Overview
Bruce D. Meyer is the McCormick Foundation Professor at the Harris School of Public Policy, University of Chicago, and a Research Associate at NBER. He is one of the leading authorities on measuring poverty and well-being using consumption rather than income, on the accuracy of survey-based transfer income measurement, and on the behavioral effects of social insurance programs. His work on transfer program under-reporting (Meyer and Sullivan 2003, 2009, 2015) established that standard survey-based poverty statistics substantially undercount government transfers, leading to systematic overestimates of poverty rates and gaps. His collaborative work with Wallace Mok (2013) on the economic consequences of disability provides the most comprehensive long-run consumption-based welfare analysis of disability onset available from U.S. panel data.
Key Contributions
- Quasi-experimental methodology (Meyer 1995): Wrote the canonical survey of natural and quasi-experimental designs in economics, published in the JBES symposium on program evaluation. Introduced Campbell's threats to validity framework to economics, advocated for multiple comparison groups and multiple time periods as diagnostic tools, and argued that the "quasi-experiment" term from psychology is more accurate than "natural experiment." Core thesis: if variation cannot be experimentally controlled, its source must be understood.
- Workers' compensation benefit quasi-experiment (Meyer, Viscusi, and Durbin 1990/1995): Used sharp increases in state maximum weekly benefit amounts — which raised benefits only for high-income workers — as a natural experiment to identify the effect of workers' compensation benefit levels on injury claim duration. The untreated comparison group (low- and average-income workers) provides the DiD control. A central example in Meyer (1995).
- Unemployment insurance research (Meyer 1990): Identified the effect of UI benefit levels and duration on unemployment spells using variation in state benefit schedules. Used the quasi-experimental variation from changes in state laws affecting specific earnings groups rather than individual earnings-history variation, which would be endogenous.
- Transfer under-reporting correction: With James Sullivan, developed program-specific reporting rate adjustments for PSID, CPS, and CEX data, showing SSDI/SSI are under-reported by ~20–40% in household surveys. These corrections are essential for accurate welfare analysis of disability programs.
- Disability onset welfare analysis (Meyer and Mok 2013): Using 42 years of PSID data, documented that Chronic-Severe disability produces −76% earnings, −28% post-transfer income, and −25% food-and-housing consumption over 10 years, with no home production substitution and deteriorating diet quality.
- Disabled women's welfare analysis (Meyer and Mok 2014): Extended the framework to 44 years of PSID data for female household heads and wives. Women's disability effects on earnings are similar to men's (Chronic-Severe: −82%), but income and consumption declines are roughly half those of men (Chronic-Severe: −10% food+housing vs. −25% for men). Documents novel nonlinear severity gradient in disability-divorce: Temporary and Chronic-Not Severe significantly raise divorce hazard; Chronic-Severe does not. No added worker effect for husbands of disabled wives; spousal care takes the form of companionship (time together watching TV) rather than instrumental home production.
- Well-being of the poor (Meyer and Sullivan 2008): Using CE Survey 1993–2003, documented the consumption-income divergence for single mothers after welfare reform: bottom-decile income fell −16% while consumption rose +7%; argued consumption is the superior welfare measure; introduced nonmarket-time break-even calculation showing welfare reform imposed real utility losses even where consumption rose.
- Well-being of the poor (series): Multiple papers with Sullivan demonstrating that consumption-based poverty rates yield substantially different (lower) trends than income-based measures, with implications for evaluating TANF, EITC, and Medicaid.
- Medicaid and mortality (Wyse and Meyer 2023): Co-author with Angela Wyse of the largest quasi-experimental Medicaid-mortality study to date. Linked the 2010 Census to IRS MAGI income records, CMS Medicaid enrollment data, and SSA Numident mortality records, constructing a universe of 37.5 million non-disabled low-income adults ages 19–59. Staggered DiD on ACA Medicaid expansion timing; discrete-time proportional hazard model; preregistered analysis plan (second nonexperimental economics study to do so). Main results: −2.5% ITT annual mortality hazard; −21% TOT for new enrollees; ~27,400 lives saved 2010–2022; $179K per life-year saved. Key policy finding: universal Medicaid closes only 5–20% of the income-mortality gap, indicating that insurance is not the predominant driver of SES health disparities. See Medicaid and Mortality.
Related
Related (additional)
Sources