Medicare Reform: Rhetoric versus Substance

It seems a shame to spend our time discussing whether a particular plan replaces Medicare or reforms it; or whether or not that plan is genuine, honest-to-goodness ‘premium support.’ But names help frame debates; and in politics framing is, if not everything, then a very big deal. Whatever else it may be, Medicare is high-stakes politics.

When Bob Reischauer and I coined the term ‘premium support,’ we did it precisely to distinguish the plan we described from ‘vouchers.’ Many people were suggesting that Congress should replace ‘Medicare’ with ‘vouchers.’ Medicare was—and is—a defined benefit program. The benefit is access to defined services at a pre-specified cost. The value of that benefit automatically rises with the cost of covered services. The value of vouchers, in contrast, is linked to an index independent of health care costs. In practice, the index is chosen because it is expected to grow less rapidly than the cost of health care. If the elderly and disabled had a voucher they could choose the insurance plan that suited them best. And if they had to pay all of any cost above the voucher, enrollees, it was hoped, would have increased incentives to enforce efficient production of health care services.

Initial voucher proposals were silent or vague on regulation. This gap signaled that sponsors didn’t think that regulation was very important. It also suggested that the resulting market would be regulated no more aggressively than were the insurance markets of the day, which is to say ‘not much.’

In short order, the term “voucher “ got a rather bad ‘rep.’ In some measure, the stigma was elicited by Medicare advocates leery of any fundamental changes. But critics raised substantive concerns. As some voucher opponents put it, the elderly and disabled would be cut off from a program—Medicare—replete with rights and protections, handed checks that would cover an ever-smaller share of their health costs, and placed at the mercy of insurers and providers who were not altogether to be trusted.

As economists, Bob Reischauer and I understood the advantages of market-based competition. We recognized that vouchers might, under the right conditions, produce important advantages. The reasoning, straight out of principles-of-economics text books, is that people have different preferences regarding the risks against which they want to protect themselves. By switching from insensitive or high-cost vendors to responsive and low-cost vendors, cost- and quality-conscious consumers goad suppliers in many markets to improve quality and hold down price, and they might do so in insurance markets as well. Insurers, in turn, would have incentives to police physicians, hospitals, and other caregivers into providing improved care at lower prices.

We were convinced, however, that plain-vanilla vouchers would not deliver these benefits. We also feared that the voucher plans then on the table were more likely to harm Medicare enrollees than to help them. We saw three changes in standard voucher plans as essential.

  • First, the cash payment should be indexed to average health care costs, not to some extraneous index. For decades, advancing medical technology had been pushing up health care spending faster than prices or incomes. It seemedlikely to continue doing so. Hence, linking a voucher to a non-health index virtually guaranteed benefit cuts that we believed the old and disabled could ill afford. To be sure, competition or some other development might conceivably drive down the cost of health care for everyone. But the fundamental purpose of Medicare was—and is—to assure that the elderly and disabled receive health care not materially different from that provided to the rest of the population. If general medical cost growth did not slow, Medicare beneficiaries should not be cut off from services available to others by some mechanical formula. Either way, the right index to use was the cost of health care, not a general index of GDP or prices.
  • Second, competition could not work in unregulated markets. Enthoven, Stiglitz and Rothschild, and others had laid out the various reasons why unregulated insurance markets, in general, and health insurance markets, in particular, can and do fail. To avoid these problems, plan offerings should be limited and standardized so that people could understand their choices. Sales information and marketing should be managed by disinterested third parties—public or private—both to help consumers understand their choices and to minimize socially wasteful competition among insurers to enroll the healthy.
  • Third, we saw as imperative the development of risk-adjusted payment algorithms to further reduce incentives to compete based on risk selection. Regulation could not entirely eliminate such incentives, as insurers can always take steps to push out enrollees who turn out to be high-cost users. Joseph Newhouse years ago suggested that the best approach might blend (ex ante) riskadjusted capitation and (ex post) adjustments based on incurred costs. Bob and I didn’t know whether adequate risk adjustment was possible, but believed that it is a precondition for successful premium support.

Against this background, it seems clear to me that all of the recent proposals for modifying Medicare that advocates call ‘premium support’ differ fundamentally from the plan Bob and I called premium support. To be sure, the term ‘premium support’ is nobody’s personal property. Language evolves and changes depending on usage [see box in PDF]. Just as those squeamish about using the word ‘sex’ might prefer the word ‘gender,’ supporters of voucher plans might out a similar discomfort prefer the term ‘premium support.’

The central issues, however, are ones of substance, not terminology. And there are two such issues. The overriding question is whether the recently advanced voucher proposals are good ideas. The secondary question, if the answer to the first question is—as I believe— ‘no,’ is whether there is a good alternative.