Claims That The ACA Would Be A Job Killer Are Not Substantiated By Research

Opponents of the Affordable Care Act (ACA) have been claiming that the law would be a job killer since its passage, and the claims have not subsided with time. Some supporters of the ACA also worry about the possibility of unintended employment consequences. In addition, the Congressional Budget Office (CBO) projected that the ACA would have negative employment effects.

However, predictions that the ACA would substantially reduce employment, and shift millions of workers from full-time to part-time work, have not been borne out. Instead we have a growing body of empirical evidence, including two new studies in the most recent issue of Health Affairs and analyses we have released, that the ACA has had little, if any, adverse effect on employment.

First, a recent update of an earlier analysis using Current Population Survey (CPS) data up to August 2015 (unpublished author tabulations) shows that the overall employment-to-population ratio is higher in 2015 than what would be expected based on pre-2014 (pre-ACA) patterns (see Exhibit 1).

The forecast model used for this analysis is very similar to what we described in an earlier paper. Contrary to predictions that the ACA would reduce employment, we find that the employment-to-population ratio is 0.9 percentage points (p<0.1) higher than expected in 2015 after accounting for trends and demographic factors. The actual employment-to-population ratio exceeds its expected value by a larger amount (1.9 percentage points, p<0.05) among adults with a high school education or less who are likely to have been more affected by the ACA’s health insurance coverage expansions.

Exhibit 1

The Congressional Budget Office (CBO) predicted that the ACA would reduce employment by about 2 million workers in 2025, although it noted substantial uncertainty around this estimate. The CBO’s assumptions about labor market effects are an important part of its dynamic scoring of the ACA’s budgetary cost. If the long-run effect predicted by the CBO is going to happen, it does not appear to have started yet.

Medicaid Expansion

Second, a series of recent papers examined whether the ACA’s expansion of Medicaid affected employment and found little evidence of an adverse effect. We conducted several analyses of this research question using data from the CPS through November 2015 and data from the American Community Survey through 2014. We focused on adults with a high school education or less and found that the ACA Medicaid expansions were associated with significant increases in Medicaid coverage, significant decreases in the proportion uninsured, and if anything, an increase in employment although the employment effects were small and not statistically significant.

A recent Health Affairs paper took a different approach to evaluate the issue and examined whether adults with family income below 138 percent of poverty, which would make them eligible for Medicaid under the ACA, were more likely to stop working in states that expanded Medicaid compared to states that did not expand. Results from this study showed that, if anything, the likelihood of exiting the labor force among this group was higher in states that did not expand Medicaid as compared to states that did expand Medicaid.

Part-Time Employment

Third, the question of whether the ACA overall, and its employer mandate in particular, has increased part-time work has been studied in several papers. Here the evidence is less clear cut as to whether there has been any effect due to the ACA, but generally suggests the effects so far are small.

In a study using CPS data through December 2014, we found no statistically significant effect for adults overall, but found that part-time employment (working less than 30 hours per week) was higher than expected by 0.5 percentage points in 2014 for adults with a high school education or less. This difference was sufficiently small, however, that we found no evidence of a change in the number of hours worked in 2014 and thus no overall change in labor supply for this group. When we updated the analysis using data through August 2015 (unpublished tabulations by the authors), we found the gap between actual and expected part-time work did not widen from 2014 to 2015, but shrank, as the employer mandate actually went into effect for employers with 100 or more workers in January 2015.

Another recent study in Health Affairs using CPS data through June 2015 also found little effect of the ACA on part-time work. Interestingly, this latest study shows the share of involuntary part-time workers just below the 30-hour threshold declined from 2014 to 2015, contrary to what would be expected if the employer mandate is having much effect.

A third paper, using CPS data up to 2014, compares trends in part-time work across industries and occupations expected to be more or less affected by the employer mandate using a difference-in-difference approach. Using results from this analysis, the authors conclude that around 1 million workers shifted to involuntary part-time status as a result of the ACA.

This finding is an outlier relative to the other studies. It is also a study that hinges on more (untested) identification assumptions than the others, specifically, that the study’s classification of affected and unaffected workers is valid and that employment in industries with more or less affected workers would be equally influenced by the Great Recession.

Given the somewhat mixed evidence on whether the ACA increased part-time employment, the decline in the part-time-work gap in 2015, ongoing ACA implementation, and the difficulties of sorting out causality in employment patterns, it’s clear that further work is needed on this topic.

Next Steps

In sum, even well into the second year of implementation of the ACA’s major provisions, setting aside one outlier study, no significant shifts in employment patterns due to the ACA have yet to materialize. Maybe there will be clear evidence of adverse effects of the ACA on employment that show up in the next year or so. And maybe some of the anecdotes about the ACA leading to reduced work hours for some workers are true even if they only apply to a small group of workers. For now, the existing body of evidence based on actual data has three messages:

• The ACA has increased health insurance coverage for many millions of Americans with no reduction in overall employment,
• The ACA’s Medicaid expansions have had no discernable adverse effects on employment patterns, and
• More study is needed on the effects of the ACA on part-time work, but the current weight of the evidence leans towards small effects so far.

Beyond further study as ACA implementation evolves, we should think about how policies can be designed to minimize any unintended consequences (other formulations of the employer mandate penalties based, for example, on payroll would minimize incentives to shift worker hours and reduce any burden on low-wage firms).

Authors’ Note

The authors are grateful to the Robert Wood Johnson Foundation for research support.

from Health Affairs Blog http://ift.tt/1X3Js6o