Code Red: Two Economists Examine the U.S. Healthcare System

July 26, 2011

Vermont’s Bold Experiment

Filed under: Budget,Health insurance,Health Reform,Health spending,Integration — David Dranove and Craig Garthwaite (from Oct 11, 2013) @ 9:02 am

I was delighted to see the lead article in Health Affairs describing Vermont’s new single payer health care financing system. Harvard Professor William Hsiao and his coauthors describe this as a “Bold Experiment” and I couldn’t agree more. It is also a very welcome experiment. For over thirty years I have heard the rhetoric that a single payer system would never work in the United States. For that matter, I have heard that a true market-based system (with vouchers) would never work either. Why not let the states experiment and find out what will and won’t work? Thankfully, the Vermont legislators and Governor Shumlin had the courage to take this leap of faith.

The biggest obstacle to implementation appears to be ERISA, which limits the extent to which states can regulate self-funded plans. Apparently, self-insured employers could object to having their tax payments used to support the plan. But Vermont can apply for an ERISA waiver under terms in the Affordable Care Act and the state hopes to begin its bold experiment in 2015.

As bold as the plan might be, Hsiao et al. might be even bolder in projecting the potential cost savings, which they peg at 25.3 percent. Academics rarely go out on a limb with projections like this that can easily be assessed in a few years time. And academics are rarely so optimistic. I wish I could share that optimism.

Let’s take a close look at the projections. Hsiao et al. expect a 2 percent reduction in expenditures from malpractice reform. This is plausible, but malpractice reform is easily severed from health financing reform – many states have already done so – and I do not see why we should attribute any resulting savings to the implementation of a single payer system.

Hsiao et al. also projecta reduction in administrative expenses of 7.3 percent, stemming from “the consolidation of insurance functions” and “reduced administrative costs for providers stemming from uniform claims administration.” This also seems plausible and the actual savings could be even higher, inasmuch as the state will be doing away with all of the marketing and medical underwriting functions of private health insurance.

From here, things get dicier. Hsiao et al project another 5 percent savings from reduced fraud and abuse. How so? The “comprehensive claims database” is supposed to make detection easier. The authors cite a 2007 FBI report as the basis for their 5 percent estimate. That report states that fraud and abuse amount to as much as 3-10 percent of total U.S. health spending, but it makes no mention of the potential cost savings from creating a comprehensive claims database. Hsiao et al. also cite a study of fraud and abuse in Taiwan but do discuss its relevance to Vermont. Will consolidating claims help stop fraud and abuse? Medicare is not a comprehensive claims database but it is awfully big, yet Medicare fraud and abuse is rampant. Large private insurers also fall victim to fraud and abuse. If Hsiao et al believe that Vermont’s state employees will do a better job fighting fraud and abuse than private insurers, good luck to them!

Hsiao et al project the biggest cost savings, 10 percent, will come from payment reform and integration of delivery systems – essentially, moving everyone into an Accountable Care Organization. (Hsiao and colleagues acknowledge that the shift to ACOs is not mandated yet include the projected cost savings as if it was a fait accompli.) As I have previously blogged, any cost savings projected from ACOs are truly speculative. And in a small state like Vermont, the shift to ACOs may backfire. To understand why, consider that single payer systems in Canada and Europe largely hold down costs by bullying the medical community into accepting low wages and restrictions on access to medical technology. The bullying works – the proof is in the cost savings. But Vermont is not large enough to support more than a handful of ACOs, each with a local monopoly. If anyone does the bullying, it will be the monopoly ACOs demanding higher rates and funding for more technology. Dominant ACOs may have even more bargaining power than the state; legislators can always be fired. (I realize that Vermont is perhaps the most liberal state in the nation and I suppose that goes for their physicians. But let’s see how liberal they are when the state decides to slash their fees by 10 percent.)

Lastly, Hsiao et al. project a savings of 1 percent in governance and administration, apparently due to “insulating major spending decisions from the political process.” They expect to keep politics out of the single payer system?

Let me reiterate. The new Vermont law is terrific. I hope they implement it as soon as possible and that it succeeds beyond my wildest expectations. But I doubt it will succeed beyond Hsaio et al’s expectations, as those truly are wild. If Vermont can reduce administrative costs and expand coverage without sacrificing quality or creating shortages, the experiment will be a success. If and when that happens, I hope more states will follow suit.

Now which state will be bold enough to experiment with a fully market-based system?


  1. Hello, David. I blog specifically about health reform in Vermont, and have these questions for you:

    1. Could you provide the citation for a provision in the PPACA that would allow VT to apply for a waiver from ERISA? This is certainly the first time I have heard of such a provision in PPACA. Even our Commissioner in charge of Health Care Administration has stated VT will never get a waiver of ERISA.

    2. Why don’t you consider the lack of waivers from the Social Security Act (for Medicaid and Medicare), the act creating Tricare (which covers a significant number of people in Vermont), the act creating the Federal Employee Health Plan (ditto the Tricare comment) equally significant barriers to a single payer? Without all of the waivers (and the flow of those federal funds to VT’s single payer), there’s no single payer.

    And an “almost single payer” is not a laughing matter (as is, for example, “almost pregnant.”) Because we’d have all of the state responsibility for a new entitlement, but none of the savings and funding assumed in Hsiao’s projections. Hsiao’s study made that clear, and when he was asked at a public meeting: Is this a menu or is it all or nothing? His answer was: all or nothing or it won’t work.

    For those of us who live here, the idea of Vermont performing this grand experiment on behalf of the rest of the nation, without a net, is a scarey proposition. And this kind of outside encouragement doesn’t help, when it’s not grounded in the reality that there’s not a chance of pulling it off without the cooperation of John Boehner and Orrin Hatch (who control Congressional action on creating some new waivers for us). What do you think our chances are?

    I’m not a lover of private insurance; I just don’t want Vermont going down the drain in the name of a Bold Experiment. This requires a federal solution. And that will take single payer advocates working Congressional district by Congressional district… (Vermont has already done our work there.)

    Comment by jeannekeller — July 29, 2011 @ 11:57 am

    • Canada has a single payer in each province. No choices, for better or worse (though I suppose one could move.)

      It was Hsiao et al who raised the issue of the ERISA waiver and did not mention any other obstacles. There is a great tradition of states experimenting with policies that ultimately were adopted nationwide. CON; rate setting, even licensing (in the 19th century). Today, Massachusetts and Utah are experimenting with exchanges. That is the best of what federalism has to offer America.

      Comment by dranove — August 3, 2011 @ 9:28 pm

  2. Excellent post Dr Dranove! Just wanted to comment that Canada is not a single payer system. It is, I think, a 13 payer system (10 provinces and 3 territories).

    Comment by Abbas Ali Mooraj — August 3, 2011 @ 9:10 pm

  3. I really do hope this payer systems succeeds. But my logic tells me why do we need to “experiment” with payer systems when we can adopt proven systems as those implemented outside of North America by the UK and France?

    Comment by CT Weber — August 25, 2011 @ 8:40 am

  4. Great article..thank you

    Comment by Vic T — September 20, 2011 @ 6:44 am

RSS feed for comments on this post.

Blog at

%d bloggers like this: