Two studies published in the most recent Health Affairs journal raise questions about the contention that the Affordable Care Act (ACA) will reduce employment, wages, and hours worked by employees.

The study by Gooptu and colleagues examined the effects of the law’s Medicaid expansion on employment and found no statistically significant effect through March 2015. A related study by Moriya and colleagues examined the subsidy structure provided to households getting health insurance through the ACA’s exchanges and similarly found no discernible effect on levels of part-time employment for employees eligible for these subsidies.

These studies provide useful new information, but they do not mean, as some reporting on them seems to suggest, that there is nothing to worry about with respect to the ACA’s effects on labor markets. Given the structure of the ACA, it would be hard to conclude the law would not eventually reduce hours worked or total compensation, although the magnitude of the resulting changes may be hard to detect in the U.S.’s large and complex labor markets.

Why CBO Concluded There Would Be An Effect On Employment

The Congressional Budget Office (CBO) issued a comprehensive assessment of the law’s likely effects on labor markets in February 2014. The agency concluded that the ACA’s “largest impacts will probably occur after 2016, once its major provisions have taken full effect and overall economic output nears its maximum sustainable level.” Taking all of the ACA’s provisions into account, CBO concluded that the law would reduce overall hours worked in the U.S. economy by 1.5 to 2.0 percent over the period 2017 to 2024 and total compensation by 1.0 percent over the same period.

There are many ACA provisions affecting employment and wages (including many new explicit taxes and employer requirements), but CBO reached its conclusion based on the following effects:

  • The ACA provides large subsidies through the exchanges to households with lower incomes to support enrollment in health insurance; those subsidies are provided independent of employment status and are reduced as incomes rise. The withdrawal of the subsidies as incomes rise represents an implicit tax on additional earned income, and therefore is likely to discourage some additional work effort in key income ranges.
  • The ACA allowed states to expand Medicaid, making it possible for more people in expansion states to secure health insurance coverage without working. Prior to enactment of the ACA, it was most often the case that a low-income individual ineligible for Medicaid would need to work to secure stable health insurance.At the same time, some people on Medicaid may work more under the ACA because the higher income eligibility threshold will allow them to stay on the program with additional wages. In addition, some people with incomes below the poverty line in non-expansion states may work more to become eligible for subsidized exchange coverage. Overall, CBO estimated that the Medicaid expansion would reduce hours worked, but less so than the implicit tax of the exchange subsidy phase-out.
  • Beginning in 2015, employers with more than 100 employees (falling to 50 this year) must pay penalties if they employ full-time workers and do not offer health insurance. Full-time work is defined as more than 29 hours per week, on average. This may encourage firms to move more workers into part-time status, although the effect of such an incentive on total hours of work is less clear.
  • Requiring insurers to offer coverage to people with pre-existing conditions and limiting premium variability provides an incentive for some older workers to retire early.
  • Providing subsidized health insurance irrespective of employment alleviates “job lock,” allowing workers to leave employment or switch to part-time jobs without fear of losing coverage. The prospect of being able to take a job without being concerned about insurance might encourage some workers to enter the workforce who would otherwise have kept their income low to qualify for Medicaid. The latter effect would not fully offset the others that would reduce hours worked.

Overall, CBO made it clear that its estimate of a reduction in total hours worked is based mainly on the law’s implicit tax on earned income for those eligible for exchange subsidies. The agency’s estimate of reduced compensation for workers is lower than the reduction of hours worked because it expects the contraction of labor supply to be concentrated among low-wage workers.

Analyzing The New Studies

The new Health Affairs studies do not find evidence that the ACA has caused a shift from full-time to part-time employment or major changes in the labor market behavior of low-income adults thus far. Those results would seem to contradict CBO’s forecast. But there are a number of reasons why this might not be so:

  • CBO assumed the effects of the law on hours worked would take some time to become discernible because labor markets were expected to remain soft through 2016, with large numbers of Americans out of the workforce compared to earlier periods. By 2017, CBO assumed labor markets would be back to normal functioning, at which point the disincentive effects for additional labor income would become more relevant. In a soft labor market, where employees’ wages are rising more slowly than normal and job opportunities are sparse, employees will find it more difficult to reach what they consider to be their optimal level of work effort and compensation.
  • The new studies examined the law’s effects in 2014 and 2015, but CBO made it clear it did not expect to see the full effects on hours worked until 2017, and then the effect would only be a reduction of 1.5 to 2 percent of total hours worked during the period 2017 to 2024. The labor market impact of the ACA in 2014 would be a fraction of that estimate. Can we reasonably expect to detect a change in hours worked that could be as low as 0.5 percent in 2014?
  • The number of workers who could be affected by the definition of full-time work is relatively small. Even among the low-income, most have jobs that are close to 40 hours per week. A Department of Health and Human Services (HHS) study shows that 63.3 percent of low-wage workers usually worked 35 hours or more per week in 2001, while only 23.5 percent worked 20 to 34 hours. Those working within a few hours of 30 per week (and thus more likely to be subject to a reduction in work hours) represent an even smaller percentage of the total. By not being able to focus on the part of the market most affected, the new Health Affairs studies understate the response.
  • The measure used by Moriya et al. to account for the impact of the ACA on hours of work differs from that used by CBO. The study estimates the change in the probability of part-time employment among employed people. CBO estimates the change in aggregate work hours. Workers who lose their jobs or leave the work force for any reason would not cause a reduction in the hours worked by those remaining employed but would reduce aggregate hours.
  • Other market adjustments may have blunted the initial impact of the ACA on hours of work and employment levels. For instance, some employers may have delayed taking actions adverse to their current workers in the hope that future improvements in the economy would allow them to better accommodate the requirements of the ACA, or those requirements might be modified. The Administration’s delay in imposing the employer mandate penalties confirms that a go-slow approach was prudent for many employers. And employers might have made some adjustments that would not show up in the Health Affairs analyses. For example, employers may have delayed or cancelled expected pay raises or halted hiring they may have considered even in a weak economy.

Fighting A Tendency On Both Sides To Jump The Gun

Proponents and opponents of the ACA are susceptible to making the same mistakes. In recent years, opponents have often predicted the law’s imminent demise based on one or two data points. Those predictions have proven to be premature. At the same time, the law’s defenders are sometimes too quick to dismiss legitimate criticisms of the law based on partial or incomplete information.

The ACA is complex, with scores of provisions that impose major changes on health insurance, labor markets, and the national economy. It is going to take some time for a clear and fully accurate picture of all of these effects to emerge.

Placed in the appropriate context, research studies such as the two Health Affairs papers provide useful information. At a minimum, they make it clear that the forecasts from CBO and others which show the ACA will eventually reduce compensation and hours worked are far from certain. But it is also too early to conclude that those forecasts were wrong. That remains to be seen.