Nicole Maestas

Overview

Nicole Maestas is a health economist at RAND Corporation (Santa Monica). Her research focuses on disability insurance, aging, and the labor market consequences of health shocks. She is best known for using examiner assignment as an instrumental variable to identify the causal effect of SSDI receipt on labor force participation — one of the cleanest identification strategies in the DI literature.

Key Contributions

• Causal effect of SSDI receipt (Maestas, Mullen, and Strand 2013, AER): exploited quasi-random assignment of 2.2M applications to ~8,700 DDS examiners as an IV for benefit receipt; LATE = 28 pp employment loss for the 23% of applicants on the margin of program entry (vs. 35 pp OLS — the gap explained by selection of more severely impaired individuals into SSDI); the disincentive is heterogeneous, ranging from ~0 pp (high-severity end of marginal group) to >50 pp (low-severity end); marginal entrants are disproportionately young and have mental disorders, not musculoskeletal impairments; processing time during appeals causes skill depreciation that partially confounds the OLS upper-bound interpretation.
• DI and the Great Recession (Maestas, Mullen, and Strand 2015, AER P&P): documented that the Great Recession produced a 6.7% application surge at its peak; used EDCS alleged onset dates to show that recession-era applicants had been impaired longer — consistent with the conditional applicant mechanism — and that DDS examiners strip this cyclicality from established onset dates.
• Program exit margin: DI-to-OA conversion (Maestas and Song 2011, MRRC WP 2010-247): Used the automatic conversion of DI beneficiaries to the OA retirement program at FRA as an exit-margin natural experiment. At FRA benefits continue unchanged but the DI SGA work rule is eliminated, dropping the implicit MTR from ~100,000% to 0% (post-2000). Using the universe of 1934–1942 birth cohorts (19.1M person-year observations, SSA administrative data 1995–2008), found that the DI work disincentive is detectable even among the oldest beneficiaries: among the ~12% with recent labor force activity, earnings reverse their downward trend sharply between ages 66 and 67. The response is concentrated on the intensive margin and is strongest among early entrants (entered DI before age 59), ruling out biological recovery. Because the experiment operates at the exit margin when general-population LFP is declining, the estimates are a lower bound on residual work capacity of the full beneficiary population. See Causal Effects of DI Receipt.
• SSDI reform proposal — work capacity measurement + partial benefits (Maestas 2019, AAPSS 686(1): 93–120): Synthesizes the research case for SSDI reform in a review article in the Annals of the American Academy of Political and Social Science. Five core findings: (1) DI participation reflects labor demand as much as health; (2) ~43% of applicants have residual work capacity; (3) capacity decays during the multi-year application process — explaining why (4) post-enrollment work incentives have universally failed (BOND: zero employment effect); (5) SSDI insurance value likely offsets distortionary costs. Proposes two interlocking reforms: replace the medical-vocational grid (last updated using 1990s Dictionary of Occupational Titles, only exertional RFC, fixed age thresholds) with a WD-FAB/ORS-based individualized system measuring function across 8 domains and matching outputs to an updated occupational requirements database; and introduce partial disability benefits (disability rating = percent earnings loss; applicants may apply while still working). Yin (2015) projects partial benefits reduce net expenditures. See Vocational Grid and Disability Insurance Program.
• Processing delay and labor force decay (Autor, Maestas, Mullen, and Strand 2015, NBER WP 20840): using separate jackknife examiner IVs for processing time (EXTIME) and allowance propensity (EXALLOW) in the DIODS universe of 1,039,221 2005 initial determinations, identified the delay-decay channel: each additional month of processing time reduces employment by 0.44–0.52 pp and annual earnings by $133 (year 3). A 5-month waiting period falsification test confirms the mechanism is human capital atrophy from LFP exit. The joint two-IV system corrects the MMS (2013) receipt estimate from −27 pp to −48 pp employment (year 3). See Causal Effects of DI Receipt.

Related

• Kathleen Mullen
• Alexander Strand
• Conditional DI Applicants
• SSA Sequential Determination Process
• Disability Insurance Program

Sources

• Maestas Mullen and Strand 2013 — Does Disability Insurance Receipt Discourage Work
• Maestas Mullen and Strand 2015 — Disability Insurance and the Great Recession
• Autor Maestas Mullen and Strand 2015 — Does Delay Cause Decay The Effect of Administrative Decision Time on the Labor Force Participation and Earnings of Disability Applicants
• Maestas and Song 2011 — The Labor Supply Effects of Disability Insurance Work Disincentives
• Maestas 2019 — Identifying Work Capacity and Promoting Work
• Carey Miller and Molitor 2025 — Why Does Disability Increase During Recessions Evidence from Medicare
• Deshpande and Lockwood 2022 — Beyond Health Nonhealth Risk and the Value of Disability Insurance
• French and Song 2014 — The Effect of Disability Insurance Receipt on Labor Supply
• Low and Pistaferri 2020 — Disability Insurance Theoretical Trade-Offs and Empirical Evidence
• Mueller Rothstein and von Wachter 2016 — Unemployment Insurance and Disability Insurance in the Great Recession
• Ruffing 2015 — Geographic Pattern of Disability Receipt Largely Reflects Economic and Demographic Factors
• Rutledge Zulkarnain and King 2019 — The Relationship Between Occupational Requirements and SSDI Activity
• Strand and Messel 2019 — The Ability of Older Workers with Impairments to Adapt to New Jobs