The Washington PostDemocracy Dies in Darkness

The Trump administration says CBO can’t be trusted because its Obamacare predictions were wrong. Are they right?

Here are the Congressional Budget Office's key estimates for how the Senate health-care plan would impact Americans' health insurance coverage and costs. (Video: Jenny Starrs, Daron Taylor/The Washington Post)
Placeholder while article actions load

The Congressional Budget Office on Monday projected Senate Republicans' bill to repeal the Affordable Care Act would leave an estimated 22 million additional Americans without health insurance in the coming decade, as well as cut federal spending by $321 billion by 2026.

That insurance prediction apparently didn't sit well with the White House, as the Trump administration put out a statement questioning the CBO's credibility, arguing that its analysis of the Affordable Care Act -- Democrats' 2010 health-care law also known as Obamacare -- was so flawed it proved the agency can't be trusted.

"The CBO has consistently proven it cannot accurately predict how healthcare legislation will impact insurance coverage," the White House said in a statement. "This history of inaccuracy, as demonstrated by its flawed report on coverage, premiums, and predicted deficit arising out of Obamacare, reminds us that its analysis must not be trusted blindly."

It has become a common talking point -- the White House and House Republicans ripped the office after it said the House Obamacare repeal bill would result in a similar swelling of the ranks of the uninsured -- but is it an accurate one?

Republicans are correct that the CBO's forecasts for Obamacare did prove inaccurate in some important respects, as well as that, on several occasions, the agency revised its initial projections. But outside the spin room, health-care experts on both sides of the aisle say that the CBO is reliably nonpartisan, and that although perfectly forecasting the consequences of complicated legislation is essentially impossible, information from the CBO is important for lawmakers to consider.

“Tensions are very high, and that there’s criticism of the CBO is not surprising,” said Alice Rivlin, a former director of the agency.

And a review of the CBO's Obamacare prediction does not suggest a partisan bias or an unreasonably optimistic or pessimistic view of the measure: The CBO's forecasts overstated both the costs and the benefits of the Democratic changes, as the measure appears likely to have a lower price tag than CBO anticipated but has also insured fewer people than proponents hoped, based on the agency's assessment.

Last year, only 21 million Americans were supposed to be uninsured, according to CBO's initial projections. The actual figure appears to be more like 27 million.

That figure is partially a result of the fact that the CBO could not have anticipated that the Supreme Court would make the expansion of Medicaid optional for states under Obamacare. Many Republican officials chose not to expand the program in their states, resulting in fewer people receiving coverage than initially envisioned.

Overall, though, the CBO's forecasts for Medicaid were approximately accurate, because greater-than-expected enrollment in other states made up for the shortfall in states that did not expand the program. Rather, unexpectedly low figures for the exchanges established by the law account for much of the discrepancy, as The Washington Post reported in March.

Read more: A detailed review of the CBO's accuracy on Obamacare

Meanwhile, although conservative opponents of Obamacare have complained about premiums under the law, premiums have been less than the CBO originally forecast.

Since the government helps pay consumers' premiums under Obamacare, the combination of premiums and enrollment being lower than anticipated has reduced the overall cost of the bill.

“There were fewer people who signed up, and the bill cost less than they had anticipated, but that’s not surprising,” Rivlin said. “These things are extremely difficult to project.”

Criticism of the CBO is routine. When Rivlin was in charge during President Ronald Reagan's administration, White House officials tried unsuccessfully to oust her after the agency produced embarrassing reports about the president's proposals.

“Their figures are phony,” Reagan said at the time.

President Trump's budget director, Mick Mulvaney, said last month that he doubted that CBO staff members were truly nonpartisan, suggesting they might rig the numbers to favor Democrats.

“At some point, you've got to ask yourself, has the day of the CBO come and gone?” Mulvaney said in an interview with the Washington Examiner last month.

His comments drew criticism from experts on both sides. Mulvaney's remarks were “really inappropriate, and frankly absurd,” said Michael Strain, an economist at the right-leaning American Enterprise Institute.

“I mean, come on,” Strain said. “The work is done with, you know, the utmost integrity. It is nonpartisan, and the analyses are sound and reasonable.”

“I have a deep faith in its commitment to nonpartisanship and the integrity of its efforts,” said Douglas Holtz-Eakin, a former CBO director and a conservative economist who supports the GOP bill.

CBO: Senate health bill would leave 22 million more people uninsured by 2026

If accurately forecasting the effects of Obamacare was a challenge for the CBO, the Republican bill could be even more difficult, former officials said.

Congress had never attempted to achieve Obamacare's goal of universal coverage, so the CBO had few historical examples to use as a guide. Yet the fact that the law set strict rules for the market and made sure there was plenty of money available for states to implement the law made forecasting easier.

Under the Republican bill, insurers would have much more leeway in terms of the kinds of insurance they can offer. Consumers, meanwhile, would have the option of not buying it at all.

And states would have to make difficult decisions about how to provide insurance for Medicaid beneficiaries with hundreds of billions of dollars less than they would have under the current system. Predicting how policymakers, insurers and consumers will respond is a daunting task.

“You have to anticipate the decisions made by, ultimately, 300 million Americans,” Holtz-Eakin said. “That's harder.”