Seeking to bat down the news in a Congressional Budget Office report about the Senate’s Better Care Reconciliation Act, President Donald Trump’s administration attempted to undermine the CBO’s credibility in a tweet.
The official White House account tweeted an image that said, "When Obamacare was signed into law, CBO estimated that 23 million people would be covered in Obamacare’s exchanges in 2017. They were off by more than 100 percent. Only 10.3 million people are covered by Obamacare."
Trump retweeted the message.
We wanted to start by addressing that CBO was "off by more than 100 percent," which wasn't very specific. The original prediction was 123 percent higher than actual marketplace enrollment.
Another way to put it: Marketplace enrollment was 55 percent lower than what CBO estimated. (This would be the most natural way to calculate the projection accuracy, said Matt Fiedler, a fellow with the Center for Health Policy at Brookings.)
In this fact-check, we wanted to look at the broader attack on CBO’s accuracy. Was the nonpartisan agency’s original look at who would gain insurance through the Affordable Care Act marketplaces off by 12.7 million people?
We found that the CBO report written after Obamacare was signed into law in March 2010 indeed estimated 23 million people would participate in Marketplace exchanges. According to the latest figures, only 10.3 million currently participate in those marketplaces.
The talking point neglects to mention, however, the CBO report’s accuracy in predicting the uninsured population at a historic low, which is what occurred. Given the information available at the time, the agency offered projections closer to reality than any other forecaster.
The Obamacare exchanges are online marketplaces for health insurance, designed to give those who are not covered by employers, Medicaid or Medicare access to coverage from competing private health care providers.
Some states set up their own marketplaces, while others defaulted to a federal-run Healthcare.gov shop.
The CBO’s estimate for exchange enrollment was too high for a number of reasons, experts told us.
The CBO originally predicted people would drop employer and non-group coverage to move into the exchanges, and that fewer people would enroll in Medicaid.
Instead, fewer employers than expected dropped coverage. And the Republican-led Congress blocked funding for the risk corridor program, leading some insurers to exit the marketplace and to increases in some premiums. More people than expected also enrolled in Medicaid.
Republican-led states also passed restrictions on some of the features designed to boost marketplace enrollment, including enrollment assistance and outreach efforts. The Trump administration continued in this vein, putting an end to all media outreach intended to boost last-minute signups prior to the Jan. 31, 2017 deadline. Already paid-for ads and emails to HealthCare.gov visitors were cut.
"It would have been hard for CBO to accurately anticipate how persistently hostile GOP state and federal politicians would be towards the ACA, which undoubtedly has helped suppress Marketplace enrollment," said Ben Sommers, an associate professor of health policy and economics at Harvard University.
The courts also threw a wrench in the CBO estimate about exchange enrollment.
When CBO researchers made their original prediction about the Affordable Care Act, they did not know that the Supreme Court would rule against the law’s mandatory Medicaid expansion in 2012.
The court determined it was up to the states to decide whether to expand coverage, and 19 states decided against the expansion.
This affected the CBO’s estimate for the uninsured population. (Instead of a reduction of 32 million uninsured by 2016, the count was closer to 20 million.) But it had ramifications for the exchanges, too.
"The Supreme Court decision slightly changed the exchange estimates and substantially changed the overall estimates," said Sherry Glied, the dean of New York University’s graduate school of public service.
One reason was that the growth of health care costs was lower than expected.
"Because of the way the subsidies were designed, lower costs meant that fewer people were eligible for subsidies (since subsidies depend on the ratio of premiums to incomes)," Glied said. "That, in turn, likely meant that fewer people switched from employer coverage to the exchanges."
The CBO made another adjustment to its exchange enrollment projection in January 2016, from 21 million predicted at that time to 13 million. The estimate followed the Obama administration’s transitional policy that allowed people to hang onto their pre-ACA health plans through 2017.
''Most of the unsubsidized people who are no longer expected to purchase insurance through an exchange are expected to purchase insurance directly from an insurer instead,'' said the adjusted 2016 report.
The Obama administration at the time said enrollment fell below projections because fewer employees went to the exchanges for insurance as more employers maintained coverage of employees.
The 13 million-enrollment estimate is closer to the most recent estimate from a June 2017 news release from the Centers for Medicare and Medicaid Services.
"The Effectuated Enrollment Report shows that 12.2 million individuals selected a plan at the end of Open Enrollment, but only 10.3 million followed through to pay the premiums necessary to maintain coverage as of March 15, 2017," according to the release, to which White House spokesman Steven Cheung referred in response to our query.
In addition to the tweet pointing at the CBO’s inaccurate projection on exchange enrollment, the White House criticized the office for a "history of inaccuracy" on "coverage, premiums, and predicted deficit arising out of Obamacare" in a statement.
While the CBO overestimated enrollment in the exchanges, it still offered projections closer to reality than any of the four other forecasters (the Office of the Actuary of the Centers for Medicare and Medicaid Services, the RAND Corporation, the Urban Institute; and the Lewin Group).
"The tweet is inaccurate in equating the exchanges with Obamacare, which implies that these numbers refer to the full coverage effects of the ACA. That ignores the role of Medicaid and employer changes, and CBO’s estimates on those were much closer," Sommers said.
It also correctly estimated a significant drop in the uninsured population, leaving it at a historic low.
"The CBO's estimates of the overall change, and of each element of it, were closer to the mark than all other researchers who released comprehensive estimates at the time of enactment," Glied said. "Nobody does it better."
The White House tweet was correct in saying that the CBO had originally estimated that 23 million people would participate in the Affordable Care Act’s exchanges, and that only 10.3 million people currently are participating in these.
The White House chose to cite a report that has been updated by the CBO following significant legislation affecting its predictions. To make its case that the CBO is unreliable, the Trump administration cherry-picked a statistic that CBO got the most wrong. The forecasters were closer to actual results on other major components of Obamacare, including Medicaid and employer changes.
We rate this statement Half True.
Editor's note: We've updated this fact-check to clarify a point about the difference between CBO's projections and actual exchange enrollment.