Why Now Is Not the Time for Medicare Premium Support

The United States faces large and growing federal
budget deficits, driven in substantial measure
by the projected growth of Medicare spending. Recently,
various groups have proposed solutions they call “premium support” or “defined
support.” A study panel of
the Bipartisan Policy Center,
chaired by former senator and
budget committee chair Pete Domenici
(R-NM) and Alice Rivlin,
former director of the Office of
Management and Budget and the
Congressional Budget Office, proposed
one variant. Senator Ron
Wyden (D-OR) and Representative
Paul Ryan (R-WI) proposed a
similar plan.

These proposals would offer Medicare beneficiaries vouchers toward the purchase of private insurance or traditional Medicare. Private-plan offerings could vary, but the actuarial value of these alternatives would have to be at least equal to that of traditional Medicare. Increases in the amount of the voucher would be capped by an index that is expected to rise more slowly than health care costs. Advocates claim that cost-conscious enrollees and competition among profit-seeking insurers would hold down program costs. But if they didn’t, the growth cap would do so by shifting costs to the elderly and disabled.

Is premium support along the lines now being proposed a good idea? Is now the time to be making fundamental changes in Medicare? We believe that the answer to both questions is no.

The idea of premium support dates from 1995. The health care reform effort of President Bill Clinton had just failed. Medicare spending was increasing faster than general health care spending. Various groups proposed replacing Medicare with vouchers in the hope that competition would hold down spending growth.

Many observers believed that simply dropping money on the Medicare population and asking them to fend for themselves in the famously dysfunctional individual insurance market was a recipe for disaster. In 1995, [Henry] Aaron and [Robert] Reischauer described elements of a newly christened plan — “premium support” — that could minimize these risks.[1] The first element was that to avoid systematic cost shifting to the elderly and disabled, the initial voucher level should be at least as generous as Medicare, and its value should grow as rapidly as overall per-person health care spending. Second, insurance offerings should be rigorously regulated. Only a few plan types should be offered so that choices would be comprehensible. Sales should occur through a government agency or nonprofit organization that would provide objective information and counseling to customers. Third, “risk adjustment” — cross-payments among health plans to offset the differences in the risk characteristics of enrollees — had to be good enough to discourage “cream skimming,” the insurance-company practice of competing by enrolling people who are expected to generate below-average costs.

Current plans do not contain these elements, and even if they did, there is little prospect in the contemporary political environment of enacting or sustaining them. Whatever virtues such a plan may have had in 1995, circumstances have changed.

For one thing, the United States has embarked on a program of health system reform that promises to control the growth of spending and that will provide additional information on whether premium support could work for Medicare enrollees. The Affordable Care Act (ACA) sets in motion a process of experimentation and change, including the implementation of accountable care organizations, bundled payments, comparative effectiveness research, a center for innovation, and an independent payment advisory board that aim to transform the financing and delivery of health care. The ACA also reduces reimbursements to certain providers, cutting Medicare spending by $500 billion over the next decade. No part of the ACA puts the burden of cost control on beneficiaries.

Moreover, current law already restricts the growth of Medicare to the same target offered by Domenici–Rivlin and Wyden– Ryan. Traditional Medicare is better structured than private plans to meet that target without harming enrollees.

And in fact, most elements of the competitive system that voucher advocates seek already exist. The results to date are not encouraging. Medicare enrollees may choose among traditional Medicare and many private plans, including health maintenance organizations. Although these private plans have been paid more than the average cost of traditional Medicare, three fourths of the enrollees have remained under the traditional system. Even if one subtracts out these extra payments to private plans, their costs have exceeded those of traditional Medicare — by 2% in urban areas, 6% in rural areas.

In addition, the ACA creates state health insurance exchanges much like the ones that premiumsupport advocates envision for Medicare. The state-based exchanges will enroll an estimated 29 million people, many receiving subsidies (vouchers) to defray part of the cost of insurance from private plans. It would be rash to move the Medicare population into exchanges before the problems of setting up the ACA exchanges have been solved. The Medicare population is larger, at nearly 50 million, and far more challenging than those covered by the ACA, since many Medicare beneficiaries have mental handicaps or some degree of dementia. Getting the ACA exchanges up and running will provide additional information to guide a decision on whether they can handle Medicare enrollees as well.

Within and beyond the health care realm, the political opposition to government regulation has intensified. There is no immediate prospect that Congress will enact and sustain the rigorous regulation — that is, limits on plan offerings and supervision of sales — necessary to make competition among insurers and traditional Medicare work efficiently and fairly. To date, insurers have been able to outfox the best riskadjustment algorithms. Consequently, even if premium-support ideas look promising on paper, the political economy of Medicare offers ample reason for caution. Cream skimming and beneficiary confusion in today’s Medicare Advantage program justify caution regarding premium support.

Advocates of premium support claim that Medicare Part D, which has a premium-support structure, shows that competition holds down spending and that beneficiaries make wise choices. Their claims are unjustified. Although Part D drug spending per enrollee is lower than was initially forecast, non-Medicare drug spending is even further below past projections.[2] Furthermore, enrollees have, on average, chosen plans that exposed them to greater financial risk than the best options available to them.[3] Most important, because Part D has no public option, it cannot provide evidence on whether private plans are better or worse than a government plan would be.[4]

So, although it’s true that Medicare is a key driver of long-term federal spending, we don’t believe that recently proposed premium- support reforms are the solution. They lack safeguards for beneficiaries. They threaten to shift costs to the elderly and disabled and force them to shop for coverage in a confusing insurance market. And the ability to run health exchanges for the Medicare population is currently in doubt.

All of this is not to say that there aren’t possible changes that could improve Medicare.[5] The addition of administrative staff could reduce improper payments and better enforce Medicare’s own coverage policy. Implementing these and other changes within the framework of the current program should make up today’s reform agenda. Serious efforts to control the growth in Medicare expenditures should begin with resolute implementation of the ACA and include measures to strengthen Medicare — both the traditional program and the existing competitive structure. Whatever the future may hold, now is not the time to expose Medicare beneficiaries to the upheaval of a shift to premium support.


[1] The Medicare reform debate: what is the next step? Health Aff (Millwood) 1995;14:8- 30.
[2] Park E. Lower-than-expected Medicare drug costs reflect decline in overall drug spending and lower enrollment, not private plans. Washington, DC: Center on Budget and Policy Priorities, May 6, 2011 (
[3] Abaluck J, Gruber J. Choice inconsistencies among the elderly: evidence from plan choice in the Medicare Part D program. Am Econ Rev 2011;101:1180-210.
[4] Frakt A, Pizer S, Feldman R. Should Medicare adopt the Veterans Health Administration formulary? Health Econ 2011 April 19 (Epub ahead of print).
Berenson RA, Holohan J. Preserving Medicare: a practical approach to controlling spending: timely analysis of immediate health policy issues. Washington, DC: Urban Institute, September 2011 ([5]