• Explaining Medicare’s slowdown

    This is a TIE-U post associated with Jonathan Oberlander’s Political Dynamics and Policy Dilemmas (UNC’s HPM 757, Fall 2011). For other posts in this series, see the course intro.

    Chapin White [1] dissects the slowdown in growth of Medicare spending in a 2008 Health Affairs paper. The abstract and charts tell the story.

    So-called excess growth [that above per capita GDP] in Medicare spending per beneficiary has varied widely and has slowed in recent years. The annual rate of excess growth fell from 5.6 percent during 1975–1983, to 2.1 percent during 1983–1997, to only 0.5 percent during 1997–2005. Changes in payment policies and regulations can help explain the observed slowdown. These include new prospective payment systems for hospitals and postacute care providers, and controls on aggregate Medicare physician spending. Competing explanations— increases in managed care enrollment, changes in Medicare cost sharing, and a systemwide spending slowdown—do not account for the slowdown.

    Of course the year-by-year data are not as clean as the three averages above might suggest. The following three charts illustrate excess Medicare growth for hospital, physician, and post-acute spending, by year and five-year averages. The key payment policy milestones are drawn as dashed vertical lines. It is these that White credits for the slowdown. This is not mere rhetoric; he does some calculations to show that the other factors mentioned in the abstract cannot explain much, if any, of the slowdown. In particular, “trends in Medicare spending growth can be and have been moderated through means other than increasing beneficiary cost sharing.” More skin in the game may not be necessary.

    Perhaps this is not really a Medicare story, though. Maybe Medicare is the beneficiary of a system-wide slowdown. White rejects this hypothesis, illustrating that excess growth in health spending for the nonelderly has increased while that for Medicare has decreased.

    I’m particularly fond of one of White’s “avenues for future research”:

    [D]o payment-driven savings in Medicare increase or decrease private-sector health spending? One theory is that reducing spending in Medicare is like “squeezing the balloon,” with Medicare savings offset by increases in other sectors.

    We know the answer to that question.


    [1] Chapin White. 2008. Why Did Medicare Spending Growth Slow Down? Health Affairs 27: 793-802.


    • For 1997-2005 we are at GDP + 0.5

      Given that FFS Mcare and volume (as opposed to prices) is an invoked culprit for systems ills, what does this say about aggregate spend? Put aside the pockets of high volume (Miami, NYC, etc), and overall, are the feds looking under the wrong stone?

      Your final sentence is the crux of course, but it is curious that on Monday, the problem is the cost shift, and on Tuesday, its the runaway costs in Mcare. Its dissonant.

      I suppose that is his point.


    • -Interesting claims. Have the policy shifts really translated into Medicare patients having “more skin in the game?” Seems like the argument is that demand curves slope downward everywhere but Medicare (e.g. this seems inconsistent with the RAND study and everything else) and that the system wide slowdown in medical spending even when controlling for changes in insurance levels hasn’t influenced seniors in the same way. Perhaps other literature will support this conclusion but I’d be surprised.

      -Given that there’s convincing data that suggests that market power increases pricing power (not sure why this had to be re-established specifically for the health-care market, but anyhow…) and evidence that providers are aggregating in response to ACA/ACO – how credible are the claims that using the ACA to shift providers into ACO’s will actually reduce system-wide (unit) cost growth?

    • Medicare spending essentially doubled between 1990 and 2000k and then doubled again from 2000 to 2010.

      This occurred despite some decent successes in cost control, especiallly compared to the freight-trein runaway growth before 1984.

      Remember that Medicare has a per-person spending amount, which is growing at less than 4-5 percent, and then has a demographic factor, the number of beneficiairies, which is growing relentlessly at 3-4 percent as more Americans turn 65 or reach 2 years of disability status.

      As a resilt, you get 7% growth in Medicare even when you do have cost comtrol.