In a post last week, I came to the conclusion that the GDP+1 growth cap in the Ryan/Wyden premium support proposal makes sense if it applies only to that program, not to Medicare as a whole. That is, the voucher level would evolve so the total cost of premium support would grow no more quickly than nominal GDP plus one percentage point. Meanwhile, under the Ryan/Wyden proposal, traditional Medicare, which would only be available to anyone currently at least 55 years old at current law levels of subsidization, would not be subject to that growth rate cap.
Except, actually, Medicare is subject to a GDP+1 growth cap. It’s in the Affordable Care Act, and it’s the job of the Independent Payment Advisory Board (IPAB) to keep Medicare within that fiscal boundary. This raises an interesting question: If Medicare is already subject to a GDP+1 growth rate cap, how much could be saved by a premium support program that is itself subject to such a cap? At first blush it seems like the answer is, “nothing.” But that’s not right.
The way the Ryan/Wyden plan works, on a market-by-market basis, voucher levels are tied to the premium of the second cheapest plan or traditional Medicare, whichever is cheaper.* In some markets the second cheapest plan will be lower cost than traditional Medicare. In such cases, premium support saves money. How much? The best estimate I’m aware of is that it would save about 5.6%, relative to current law. However, that’s a snap-shot, an estimate of the one-time savings of premium support. It doesn’t tell us anything about how the cost of premium support evolves over time, relative to traditional Medicare or current law.
So, the question becomes, in how many markets will the second cheapest plan be lower cost than traditional Medicare, not right now but over time? That question is answerable in detail with some assumptions and some data, which I do not have. But I can say this much, the number of markets in which this would occur would shrink over time if the provider payment provisions of the ACA remain in place. Those payment provisions are designed to drive Medicare payments further and further below that of private plans, on average. (See this in charts here.)
Eventually, when Medicare payments get low enough, it is not likely any private plan would be able to match Medicare costs in any market. Traditional Medicare would be the cheapest plan everywhere, and premium support would not save anything relative to traditional Medicare. On the other hand, based on the way vouchers would be set (as described above), it wouldn’t cost the government anything relative to traditional Medicare either. The extra costs of private plans would be borne by beneficiaries who enrolled in them.**
Thus, Ryan/Wyden (or similar) premium support could save money relative to traditional Medicare in the near term, but after some time it would not. This conclusion hinges on the Medicare provider payment provisions of the ACA staying in place. Another way to credit premium support with even greater savings relative to, well, something, is to assume that current law does not hold. In particular, assuming the payment provisions of the ACA do not hold, premium support saves more money (relative to that failed-ACA world) for longer. In other words, premium support seems like an even better deal for taxpayers if one assumes the ACA fails.
But notice what that means. If the ACA fails and traditional Medicare does not become the cheapest plan everywhere then some otherwise more expensive private plan sets voucher levels. That is, the plan that sets the voucher levels is one that would be more expensive than traditional Medicare if the ACA’s payment provisions did not fail (which we are now assuming they do). With voucher levels tied to such a plan, the taxpayer pays even more, not less, than if the ACA’s payment provisions stay in place. So, the best deal for the taxpayer (leaving beneficiaries aside for the moment) is if the ACA succeeds. That is also the circumstance in which premium support saves the least relative to current law.
Consequently, from a taxpayer (or federal budget) perspective only, whether you are pro or anti premium support, you should want the ACA to succeed. Of course if you’re pro premium support for other reasons, it’s advantageous to argue that the ACA will fail and use that to set the baseline against which to measure the savings from premium support. But make no mistake, assuming ACA failure as a means of justifying premium support is actually an argument for a more expensive premium support proposal. Who would be for that?
* Plans must be actuarially equivalent to traditional Medicare. See my Ryan/Wyden summary.
** The persnickety may notice that I’m dodging the possibility that traditional Medicare is cheaper than the second cheapest private plan but not cheaper than the first cheapest private plan. This post is complicated enough without that wrinkle. And I don’t think that wrinkle changes the overall thrust of my points. So, yeah, I’m dodging it in service to simpler exposition.