Critics had predicted that, under Obamacare, employers would shift workers to part-time schedules to avoid a requirement that they provide insurance to full-time employees. Critics also predicted people might opt to work less because they could obtain insurance through the marketplaces or expanded Medicaid coverage.
Last month, the Congressional Budget Office renewed its warning that Obamacare would discourage the equivalent of 2 million Americans from working by 2025, since they can get health insurance without relying on an employer.
While the new studies don’t rule out a dim long-term forecast, they do offer reason for optimism. The predictions that the ACA would have broad economic consequences weren’t borne out in researchers’ data: Part-time jobs didn’t significantly increase, and states that expanded Medicaid did not have significant shifts in employment patterns.
"There were a lot of stories about employers, with anecdotes about employers shifting jobs to part-time," said Larry Levitt, an economist at the Kaiser Family Foundation not involved in the new research. "You can't deny those stories — they’re real. They’re just not generalizable. It’s not what is mostly happening, based on this study and others."
That isn't to say that individual people and employers haven't made changes because of the law. White Castle, a company that had warned it would reduce some employees' hours a few years ago, has changed its scheduling practices so more employees work fewer than 30 hours a week, the threshold at which large companies don't have to provide health coverage. Vice President Jamie Richardson said the company had 1,400 employees working 30 to 35 hours before the law and very few do today.
"In practice, if you're part time, you're going to be scheduled to work less than 30 hours per week," he said.
At other firms, though, owners ultimately decided not to restrict hiring.
In 2013, Jody Manor, who owns Bittersweet Catering, Cafe and Bakery in Alexandria, Va., put his expansion plans on hold, fearing he would have to pay for pricier comprehensive health coverage if he expanded his small business to more than 50 employees.
Instead, Manor found out he could get a better deal from the bakery’s health insurer with a larger payroll. He opened a new cafe later that year, and employees’ premiums declined by 3 percent, the only decrease he can recall since Bittersweet opened in 1983.
The results of the new research suggest that Manor’s approach was the more common response to new health insurance requirements.
One study, published Tuesday in the journal Health Affairs, focuses on a widely criticized provision of the law that requires employers with at least 100 full-time workers to offer health insurance as of the beginning of 2015. (This year, that requirement is being extended to firms with more than 50 employees.) Critics worried that the provision would deter small businesses from expanding and would encourage them to cut their workers' hours. But this study is the latest in a series to conclude that Obamacare did not, in fact, widely result in more firms asking employees to work part time.
The authors examined Census data and found no increase in the likelihood of working part time, except for a 0.18 percentage point increase in the likelihood of working 25 to 29 hours per week between 2013 and 2014 — a trend that the authors say predated the ACA.
"These results are surprising to me, given that there was a concern that employers would make adjustments in employee work hours," said Asako Moriya, a service economist with the governmental Agency for Healthcare Research and Quality who led the work.
The study included another piece of evidence that appears to contradict the notion that the law would cause an increase in part-time work: The number of people working 25 to 29 hours a week in firms not subject to the mandate increased between 2012 and 2015, while the number of people at firms subject to the mandate slightly decreased.
Two groups of people were slightly more likely to work part time: workers with little education and older workers between 60 and 64 years old. There was suggestive, but not conclusive, evidence that the shifts were voluntary and not caused by employers cutting hours.
One reason that some workers might be voluntarily working fewer hours is to reduce their earnings in order to qualify for subsidized insurance from the federal government — either through the insurance marketplaces or through Medicaid.
The new research on the effects of the Medicaid expansion does not suggest that people are looking to limit their earnings. On the contrary, some might even be looking to work more.
"In general, there's been no evidence that the ACA has done anything significant to employment," said Robert Kaestner, an economist at the University of Illinois at Chicago and one of the authors of the working paper, using the abbreviation for the Affordable Care Act. "We don't have to worry too much."
Previously, only people living in poverty were eligible for Medicaid, but the law expanded that group to people with incomes up to 138 percent of the federal poverty level. The Supreme Court, though, decided that the federal government couldn't require states to expand Medicaid.
That decision created an opportunity for researchers, such as Moriya — who is also one of the authors of the second article in Health Affairs — and Kaestner and his colleagues at the University of Illinois.
Following the decision, some states chose to expand Medicaid to a less impoverished group of people, while others did not, allowing the researchers to compare the change over time in the numbers of people working and their schedules in these two groups of states.
Obamacare's detractors had said that with access to Medicaid, some people earning between the poverty level and 138 percent of that level would quit working. They might have been working in jobs that did not pay well but offered health insurance sponsored by the employer. With insurance provided by Uncle Sam, people in that category might have decided going to work wasn't worth it.
At the same time, some people in that group who did not subscribe to insurance through their employers might have turned down additional hours or promotions in order to keep their incomes below 138 percent of the poverty level and maintain their newly acquired eligibility for Medicaid.
Moriya's group found that if some who gained Medicaid worked less as a result, their numbers were too small to measure reliably. Although they found that people were about 0.6 percentage points more likely to leave a job in the states that expanded Medicaid than in the states that maintained the program as it was, the difference was slight enough that it could have been due to chance.
In fact, according to the calculations by Kaestner's group, the labor market actually looked a little bit stronger in the states that expanded Medicaid.
That might be because some people whose incomes were below the poverty level before the expansion might have been working less to maintain their eligibility, and took on a second job or additional hours when the limit on income in their state increased to 138 percent of the poverty level.
"People can work more and earn more," Kaestner said. "It frees people to actually be able to work without getting penalized."
Moriya and her colleagues examined only the number of people leaving a job or working fewer hours, not the number taking on a new job or working more. As a result, their analysis might not have accounted for this positive aspect of the expansion, yielding more pessimistic numbers.
Yet neither group of researchers produced statistically significant results, suggesting that any effect on the labor market was likely very small.
There will be real economic effects of easier access to health insurance — people who retire earlier or stay home to take care of a child because they can get health insurance on their own. But as the data suggest, these changes are happening on the fringes.
So far, trying to detect the effect of the ACA on the job market is like "finding a needle in a haystack," Levitt said.
More from Wonkblog: