• Premium support proposal and critique: Proposal

    This post is part of a series. If you haven’t read the prior posts in the series, you really should. The introduction explains what I’m doing and links to all posts to date. This post is a proposal for a specific type of Medicare premium support program. I will answer questions about it and critique it later in the series.

    Any premium support that satisfies the goals listed at the end of last post must have the following specific features.

    Inclusion of private plans and a public option (traditional Medicare, TM). There is no reason to exclude TM from a premium support program. The whole point is for many plans to compete (in a sense defined below). TM has many features that make it an attractive player in a competitive market. By virtue of its large market share today (75% of beneficiaries) in some markets it can set prices below those of all private plans without alienating very many providers. (In others, some private plans obtain lower prices.) That translates into relatively lower taxpayer cost and cost growth, a virtue that should be maintained. Hereafter, unless otherwise specified, “plan” means either TM or a private plan. In other words, TM is “just” another plan. (Further reading related to these points here and here.)

    All plans must provide at least the standard TM benefit. Plans may innovate within whatever set of boundaries deemed sensible (I’m not defining them here), but must at least provide the standard TM benefit. Plans may offer additional benefits beyond the minimum standard. That benefit standard must include some notion of network adequacy. Plans can’t set payments so low so as to exclude too many providers. This holds for TM as well. Note that this puts a constraint on how low prices can go for any plan. Requiring all plans to meet one minimum standard maximizes beneficiaries’ ability to meaningfully compare plans. (Further reading on this point here and here.)

    Replace the current set of arbitrary premium support levels for TM and MA with a competitive bidding system. It’s important to recognize that we have a type of premium support today. I know of nobody who can say very many good things about it. In contrast to the administered pricing system that exists today, the level of premium support should instead be set, market-by-market, by some function of bids submitted by TM and private plans (e.g. minimum bid, second lowest, average). It’s simplest to think of the support level set at the minimum bid, but that’s not an absolute necessity. Setting it at the minimum bid protects taxpayers from overpaying for benefits, as some think they do today for MA plans and others think they do for TM. Setting support at a market, rather than national, level (i.e. tying it to the lowest bid locally, not nationally) accounts for geographic variation in health care costs.

    Bids reflect the plan’s cost of providing the standard TM benefit. Note that under any version of this bidding system, there is at least one plan that is as at least as affordable to beneficiaries as is TM today, however that plan may not be TM in all markets. Plans that cost more than the minimum bid or offer benefits richer than TM must charge beneficiaries a premium for that additional cost or coverage. Thus, inefficient plans suffer in the marketplace. Benefits that are of value to beneficiaries are likely to be offered even if they are not available from TM. However, beneficiaries would pay out of pocket for additional benefits. If this is considered unfair to beneficiaries or to TM then it suggests that the minimum benefit should be enriched. This brings front and center a debate about what the minimum benefits of Medicare should be, a stark contrast to the arbitrary variation of benefits available through MA today and the slow-to-update set of benefits available through TM. (More on these points here.)

    Provide low income subsidies. No beneficiary should be unable to obtain coverage for necessary health care due to income. Thus, the level of premium support should vary by income so that lower income beneficiaries can afford the coverage they need.

    Risk adjust premium support. Plans that enroll beneficiaries of higher (or lower) health risk should not be penalized (rewarded) for doing so. Hence, premium support should be risk adjusted. There is great concern among many that risk adjustment will be insufficient to prevent private plans from cream skimming. This is a serious and legitimate issue to which I will return later in the series. Whatever body oversees the program should use every tool necessary to address it including all manner of prospective and retrospective risk adjustment, risk corridors, risk sharing, reinsurance, auditing to check that plans’ actual costs are within reasonable bounds of risk-adjusted payment, etc. Plans found to be gaming the system on selection should be ejected from the program.

    Additionally, greater transparency should be required of private plans. The same level of information about utilization that is available from TM for the beneficiaries it covers should be made available by private plans. Researchers should have access to those data to they can independently check performance of the program, including analyses of selection (with all due privacy protections, of course). If a plan is receiving public funds, as they all would, the data that documents the use of those funds should be just as public as that of TM.

    TM must be afforded flexibility. Because it is governed by Congress to a larger extent than private plans, TM is unable to innovate in the same ways private plans can. This is unfair and unnecessary. TM should be liberated from such constraints and shielded from politics. The IPAB is a good model and something like it should govern TM, if not the entire premium support program. Moreover, all other provider payment reforms embodied in the ACA and any future ones dreamed up by the IPAB should be permitted to go forward. Despite what many pundits claim, those payment reforms are not the antithesis of a competitive program. That’s a false distinction. They are what will help TM be a competitive, cost efficient plan, one that could give private plans a run for their money. That should be encouraged. (More on these points here.)

    AF

    Share
    Comments closed
     
    • I think that this could be an excellent way forward, though there are lots of details that can make or break it. I will suggest one wrinkle, based on my experience with the Swiss insurance system. Like your proposal, plans in Switzerland are required to offer a fixed minimum package of benefits and are allowed to offer additional benefits in return for additional premiums. The wrinkle is that if you wish to purchase additional benefits, you actually purchase separate, supplemental policies for various groupings of benefits. I have found this to be extremely helpful in helping to understand exactly what I am paying for, and in judging the cost/benefit of the various options. Helping my mother sort through MA options, I have found it very difficult to understand exactly what she is buying.

    • I disagree that the optimal system should include only plans that provide the standard benefit. If the standard benefit were actually minimal, I might agree, but as it is, every year the governments (both national and state) add more and more requirements to insurance plans forcing them to cover more and more that only helps subsets of people. Is there another advantage to this requirement other than it “…maximizes beneficiaries’ ability to … compare plans.” Why is reducing the cost of comparison shopping more important than reducing costs, or providing an insurance plan that better fits their needs?