• New savings estimate of Medicare competitive bidding

    Roger Feldman, Robert Coulam, and Bryan Dowd* previously had estimated that a competitive bidding premium support program for Medicare could save 8 percent of Medicare spending (pdf). Their new estimate, described in a recently published AEI paper (pdf), incorporates increased enrollment into Medicare Advantage (MA) as well as revisions in MA payment policy embedded in the ACA. (Technical point: Due the nature of the plan bid data available to the researchers, the new estimate is based on comparing the 25th percentile of MA bids of per enrollee costs (not the lowest or second lowest) to traditional per beneficiary Medicare spending. See the paper for details.)

    The bottom line from Feldman, Coulam, and Dowd:

    Using data from 2009, we estimate that fully-implemented competitive bidding would save 9.5 percent of Medicare spending, which is more than our previous estimate. However, the Affordable Care Act of 2010 (ACA) is estimated to save 4.2 percent if it is implemented as the law requires, and so competitive bidding would save 5.6 percent more than the ACA legislation.

    That’s a one-time savings or shift in the spending curve. It’s a large shift, so I’m not dismissing it as unimportant. Far from it. We should keep options of this type on the table for discussion. But other things should be on the table too. Beyond the scope of their work is the question of how a competitive bidding premium support would alter the rate of growth in Medicare spending. Though it would clearly shift the curve, would it bend it? This is a crucial question.

    Not all curve bending (or shifting) is the same. What we should want is bending/shifting driven by increases in efficiency in the delivery of health care and health. It’s not a big leap from this ambition to consideration of comparative effectiveness. After all, it’s the very definition of inefficient to employ a treatment or mode of delivery that costs more for no more health than another. Comparative effectiveness research (CER) is intended to discover just when, where, and for whom this occurs. Big money is at stake and substantial growth. For instance, ten percent of the growth in Medicare since the mid-1990s is due to percutaneous coronary intervention (coronary angioplasty) a procedure known to be employed in many cases for which a cheaper medical therapy would do. We know this from CER, the COURAGE trial.

    An important feature of CER is that it’s a public good. Consequently, it will be (and is) underprovided by the market. The infrastructure to support and translate it into practice is a proper role for government. Funding for such an infrastructure is provided by the Affordable Care Act. The key question, again well beyond the scope of Feldman et al.’s work, is what a premium support program — or the legislation in which it is embedded — would do to that nascent public infrastructure.

    It’s this question, among a few others, that gives me pause about premium support. It’s not that it isn’t a worthy idea, yet no idea is perfect. It’s that other ideas are worthy too and some of them are already written into law and are just being implemented. Often, proponents of premium support also support repeal of that law, suggesting it’s one or the other. It’s actually a false choice. Our problems in health care are broad. We ought not discard potential solutions before they’re even attempted.

    More in my series on premium support, in the FAQ, and my NEJM paper with Henry Aaron.

    * FULL DISCLOSURE: I work or have worked with all of the authors.


    • Great post Austin. Out of curiosity, would it be reasonable to expect competitive bidding (with the addition of a public option) to generate even greater savings within the ACA framework?

      From my recollection a major conservative argument against a public option for health reform was that private insurers wouldn’t be able to compete against the government, and would go out of business. If my recollection is correct, wouldn’t this also be a problem if it were implemented in Medicare?

      Is there some serious inconsistency here or am I just missing something? Thanks, and keep up the great work.

    • There is a big lacuna in this analysis. What is the impact on benefits, plan networks and plan margins in order for this to happen? Whether the expected savings are 8% or 9% or 5%, health plans don’t have margins large enough to lower their premiums by that much across the board. The lowest priced plan in a particular region (used in this analysis) might be placing a temporary aggressive bid that can’t be sustained, so there may be an illusory quality to pegging the savings to the lowest or second-lowest bid.

      In any case, the lowest two bids in a given county are that way for a reason: lower benefits, smaller networks, lower pay to providers, etc. I’m not necessarily opposed to any of those things, but we can’t just ignore the mechanics of how a sustained price drop of 5-10 points is possible.

      • It’s a price drop relative to the lavish payments MA plans currently receive, which fund additional benefits. When pared back to the standard Medicare benefit, things are cut, hence savings. Moreover, by setting premium support at the lowest bid, anyone who wants to enroll in a plan that costs more has to make it up out-of-pocket. In that sense, it’s a shift to beneficiaries and away from the government. But, in every market, the standard Medicare benefit is available at no additional premium (beyond the standard Part B premium). Make sense? (I’ve been through all this many times in prior posts, to which I linked. So I didn’t spell it all out here.)