The Washington Post

Congressional Budget Office data taken out of context as ‘job’ losses tied to health-care law

“How many workers will have to lose their jobs?”
— voiceover of attack ad against Sen. Kay Hagan (D-N.C.) by the campaign of Senate hopeful Thom Tillis (R), North Carolina speaker of the House

“Hagan’s Support of ObamaCare Will Reduce Employment in North Carolina: Equivalent of 74,469 Jobs Could Be Lost”
— National Republican Senatorial Committee news release, citing a study by Americans for Tax Reform

The Congressional Budget Office report on the impact of the Affordable Care Act was released on Tuesday, and already we have the first attack ads.

The Fact Checker devoted one column to explaining what the CBO’s report actually meant. But we have long learned that all the fact checks in the world won’t stop politicians if they think an attack line moves voters. So how do these first attacks fare?

The Facts

In its report, the CBO said the Affordable Care Act, a.k.a. Obamacare, would reduce the number of hours worked by the equivalent of 2.5 million full-time workers by 2025. That means that workers will decide to reduce their hours, not that employers are reducing the number of jobs.

A fierce dispute has erupted between liberal and conservative economists about what this figure means, and depending on your political perspective, both sides make compelling cases. (Conservatives decry the smaller economy that will result; liberals celebrate that workers will not be bound to their jobs because of health insurance.) But that’s not The Fact Checker’s business. Our concern is whether the CBO’s report is being described correctly by politicians.

The Thom Tillis ad flashes this language when the voiceover asks, “How many workers will have to lose their jobs?” because of Hagan’s support for the health-care law: “Congressional Budget Office estimates 2 million lost jobs due to Obamacare.”

But that’s not what the report says. Don’t take our word for it. Here’s what CBO Director Douglas W. Elmendorf said in congressional testimony when he was asked about claims that jobs are being lost because of the law:

“The reason that we don’t use the term ‘lost jobs’ is there’s a critical difference between people who like to work and can’t find a job or have a job that was lost for reasons beyond their control and people who choose not to work,” he said, giving the example of someone who was laid off versus someone who decided to spend more time with their family or retire early.

The Tillis ad gets in trouble by using language that asserts that the CBO says there will be “2 million lost jobs” because of the law. First of all, it’s not jobs but workers. Second, it is lacking context, because that is off a base of more than 160 million workers (i.e., less than 2 percent.)

Now, all things being equal, a reduction of 2 million full-time workers eventually would mean 2 million fewer jobs. But the CBO estimate mixed full-time and part-time workers, and then the total number of hours expected to be lost was converted to a full-time equivalent. Some workers might reduce their weekly hours by only an hour or two, so the impact on overall jobs would be muted.

“Whether faulty government policies drive people to quit work, or if a person is fired, or if a person is not hired in a job they’d otherwise get because companies can’t expand, that is a job that is lost,” said Jordan Shaw, campaign manager for Tillis.

The National Republican Senatorial Committee news release is also an interesting example. The headline actually tries hard to be restrained and relatively accurate. It says that the law will “reduce employment” and then carefully notes this is the “equivalent” of a number of jobs. The release further says “fewer people will be employed and fewer hours worked because of the unpopular law.” That’s phrased correctly.

Brad Dayspring, a spokesman for the committee, said the release purposely avoided saying the law would “cost” jobs. “We did that with cognizance because we want to be factual while pointing out the potential harm of what we believe to be a terrible policy,” he said.

The problem is the number of jobs, 74,469, which is derived from an analysis by the Americans for Tax Reform titled “How Many Jobs Might Obamacare Cost Your State?” The anti-tax group did a simple calculation based on how many jobs each state currently has.

But that is inherently misleading. Not every state expanded Medicaid, so the effects are different. Indeed, North Carolina is one of those states that did not expand Medicaid, so right away, that means the impact on worker participation will be lower than in other states. Moreover, every state has a different distribution of income, so there can be vast differences in the percentage of people who will be affected by the subsidies for low-income workers and the law’s taxes on the wealthy. That could also significantly change the result, state by state.

Finally, the emphasis on “jobs” once again takes the claim further from the CBO’s initial emphasis on workers. As we noted, because the estimate of hours is a mix of full-time and part-time work, there is no 1-1 relationship between the reduction in full-time equivalent workers and the decrease in jobs that will exist in the economy.

“The table does not make any firm claims that a state will lose x or y number of jobs,” said Ryan Ellis, the tax reform group’s tax policy director, describing the study as a “starting point for research.”

He defended the “cost jobs” language, saying: “We would say that if a person chooses not to work because Obamacare has made their working a prohibitive financial choice for them, then that’s a job killed by Obamacare.”

The Pinocchio Test

We are going to excuse the National Republican Senatorial Committee from the Pinocchio Test, because it is clear they tried hard, within the confines of a news release, to accurately describe the CBO report. The poison root was the tax reform group’s analysis, with its misleading headline.

Both the Tillis campaign and that organization fall into the trap of saying a specific number of jobs will be lost. This is not what CBO said, as made clear by Elmendorf’s testimony. For non-economists, the use of the phrase “jobs” is especially confusing because it sounds like a decision made by employers rather than workers.

There is certainly damaging material in the CBO report, ripe for plucking. Keith Hennessey, a former Bush administration official, has offered some variations of possible attack lines, such as “Obamacare will shrink our economy by driving millions of moderate income people to work less, and discouraging some of them from working at all.” That’s a negative interpretation — liberals would emphasize the benefits—but it also is a statement that accurately reflects what the CBO said.

But in the meantime, “costs jobs” attacks are going to keep earning Three Pinocchios.

To read previous Fact Checker columns, go to

Glenn Kessler has reported on domestic and foreign policy for more than three decades. He would like your help in keeping an eye on public figures. Send him statements to fact check by emailing him, tweeting at him, or sending him a message on Facebook.

The Freddie Gray case

Please provide a valid email address.

You’re all set!

Campaign 2016 Email Updates

Please provide a valid email address.

You’re all set!

Get Zika news by email

Please provide a valid email address.

You’re all set!
Show Comments
New Hampshire has voted. The Democrats debate Thursday. Get caught up on the race.
The big questions after New Hampshire, from The Post's Dan Balz
Can Bernie Sanders cut into Hillary Clinton's strength in the minority community and turn his challenge into a genuine threat? And can any of the Republicans consolidate anti-Trump sentiment in the party in time to stop the billionaire developer and reality-TV star, whose unorthodox, nationalistic campaign has shaken the foundations of American politics?
Clinton in New Hampshire: 2008 vs. 2015
Hillary Clinton did about as well in N.H. this year as she did in 2008, percentage-wise. In the state's main counties, Clinton performed on average only about two percentage points worse than she did eight years ago (according to vote totals as of Wednesday morning) -- and in five of the 10 counties, she did as well or better.
What happened in New Hampshire
Upcoming debates
Feb. 11: Democratic debate

on PBS, in Wisconsin

Feb 13: GOP debate

on CBS News, in South Carolina

Feb. 25: GOP debate

on CNN, in Houston, Texas

Campaign 2016
What happened in N.H.

To keep reading, please enter your email address.

You’ll also receive from The Washington Post:
  • A free 6-week digital subscription
  • Our daily newsletter in your inbox

Please enter a valid email address

I have read and agree to the Terms of Service and Privacy Policy.

Please indicate agreement.

Thank you.

Check your inbox. We’ve sent an email explaining how to set up an account and activate your free digital subscription.