“Joe said that Medicare-for-all would cost over $30 trillion. That’s right, Joe. Status quo over 10 years will be $50 trillion.”

— Sen. Bernie Sanders (I-Vt.), remarks during the third Democratic presidential debate, Sept. 12, 2019

Sanders made this comment after former vice president Joe Biden criticized the senator’s Medicare-for-all plan for potentially adding “$30 trillion, $3.4 trillion a year” to the federal budget. In his response, Sanders favorably compared that $30 trillion to what he called the “status quo” for health-care spending — $50 trillion. He added: “Every study done shows that Medicare-for-all is the most cost-effective approach to providing health care to every man, woman and child in this country.”

The implication was that Medicare-for-all would reduce national health expenditures from the “status quo.” When we queried the Sanders campaign during the debate for the source of the $50 trillion figure, we were directed to posts that were written by Paul Waldman for The Washington Post’s “The Plum Line,” an opinion column.

Waldman compared a government projection that national health spending would amount to about $50 trillion over the next decade to an estimate that projected the federal cost of Medicare-for-all as $32.6 trillion. In one column cited by the campaign, Waldman wrote: “So if Medicare-for-all actually costs $40 trillion, we would save $10 trillion. Hooray!”

Waldman told the Fact Checker he has not read academic studies that looked closely at the impact of a Medicare-for-all plan on national health expenditures. All but one of five major studies, from the left to the right, predict the Sanders plan would increase health spending, not reduce it.

During our reporting for this fact check, a spokesman for the Sanders campaign said it was not endorsing Waldman’s analysis but was simply supplying a source for the $50 trillion figure that Sanders cited in the heat of the debate. But he declined repeated queries to explain the point Sanders was making with his comparison.

The Facts

We will try not to get too much in the weeds, but this debate involves some big, confusing numbers.

The $50 trillion figure comes from an estimate of national health expenditures (NHE) produced by the Centers for Medicare and Medicaid Services for the years 2018 to 2027.

The “over $30 trillion” refers to a 2018 study by the Mercatus Center at George Mason University, which has been funded by entities associated with the Koch family, that calculated new federal spending on health care if the Sanders plan were enacted. The number $32.6 trillion is often used, but the study actually offers a possible range of $32.6 trillion to $38.8 trillion for the years 2022 to 2031.

Note that the $50 trillion and the $30 trillion involve different 10-year time frames and thus are not directly comparable.

The Sanders campaign has cited this study to claim his plan would lower health-care costs, but the author, Charles Blahous, has publicly and persistently insisted the campaign has misinterpreted his findings. In particular, he says the “lower-bound” estimate is unrealistic because it assumes Sanders would be successful in reducing payments for treatments now covered by private insurance by 40 percent.

Roughly speaking, there are two parts of the $50 trillion in national health expenditures that would be affected by a Medicare-for-all plan: $17 trillion in current government spending on Medicare, Medicaid, Obamacare and the like, and $26 trillion in spending by households, employers and states. (Another $7 trillion would not be affected.) Medicare-for-all would essentially take over the state, employer and household spending, so that’s the key variable.

Using the same time frame (2022 to 2031) so we can compare apples to apples, the “status quo” for health expenditures in the 10-year period covered by the Mercatus paper would be $59.7 trillion. Under the Mercatus range of estimates, total health expenditures under Medicare-for-all would be $57.6 trillion to $63.8 trillion over 10 years. On average, that’s slightly higher than under the current path.

“Bottom line is that it is incorrect to say that the plan is cheaper than current law, even on a national basis,” Blahous said. “Most of the range of possible outcomes would have total NHE increasing under Medicare-for-all. The increase occurs primarily because of the increase in the numbers of those covered as well as in the generosity of coverage, both of which would spur additional demand and, other things being equal, additional spending.”

The Mercatus study is not the only detailed look at the potential impact of a single-payer plan on federal spending. In a 2016 report, the Urban Institute said that federal spending would increase by about $32 trillion between 2017 and 2026 and national health expenditures would increase by $6.6 trillion on top of the status quo in that period.

“The increase in federal spending is so large because the federal government would absorb a substantial amount of current spending by state and local governments, employers, and households,” the report said. “In addition, federal spending would be needed for newly covered individuals, expanded benefits and the elimination of cost sharing for those insured under current law, and the new long-term support and services program.”

Kenneth Thorpe, chairman of the health policy department at Emory University, in a 2016 study also predicted an increase in spending, as did a Rand Corp. study in 2019 led by Jodi L. Liu.

Thorpe, a skeptic of the Sanders plan, said he was surprised when Sanders made the comparison during the debate. The two figures are “apples and oranges and not comparable,” he said.

Democratic presidential hopefuls say Medicare-for-all would be less expensive than the current health-care system. But it's not that straightforward. (Atthar Mirza/The Washington Post)

“Medicare-for-all would increase the dollars paid by the federal government, if private premiums are largely eliminated and new taxes are established to fund a national single-payer system,” Liu said. “How this impacts national health expenditures depends on a variety of assumptions. In our RAND report, we projected a 1.8 percent increase in national health expenditures if Medicare-for-All was implemented in 2019.”

The New York Times Upshot earlier this year helpfully illustrated the impact envisioned by the five studies — Mercatus, Urban, Rand, Emory and the University of Massachusetts at Amherst — for national expenditures under Medicare-for-all, compared to current law. Only one of the five studies showed a significant decline.

That study, by Gerald Friedman, an economics professor at UMass-Amherst, found health spending would decline over a 10-year period from a baseline of $46.8 trillion in health spending between 2019 and 2029 under six scenarios. The most optimistic model, assuming Americans shared some costs with the government, shows a decline in spending to about $34 trillion, or 28 percent. But a model with no cost-sharing — Sanders’s plan envisions no out-of-pocket spending except for prescription drugs — would have national health expenditures of nearly $41 trillion, or a decline of 13.2 percent.

“Bernie is overstating things,” Friedman said. “Even with optimistic assumptions about savings realized and a slowdown in growth to Canadian or European growth rates and an actuarial value of 87 percent (not the 96 percent in his program), spending would not decline by $20 trillion, or 40 percent. I am perfectly comfortable in saying that there will be financial savings, savings over-and-above the improvements in health that would come with greater access and improvements in financial stability with universal quality insurance. The benefits of Medicare-for-all are substantial enough that there really is no need to exaggerate.”

The Pinocchio Test

Sanders implied during the debate that his Medicare-for-all plan would reduce health-care spending compared to the “status quo” of $50 trillion over 10 years. His use of the “more than $30 trillion” figure might have led listeners to assume he was comparing the two figures. Certainly that’s how the authors of these studies interpreted these comments.

That might have been a logical assumption. During the debate, his staff shared, as the source of the $50 trillion figure, a pair of Washington Post opinion columns that made a similar comparison. But a spokesman has since said that the campaign was not endorsing the analysis in those columns. Instead, he says Sanders was simply pointing to a government estimate for health expenditures over a 10-year period.

Since the spokesman declined to explain why Sanders made the comparison, we are left with a conundrum. It certainly sounded as if he was saying his plan would reduce health-care spending from $50 trillion over 10 years to $30 trillion, even though most studies predict spending will increase. The Washington Post columns originally cited as a source also framed the issue that way.

We can understand a misfired talking point in the heat of a debate. Regular readers know that we always offer the opportunity to politicians to explain that he or she made an error and then we leave the claim unrated. We don’t try to play gotcha and simply want to inform readers about the facts. (For instance, in response to our queries about a fact check, Biden recently admitted he misspoke about his opposition to the Iraq War.) Similarly, we offered multiple opportunities to the Sanders campaign to explain his statement or acknowledge an honest mistake. The campaign did not respond to that request.

On balance, as uttered during the debate, the statement is mostly false. Four of the five key studies on the impact of the Sanders plan estimate that national health expenditures will rise over 10 years, while the author of the fifth predicts a decline but says Sanders’s statement is exaggerated.

Three Pinocchios

Send us facts to check by filling out this form

Sign up for The Fact Checker weekly newsletter