The Effect of Longevity on Retirement and Un-retirement

An efficient retirement plan requires a precise prediction of one’s own longevity.

An efficient retirement plan requires a precise prediction of one’s own longevity.

Zhendong Zhao is PhD candidate in Economics at the City University of New York (CUNY) Graduate Center. He is currently working as a Graduate Teaching Assistant for PhD courses at Economics Department.

He has an extensive research experience in labor and health economics, specialized in retirement, Social Security benefits, human capital theory, applied econometrics, and dynamic programming. He is familiar with large datasets from Health and Retirement Study, Current Population Survey, Medical Expenditure Panel Survey, and Panel Study of Income Dynamics.

In the context of the research domains study in our department “Labour Market”, Zhendong Zhao has been invited to present his scientific article titled: “The Effect of Longevity on Retirement and Un-retirement ». The expectation of remaining lifespan plays a key role in the sequential decisions of when to retire, and whether to re-enter the labor market after full-time retirement. Sufficient and efficient provisions for retirement could save an individual from outliving his or her retirement wealth, or not consuming enough and leaving an unnecessarily large amount of wealth at death. An efficient retirement plan therefore requires a precise prediction of one’s own longevity.

In his article Zhendong Zhao examines the effect of longevity on older Americans’ labor supply decisions of retirement and un-retirement. Instead of using self-rated survival probabilities as the proxy of longevity expectations, he uses data from the Health and Retirement Study to predict longevity from a survival model with a rich set of variables including parental mortality information, current health and socio-economic variables. He finds that the predicted longevity fits actual longevity better than subjective survival rates.

Using predicted longevity as one of the independent variables in a sequential logit model of retiring and un-retiring, he finds that individuals retire and re-enter the labor market as if they knew their true potential longevity regardless of their wrong subjective survival expectations, i.e., individuals with higher predicted objective longevity retire later, and are more likely to return to the labor market after initial retirement.

Moreover, he finds that many individuals misperceive their future survival probabilities, and the mismatch of subjective and objective longevity induces suboptimal saving behaviors among retirees.