There’s some disagreement in the literature on the effects of disability insurance on labor force participation that I thought I’d post about given this recent NPR Unfit for Work feature. Till von Wachter, Jae Song, and Joyce Manchester have a nice paper that uses great social security earnings data to weigh in on this important debate about DI and labor force participation.
- Large Effects of DI on LFP: “A growing body of research suggests that some workers in difficult economic conditions exit the labor force to apply for and often receive DI benefits (e.g., Rupp and Stapleton 1995; Black, Daniel, and Sanders 2002). That pattern has intensified since the mid-1980s (Stapleton, et al. 1998; Autor and Duggan 2003). Because such economically motivated applicants on average should have less severe disabilities, some of them might have worked in the absence of DI.”
- Small Effects of DI on LFP: “Using data covering the mid to late 1970s, [Bound] found that the employment rate of older male rejected DI applicants was quite low. That finding has been replicated for the early 1990s and extended by work exploiting different features of the DI system to obtain more precise counterfactuals (e.g., Bound, Burkhauser, and Nichols 2003, Chen and van der Klaauw 2008, Maestas and Yin 2008). “
- Their paper: “We first replicate Bound’s result for male applicants ages 45-64 and show that his main conclusion is stable over time and robust to many alternative specifications. We then extend Bound’s analysis to male applicants ages 30-44. Those younger applicants constituted a small fraction in Bound’s sample, but they grew to almost 40 percent of new DI beneficiaries in the early 1990s and accounted for about 30 percent of new beneficiaries and more than half of rejected applicants in 2007. For young rejected applicants we find significant post-application employment.”
Bottom Line Findings: “We provide evidence that younger rejected male DI applicants exhibit substantial labor force attachment. Similarly, applicants with low-mortality impairments such as back pain and mental health problems exhibit substantial labor force attachment. We show that continuing increases in the share of younger beneficiaries or beneficiaries with low-mortality impairments will further raise the potential employment of workers receiving DI benefits. Since younger new DI beneficiaries are on average on the DI program longer than older new beneficiaries, the higher potential employment rates we find imply significant potential losses in lifetime employment and earnings. These losses are substantial compared to the higher lifetime value of DI benefits for younger new beneficiaries. We also provide new results on the level and dynamics of earnings of different applicant groups before and after application. Our results confirm that an increasing number of individuals may have applied for DI because of worsening economic conditions. Such economic inducement is consistent with our finding of substantial employment among beneficiaries whose application was rejected at the DDS level.”