Health Insurance and Early Retirement: Evidence from State Exchanges and Medicaid Expansion

Policies implemented under the Affordable Care Act (ACA) to increase the number of insured persons may affect labor supply through “job lock” or “retirement lock” mechanisms. These include the expansion of state Medicaid programs and federal subsidies on the exchange marketplace. In this paper, I study whether these policies encourage early retirement, examining persons aged 62-64 who are eligible for Social Security benefits. I use a quasi-experimental differences-in-differences framework to compare individuals eligible for subsidies/Medicaid to those not eligible for either within a given state.

Prior work on the topic focuses largely on Medicaid expansions, ignoring the exchange subsidies offered to individuals in non-expansion states. These subsidies may also encourage early retirement, even at moderately higher incomes where employer-sponsored insurance is more common and there is greater potential for “job-lock”. Similarly, Supplemental Security Income and portions of Social Security are not counted towards income for Medicaid eligibility, so a larger portion of individuals in expansion states also face incentives to retire early.

The empirical strategy is a differences-in-differences model with state fixed effects. I bin individuals into various thresholds of the poverty line (e.g., 400-301%, 300-201%, 200-139%, 138-100%…) to compare those eligible for Medicaid expansions or large exchange subsidies to those at higher percentages of the poverty line (401-500%) neither eligible for Medicaid nor subsidies. A critical assumption is that insurance coverage and labor supply trends are parallel across these groups in the pre-ACA period (a testable assumption) and would continue to be in the absence of exchange subsidies or Medicaid. While the later is not testable, I take care to bound my estimates under different scenarios where this might fail. I estimate the model separately for Medicaid expansion states and states that only offered exchange subsidies to compare effects of the policies across poverty line thresholds.

Data comes from the American Community Survey (2010-2016) which has information on employment, benefit uptake, and health insurance for over 200,000 individuals aged 62-64. (I also use the CPS to confirm my labor market results.) I carefully consider the channels through which one retires: individuals may leave their employer and employer-sponsored insurance, but unemployed persons may also stop their job search to retire as well. Individuals may also choose to transition to self-employment or part-time work as a “bridge” to retirement. Finally, by focusing on this limited age group, I can use take up of Social Security benefits reported in the surveys to better measure retirement.

Preliminary results show both expansion and non-expansion states increased insurance coverage among low income individuals closer to the poverty line. In expansion states, this is driven by Medicaid enrollment; among non-expansion states, this is driven by insurance purchased through exchanges with large subsidies. The same populations that experienced increases in insurance coverage are also more likely to drop out of the labor force and take up Social Security benefits. Roughly half of of this effect is driven by unemployed individuals ending their job search. I find no changes in self-employment or part-time work.