“If I am in Good Health”: Using Subjective Conditional Expectations Relating Health to Retirement with Lesson for Social Security Claiming and Sustainability

Published: 2017
Project ID: UM17-06


How much does health drive retirement? Recovering this relationship is challenging, since for each retiree we typically observe the age at which they retired given the health path they actually experienced. We cannot directly observe at what age they would have retired had their health been different. Hence, inferences that affect the sustainability of Social Security for potential changes in the trajectory of population health must necessarily be model-based. This project provides a novel strategy for assessing the effects of changing health. Its strategy is to ask older working individuals to report the likelihood that they will be working to specified horizons under alternative scenarios regarding their own health. Our analysis will deliver novel, individual- and aggregate-level, estimates of the ex ante effects of health on retirement age. Such estimates are a key ingredient of any forecast of the population labor supply and, thus, are of great relevance to the Social Security Administration.