The effects of disability programs on financial outcomes in the United States
In his Pulitzer-Prize-winning book, Evicted: Poverty and Profit in the American City, Matthew Desmond documents the instability in the lives of low-income tenants in Milwaukee, who are frequently uprooted due to eviction. He notes one exception: Renters receiving disability income from the Supplemental Security Income program have a more reliable income than low-wage workers and are therefore more likely to pay rent on time. Their disability income translates into greater housing stability.
This program, administered by the U.S. Social Security Administration, serves 6.3 million low-income Americans with disabilities, while the the agency’s Social Security Disability Insurance program serves 9.5 million American workers with disabilities. Both programs provide monthly income and health insurance to individuals with qualifying disabilities.
The expansion of these programs in recent decades has sparked public debate about who is considered disabled and who should be required to work for a living. Economists have largely focused on the effect of these programs on how much people work. Several studies find that the two programs reduce labor force participation by around 30 percentage points, meaning that these programs cost society some productivity.
Yet, as Desmond’s account suggests, disability programs may also have substantial benefits to recipients and to society. To date, these benefits have been difficult to study quantitatively due to the absence of data on consumption and well-being for the population of disability recipients. Understanding both the costs and benefits of Social Security programs for disabled Americans is critical to determining the appropriate generosity of these programs: If the benefits outweigh costs, then this argues for relatively generous transfers, while high costs relative to benefits argue for smaller transfers.
In a recent working paper published by the Washington Center for Equitable Growth, we study the effects of the Social Security Disability Insurance and Supplemental Security Income programs on several markers of financial distress. We construct the first nationwide dataset of Social Security disability applicants—about 44 million applications between 2000 and 2014—linked to bankruptcies, foreclosures, evictions, and home transactions. We use a research design based on the Social Security Administration’s decision-making process to ensure that we are capturing the causal effects of the programs, rather than simply correlations.
In particular, we use a Social Security Administration rule that instructs government examiners to use more lenient standards for applicants who are above age 55 relative to those who are below age 55. A similar change in rules occurs at age 50. We show that, before they apply, applicants who are above the cut-off age at the decision date look similar to those who are below the age cut-off. This means that we can attribute any differences in financial distress after the decision to the higher likelihood of being approved for disability benefits above the age cut-off.
We find that the Social Security disability programs reduce rates of financial distress substantially. Being approved for disability benefits at the initial stage (before appeals) reduces bankruptcy rates by 30 percent over the next 3 years. Among homeowners, approval reduces foreclosure rates by 30 percent and reduces rates of home sales by 20 percent over the next 3 years. (We also find a 10 percent decline in eviction rates over the next 3 years, consistent with Desmond’s account, but the estimates are not statistically significant.) In addition, we find that being approved for disability benefits increases the likelihood of purchasing a home by 20 percent over the same time period.
These results indicate that many recipients use their disability benefits to stay in homes that they would have otherwise lost to foreclosure or sale. More generally, they suggest that disability programs reduce rates of extreme consumption losses among recipients.
These effects are large relative to the size of disability benefits—on the order of $9,000 annually for the Supplemental Security Income program and $15,000 annually for the Social Security Disability Insurance program. Why are the effects so large? Two descriptive facts from our paper provide a clue. First, we find that rates of bankruptcy, eviction, and foreclosure and home sale (for homeowners) are higher among disability applicants than the general population. Second, we find that disability applicants apply for these programs after several years of increasing financial distress. At the time they apply, a large fraction of applicants are at risk for bankruptcy, foreclosure, and distressed home sale. It is perhaps not surprising, then, that being approved for disability benefits reduces the likelihood of these events.
This evidence on the effect of disability programs is consistent with evidence from other social safety net programs. Recent studies have found that public health insurance reduces medical out-of-pocket spending, medical debt, and mortgage delinquency rates. Another study found that unemployment insurance drastically reduces foreclosures.
The Social Security Disability Insurance and Supplemental Security Income programs may also have spillover benefits to nonrecipients. By reducing foreclosures among disability recipients, for example, these two programs probably increase the property values of neighboring homeowners.
How do these results inform the public policy debate over disability programs? The results do not speak to how disabled recipients are or whether they could work in the absence of benefits. Instead, they indicate that the two disability programs cut rates of financial distress among recipients substantially, providing some of the first evidence on the benefits of these programs to recipients.
—Manasi Deshpande is an assistant professor at the Kenneth C. Griffin Department of Economics at the University of Chicago. Tal Gross is an assistant professor of markets, public policy, and law at Boston University’s Questrom School of Business. Yalun Su is a research assistant at the Kenneth C. Griffin Department of Economics and the Becker Friedman Institute at the University of Chicago.