Republicans pushing to repeal President Barack Obama’s health care overhaul warn that 650,000 jobs will be lost if the law is allowed to stand.
But the widely-cited estimate by House GOP leaders is shaky. It’s the latest creative use of statistics in the health care debate, which has seen plenty of examples from both sides.
A recent report by House GOP leaders says “independent analyses have determined that the health care law will cause significant job losses for the U.S. economy.”
It cites the 650,000 lost jobs as Exhibit A, and the nonpartisan Congressional Budget Office (CBO) as the source of the original analysis behind that estimate. But the budget office, which referees the costs and consequences of legislation, never produced the number.
What follows is a story of how statistics get used and abused in Washington.
What CBO actually said is that the impact of the health care law on supply and demand for labor would be small. Most of it would come from people who no longer have to work, or can downshift to less demanding employment, because insurance will be available outside the job.
“The legislation, on net, will reduce the amount of labor used in the economy by a small amount –roughly half a percent – primarily by reducing the amount of labor that workers choose to supply,” budget office number crunchers said in a report from last year.
That’s not how it got translated in the new report from Speaker John Boehner, R-Ohio, and other top Republicans.
CBO “has determined that the law will reduce the ‘amount of labor used in the economy by roughly half a percent,’ an estimate that adds up to roughly 650,000 jobs lost,” the GOP version said.
Gone was the caveat that the impact would be small, mainly due to people working less. Added was the estimate of 650,000 jobs lost.
The Republican translation doesn’t track, said economist Paul Fronstin of the nonpartisan Employee Benefit Research Institute. “People voluntarily working less isn’t the same as employers cutting jobs,” he explained.
For example, CBO said some people might decide to retire earlier because it would be easier to get health care, instead of waiting until they become eligible for Medicare at age 65.
The law “reduces the amount of labor supplied, but it’s not reducing the ability of people to find jobs, which is what the job-killing slogan is intended to convey,” said economist Paul Van de Water of the Center on Budget and Policy Priorities. The center advocates for low-income people, and supports the health care law.
In theory, any legislation that increases costs for employers can lead to job loss. But with the health care law, companies can also decide to pass on added costs to their workers, as some have already done this year.
To put things in perspective, there are currently about 131 million jobs in the economy. CBO projects that unemployment will be significantly lower in 2014, when the law’s major coverage expansion starts.
A spokeswoman for House Ways and Means Committee Republicans pointed out that CBO’s report did flag that some employers would cut hiring. “The CBO analysis does not claim that the entire response is people exiting the labor market,” said Michelle Dimarob.
The law’s penalties on employers who don’t provide health insurance might cause some companies to hire fewer low-wage workers, or to hire more part-timers instead of full-time employees, the budget office said. But the main consequence would still be from more people choosing not to work.
That still doesn’t answer the question of how Republicans came up with the estimate of 650,000 lost jobs.
Dimarob said staffers took the 131 million jobs and multiplied that by half a percent, the number from the CBO analysis. The result: 650,000 jobs feared to be in jeopardy.
“For ordinary Americans who could fall into that half a percent, that is a vitally important stat, and it is reasonable to suggest they would not characterize the effect as small,” she said.
But Fronstin said that approach is also questionable, since the budget office and the GOP staffers used different yardsticks to measure overall jobs and hours worked. The differences would have to be adjusted first in order to produce an accurate estimate.
Said Van de Water, “The number doesn’t mean what they say it means.”