Reform of the health care system in the United States has reached the top of the national policy agenda. A central issue in the debate is the status of health benefits for retirees, particularly those under age 65. Currently approximately 35 percent of workers are offered retiree health benefits from their employers. Any health care proposal that provides access to health insurance for all workers who retire before Medicare eligibility may significantly alter the set of retirement incentives facing older workers. The demand for health insurance among older Americans further determines the impact of reform efforts that rely upon premium subsidies to increase health insurance coverage or offer choices among health care plans. The behavior of older Americans, both their labor force and insurance decisions, is likely to affect the cost of health care reform. Currently, efforts to assess the behavioral and financial implications of health care reform efforts are limited by the lack of empirical research addressing these issues. This proposed research has two goals: (l) assess the employment effect of changes in access to retiree health benefits; and (2) estimate the demand for health insurance among older workers (those aged 50 to 64). To assess the effect of changes in access to retiree health benefits, we need to understand the relationship between health benefits and retirement behavior. The health and retirement literatures, however, have been able to provide only preliminary estimates of this effect, due largely to data limitations. We propose to use three relatively new data sources to address this issue: the 1987 National Medical Expenditure Survey, the Panel Study of Income Dynamics, and the Health and Retirement Survey. These three data sources, only recently available, will be used to estimate static and dynamic models of retirement behavior to assess the relationship between access to health insurance and retirement behavior. We will also test hypotheses regarding the relationship between retirement and other factors such as health status, economic variables and individual characteristics. Estimates from the model will be used to simulate the effect of different health care reform proposals on retirement behavior. The second goal of the proposed research is to study demand for health insurance among older Americans. The price elasticity of demand for insurance is another key parameter for assessing the impact of health reform proposals. While an extensive literature exists, there has been little effort to study demand for health insurance among the older population. We propose to use data from the 1987 National Medical Expenditure Survey to study demand for health insurance among workers aged 50 to 64. By estimating the price elasticity of demand among older workers and the demand across plan types (traditional versus HMO), we can inform policy about the health insurance choices older Americans would make under
