Health and Retirement Study

Overview

The Health and Retirement Study (HRS) is a biennial longitudinal survey of Americans aged 50 and older, conducted by the University of Michigan with funding from the National Institute on Aging and the Social Security Administration. It began in 1992 and resurveys respondents every two years, adding new cohorts periodically. As of the early 2010s, the HRS contains five entry cohorts, with total N across waves in the tens of thousands. The HRS is the primary data source for empirical research on health, wealth, and retirement behavior in the United States.

Key Features

• Longitudinal panel: Biennial resurveys allowing tracking of health trajectories, asset drawdown, and economic outcomes from pre-retirement through death.
• Comprehensive wealth data: Collects total non-annuity wealth (home equity, financial assets, IRAs, Keoghs), plus annuity streams (Social Security, DB pensions).
• Health data: Rich multi-domain health battery; RAND HRS processed data constructs a PCA-based health index from 27 variables.
• Multiple entry cohorts: Original HRS cohort (born 1931–1941, entered 1992) plus AHEAD, War Baby, Early Baby Boomer, and Mid Baby Boomer cohorts.
• Linked administrative data: SSA earnings records and Medicare claims linkage available for approved researchers.

Key Uses in This Wiki

• Poterba et al. (2013a): Education-wealth gap decomposition; PCA health index; portfolio return gradient; N=23,952 respondents from first three cohorts, 1992–2008.
• Benítez-Silva et al. (2004): Waves 1–3 (1992/93, 1994/95, 1996/97) used to identify 393 DI applicants in a ±6-month window around interview dates; linked to SSA administrative records to test the Rational Unbiased Reporting hypothesis. The hlimpw variable ("Does a health problem keep you from working altogether?") is the key self-reported disability measure tested as an approximate sufficient statistic for SSA ultimate award decisions.
• Conti, Berndt, and Frank (2009): Birth cohort 1931–1941, waves 2–5 (1994–2000). First systematic use of the 8-item CES-D depression screener (cutoff ≥4) from HRS for DI/SSI application outcome modeling. Also uses HRS's 39-condition health battery. CES-D ≥4 captures ~79% of CIDI DSM-III major depression cases but overidentifies transient episodes, creating downward bias in estimated depression effects.

Related

• James Poterba
• Steven Venti
• David Wise
• Hugo Benítez-Silva
• Rena M. Conti
• Ernst R. Berndt
• Richard G. Frank
• Income-Mortality Gradient
• Rational Unbiased Reporting
• Social Security Administration

Sources

• Poterba Venti and Wise 2013a — Health Education and the Post-Retirement Evolution of Household Assets
• Benítez-Silva et al 2004 — How Large Is the Bias in Self-Reported Disability
• Conti Berndt and Frank 2009 — Early Retirement and DI-SSI Applications Exploring the Impact of Depression
• Armour and Knapp 2021 — The Consequences of Claiming Social Security Benefits at Age 62
• Banks Marmot Oldfield and Smith 2006 — Disease and Disadvantage in the United States and in England
• Benítez-Silva et al 1999 — An Empirical Analysis of the Social Security Disability Application, Appeal, and Award Process
• Bound Caswell and Waidmann 2013 — Insurance Value of Disability Insurance for Individuals with Mental Health Impairments
• Coe and Rutledge 2013 — How Does the Composition of Disability Insurance Applicants Change across Business Cycles
• Cutler Meara and Richards-Shubik 2012 — Unemployment and Disability Evidence from the Great Recession
• Cutler Meara Powell Richards-Shubik and Stewart 2015 — Why Do Disability Insurance Claims Increase During Recessions
• Deshpande and Dizon-Ross 2022 — The (Lack of) Anticipatory Effects of the Social Safety Net on Human Capital Investment
• Poterba Rauh Venti and Wise 2006 — Lifecycle Asset Allocation Strategies and the Distribution of 401(k) Retirement Wealth
• Poterba Venti and Wise 2010 — Family Status Transitions, Latent Health, and the Post-Retirement Evolution of Assets
• Poterba Venti and Wise 2017 — The Long Reach of Education Health, Wealth, and DI Participation
• Singleton 2024 — Health Information and the Timing of Social Security Benefit Entitlements