Summary
Hemmeter and Bailey (2016) examine the earnings of disability insurance (DI) beneficiaries who undergo a Full Medical Review (FMR) — a continuing disability review (CDR) conducted by state Disability Determination Services — comparing those whose eligibility is "ceased" (terminated after medical review) to those "continued." Using Social Security Administration (SSA) administrative records linked to earnings data (1998–2008 FMR cohort, N = 2,124,835 DI workers), they estimate post-cessation earnings as an upper bound on what continued beneficiaries would earn if their benefits were similarly terminated. The majority of ceased beneficiaries work after program exit, but average earnings are low (~$13,000/year) and few sustain earnings above standard sufficiency thresholds across all five post-FMR years.
Key Claims
- The overall cessation rate after all appeals is 5.7% for DI workers undergoing FMRs, with large variation: 13.4% for ages 18–30 vs. 2.8% for ages 50–59; 17.6% for neoplasm diagnoses vs. <4% for congenital/endocrine conditions; only 21.5% of CDR-selected beneficiaries ever reach FMR stage
- Mean post-FMR earnings for ceased beneficiaries: ~$12,819/year (~70% of $19,331 pre-DI earnings); for continued beneficiaries: $1,477/year
- ~70% of ceased beneficiaries have some earnings in the 5 years after FMR; ~25% have no post-cessation earnings at all
- Only 37% have earnings in all follow-up years; only ~20% maintain earnings above Substantial Gainful Activity (SGA) or poverty threshold in every post-FMR year
- Cessation effect (upper bound): +$11,342/year and +43 percentage points (pp) probability of having earnings above SGA, poverty line, or full-time minimum wage (all significant at <1%)
- The exit margin shows a substantially larger potential employment effect than entry-margin estimates: Maestas, Mullen, and Strand (2013) found ~28 pp at the entry margin; Hemmeter and Bailey find ~43 pp at the exit margin — reflecting that FMR cessation selects the healthiest end of the current beneficiary distribution
- Expanding CDRs to more marginal cases (Medical Improvement Not Expected [MINE] diary types, low CDR profile scores, mailer recipients) would yield lower average earnings among newly ceased beneficiaries than current cessation populations, since those groups already show worse labor market outcomes
- About 20% of former DI workers return to the program within 8 years (Hemmeter and Stegman 2013 companion)
- SSA's CDR return is $13.2 per $1 spent on average (fiscal year [FY] 2011), but the Congressional Budget Office (CBO) estimates only ~$3 at the margin above current spending — consistent with marginal cases having worse cessation outcomes
Concepts Introduced or Extended
Entities Mentioned
Quotes
"While over half of ceased DI workers earn above the SGA level or the individual poverty threshold in at least one of the 5 years after program exit, only about 20% consistently earn over SGA or the poverty threshold in all post-program years."
"Increased program integrity, while appropriate for Social Security from a fiduciary perspective, may also increase the number of people requiring other supports."
My Take
This paper makes a clean methodological contribution by extending Bound's (1989) comparison-group approach to the exit margin — something rarely done because post-cessation earnings data for a large CDR population requires SSA administrative access. The ~43 pp upper bound is striking but must be interpreted carefully: the FMR population is pre-selected by SSA's profiling model for high likelihood of medical improvement, making it non-representative of the full CDR population. The paper's honest accounting of what expanding CDRs would yield — substantially worse outcomes for marginal cases — directly challenges the "more CDRs = proportional savings" assumption implicit in program integrity advocacy.