“The Medicare reduction was aimed at cost efficiency, while the Medicaid reduction was aimed at reducing the number of enrollees.”
—PolitiFact’s Angie Drobnic Holan, during a May 9, 2018 Reddit AMA
“The Senate bill takes several steps to restrain Medicaid spending.
It rolls back who is eligible.”
—PolitiFact fact check of a Kellyanne Conway claim about Medicaid, written by Jon Z. Greenberg
PolitiFact’s Angie Holan overstated the case when she said the BCRA aimed to reduce Medicaid enrollment. The BCRA itself did nothing to limit enrollment, but its lower Medicaid reimbursement rates and repeal of the individual mandate would result in states and individuals making decisions that would lower enrollment.
Greenberg was simply wrong to say the BCRA rolled back Medicaid eligibility for similar reasons. States might limit eligibility in response to lower reimbursement rates, but the BCRA itself did not touch Medicaid eligibility.
In 2017, Zebra Fact Check charged each of the “elite three” fact checkers with reporting inconsistently on budget cuts to projected future spending.
We made attempts to get responses from those fact checkers, PolitiFact, the Washington Post Fact Checker and FactCheck.org. When we heard about PolitiFact’s participation in a May 9, 2018 Reddit AMA (“Ask Me Anything”) we decided to press the issue yet again, segueing from PolitiFact writer Jon Z. Greenberg’s statement about how PolitiFact deals with charges of bias.
Zebra Fact Check (edit notation in the original):
But what about PolitiFact’s blatant inconsistency on making budget cuts to the projected growth of programs like Medicare and Medicaid? When the Affordable Care Act cut the growth of Medicare, PolitiFact insisted that it wasn’t a true cut because spending continued to grow, albeit at a (EDIT: slower) pace. But when the proposed Trump budget cut the growth of Medicaid, PolitiFact discovered that slowing the growth of spending was, indeed, a cut (the inconsistent approach dictates the bulk of the budget cut ratings).
This is a longstanding pattern in PolitiFact’s work.
The comment drew a response from PolitiFact Editor Angie Drobnic Holan:
Angie here … We talked about the reasons for that in the Medicaid check. The Medicare reduction was aimed at cost efficiency, while the Medicaid reduction was aimed at reducing the number of enrollees. Go back and look at the Medicaid checks and you should see that. With as many checks as we do — more thant 13,000 — there might be inconsistencies from time to time, but I disagree that there’s any longstanding pattern.
We reviewed PolitiFact’s fact checks on recent GOP proposals reducing the growth of Medicaid. An example from June 26, 2017, authored by Greenberg and edited by Holan, appeared to fit Holan’s description.
The Senate bill takes several steps to restrain Medicaid spending.
It rolls back who is eligible. Before Obamacare, Medicaid covered low-income children, pregnant women, elderly and disabled individuals, and some parents, but excluded other low-income adults. Under the Affordable Care Act, 31 states and the District of Columbia exercised the option to make Medicaid available to anyone making up to 133 percent of federal poverty. For a family of three, that would be about $27,000 a year. The Senate bill eliminates that option as of January 2018.
This shuts out the Medicaid option for over 2.5 million people in states that didn’t expand eligibility.
Greenberg sharpens the point later in the fact check (bold emphasis added):
When Democrats passed the Affordable Care Act in 2010, it included less money for Medicare over the years. In a mirror image of today’s debate, Republicans, notably House Speaker Paul Ryan, accused Democrats of cutting a vital health care program for the elderly. Democrats argued back that spending increased.
We did a series of fact-checks about the back-and-forth and generally rated the Republican attacks Half True.
Then, Democrats reduced payment levels to health care providers. Now, Republicans propose reducing payments to states. But at least in terms of the money trends, both situations follow the same general outlines — with one key difference:
The Democrats didn’t change the eligible population, and the Republicans do.
Our fact check will address three questions. Was the BCRA aimed at cutting Medicaid enrollment? Does Holan’s explanation for PolitiFact’s inconsistency withstand scrutiny? And was PolitiFact right that the Better Care Reconciliation Act would have cut Medicaid eligibility?
Analyzing the Rhetoric
We will look at Greenberg’s claim about Medicaid eligibility first, as it serves as the apparent premise of Holan’s later assertion about the Senate bill targeting Medicaid enrollment.
Did the BCRA Cut Medicaid Eligibility?
We failed to find evidence supporting the claim in Greenberg’s fact check that the BCRA cut off states’ option to expand Medicaid. The sources Greenberg listed in the fact check were no help to us.
We found that the well-respected fact-checkers at FactCheck.org concluded the BCRA did not end states’ ability to opt for the ACA’s Medicaid expansion:
The Senate bill isn’t “ending the Medicaid expansion” eligibility — it’s ending the enhanced federal funding of the expansion-eligible population under the ACA.
We used Twitter and email to ask Greenberg how to resolve the discrepancy. We will update this item if we receive any response.
Lacking any evidence in support of PolitiFact’s claim, we stick with FactCheck.org’s judgment: The BCRA does not preclude states from expanding Medicaid. It simply makes it more expensive for states to pay for the expansion.
Cutting Eligibility vs Cutting the Growth of Eligibility
Though the BCRA does not strictly prevent states from expanding Medicaid, CBO predicted states that would have opted for expansion without the BCRA would not expand Medicaid if the BCRA shrank the favorable reimbursement rates the ACA had promised.
If those states expanded Medicaid then Medicaid eligibility would grow. It follows from CBO’s reasoning that the BCRA’s lower reimbursement rates would result in slower growth of Medicaid eligibility.
Additionally, lower reimbursement rates may influence current Medicaid expansion states to reverse course.
CBO … anticipates that some states that have already expanded their Medicaid programs would no longer offer that coverage, gradually reducing the share of the newly eligible population residing in a state with expanded eligibility as the matching rate for that population declined—with that share reaching about 30 percent in 2026.
CBO projected that the percentage of the eligible Medicaid expansion population in Medicaid expansion states will drop from about 50 percent down to 30 percent by 2026.
Though missing from PolitiFact’s reporting, CBO also says repeal of the individual and employer mandates results in lower Medicaid enrollment. That adds another wrinkle to our attempts to estimate the BCRA’s effect on Medicaid enrollment. CBO’s Medicaid enrollment estimates for a proposed bill that would primarily end the individual and employer mandates suggest that the repeal of the mandates would account for nearly half of the drop in Medicaid enrollment for 2026 (7 million) predicted for the BCRA. We note that some of that number may not have had eligibility anyway if their states opted out of the Medicaid expansion. But in such cases removal of eligibility does not fully account for the drop in enrollment. CBO says about 7 million eligible persons would not have enrolled if not for the threat of penalty.
Though PolitiFact’s fact check appears to wrongly conclude that the BCRA prohibits Medicaid expansion, CBO says lower reimbursement rates make future expansion unlikely and may influence some states to end their participation in the Medicaid expansion.
The BCRA would not, strictly speaking, cut Medicaid eligibility. But the new lower reimbursement rates would likely prompt states to lower Medicaid eligibility from its current levels.
Greenberg’s fact check did not mention the term “enrollment.”
Was the BCRA Aimed at Cutting Medicaid Enrollment?
The Better Care Reform Act clearly aimed at cutting the growth of Medicaid spending. And any bill aimed at cutting the growth of spending has to cut spending growth or else it will not realize that aim. For a program like Medicaid, legislators might curb spending growth by cutting reimbursement rates, cutting the number of enrollees or both.
Through its adjustments to Medicaid reimbursement, the BCRA clearly takes the first avenue. Does it also take the second?
Taking the BCRA as an attempt to cut Medicaid enrollment, let us consider the effectiveness of the measures it used.
CBO identified three primary ways the BCRA might affect Medicaid enrollment:
- States might cut back their existing Medicaid expansion (in response to lower reimbursement rates).
- States might stop future plans to expand Medicaid (in response to lower reimbursement rates).
- People eligible for Medicaid might not enroll in Medicaid without the threat of the individual mandate penalty.
The first two clearly stem from the BCRA’s adjustments to reimbursement rates and are not stipulations of the law. Rather, they represent the projected responses of state governments to the prospect of sharing a greater percentage of Medicaid costs.
The third curb on Medicaid enrollment does not stem from adjustments to reimbursement rates. It stems from eligible individuals choosing not to enroll in Medicaid without the threat of a penalty. And it’s worth noting that Medicaid-eligible persons may retroactively receive Medicaid benefits after enrolling.
Medicaid.gov (bold emphasis added):
Once an individual is determined eligible for Medicaid, coverage is effective either on the date of application or the first day of the month of application. Benefits may also be covered retroactively for up to 3 months prior to the month of application, if the individual would have been eligible during that period had he or she applied. Coverage generally stops at the end of the month in which a person no longer meets the requirements for eligibility.
We think Holan’s claim that the BCRA aimed to cut Medicaid enrollment best counts as hyperbole. Of the two methods available for slowing the growth of costs the BCRA made clear use of one of them: cutting reimbursement rates. The other factor slowing the growth of costs, lower enrollment, stems from state actions in response to greater cost sharing, along with individuals not enrolling if not threatened with a penalty.
One would more accurately say the BCRA aimed to slow Medicaid spending by cutting reimbursement rates, with states likely responding with restrictions on Medicaid eligibility.
The BCRA’s Indirect Effects on Enrollment
Congressional Research Service reports on projected Medicaid enrollment, when combined with CBO’s estimates of the BCRA’s (indirect) effects on enrollment, lead us to conclude that Medicaid enrollment would likely drop below current levels with passage of the BCRA. The projected effects of the BCRA, in other words, would not merely slow the growth of enrollment.
Does PolitiFact’s Excuse Hold Up?
Does PolitiFact get itself off the hook for treating budget cut fact checks inconsistently by blaming the harsh rating of Conway on cuts to enrollment?
Two factors make Holan’s rationale count as rationalization.
First: PolitiFact tries to justify the differing treatment of the ACA’s cuts to Medicare and GOP cuts to Medicaid by saying the latter results in a cut to benefits through the exclusion of some beneficiaries. But PolitiFact conceded in 2009 that the proposed health care law was likely to cut some Medicare benefits. It simply chose to ignore its own finding while it was issuing harsh ratings to Republicans who said the ACA was cutting Medicare.
PolitiFact, Aug. 14, 2009 (bold emphasis added):
That leaves us with Obama’s claim that, under the health care reform proposal, Medicare benefits will not be cut. He’s right that the bill does not directly trim Medicare benefits; instead, the government is proposing ways to slow or eliminate some Medicare spending to beef up other aspects of the plan. But experts told us it’s conceivable or even likely that those financial changes could lead to reduced benefits, particularly for people in the Advantage program. From that perspective, it’s a stretch for Obama say that Medicare patients won’t see changes in their plans as a result. We give Obama a Half True.
PolitiFact, on balance, simply did not care about its own ruling that the health care law would cut Medicare benefits. It ignored its “Half True” ruling of Obama’s claim repeatedly while issuing harsh ratings to Republicans who dared to claim the ACA cut Medicare.
Second: Greenberg’s fact check of Kellyanne Conway effectively ignores the context of Conway’s statement. PolitiFact ignores the context despite reporting it (bold emphasis added):
In an interview Sunday, ABC news host George Stephanopoulos brought up the bill’s projected $800 billion in Medicaid savings and asked Conway if that undermined the president’s campaign promise to spare Medicaid from cuts.
“These are not cuts to Medicaid, George,” Conway said on This Week on June 25. “This slows the rate for the future.”
Stephanopoulos did not ask Conway if lower enrollment or eligibility undermined the president’s promise. He asked if the projected $800 billion in savings undermined that promise. PolitiFact, before Trump took office, repeatedly said slowing the growth of spending was not a cut, rating that association “Half True” or worse—usually worse— when applied to Medicare. Yet when Conway tried to use PolitiFact’s past reasoning in response to Stephanopoulos, PolitiFact rated her claim “Mostly False.” Taking Conway’s statement to apply to enrollment or eligibility and not specifically the dollar savings Stephanopoulos mentioned counts as taking Conway’s statement out of context.
PolitiFact reported falsely, took Conway’s claim out of context and justified giving Conway a poor rating using a rationale that it might have also applied to claims about the ACA’s slowing of Medicare spending.
“The Medicare reduction was aimed at cost efficiency, while the Medicaid reduction was aimed at reducing the number of enrollees.”
Given a maximally charitable interpretation, we can count Holan’s claim as true. One could plausibly claim the ACA’s slowing of Medicare spending was aimed at cost efficiency. At the same time, one could plausibly claim that slowing Medicaid spending was aimed at reducing enrollment. In truth, both budget strategies aimed at cost efficiency. The ACA ended services some Medicare beneficiaries received through Medicare Advantage and lower reimbursement rates under the ACA would likely drive some providers out of the market (reducing delivery of services). The BCRA, by comparison, shifted the costs of Medicaid expansion toward the states, making it likely the states would move to restrict Medicaid eligibility, It also capped reimbursement in a way likely to end up squeezing Medicaid providers (and reducing delivery of services).
In both cases, lower reimbursement rates would likely affect the availability of medical services. PolitiFact exposed the effects on services in one case but kept them largely unmentioned in the other.
Holan’s phrasing, intentionally or not, obscured the similarity of the two cases with a false implication of dissimilarity.
“It rolls back who is eligible.”
This key claim from PolitiFact’s fact check, which we attribute to writer Greenberg, was literally false. The BCRA did not directly cut Medicaid eligibility, contrary to what Greenberg reported. At the same time, the BCRA’s changes to Medicaid reimbursement provided an incentive for states to restrict Medicaid eligibility. Making the latter point should not require a misstatement of the facts like we see in PolitiFact’s fact check.
White, Bryan W. “Why the ‘Elite Three’ Fact Checkers Miss out on Public Trust.” Zebra Fact Check, Zebra Fact Check, 6 Sept. 2017
“We Are PolitiFact, the Largest Political Fact-Checking Newsroom in the United States. Ask Us Anything!” Reddit, Reddit Inc., 8 May 2018
Jon Greenberg. “Conway’s Mostly False Claim GOP Bill Doesn’t Cut Medicaid.” PolitiFact.com, PolitiFact, 26 June 2017
Robertson, Lori, and Brooks Jackson. “GOP Medicaid Spin.” FactCheck.org, FactCheck.org, 13 July 2017
White, Bryan W. “Hey @JonZGreenberg …” Twitter, Twitter, 10 May 2018
White, Bryan W. “Email Outreach to PolitiFact’s Angie Holan and Jon Z. Greenberg.” Zebra Fact Check, Zebra Fact Check, 21 May 2018
“CONGRESSIONAL BUDGET OFFICE COST ESTIMATE H.R. 1628 Better Care Reconciliation Act of 2017.” CBO.gov, U.S. Congress, 26 June 2017. Pages 18,19
Ponnuru, Ramesh. “PolitiFact, Wrong Again on Health Care.” National Review, National Review, 10 Oct. 2017
Park, Edwin, et al. “Frequently Asked Questions About Medicaid.” CBPP.org, Center on Budget and Policy Priorities, 29 Mar. 2018
“2016 Actuarial Report on the Financial Outlook for Medicaid.” CMS.gov, Centers for Medicare & Medicaid Services, 2016. Page 10
“2016 Actuarial Report on the Financial Outlook for Medicaid.” CMS.gov, Centers for Medicare & Medicaid Services, 2016. Page 27
“Estimate of Direct Spending and Revenue Effects of H.R. 1628, the Healthcare Freedom Act of 2017, an Amendment in the Nature of a Substitute [S.A. 667].” CBO.gov, U.S. Congress, 27 July 2017. Page 17
Richert, Catharine. “Obama Claims Medicare Benefits Will Not Be Cut under the Health Care Bill.” PolitiFact.com, PolitiFact, 14 Aug. 2009
White, Bryan. “PolitiFact Revises Its Own History.” PolitiFactBias.com, PolitiFact Bias, 4 July 2017