Association for Public Policy Analysis & Management

Work status, wages, and early Social Security claiming are important determinants of financial security during retirement. In the last two decades, many states have expanded their Medicaid programs to reach adults at higher income levels and eliminated categorical eligibility requirements such as disability or responsibility for a dependent child. The availability of publicly subsidized health insurance for non-elderly adults may affect retirement-related decisions among older Americans, particularly those in poor health who may value the health benefits highly.

In this paper, we examine the behavioral effects of increased Medicaid availability in older adulthood on retirement decisions, including employment status, retirement income, and Social Security claiming, which are important determinants of financial security in retirement. We use the natural experiments created by state decisions to expand Medicaid to estimate the effect of expanded public health insurance coverage on retirement decisions in the nationally representative Current Population Survey Annual Social and Economic Supplement (CPS ASEC) and a restricted version of the Medical Expenditure Panel Survey (MEPS). We combine the CPS and MEPS with detailed data we collected on the numerous state expansions of public health insurance that took place from 1996 – 2013. We are able to assess a broad set of retirement-related outcomes including overall labor supply, the decision to retire, and Social Security income. Finally, because the MEPS includes multiple measures of health and function, we are able to examine whether these effects differ across the distribution of health.

Two streams of research support the hypothesis that these Medicaid expansions may influence retirement decisions, Social Security claiming and retirement income. First, recent studies indicate that Medicaid expansions to non-elderly adults without dependent children may cause larger labor supply responses among those close to retirement age (Dague et al., 2017; Garthwaite et al. 2014). Second, a large empirical literature on health insurance and retirement generally finds that the availability of health insurance, independent of employment (e.g., Medicare or retiree health insurance), hastens retirement. Medicaid, like retiree insurance, can bridge the time between leaving work and eligibility for Medicare. This relationship between health insurance and retirement may be particularly salient for Medicaid-eligible adults who have low-incomes and report poor health relative to their privately insured peers.

The long-run financial health of publicly funded retirement benefits like Social Security depends critically on decisions about when and how people retire, and evidence from this study can help inform such crucial policy decisions.