Academic journal article Federal Reserve Bank of St. Louis Review

The Effects of Health and Wealth Shocks on Retirement Decisions

Academic journal article Federal Reserve Bank of St. Louis Review

The Effects of Health and Wealth Shocks on Retirement Decisions

Article excerpt

Both health status and net worth can affect retirement decisions. In some cases, early retirement may be precipitated by a shock to an individuals health and/or economic status. The authors examine how health and wealth shocks affect retirement decisions. They use data from the Panel Study of Income Dynamics to estimate a first-differences model of health and wealth shocks on retirement over the course of the 2000s in the United States. Their results suggest that acute health shocks are associated with labor market exits for older American men but not women. These results appear particularly strong for blacks, whose labor force participation seems particularly sensitive to health status, which may be due to different occupations for blacks and whites. (JEL J26, 112, D91)


Retirement decisions both affect and are affected by health status. Health status, in turn, has been linked to net worth. And according to the life cycle model of savings, retirement has an important effect on net worth because retirees begin to expend their assets to maintain consumption once they leave the labor force. Given the multidirectionality of all these influences at this three-factor nexus, it has been difficult to separate the direct impacts of health and wealth on retirement from the reciprocal effect of retirement on health and wealth. This is our goal in the present article.

Many studies have found that health shocks predict retirement decisions (see, e.g., Hagan, Jones, and Rice, 2009). For example, using fixed effects estimators and instrumenting subjective health by "health stock," Disney, Emmerson, and Wakefield (2006) find that ill health strongly predicts early retirement among respondents older than 50 years of age in the British Household Panel Survey. Health limitations have also been shown to have a similar impact on early retirement decisions. However, the evidence does not completely support the claim that health shocks lead to exit from the labor force. For example, French (2005) finds that health is not among the more important determinants of job exit at older ages.

In addition to this ambiguity, we also know that retirement may adversely affect health. U.K. panel data have been marshaled to show that labor force participation has a gender-specific effect on health. Specifically, ongoing labor force participation is detrimental to male health but positive for female health (Cai, 2010). However, the evidence is more consistent in the United States: Thanks to the natural experiment of the Social Security notch, which applies to individuals born between 1917 and 1921, we know that retirement is not good for the health of septuagenarians. Specifically, individuals born before January 1, 1918, received lower Social Security benefits compared with their counterparts born on January 1, 1918, and thereafter. Lower payments resulted from a 1977 correction of a previous flawed calculation. (1) On average, those who received the higher payments thanks to a change in cost-of-living indexing during the 1970s left the labor force earlier than those who received the lower payments. However, despite the greater transfer income, Snyder and Evans (2006) found that those who retired earlier had a shorter life expectancy.

Meanwhile, just as health shocks can affect an individual's decision to continue working, so also can financial shocks. In addition to unexpected job loss, other shocks to individual net worth include investment performance, unexpected costs, family transitions (such as divorce), or other dynamics such as tax law changes. According to economic theory, smoothing consumption over the life course would require a delay in retirement when a negative shock to retirement savings occurs, just as a positive shock would allow an early exit from the labor market. (2)

Indeed, empirical research has consistently found a positive wealth effect on retirement exits from the labor market. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.