Association for Public Policy Analysis & Management

One of the central goals of the most recent health care reform was to reduce the number of uninsured people and improve access to care, especially among those who lacked access to group insurance. The evidence has suggested that the expansions of insurance options were associated with the reductions in uninsurance among the target population, especially among low-income individuals; however, a key remaining question is whether and to what extent these expansions affected coverage, access, and utilization for workers, particularly those without access to group insurance.

In this study, we assess post-2014 changes in insurance coverage, access to health care, and utilization of health care separately for various categories of workers including the self-employed and wage earners with and without offers of employer-sponsored insurance (ESI) through either their own or their spouse’s workplace. We use individual-level data for the period 2010-2016 from the National Health Interview Survey (NHIS), which provides detailed information on socio-economic characteristics and our outcomes of interest for a representative sample of the U.S. noninstitutionalized civilian population. We restrict our sample to individuals ages 26-64, which results in a sample size of 231,424 non-elderly adults.

We estimate interrupted time series models that control for a linear time trend and demographic and socioeconomic characteristics, and compare the post-2014 change in each outcome for workers to non-workers. We begin by confirming that there were no changes in the composition of various categories of workers after the reform, which validates our empirical approach. Then we document that prior to 2014, workers without ESI offers were less likely to have insurance coverage, access to insurance, and utilization than non-workers. For example, the self-employed and wage earners without ESI offers (47.0 percent and 57.1 percent, respectively) were more likely to be uninsured than non-workers without ESI offers (30.4 percent). Also, the self-employed and wage earners without ESI offers (65.9 percent and 61.8 percent, respectively) were less likely to have at least one doctor visit within a year compared to non-workers without ESI offers (80.4 percent).

Our regression results reveal that each of these three employment groups experienced nearly a one-third decline in their rate of uninsurance. Further, our results suggest that coverage gains for non-workers without offers came primarily from Medicaid while workers without offers benefitted from Marketplace coverage. Finally, we find that improvements in access to care and the chance of having had a recent doctor visit for the self-employed and wage-earners without ESI offers were equal to, or greater than, improvements for non-workers.

To our knowledge, we are the first to describe the heterogeneous effects of the recent reforms on coverage, access and utilization for workers. We found that the insurance expansions disproportionately benefitted workers without access to group insurance that have historically experienced high rates of uninsurance and barriers to accessing health care. Our findings underscore the degree to which some groups of workers could be affected by policies that pertain to Marketplace exchanges.