Job Demand and Early Retirement
by Sepideh Modrek, Stanford University
Policy initiatives to address the long-term deficit have suggested that further increases to the full retirement age (FRA) may be necessary. Yet, there are subpopulations that may be adversely affected by such changes. We explore one such population, those in physically demanding jobs. We intend to examine whether and how physical job demand relates to the early retirement decisions while considering underlying health, injury history, and wealth accumulation in a population of aging manufacturing workers. We propose to use a rich set of administrative dataset, including detailed personnel records; externally rated physical job demand; real-time injury, health, pension, and wage data on a cohort of approximately 2,000 male Alcoa employees ages 45-50 in 1985 (birth cohorts 1935-1940) followed forward to 2009. We propose to use standard survival techniques to first descriptively examine whether and to what extent externally rated physical job demand at middle age, a decade or more before retirement, or the usual onset of chronic disease, is related to early retirement. Then we propose to use difference-in-difference methods exploiting the exogenous variation created by the 1983 Social Security reform, which increased the FRA from 65 to 66 in two-month increments per year of birth for cohorts born from 1938 to 1943, to examine if changes to the FRA had any effect on retirement age for those with more physically demanding jobs. Third, we intend to document the progression of physical job demand and to examine endogenous transitions to less-demanding jobs based on injury history or other health considerations. Finally, we hope to use propensity score methods to circumvent endogenous transition to less physically demanding jobs and to ultimately examine whether transitions to less-demanding jobs help delay retirement.