Absence of conflict of interest.
Citation
Highlights
-
The study’s objective was to examine the impact of the Patient Protection and Affordable Care Act’s (ACA) Medicaid expansion on transitioning from employment to unemployment, from full-time to part-time employment, from unemployment to employment, and on switching jobs. The authors also investigated the ACA and employer-sponsored insurance, the profile of which can be found here
-
The study used a difference-in-differences design to compare employment outcomes pre- and post-ACA for individuals in states that expanded Medicaid versus those in states that did not expand Medicaid using data from the Current Population Survey (2005 – 2018).
-
The study suggested that there was no association between living in a state that expanded Medicaid access and the labor market outcomes the study examined.
-
The quality of causal evidence presented in this report is low because the author did not ensure that the individuals being compared were similar on important characteristics before the intervention. This means we could not be confident that any estimated effects were attributable to Medicaid expansion and not to other factors. However, the study did not find statistically significant effects.
Intervention Examined
Medicaid Expansion under the Affordable Care Act
Features of the Intervention
Prior to the ACA, only select groups of adults were eligible for health insurance coverage under Medicaid. The ACA expanded Medicaid coverage for nearly all working-age adults with household incomes below 138% of the federal poverty level, but, in 2012, the Supreme Court ruled that states could opt out of expanding Medicaid coverage. Since implementation of the Medicaid expansion began in 2014, more than half of U.S. states have expanded Medicaid coverage for low-income, working-age adults.
Features of the Study
The study uses a difference-in-differences design to compare labor market outcomes before and after the ACA for low-income, working-age adults in states that expanded Medicaid (the treatment group) versus states that did not expand Medicaid (the comparison group). Low-income adults are defined as those with family income in the baseline year below 138 percent of the federal poverty level, while working-age is defined as individuals 19-64 years old.
The study uses data from the Current Population Survey (CPS) to measure labor market outcomes for survey respondents at monthly intervals from January 2005 through March 2018. The CPS uses a cross-sectional design in which new respondents are added monthly and each respondent remains in the sample for 16 months.
The study examines the impact of Medicaid expansion on four outcomes: 1) whether a respondent employed at baseline transitioned to unemployment across a 16-month period; 2) whether a respondent switched jobs during a 16-month period; 3) whether a respondent transitioned from full- to part-time employment across a 16-month period; and 4) whether a respondent who was unemployed at baseline gained employment across a 16-month period. The analytic sample size varies by outcome, from 106,199 individuals to 223,000.
Findings
Employment
- The study suggested that the ACA’s Medicaid expansion was not associated with any of the labor market outcomes the study examined.
Considerations for Interpreting the Findings
The study accounted for some preexisting differences between the groups, using statistical models that controlled for individual characteristics like age, gender, and race/ethnicity as well as some state-level characteristics like the state’s unemployment rate. However, the study did not control for individuals’ prior history of job switching, unemployment, or part-time work or demonstrate that the groups were comparable at the individual level on pre-intervention measures of the outcomes. The author does show that pre-intervention trends across the two groups (expansion versus non-expansion states) were comparable for three of the four outcomes but this is not sufficient given that analyses are at the individual level and the individuals included in the analysis change from pre- to post-intervention. Therefore, preexisting differences between the groups—and not Medicaid expansion—could explain any observed differences in outcomes.
Causal Evidence Rating
The quality of causal evidence presented in this study is low because the author did not ensure that the individuals being compared were similar on important characteristics before the intervention. This means we could not be confident that any estimated effects were attributable to Medicaid expansion and not to other factors. However, the study did not find statistically significant effects.