Understanding the link between health, wealth, and retirement is one of the central challenges for economic policy. In the United States, as in all developed countries, retirement ages have trended downward over time, even as length of life has expanded. This trend, along with lower fertility rates and higher costs of medical care has created challenges for public and private retirement programs. One obvious solution to the aging problem is to have people work until older ages. But consideration of policy for the elderly has been hampered by a lack of good retirement models. Why are people retiring earlier? Will continued health improvements help stem the trend towards earlier retirement? What impact would Social Security or Disability Insurance reform have on retirement rates? These questions are fundamental in the design of appropriate public policy and are central to our proposed research. The three aims of this proposal are 1) To develop a structural model of retirement that will allow us to predict the impact of potential policy changes and illustrate the social welfare tradeoffs of different policy options;2) To develop a leading indicator model of future retirement and disability-claiming patterns that can be used to predict the future retirement and disability-claiming behavior of today's middle-age generations;3) To explore what might be left out of existing economic models of disability and retirement.