Henry Aaron and me on premium support

I am honored to join Henry Aaron on the by-line of this, just published, NEJM commentary on premium support. It is short and ungated, so I hope you will read it. It offers a negative view of premium support, particularly the variants of the concept that have been recently proposed. I am certain this will delight some readers. I am just as sure many will be disappointed. Some may be surprised. There’s no pleasing everyone. The best I can do is explain my view as clearly as I can, as well as the process by which I came to it.

(Joe Antos also has a commentary on premium support in NEJM today. I discussed another paper of his on this topic late last year.)

If you read our paper closely, you will see that we have some specific and principled concerns about recent premium support proposals (Domenic/Rivlin and Wyden/Ryan). However, putting those aside, we have even more concerns about the political climate surrounding and timing of consideration of premium support. Fundamentally, and in light of the reforms that are expected to be implemented under the ACA, we argue that now is not the right time for premium support. There is too great a risk of a poor outcome and too much potential in the ACA to make such a fundamental course change right now. Even passing legislation for implementation years from now threatens the potential of the ACA by weakening political support for it.

I assure you that I didn’t come to the conclusion expressed in the paper easily. As regular readers of this blog know, I have spent many months and many posts considering premium support, mostly focused on a variant of “competitive bidding“, explaining how its myriad variations could work, and illuminating the advantages it offers. I have also taken seriously and written about its limitations and the legitimate concerns about it expressed by many health policy experts. I have walked a mile in its shoes, so to speak. (If I’ve walked that far, Henry has walked a marathon.) At the end of the journey, I have reached a conclusion, as expressed in the commentary and shared by Henry.

Naturally, reasonable people can disagree. There are no hard and fast right answers in health policy. Judgements have to be made with incomplete information and accepting the virtues and limitations of the consequences. To that extent, the piece is an expression of opinion. It’s grounded in evidence, experience, and fact, but still it is only opinion that can bring it to a normative conclusion. You have been warned.

Below I quote a few points made in the paper and elaborate on them or link to posts that do. I suggest you read the paper first and then come back to read the rest of this post and follow the links if you want to see the supporting argument and/or evidence. Also, other than the quoted bits in bold, the following are my words and do not necessarily reflect Henry’s view.


“[Voucher] value should grow as rapidly as overall per person health care spending.” Ever since his 1995 paper with Robert Reischauer, Henry Aaron has been consistent in his position that Medicare  premium support payments should keep pace with health care costs and not be tied to an arbitrary index.

“Only a few plan types should be offered so that choices would be comprehensible.” As have others, I have argued the merits of requiring participating plans to adhere to specified benefits at least as generous as traditional Medicare. This is in contrast to an actuarial equivalent standard, which permits great leeway in plan design that could be used by plans to compete on risk selection.

Both of the forgoing criteria are consistent with the premium support plan I described in my recent series and are at odds with recent bipartisan proposals (Domenici/Rivlin, and Wyden/Ryan).

“Current plans do not contain these elements, and even if they did, there is little prospect in the contemporary political environment of enacting or sustaining them.” This was  precisely my conclusion to my premium support series. Political reality is a serious threat premium support with a level playing field between private plans and traditional Medicare.

“The Affordable Care Act (ACA) sets in motion a process of experimentation and change.” The reforms in the ACA deserve to be tested. The greatest threat to their success is political, not technical, so replacing them with an equally politically unstable premium support concept is not a clear improvement. The Wyden/Ryan proposal is silent on the ACA reforms; the Domenici/Rivlin one strengthens some of them (e.g. extending the IPAB’s purview to Medicare benefits) and leaves others as is.

“Traditional Medicare is better structured than private plans to meet that [GDP+1% growth] target without harming enrollees.” What’s ultimately needed is a body that will make the hard decisions on how much to cover new medical technologies, informed by the best science (comparative effectiveness research, if not cost effectiveness analysis). Medicare, under guidance by the IPAB, could be that body. Private plans do not appear to be well suited for the role.

“[M]ost elements of competitive system that voucher advocates seek already exist [in Medicare Advantage].” What does not exist in MA is the linking of plan bids to a level of premium support paid to all plans. That is, it lacks a competitive bidding structure, which some premium support  proposals would insert. It is a good idea, particularly if traditional Medicare is protected while the ACA reforms can play out. That is precisely the version of competitive bidding that was included in the ACA before it was amended by the reconciliation bill. That it was so quickly overturned, and the history of MA and competitive bidding, suggest that Congress has a great deal of difficulty sticking to a sound, competitive design for comprehensive Medicare plans. It is not clear why we should expect anything different under a new flavor of premium support.

“[Recent premium support proposals] threaten to shift costs to the elderly and disabled.” Both the Domenici/Rivlin and Wyden/Ryan proposals rely on a GDP+1% growth rate cap to control costs if competition doesn’t suffice. That’s the source of the cost shift threat. To be fair, the Wyden/Ryan proposal says that cost growth above the cap will be brought in line by reducing “support for the sector most responsible for cost growth, including providers.” However, this is vague to the point of meaninglessness. If health sectors could be so controlled by Congress, traditional Medicare, under guidance by Congress, would do so, which would make the GDP+1% rate growth cap nonbinding. If the cap is binding, it is Congress that needs to spread the pain around the responsible health sectors anyway. So, pending further detail, I don’t buy that beneficiaries won’t be harmed.

Having said all that, premium support concepts based on competitive pricing of Medicare plans are worth considering, but not at the expense of other cost control ideas and not without additional safeguards for beneficiaries. Premium support, in some form, is an arrow we should keep in the quiver. But it is my personal view that it is not one we should fire today.


Hidden information below


Email Address*