COVID‐19‐Related Unemployment and Health Insurance Coverage in Medicaid Expansion and Non‐Expansion States

RESEARCH OBJECTIVE: Job losses related to the COVID‐19 pandemic caused the United States' unemployment rate reached just under 15%—far eclipsing peak unemployment in the 2007–2009 Great Recession (10%). However, expanded access to non‐employer sponsored health coverage created by the Affordable Care Act's (ACA) major coverage expansions—Medicaid expansion, in particular—may insulate some households from longer spells of being uninsured while unemployed. In this study, we explore whether net health insurance losses were larger in states that did not expand Medicaid under the ACA. STUDY DESIGN: We apply cross‐sectional differences‐in‐differences regressions to compare differences in health insurance coverage source and status linked to recent lob losses attributable to the COVID‐19 pandemic in states the expanded Medicaid against states that did not expand Medicaid. POPULATION STUDIED: 702,755 Non‐elderly adults aged 19–64 in the Census Bureau‘s experimental Household Pulse Survey (HPS). Our results are based on analysis of the Phase I HPS which includes adults surveyed between April 23 and July 21, 2020. PRINCIPAL FINDINGS: 55% experienced a recent job loss—16% reported experiencing a job loss associated with the COVID‐19 pandemic (e.g. layoff, business closed, furlough). Adults reporting COVID‐related job losses were 21 percentage points (p < 0.01) less likely to have employer‐sponsored health insurance (ESI) at the time of their interview (relative to 78 percent among those not experiencing a job loss). People experiencing job losses were more likely to report having non‐group coverage plausibly from the exchanges, although there were no differences states that did not expand Medicaid and those expanding Medicaid as of March 2020. Pandemic‐linked associated enrollment in Medicaid was larger among residents of expansion states, however the association was marginally stronger among middle‐income (138–400% of the Federal Poverty Limit [FPL] based on their 2019 incomes used for tax filing) households (8.3 PP, p < 0.01) than among lower‐income households with incomes under 138% FPL (6.1 PP, p < 0.01). COVID‐related job losses were negligibly associated with uninsurance among low‐income adults in either expansion or non‐expansion states; however, middle‐income (138%–400% FPL) adults were almost 5 PP (p < 0.01) less likely to be uninsured at the time of the interview if their state had expanded Medicaid in place relative to their counterparts in non‐expansion states. CONCLUSIONS: Our exploration suggests that households negatively affected by the pandemic are utilizing Medicaid to insure themselves against potential health risks they would incur while being unemployed. Although our results come from a single cross‐section, we highlight that the strongest association among COVID‐related job losses, Medicaid enrollment, and potential protections from long uninsured spells was concentrated among people that would be ineligible for Medicaid in 2019—just prior to the pandemic. IMPLICATIONS FOR POLICY OR PRACTICE: This study investigates an underexplored component of the Medicaid program is its ability to act as a safety net program for households affected by job loss. At present, 39 states expanded Medicaid under the ACA. Our findings suggest states with expanded Medicaid eligibility may allow households to partly stabilize health care needs should they become detached from private health coverage due to a job loss during the pandemic. PRIMARY FUNDING SOURCE: The Robert Wood Johnson Foundation.