Health Status and the Retirement Decision among the Early-Retirement-Age Population

Shailesh R. Bhandari, U.S. Census Bureau

This paper uses the SIPP Wave 1-6 to addresses two questions relating to the role of health status in early retirement decisions. The research questions are: What are the alternative ways of measuring health status? And, how do the alternative measures of health status explain the odds of early retirement among people aged 50 to 64? This paper considered alternative measures of health status in its regression models. Preliminary results show that the self-reported health-status measure, the doctor-visit frequency measure, and the hospital night frequency measure do not help much. Alternatively, results using a dummy variable representing whether or not the respondent was taking prescription drugs appeared plausible; and results using “change in health status” as an explanatory dummy variable produced more robust results than the other model specifications.