Medicare Advantage benchmark
August 15, 2012 5 Comments
Reihan Salam has a nice post looking more closely at the cost of the current Medicare Advantage program and describing how premium support with competitive bidding in Medicare could work better to lower costs. He quotes Ezra Klein as saying competition hasn’t worked well to hold down costs in Medicare, and Reihan rightly points out that the way in which Medicare Advantage plans are paid by Medicare muddles this story. I would restate what Ezra said as private insurance options in Medicare have not worked well to hold down costs in Medicare. And it has everything to do with how Medicare Advantage plans are paid by Medicare.
Reihan notes that currently there is a county level benchmark that serves as the basis for how much Medicare pays private insurers who offer Medicare Advantage (MA) plans (if a plan offers a premium that is lower than the benchmark, patients get a rebate; if higher, they must pay the difference). So, there could be some competition in MA, but it is set around an administratively set payment. It is important to note that the county-level benchmark levels are based on historical Medicare Fee cost of Medicare in a county.
For the early history of private plans in Medicare (starting around ~1982, I believe implemented via a TEFRA 82 demonstration but I can’t dig it up right now) the payment levels from Medicare to private insurers were set at 95% of the average adjusted per capita cost of FFS Medicare in a given county. This did not save Medicare money because it turned out that generally healthier persons choose private insurance options, and were beneficiaries who would have cost less than 95% of the average cost in their county had they stayed in FFS Medicare. As you might guess, private plans proliferated in high cost areas, and often didn’t offer coverage in low cost ones. Payment rates from Medicare to private plans were boosted over time, with a goal of stimulating private insurance options and enrollment in same. Per MEDPAC on the current benchmark:
The benchmark is a bidding target. The local MA benchmarks are based on the county-level payment rates used to pay MA plans before 2006. (Those payment rates were at least as high as per capita FFS Medicare spending in each county and often substantially higher because the Congress set floors to raise the lowest rates to stimulate plan growth in areas where plans historically had not found it profitable to enter.)
The primary reason I have changed my mind and am willing to support a move toward a comprehensive competitive bidding option in Medicare (so long as it is in conjunction with implementing the ACA or a broader deal) is that I believe that private insurance options in Medicare will never go away for political reasons, and that the current MA approach systematically over-pays private insurance companies. Given my understanding of reality, competitive bidding has a reasonable chance of improving on the current (and longstanding) overpayment to private plans. Further, if we are going to do this, I see no reason to wait 10 years to start, especially given movement ahead on setting up insurance exchanges via the ACA, since the infrastructure necessary to implement both are similar. Perhaps a policy of saying competitive bidding in Medicare can start 2 years after a state sets up an operational ACA exchange could provide some incentive for “cold feet” states to move forward.