Jae Song

Overview

Economist at the Social Security Administration. Works on disability insurance, labor market outcomes of DI applicants and beneficiaries, and structural models of the DI determination process.

Key Contributions / Features

• Causal effect of DI receipt on labor supply (French and Song 2014): Co-authored the ALJ judge lottery study establishing a 26 pp causal LFP reduction from DI receipt, and documenting that over 60% of ALJ-denied applicants eventually receive benefits within ten years. See DI Denied Population.
• Causal effect of DI allowance on mortality (Black et al. 2024): Co-authored the extension of the judge IV to 10-year mortality (N = 610,231, ages 55–64, 1995–2004). LATE = +2.8 pp for marginal appellants; MTE decomposition shows inframarginal (sicker) recipients benefit while marginal (healthier) recipients experience increased mortality via work disincentives. See Causal Effects of DI Receipt and DI Beneficiary Mortality.
• Employment and earnings of allowed and rejected DI applicants (von Wachter, Song, and Manchester 2011): Co-authored the AER longitudinal analysis (101(7): 3308–3329) linking SSA application records (1981–1999) to W-2 earnings data (1978–2006). Key findings: rejected applicants earn $10,000/year median (ages 45–64) vs. ~$35,000 for matched non-applicants and ~$3,000 for DDS-allowed beneficiaries; hearings-allowed applicants have higher employment (~24.5$5,000) than DDS-allowed (17.9%; ~$3,000); pre-application earnings dip is multi-year and gradual for rejected applicants but concentrated in the year before application for allowed applicants; foregone earnings of rejected applicants ≈ one-third of total DI liabilities. See DI Denied Population and Causal Effects of DI Receipt.
• Structural model of DI using matched SIPP/admin data (Lahiri, Song, and Wixon 2008): Co-authored a structural model of the DI determination process incorporating econometric selection correction.
• Program exit margin — DI-to-OA conversion (Maestas and Song 2011, MRRC WP 2010-247): Used automatic conversion of DI beneficiaries to OA at FRA as an exit-margin RD. Universe of 1934–1942 birth cohorts (19.1M person-year observations); disincentive effect is significant among the ~12% with recent labor force activity — annual earnings reverse downward trend sharply at ages 66–67 post-conversion. Strongest for early entrants (entered DI before age 59). Interpreted as lower bound on residual work capacity of the full beneficiary population. See Causal Effects of DI Receipt.
• FRA behavioral substitution into DI (Duggan, Singleton, and Song 2007): Co-authored the cohort-analysis study identifying the behavioral channel by which the 1983 FRA increase expanded DI rolls. The FRA rise cuts early OA at 62 from 80% to 70% of PIA (born 1960+) while leaving DI unchanged, raising the expected PV advantage of DI over early OA from ~$37K (born 1937) to ~$55K (born 1960). Using SSA CWHS data for men born 1935–1945, found ~42,000 additional male DI recipients in the long run (1.3% of male rolls), concentrated among men ages 63–64 and low-AIME workers. The induced DI increase offsets only 3.9% of projected OASDI savings from the FRA rise. See DI Growth Decomposition.

Related

• Eric French
• Bernard Wixon
• Till von Wachter
• Joyce Manchester
• Disability Insurance Program
• DI Denied Population
• Causal Effects of DI Receipt
• Bernard Black
• DI Beneficiary Mortality
• Mark Duggan
• Perry Singleton
• DI Growth Decomposition

Sources

• French and Song 2014 — The Effect of Disability Insurance Receipt on Labor Supply
• Von Wachter Song and Manchester 2011 — Trends in Employment and Earnings of Allowed and Rejected Applicants to the Social Security Disability Insurance Program
• Black French McCauley and Song 2024 — The Effect of Disability Insurance Receipt on Mortality
• Maestas and Song 2011 — The Labor Supply Effects of Disability Insurance Work Disincentives
• Duggan, Singleton, and Song 2007 — Aching to Retire The Rise in the Full Retirement Age and Its Impact on the Disability Rolls