Bernie Sanders Cherry-Picks New and Flawed Study to Advocate for Medicare for All

During the Democratic primary debate last Tuesday, forerunner Bernie Sanders had a rough night because the other candidates were targeting Sanders. As Pete Buttigieg brought up, "Senator Sanders at one point said it was going to be $40 trillion, then it was $30 trillion, then it was $17 trillion. That's an incredible shrinking price tag." Of course, Sanders wants to make Medicare for All (M4A) appear as inexpensive as possible. That is why during the debate, Sanders mentioned a study from Lancet magazine, which is one of the most prestigious medical journals in the world. It is true that such a study exists in Lancet (Galvani et al., 2020), and it is true that the study claims that it would save $450 billion annually in health care expenditures. This study sounds like a win for the Sanders campaign. But is it? Let's forget for a second that the primary co-author, Prof. Alison Galvani, was an unpaid advisor for the Sanders campaign. The study itself is riddled with flaws that make the claim quite spurious. Kaiser Health News (KHN) and Politifact co-authored an analysis of Sanders' claim. I will integrate those findings with other points I found upon research to show why the Lancet study is greatly flawed.

• Estimation of payment rates to hospitals. This is the first flawed assumption made under the Lancet study, which is that it assumes that Sanders' M4A plan can pay hospitals Medicare rates across the board. First, the hospital lobby is a powerful lobby in the United States that has fought off payment reductions in the past, which would put political feasibility into question. Second, this assumption has not played out in the past. As I brought up when analyzing Elizabeth Warren's M4A plan in November, there is one state that has a plan similar to Sanders' M4A plan: Maryland's all-payer system. Under Maryland's system, they could not keep prices down (Pope, 2019). In Washington state, they passed a public option. They could not keep prices down at Medicare rates; they ended up being 174 percent higher than Medicare rates.
• Estimation of utilization rates. As one can imagine, some people skip treatment if they cannot afford it. The authors of the Lancet study assumed that there would be an increase of consumption of health care goods and services for the 24 percent that are uninsured or underinsured. The problem with the assumption that the Lancet study makes is that the remaining 76 percent are not going to increase their utilization as a result of M4A. Harvard health economist Adrianna McIntyre was so critical of this assumption that she wondered how the Lancet study ever made it past the peer review process. And I'm not at all surprised at McIntyre's criticism. Other countries that have implemented single-payer healthcare could not keep down utilization rates, which contributed to single-payer costs.  
• Estimation of administrative costs. By removing the middle man (i.e., the insurance company), certain costs are removed, such as co-pays. There is some intuition to that notion, which is why the Lancet study estimates a savings of $219 billion annually. Medicare has its administrative costs at 2 percent, and the Lancet study assumes that M4A can maintain those rates. I remain skeptical about these cost savings. First, Medicare is partially administered by the IRS [for tax collection] and the Social Security Administration [for collecting premiums]. Also, Medicare's administration is tax-exempt, which cannot be said for private insurers. Second, when you look at cost per beneficiary as opposed to percent of budget, it turns out that Medicare pays more than private insurers (Book, 2018). Third, as brought up in the KHN/Politifact analysis, while M4A would require a smaller back-end staff, it would still not be able to fully reduce the costs at Lancet estimations due to need for such expensive items as electronic health records. This helps explain why Politifact ranked Sanders' "administrative cost savings" claim as half-true.
• Long-term care. Sanders wants to be more comprehensive in M4A than even the Lancet authors assume. Sanders wants to cover for long-term care, which is estimated to cost $4 trillion over a decade. This expenditure is not covered at all in the Lancet study.  

Postscript: It's no wonder why Sanders picked the Lancet study to make M4A look good: it comes with quite rosy and unrealistic assumptions. Sanders' claims ignore how other countries could not contain their costs. More importantly, he claims that all studies, whether "conservative or progressive" show savings. The truth is that reputable estimates independent of the Sanders campaign put the cost of M4A anywhere from an additional $26 to $35 trillion over a decade, which is significantly higher than his most current estimation of $17 trillion. The centrist Rand Corporation estimated that M4A could increase costs by 9.8 percent in a year. The left-of-center Urban Institute estimated that a) federal spending would increase by $34 trillion over ten years, and b) even factoring in the net spending of private and public spending, it would still cost a net increase of $7 trillion over ten years. Urban's conclusion was that "the increase of spending for people with this generous new coverage would outweigh the savings from lower prices for health care providers and and lower administrative costs." And this says nothing of the libertarian Mercatus Center, provided a high-bound estimate of a net increase of $32 trillion in new costs.

Yes, it is true that there is a study from Lancet claiming how M4A could theoretically save money. Yet it is equally true that it has such flawed assumptions that it ignores the reality that M4A comes with too high of a price tag, far higher than Sanders claims. If Sanders were to confront this reality, he would not only realize that he would have to considerably raise taxes on everyone (and not just the 1 percent), but that he truly does not know how he would pay for it.