Retirement Research Consortium

Although government expansion of health insurance to older workers leads to labor supply reductions for recipients, there may be spillover effects on the labor supply of affected spouses who are not covered by the programs. In the simplest model, health insurance on the job is paid for in terms of lower compensation on the job. Receiving health insurance exogenous to employment is akin to a positive income shock for the household, causing total household labor supply to drop. However, it is not clear within the household whether this decrease in labor supply will be borne by both spouses or by a specific spouse. We use a mid-1990s expansion of health insurance for U.S. veterans to provide evidence on the effects of expanding health insurance availability on the labor supply of spouses. Using data from the Current Population Survey, we employ a difference-in-differences strategy to compare the labor market behavior of the wives of older male veterans and non-veterans before and after the VA health benefits expansion to test the impact of public health insurance on these spouses. Our findings suggest that although household labor supply may decrease because of the income effect, the more flexible labor supply of wives allows the wife's labor supply to increase, particularly for those with lower education levels.

This study considers nonworking older adults and their channels of support before qualifying for Social Security benefits. Using 18 years of data from the HRS, we analyze nonearners' characteristics, including demographics, health status, lifetime labor force attachment, along with the levels and sources of their income and assets. We explore the effects of various factors on the likelihood of being a nonearner and observe the consequences of not working during one's 50s with regard to poverty, age of Social Security claiming, and overall retirement satisfaction. Finally, we analyze how these relationships have changed over time, particularly after the Great Recession.

The paper uses micro-survey data from successive waves of the Panel Study on Income Dynamics to investigate the distribution of wealth and job losses during the 2007–09 recession for different segments of the population, and the effect of the recession on retirement decisions of older workers. Estimates of wealth loss are constructed for major socioeconomic groups and compared with those of the Survey of Consumer Finances. The panel dimension of the data is used to measure change in the labor force status of workers and to estimate the determinants of the decision to transition from participation in the labor force to retirement. The study concludes that retirement decisions are influenced both by variations in labor market conditions and the value of household wealth, but the labor market exerts a larger impact.

The increasing cost of employer contributions for employee health insurance reduces the percentage of compensation that is subject to the payroll tax. Rising insurance contributions can also have a more subtle effect on the Social Security tax base because they influence the distribution of money wages. Workers bear most of the burden of employer health contributions through lower money wages. Any change in the average cost and distribution of costs of employer health plans can have an effect on the distribution of wages and the percentage of wages subject to the payroll tax. This paper uses the Medical Expenditure Panel Survey (MEPS) to analyze trends in the cost of employer health contributions and the cross-worker distribution of health contributions. Our analysis shows that the 1996–2008 increase in employer health premiums was faster than overall compensation increases but only slightly faster among workers below the taxable maximum compared with those above the maximum. However, because employer health insurance premiums represent a much higher percentage of compensation below the maximum taxed earnings amount, the effect of health cost trends exerted a disproportionate downward pressure on money wages below the taxable maximum, reducing the percentage of compensation subject to the payroll tax. We simulated the implications of the health reform law on the trend in employer health costs around 2016. We find only slight effects on the fraction of worker compensation that will be subject to Social Security taxes. The higher insurance costs faced by employers who will be required to offer health plans will be approximately offset by lower health costs on the part of employers who will see some insured employees accept subsidized health insurance outside of an employer plan. The main long-term impact of reform on the taxable wage base is likely to be through its effect on the trend in underlying health care costs.

This study examines Social Security claiming behavior, which has important implications for older Americans and the system itself. Retirees may begin collecting benefits as early as age 62, but early claimants receive lower monthly benefits for the rest of their lives. Our data come from Survey of Income and Program Participation files from 1984 to 2009 linked to administrative records on earnings and benefits. The sample is restricted to respondents with 40 quarters of covered employment who did not claim benefits before age 62. Results indicate that early claiming declined over the past decade, after increasing over the previous 10 years. For men, the share claiming at age 62 fell from 55.3 percent in the 1930–34 birth cohort to 46.4 percent in the 1940–44 cohort. Over the same period, the share of women claiming at 62 fell from 59.3 to 49.0 percent. The recent trend toward delayed claiming is evident among all educational groups, not just college graduates. Hazard models show that high unemployment boosts Social Security claiming among men with limited education. A 1 percentage point increase in the state unemployment rate is associated with a 0.4 percentage point increase in the likelihood each month that men who never attended college claim benefits, a relative increase of 6 percent. This estimate implies that the Great Recession increased claiming for men with limited education by about 40 percent. Claiming behavior among women and well-educated men is not significantly correlated with the state unemployment rate, however.