CMS currently pays Medicare Advantage health plans on a method that relies on the difference between a benchmark for a county based on fee-for-service Medicare costs in that county, and the plan’s bid. If a plan bids below the benchmark, it actually gets a rebate which it must use to provide either lower premiums or more benefits to beneficiaries. Plans increasingly are offering zero premium plans to entice beneficiaries. A Commonwealth Fund analysis looks at whether CMS may be paying plans more than it has to. (Comm. Brief) CMS and Congress have already reformed the plan reimbursement once to try to limit payments by reducing the benchmarks. The Commonwealth Fund researchers focused on whether this change seems to have actually reduced payments to a true competitive level. What they found was that in areas with higher benchmarks, plans seem willing to accept somewhat higher input costs, but still are able to make more in rebates, and probably profits. Out of every dollar in a higher benchmark, plan costs are 32 cents higher and rebates are 52 cents greater. High-cost fee-for-service markets see this effect most strongly, as we would expect. It appears that MA plans, even after adjusting for different input costs like hospital wages in different areas, pay more for care in higher benchmark regions. I suspect this is really due to hospital market power in those regions, which may be partly, but not completely offset by plan market concentration.
The authors suggest maybe using a national benchmark to reduce some of this gamesmanship. It seems to me that the simplest thing for CMS to do is to move to a pure competitive bid system. The lowest bid by an MA plan in a geographic area would in essence become the benchmark and all plans would be reimbursed based on that bid. That is the best way to ensure that plans do everything possible to lower their input costs and to minimize what Medicare pays. It would require that beneficiaries have adequate information to compare plan packages and premiums. The other approach would be to voucherize the whole Medicare program. Give every beneficiary a voucher based on average costs, adjusted for health status, and the value of the voucher increased at no more than general inflation from year-to-year. They could use the voucher for an MA-type plan, or they could enroll in a fee-for-service plan administered by one of the Medicare carriers or intermediaries, but they would be responsible for costs over the voucher. This would encourage more beneficiaries to go to an MA plan, where quality and care management is generally superior to that in fee-for-service Medicare.