DESCRIPTION (provided by applicant): This project has three main objectives. First, it aims to identify the entire retirement processes of individuals and couples, defined by labor force exits and reentries, Social Security and pension receipt, and retiree self definitions. Second, it assesses the effects of choice (disability, job displacement, caregiving, and spouse control), instrumental, and cognitive utilities (reflecting social, couple, family and personal expectations about the timing of the retirement transition) on retirement transition processes. The third objective is to assess the outcomes (retirement plan implementation, benefit optimization, retirement satisfaction, change in income, food security, depressive symptoms, health risk behaviors, self-rated health, health insurance coverage, and life expectations) of specific retirement transition processes for individuals and couples. The analyses will rely on secondary analyses of the Health and Retirement Survey (waves 1992-2004). Main analyses techniques employed will include multistate life tables, regression techniques for censored data and interdependent observations (couples), structural equation modeling, and general estimation procedures. The project has important health-related outcomes. It will demonstrate to what extent health limitations preclude choice over retirement transitions and whether such limited choice significantly reduces retirement benefits as well as emotional and physical well-being and access to health insurance. The project will further assess whether negative outcomes of initial retirement transitions can be reduced through subsequent processes such as reentry into the labor force. In addition to its importance for health-related outcomes, the project also provides theoretical and methodological contributions to the retirement literature. Major innovations include the identification of retirement processes, the emphasis on control and cognitive utility, the measurement of post-retirement financial well-being in terms of the difference between actual and optimal benefits, and the assessment of cross-spouse influences and outcomes.