in the Wa West district of the Upper West Region of Ghana to (a) understand the reasons for low enrollment in SHIs; (b) estimate the effects of insurance coverage on utilization of healthcare services, financial protection and health outcomes, and c) learn about how resource-constrained households allocate health resources among its members. The interventions were increased convenience of signing for insurance, an education intervention that provided information about the insurance program, and a subsidy intervention that included varying levels of subsidies for insurance premiums.;The first chapter deals with objectives (a) and (b). The results show that inadequate information about the insurance program, and insurance premium and fees affect enrollment. The results also show that the demand for insurance is price elastic in the sense that small subsidies generate substantial enrollment effects. Insurance coverage leads to increased utilization of healthcare services, reduced out-of-pocket payments among individuals with prior positive expenses and ,moderate improvement in health outcomes. The results suggest strong complementarities between providing information and providing subsidies in utilization and health outcomes, an indication of the importance of the combined interventions for achieving changes in health-seeking behavior and outcomes.;The second chapter focuses on objective (c): intra-household allocation of health resources among resource-constrained households. The analysis in this chapter is based on households who were assigned to receive subsidies only and the pure control group. Two types of vouchers were issued to households who did not receive full subsidies: one that allowed households to decide how to allocate subsidy among its members and one in which they had no control over the allocation. This chapter compares within household enrollment patterns across these two vouchers. The results suggest that households prioritize children in the presence of resource constraints. Among children, households who were allowed to determine allocation of subsidy amounts enroll 11.7 percentage or 18% more boys than girls. The results suggest that these patterns of allocation cannot be explained by baseline health conditions or expected health. The chater presents supporting evidence that differential labor market participation is a likely explanation for the differential allocation by gender among children: among children aged 7-17 years, labor market participation is 3.6 percentage points higher for boys than girls.;The third and final chapter is coauthored with Ayaga A. Bawah and James F. Phillips. The chapter seeks to explore how the quasi-experimental introduction of reproductive and family planning services affects the fertility behavior of different socio-economic groups in a rural African setting. We combine a quasi-experimental introduction of reproductive and family planning services in the Kassena-Nankana districts in the Upper East Region of Ghana with longitudinal data from the Navrongo Health and Demographic Surveillance System to quantify the differential fertility effects of the interventions by socio-economic status (as measured by woman's education status, her husband's education status and wealth). We track the fertility behavior and outcomes of more than 24,000 women in their reproductive age (15-49) over a period of eighteen years. Our results show that before the interventions educated women did not have significantly fewer children, but desired lower family sizes and were more likely to use modern contraceptives. However, husband's education was associated with lower fertility especially when their wives were also educated. Wealth was associated with higher fertility, reflecting a higher child survival rate in wealthy families. Moreover, controlling for wealth does not affect the effect of education on fertility. We find that the reproductive health interventions affected both educated and uneducated women but the effect on educated women was stronger, leading to the emergence of an education-fertility differential 16 years after the introduction of the interventions. Our results suggest that in settings where men dominate reproductive decision-making, their education status may have a stronger effect on fertility than the educational attainment of women.