Session: Disability
Room: Hollister McManus Lounge
Time: Mon 15:00-16:30

Presenter: Jody Schimmel (Mathematica Policy Research. )

Discussant: Denise Whalen (University of North Carolina, Chapel Hill)

Abstract

After the onset of a health condition that limits one’s ability to work, many suffer large losses in earnings and household income. These losses may be most prevalent among older workers, who are more at risk for the onset of new health conditions. Some of the resulting earnings losses may be offset by receipt of unemployment insurance, worker’s compensation, or public programs such as Social Security Disability Insurance (SSDI), Supplemental Security Income (SSI), and Social Security retirement (SSR). However, the extent to which these other sources are able to offset earnings losses and protect older workers from poverty is not fully known.
Evidence from previous work indicates that the onset of health-related work limitations has significant negative effects on labor supply, leading to depressed earnings, lower household income despite increases in public and private benefits, and decreased wealth accumulation. But, the majority of earlier studies rely on fixed effects methods to identify how an individual’s income deviates from its pre-onset trajectory. As individuals near retirement age however, one’s pre-onset earnings trajectory may not accurately reflect the path that earnings would have followed in the absence of work-limitation onset. For example, if households had been planning on retiring but then experienced a health-related work limitation, estimates from the existing studies would overstate the effect of work-limitations on earnings and household income losses by assuming the pre-onset trajectory continued.
To better measure the counterfactual of how individual earnings and household incomes would have evolved without work-limitation onset, we carefully constructed a comparison group using propensity score matching methods. Using data from the Health and Retirement Study (HRS), we identify individuals who experience work-limitation onset after their baseline interview and match them to those who do not experience onset. Our matching method uses a battery of demographic, socioeconomic, and employment-related characteristics that may affect the likelihood of experiencing onset and income trajectories to predict the time to disability onset. Using the long panel of data in the HRS, we are able to use our matched treatment and comparison samples to assess how various sources of income evolve after work-limitation onset compared to what they would have been in the absence of onset.
Preliminary findings show that earnings losses after work-limitation onset are large and are not fully offset by other sources of income. Holding sex, baseline marital status, and eligibility for Social Security Retirement (SSR) constant, mean earnings four years after onset for the work limitation onset group are 70 percent lower than for the comparison group. Mean unemployment, workers’ compensation, disability insurance, and retirement benefits for the work limitation group are higher after onset than in the comparison group, but on average they offset less than 30 percent of the earnings difference. Post-onset poverty is 6 to 12 percentage points higher among those experiencing onset; eligibility for SSR at age 62 buffers some of this increased risk of poverty.

Key Terms

Disability, work limitation, income, older workers

Authors:

Jody Schimmel (Mathematica Policy Research)