How well do fact checkers fact check?
Each of the “elite three” mainstream media fact checkers, PolitiFact, the Washington Post Fact Checker and FactCheck.org, has had the opportunity to weigh in on the effects of the GOP’s health care reform proposals.
The “Affordable Care Act” passed under the administration of President Barack Obama applies a tax to individuals and families who fail to get health insurance coverage. The individual mandate thus encourages the public to get health insurance.
Each of the “elite three” fact checkers has struggled to accurately describe the Congressional Budget Office’s view of the effect of repealing the individual mandate. Some of that blame falls on the CBO for failing to communicate clearly. But the fact checkers failed to notice their reporting did not add up, and magnified that failure by later neglecting to correct false statements and omissions in their past fact checks.
A “don’t care” attitude on the accuracy of a fact check ought to damage fact checker credibilty. Accordingly, this article aims to shame the fact checkers into correcting their past errors and omissions or else help appropriately damage the reputation of fact checkers who fail to correct their past work.
As we noted in an earlier post focusing on the Washington Post Fact Checker, the bulk of the credit for publicizing the CBO’s views on the effects of repealing the individual mandate belongs to Avik Roy and Ramesh Ponnuru.
How did each of the “elite three” mainstream media fact checkers handle the issue? We will look at each in turn and then look at how all three have gone on to publish fact checks that conflict with their earlier work.
PolitiFact’s cornerstone fact check on GOP health insurance reform came from a June 28, 2017 fact check written and researched by Jon Greenberg.
Ponnuru correctly noted Greenberg failed to take the individual mandate’s effect on Medicaid coverage into account in his analysis.
We identified a number of false statements in Greenberg’s fact check. We will focus on two of them. First:
The CBO said that compared with the laws in place today, 15 million fewer people who need insurance would be able to get it through Medicaid or anywhere else.
But the CBO explicitly said, in a report Greenberg cited, that the repeal of the individual mandate would result in fewer people signing up for Medicaid. It follows that people who do not sign up for Medicaid because of the threat of a tax penalty are not making their decision based on affordability. Medicaid, after all, carries little to no cost for beneficiaries. People who did not sign up for Medicaid because the lack of a penalty could get insurance through Medicaid, contrary to what PolitiFact claimed the CBO said.
The CBO said two-thirds of the 22 million without insurance would lose out through cuts in the Medicaid program.
This statement misleads more subtly than the first.
The CBO said about two-thirds of the 22 million more projected not to have insurance were from the group of people insured under Medicaid. But is that the same as saying the two-thirds would not have insurance thanks to “cuts in the Medicaid program”?
It is not the same thing.
In fact, states receive Medicaid dollars based on enrollment figures. When fewer people enroll in Medicaid, the states receive less money.
It’s true that some would lose eligibility for Medicaid because of states declining the ACA’s option of Medicaid expansion, but as PolitiFact pointed out, about two-thirds of the drop in projected Medicaid enrollment comes from persons now eligible for Medicaid. PolitiFact implied that group would lose Medicaid insurance because of cuts to Medicaid.
The CBO report describes two ways the proposed GOP legislation would decrease Medicaid spending. First, the BCRA places a cap on per capita spending for most Medicaid beneficiaries. Second, the BCRA changes the federal share of state Medicaid spending. Together, those changes may discourage states from choosing to expand Medicaid under the ACA. Lower enrollment would also lead to lower spending on Medicaid.
PolitiFact, as Ponnuru pointed out, simply avoids any mention of the CBO’s assessment that repealing the individual mandate impacts Medicaid enrollment. Thus, the role of the individual mandate in lowering Medicaid spending fails to make it into PolitiFact’s fact check.
That information was certainly included in the CBO report, though the CBO did not prominently emphasize it (bold emphasis added):
This legislation would, upon enactment, eliminate penalties associated with requirements that most people obtain health insurance coverage (also called the individual mandate) and that large employers offer their employees health insurance coverage that meets specified standards (also called the employer mandate). As a result, fewer people would enroll in health insurance obtained through the nongroup market, in employment-based health insurance coverage, and in Medicaid.
The CBO later expanded on that point (bold emphasis added):
Under current law, the penalty associated with the individual mandate applies to some Medicaid-eligible adults and children. (For example, it applies to single individuals with income above about 90 percent of the FPL.) In addition, some people apply for coverage in the marketplaces because of the penalty and turn out to be eligible for Medicaid. And some who are not subject to the penalty think they would be if they did not enroll in Medicaid.
The agencies do not expect that, with the penalty eliminated under this legislation, people enrolled in Medicaid would disenroll. However, among people who would become eligible for Medicaid under the legislation or who would need to recertify their eligibility, the proportion of people who enroll in the program would, by CBO and JCT’s expectations, be lower—closer to the proportions observed for those groups prior to the institution of the penalty.
In a Jan. 15, 2017 fact check the same PolitiFact writer, Jon Greenberg, wrote that many of the 20 million who supposedly gained insurance thanks to the ACA were eligible apart from the ACA’s Medicaid expansion:
About 20 million people gained coverage and about 14.5 million of those were under Medicaid or CHIP. But a sizeable fraction of that 14.5 million were eligible before the Affordable Care Act took effect. One estimate said about a quarter of them were previously eligible. Another estimate put it as high as half.
It follows that about 3.6 million to 7.3 million fewer people will have insurance through Medicaid as a result of repealing the individual mandate. It also follows that those millions remain eligible for Medicaid but choose not to sign up for that free program, much as they did before the ACA penalized them for not having insurance.
This renders false Greenberg’s claim that “15 million fewer people who need insurance would be able to get it through Medicaid or anywhere else.” Millions would be able to get insurance through Medicaid or other means, contrary to what Greenberg and PolitiFact claimed.
We fault Greenberg and PolitiFact for not picking up on the discrepancies he passed along to his readers. And we fault them doubly for not fixing the problems with this fact check in the months since criticisms like Ponnuru’s have appeared.
PolitiFact and Greenberg owe their readers either a correction or a justification.
We address the Washington Post Fact Checker on Page 2.