• How voucherizing Medicare is not like Part D

    Fred Barnes, executive editor of the Weekly Standard, wrote an opinion piece for NPR earlier this week in which he made the case for voucherizing Medicare. To the extent that I find the competitive bidding version of vouchers a sensible idea, I agree  with Barnes. However, that’s not what he’s talking about, or is it? His piece is a bit confused on that point and a few others. I’ll explain.

    There are some things Barnes and I seem to agree on too. Let’s start at the beginning. His column opens,

    Social Security’s looming deficit can be handled, for the time being, by adjusting benefits a tad downward. Medicaid’s runaway spending can be restrained by giving state governors more flexibility in administering the program. These are modest solutions. Medicare is different. It needs a big solution.

    OK, I agree that Social Security is an insignificant problem relative to Medicare. No doubt it could be solved by tinkering with benefits, though that is not the only way to solve it, as Don has been explaining. I’ll leave the details on SS to Don. I’ve written a lot already about Medicaid, but Harold Pollack has written more, and specifically on what “flexibility” will do. Medicare is different and it does need a big solution. No argument there.

    But what is the solution? Barnes says the only solution (the only one) is premium support or defined contribution, also known as voucherization. Well, it isn’t the only solution, but it is a solution. Anyway, he wants to set a cap on the total amount the government would spend on vouchers, essentially the Ryan-Rivlin plan. Meanwhile, he completely dismisses the need to specify how that cap will be achieved and disregards entirely the political reality of attempting to do so. “A policy statement on how that level of spending would be achieved is not required,” he writes. He follows a bit later with,

    Supporters of President Obama’s health care law tout its pilot projects to reduce Medicare’s cost. Nothing wrong there. But we’ve seen hundreds of such projects over the past 40 years — with minimal impact on Medicare. Few resulted in cost savings.

    So how is a voucher plan with a cap set by fiat any different? In particular, how is it different from Medicare Advantage, the voucher program that exists today, the costs of which have exploded? Barnes doesn’t say. Without a policy statement that explains it, which he rejects, he can’t.

    He does, however, point to a voucher plan within the current Medicare program that works very well, Part D.

    What has worked is competition. The Medicare prescription drug benefit program, enacted in 2003, has cost 40 percent less than projected. This is due to competition among providers for the business of millions of seniors.

    Yes. Yes. One thousand times, yes! Part D is a good model, with protections for taxpayers and beneficiaries. Guess what it has that Medicare Advantage and the Ryan-Rivlin plan do not. (Wait for it …) Competitive bidding. That’s the source of its protections. That’s the crucial way in which it differs from other voucherization ideas.

    All Part D plans (PDPs and the drug portion of MA plans) submit bids for the cost of standard coverage. From these bids a nationwide average is computed and a statutorily determined fraction of this average cost is set as the “base premium.” Finally, a plan’s premium (charged to each enrolled beneficiary) is the base premium plus the difference between that plan’s bid and the national average bid. Thus, ignoring adjustments for beneficiary risk and other details (found here) a plan is paid by Medicare a fixed monthly per-beneficiary rate equal to the national average bid less the base premium. Because the payment is tied to national average bids (a market signal) this is a form of competitive bidding.

    No such bidding system would exist under the Ryan-Rivlin plan. Thus, Barnes preferred method of voucherizing Medicare is not like Part D. Yet he points to Part D as a justification, a model, for it. Well, if Part D is so laudable, let’s follow that model.

    How voucher levels are set is the most important aspect of a premium support plan. It’s crucial to understand the mechanics. Is it by declaring, “Thou shall not raise subsidies faster than GDP + 1 percent,” or is it based on an apolitical market mechanism? Don’t be fooled by the slight of hand Barnes has employed. Competitive bidding is different. Part D does it. Medicare Advantage and the Ryan-Rivlin plan do not.

     

     

     

    Share
    Comments closed
     
    • Hi:

      Let’s try that again.
      Please use the <strike> tag on the Rivlin part of the name of the plan. I.e., Ryan-Rivlin plan.

      Alice Rivlin has publicly dissociated herself from Ryan’s plan.

      Thanks

    • @Richard – The bidding is local in the exchanges too. Part D, being Rx drugs is different. The cost of providing a drug benefit does not vary that much by region.