Do stronger employment discrimination protections decrease reliance on Social Security Disability Insurance? Evidence from the U.S. Social Security reforms

Abstract

This paper examines spillovers onto Social Security Disability Insurance (SSDI) that occurred due to the Social Security Amendments of 1983, which, among other changes, gradually increased the retirement age for full benefits from 65 to 67. We determine whether the spillovers onto SSDI were different in states with age and disability discrimination laws that were broader (covered more people) or stronger (allowed for more damages for plaintiffs) than the federal Age Discrimination in Employment Act and the Americans with Disabilities Act. Our paper uses three sources of data: (1) counts of the universe of SSDI applications and receipts by state, age group, sex, and year; (2) the Health and Retirement Study, merged with restricted-access state identifiers; and, (3) the Health and Retirement Study, merged with restricted access state identifiers and Social Security Administration Form-831 disability records. To quantify the moderating impact of existing state laws on spillovers onto SSDI applications, receipts and employment, we use a difference-in-differences approach, comparing age cohorts who were affected by the reforms to similar age cohorts who were unaffected, and then this comparing this affected-unaffected difference across states by state law. Using the Health and Retirement Study data, we also conduct heterogeneity analysis to determine if effects differed for different age groups (ages 55-61, ages 62-64, ages 65 to the full benefits retirement age), those with or without disabilities, and by sex. The paper found that: Spillovers to SSDI application and receipt occurred mostly at ages 55-61 rather and ages closer to traditional retirement ages. These spillovers almost exclusively occurred for those with disabilities only. The employment effects from the Social Security Amendments instead affected those without disabilities and those closer to traditional retirement ages more. Women faced larger spillovers to both SSDI application and receipt and to employment. States with a medical definition of disability, which much more broadly covers individuals by state employment discrimination laws compared to the restrictive federal Americans with Disabilities Act, had reduced spillovers to SSDI application. In some cases, this was also associated with reductions in SSDI receipt, but our strongest evidence suggests just a decrease in SSDI applications. There is weak evidence that other broader or stronger features of state disability discrimination laws reduce SSDI application and receipt. We do not find any clear moderating effect of disability discrimination laws on employment. We do not find any moderating effect of broader or stronger features of state age discrimination laws and SSDI application and receipt, but we do find evidence that these features boost employment for those approaching traditional retirement age. The policy implications of the findings are: A broader definition of disability under discrimination law, like the medical definition of disability, seems to be associated with reduced SSDI applications. These foregone SSDI applications, which we argue in the paper were likely unsuccessful, reduced costs both for applicants and for SSA, given the large and indirect costs of SSDI application, especially given the long wait times for a final decision. Other stronger and broader features of disability discrimination laws may also reduce reliance on SSDI. There is not clear evidence that disability discrimination laws boost or reduce employment, suggesting that they may not have a strong role in affecting employment outcomes for individuals with disabilities and/or older workers.