EMPLOYER-SPONSORED HEALTH INSURANCE: COSTS, COVERAGE AND CONTRIBUTIONS (HEALTH INSURANCE, INSURANCE COSTS).

摘要

This dissertation focuses on three issues related to employer-sponsored health insurance: employers' costs, recent trends in who is covered, and the financing of insurance by contributions from employees.; Chapter one, which is joint work with Alan Krueger, analyzes the components of changes in employers' health care costs over the 1992-94 and 1987-93 periods. We find that employer costs have decreased primarily as a result of a steady decrease in the fraction of workers with coverage and a decrease in the rate of growth of insurance premiums. We conclude that the shift to managed care does not appear to be directly responsible for significant cost savings because managed care premiums are almost as high as those for fee-for-service plans, on average.; Chapter two, which is joint work with Henry Farber, analyzes trends in health insurance coverage over the period 1979-1997 for workers on different types of jobs. We find that while coverage declined for all workers, declines for workers in part-time and new jobs were larger than for long-term full time workers. They were also more likely to be due to decreased eligibility for health insurance, rather than a decline in enrollment conditional on eligibility, which was the main source of the decline for long-term full-time workers.; Chapters three and four explore possible explanations for why firms require workers to contribute explicitly to their health insurance premiums. In chapter three I consider the possibility that contributions arise because workers do not sort perfectly across firms on the basis of demand for health insurance, so employee contributions allow firms to identify which workers want health insurance. Using data from a 1993 survey of 22,347 establishments, I find some empirical support for this hypothesis. Chapter 4 considers two alternative explanations: first, that firms use employee contributions to cut total compensation in response to premium increases, and second, that employee contributions represent a mechanism for firms to share nondiversifiable risk of health expenditures with workers. Using panel data from a 1992-1995 survey of 395 large firms, I find some empirical support for the former hypothesis and no conclusive support for the latter.