Mr. Chairman and Members of the Committee:

Thank you for inviting me to appear before you. I am Joseph Antos, the Wilson H. Taylor Scholar in Health Care and Retirement Policy at the American Enterprise Institute. I am also adjunct professor in the School of Public Health at the University of North Carolina at Chapel Hill.  I have previously served as the assistant director for health and human resources at the Congressional Budget Office (CBO), and earlier held several research and management positions in the Health Care Financing Administration, the precursor to the Centers for Medicare and Medicaid Services (CMS). The views I present today are my own and do not represent the position of the institutions with which I am associated.

 The chances of hammering out an agreement on reforming Medicare and providing a prescription drug benefit to 40 million seniors and disabled people seem to diminish every day. Democrats maintain that the prescription drug benefit is just not good enough, while Republicans raise the specter of program meltdown if we spend too much. Republicans see the promise of lower cost and better performance through competing health plans, while Democrats fear that competition will jeopardize the traditional Medicare program and harm millions of beneficiaries.

 There has been growing speculation in the press that the Medicare conference will not be able to reach a compromise and will need an exit strategy.  That strategy could focus on the needs of low-income beneficiaries--those with no prescription drug coverage, some of whom have very high drug costs.  A carefully designed drug benefit targeted on those most in need could be a very good investment of taxpayer dollars.  But such a program could be as controversial as the bills that are currently under consideration.

Chairman Burton has developed the Medicare Safety Net Prescription Drug Act, which provides a drug benefit to low-income beneficiaries.  In broad terms, the proposal is similar to other targeted benefit approaches.  Beneficiaries would have access to discounted drug prices, their purchases would be subsidized through a personal account accessible by a debit card, and catastrophic insurance would be provided.  The benefit would not be open to all Medicare beneficiaries, and federal outlays would be capped. The proposal is limited to a drug benefit, and does not include broader reform measures.

The Chairman’s proposal has several desirable features but also significant flaws that are shared by other similar proposals. My testimony will address those design aspects and suggest other approaches that could be more effective in reaching policy goals.

Two key points emerge.  First, full consumer choice and strong competition among health plans are necessary to assure that beneficiaries receive the best value from a Medicare prescription drug benefit.  Second, a targeted drug benefit is likely to mushroom into an expensive entitlement within a few years through future legislative expansions.  To ensure that the Medicare program will be able to accommodate future fiscal shocks, any prescription drug proposal should include elements that can form the basis for future reforms.

Designing a Low-Income Drug Benefit

The first principle in designing a Medicare prescription drug benefit is that someone will be unhappy, no matter what you do. The plethora of competing, and often contradictory, policy objectives cannot all be satisfied. For example, the House and Senate bills carve a “doughnut hole” out of the middle of the drug benefit, not because that is good policy but to keep the CBO cost estimate down to $400 billion.  This budgetary legerdemain results in a peculiar kind of drug benefit that some have labeled unfair to the poor.  However, everyone eligible for Medicare would participate.  The bills’ sponsors struck a balance between benefit generosity, budget cost, and beneficiary participation that others would dispute.  A narrower proposal like the Burton bill shifts that balance--creating new winners and losers among beneficiaries, gaining some votes and losing others in Congress.

I will focus on five major design features of the Burton proposal:  eligibility, benefit structure, benefit administration and competition, the budget cap, and prescription drug reimportation.  Although the proposal contains some innovative elements, it does not stray far from the regulatory model of traditional Medicare.  Other proposals, including variants of the bill proposed earlier this year by Congressman Cal Dooley (D-Cal.), are more promising but also fall short of the mark.

Eligibility

More than three-quarters of Medicare beneficiaries have some prescription drug coverage (see Table 1). Perhaps surprisingly, that is true at all income levels. Lower-income people are more likely to have coverage through Medicaid, while higher-income people primarily have private coverage (see Table 2).

Proposals for unrestricted eligibility for a Medicare drug benefit, such as the House bill, would displace much of the existing coverage and substitute federal taxpayer dollars for other funds that are now being spent.  The Senate bill excludes people (the “full duals”) who are eligible for both Medicare and full Medicaid benefits from the Medicare benefit.  That reduces federal outlays (which the Senate bill spends elsewhere), but leaves states with a growing liability.

A targeted benefit gives larger subsidies to low-income beneficiaries, rather than distributing that money to everyone.  The Burton proposal goes further, limiting the benefit to those who are not eligible for any other prescription drug coverage. If it could be implemented, that would be a stronger restriction than excluding those who are enrolled in some other drug benefit.

It is possible to verify the incomes of most beneficiaries through income tax forms.  But it is extremely difficult to determine if someone does not have prescription drug coverage and is not eligible for such coverage from some source. The absence of insurance does not leave a paper trail.

Table 1.  Prescription Drug Coverage of Medicare Beneficiaries by Poverty Level, 2000

% Of Federal Poverty Level  % of Medicare Beneficiaries  % With Rx Coverage

Less than 100%           24%      77%
100% to 199%            31%      75%
200% to 299%            19%      82%
300% or higher           26%      81%

All income levels         100%      79%

Note: Includes only Medicare beneficiaries living in the community.

Source:  Tabulations of the Medicare Current Beneficiary Survey from Becky Briesacher, University of Maryland.

Table 2.  Sources of Prescription Drug Coverage for Medicare Beneficiaries by Poverty Level, 2000

% Of Federal  Employer-              Other
Poverty Level   Sponsored      Medigap     Medicare HMO     Medicaid     Public

Less than 100%     12%     7%      13%        42%         10%
100% to 199%      28%    12%      19%          9%  8%
200% to 299%      45%   14%      20%          2%  4%
300% or higher     49%   15%      14%          1%  4%

Note:  Beneficiaries may have more than one source of coverage.  Includes only Medicare beneficiaries living in the community.

Source:  Tabulations of the Medicare Current Beneficiary Survey from Becky Briesacher, University of Maryland.

Even if we could enforce such eligibility restrictions, we might not want to.  Some people may be eligible for private coverage but cannot afford to participate.  Others may have limited coverage that provides no catastrophic protection.  Some people with incomes too high to qualify for the Burton program might have high prescription drug expenses and no coverage.  Future Congresses would be tempted to loosen the eligibility limits to accommodate many of those people, just as they would be tempted to expand benefits and fill the doughnut hole under the Senate and House proposals.

Benefit Structure

The House and Senate bills offer traditional first-dollar coverage for prescription drugs.  After a modest deductible ($250 in H.R. 1 and $275 in S. 1), beneficiaries would have a significant fraction (80 percent in H.R. 1 and 50 percent in S. 1) of their prescription drug costs paid by the government.  This is a use-it-or-lose-it benefit, and enrollees will have a powerful incentive to use it after they’ve paid $420 or more in annual premiums.

The Burton proposal changes that incentive.  Instead of first-dollar insurance coverage, beneficiaries would be required to pay a high deductible (perhaps $3,000) before catastrophic insurance covers their prescription drug expenses.  To help them pay the deductible, beneficiaries would be given a cash subsidy paid into individual accounts and accessible using a debit card.  Under this approach, beneficiaries would be sensitive to prescription drug costs and still be protected financially. Beneficiaries would use their own money (from their accounts and out-of-pocket payments) to pay for drugs until they met the deductible.  Any amount left in their accounts would roll over for use in the next year.

The drug account would not become a permanent asset for beneficiaries, however.  If an enrollee lost eligibility for the benefit or passed away, his cash balance would revert to the Treasury.  Although that may seem fiscally prudent, it undercuts the beneficiary’s incentive to limit unnecessary spending.  Beneficiaries might not be as diligent in selecting lower-cost pharmaceuticals if they felt that the account balance could be taken away at any time.

Benefit Administration and Competition

The prescription drug benefit under the Burton proposal would operate much like a Part B benefit.  That is, a division of the U.S. Department of Health and Human Services (HHS) would determine what drugs would be covered, set their prices, certify participating pharmacies, determine the eligibility and level of subsidy available to beneficiaries, establish the personal drug accounts, and issue the debit card.  Private entities would be contracted to handle administrative functions and pay the bills, just as large insurance companies acting as Part B carriers do today.

The proposal asserts that the Secretary of HHS would negotiate pharmaceutical prices, but those negotiations would quickly become rate setting exercises similar to the way physician fees are set.  It is clear that any negotiations that did occur would be the exception rather than the rule.  Since Medicare beneficiaries use every available pharmaceutical, that means negotiating tens of thousands of prices for products in every dosage form, strength, and packaging.  Prices for drugs would be set in an unspecified manner the first year.  Almost certainly, HHS would establish an inflation factor that would ratchet up the entire price structure in subsequent years.

Negotiations would be necessary whenever a new drug appeared on the market.  The Secretary would be able to withhold access to any new pharmaceutical, a powerful threat that could lead to low prices for new drugs under Medicare.  However, there are bad side effects with this policy prescription:

• Competition from generics and therapeutically similar drugs would no longer force down prices of branded drugs under a rigid federal price structure.
• Delaying the entry of a new drug onto the federal formulary would be politically difficult and could hurt some patients.
• If Medicare set prescription prices at very low levels, manufacturers are likely to raise prices to private purchasers, including most people under age 65.  The proposal includes a reimportation provision, discussed below, to limit that possibility.
• The threat of a low launch price would deter the research and development of potentially valuable or life-saving drugs, particularly those that treat illnesses associated with older age groups.

The budget savings from top-down regulation are immediate and seductive.  But the consequences of such an approach are long-term and serious, discouraging the research and development that could lead to more effective and potentially cost-saving drug therapies.  Even in the near term, lower prices for Medicare could mean higher prices for everyone else.

A competitive approach can strike a better balance between lowering prices and promoting innovation.  This is the conceptual basis of the House prescription drug provisions, and it relies on the proven ability of competing private plans to negotiate substantial discounts and manage the cost of the benefit.

If private drug plans are placed at risk for the cost of providing prescription drugs to their Medicare enrollees, they have a strong incentive to limit cost growth.  The plans can act on that incentive if they are given the flexibility to manage the benefit aggressively.  With a wide choice of plans, beneficiaries will be able to select a plan that meets their needs—and change plans if they are dissatisfied.

A number of proposals, including the Medicare Rx Now Act proposed this year by Congressman Dooley and the Medicare Rx Drug Discount and Security Act proposed last year by Senator Chuck Hagel (R-Neb.), rely on private plans to deliver a benefit structured like that of Chairman Burton’s bill.  The earlier proposals would make drug benefits available to all Medicare beneficiaries, rather than targeting those with low incomes.  They merit discussion because they use a modified form of plan competition.   

The Dooley and Hagel bills would offer the new benefit through many competing plans, but those plans would not be liable for excess costs for beneficiaries who exceed the catastrophic spending limit.  Instead, the federal government would pay all costs of catastrophic coverage on a fee-for-service basis.

Although that might seem to be a reasonable split of private and federal responsibilities, such an approach is a short step from the situation posed by the Burton bill.  The federal government could pay whatever each plan asked, but then some plans would be paid more than others for the same prescription drug purchases.  To prevent unfairness and potential fraud, HHS would probably establish a federal price list for all pharmaceuticals.  Reimbursement based on federal, rather than actual, prices would potentially ignore real cost differences faced by the private plans.  The likely result would be an increasingly complex pricing system, as the price schedule is modified to take account of special circumstances.  If federal prices lagged behind actual prices (likely if Congress faces budget pressures and holds down Medicare reimbursements), private plans would drop out of the program and the demand for a fully federalized system might be irresistible to policymakers.

An alternative approach would solve many of the problems posed by proposals like the Dooley bill.  First, the restrictions on how private plans could manage their drug benefit should be relaxed.  The Dooley bill assumes that many kinds of plans would participate, including employer-sponsored retiree plans.  However, the regulatory requirements imposed on such plans would force such plans to revamp their benefits and methods of operation, making participation in Medicare impractical at best.  Second, plans should be placed at financial risk for the catastrophic insurance benefit.  To assure a stable system and encourage plan participation, a national reinsurance pool run by the participating plans could be organized for all Medicare drug plans.  Such a pool would spread excess costs among all beneficiaries, eliminating the incentive to avoid enrolling sicker seniors.  Premiums would include the cost of those high expenses averaged over all Medicare beneficiaries.  That would protect private plans from the problems of adverse selection and provide greater incentive for plans to participate in the program.  This arrangement would offer the advantages of a national risk pool without the threat of price controls and limits on beneficiary choice posed by fully federalized benefits.

Budget Cap

The prescription drug benefits proposed in the House and Senate bills are open-ended entitlements and are likely to cost far more than $400 billion by 2013.  CBO has produced its best estimate of federal outlays, but the likelihood of higher spending is greater than the reverse.  Moreover, political pressure to fill the doughnut hole in the benefit is likely to be irresistible, and future benefit expansions could easily double the level of actual outlays.

The Burton proposal includes a hard cap on spending in the hope of preventing higher outlays resulting from either estimating errors or legislative action.  If spending was projected to exceed the cap, the Secretary would increase the catastrophic coverage threshold or reduce the federal subsidies provided through the prescription drug accounts.

Spending caps do not work.  Nothing can prevent a future Congress from changing the law and wiping out the cap.  If the cap was binding, that would reduce the value of the drug benefit to low-income people--a difficult situation to sustain politically, particularly if further reductions would be necessary year after year.

Medicare currently has a spending cap on physician payments called the sustainable growth rate.  If physician spending rises above that growth rate, fees are to be reduced. In 2002, Medicare reduced physician fees by 5.4 percent across the board.  That caused a reduction in access to physician services in some parts of the country and complaints from doctors everywhere.  A further reduction was scheduled for 2003.  In February, Congress modified the payment formula to give physicians an increase, and there is substantial sentiment for additional relief in future years.

This is a clear case in point.  The sustainable growth rate was popular only when it was not acted upon.  We can expect no more from a spending cap on prescription drugs.  Rather than trying to control spending with caps, we should design a program that gives beneficiaries more control--and more personal responsibility--over their health care.  Personal accounts and high deductible insurance provide some incentive for prudent purchasing, but other restrictions in the Burton bill limit their impact.  More plan options and effective competition are needed if we expect to limit spending without limiting the value of the benefit.

Reimportation

The Burton bill would apply additional pressure on the pharmaceutical industry by authorizing the importing of drugs (often called reimportation).  Importers who resell the drugs would be responsible for ensuring that the imported products are genuine and safe.  This provision is intended to lower prices of drugs in the market generally, not just those dispensed under Medicare.

Supporters of a reimportation provision point out that U.S. residents pay the highest drug prices in the world, exceeding the prices found in Germany, France, and other developed countries.  Those countries threaten to produce their own versions of branded drugs (“compulsory licensing”) unless pharmaceutical manufacturers agree to sell products at very low prices.  Since the cost of developing a drug is very high (perhaps as high as $800 million) and the cost of manufacturing it is usually fairly low, manufacturers are ahead in the short term as long as their costs of production are covered.  Such prices would not compensate for the huge costs of research and development required to get a new drug to market, thus discouraging future research.

The U.S. government also negotiates low prices for prescription drugs, but those prices apply only to federal programs such as the Veterans Affairs health program.  Reimportation is an indirect way of establishing price controls in the U.S. for all pharmaceuticals sold in all markets.

The problems with price controls were discussed above, and those comments apply in full to reimportation.  The likely impact of reimportation may be disappointingly small, however.  Middlemen are likely to absorb most of the price reduction that might be possible by importing drugs at a lower price.  People with employer-sponsored health plans are unlikely to see reductions in their drug costs because such plans typically have multi-tiered copayment systems, and savings that might occur are likely to be retained by the employer.  People without drug coverage stand to benefit more, but even in this case we should expect to see an uneven pattern of discounts rather than across-the-board reductions of any significant magnitude.

Conclusion

A targeted low-income benefit could be a feasible alternative if the Medicare conference stalls.  Such a benefit would avoid displacing the good coverage that many beneficiaries now have and are fearful of losing.  It could provide important financial support for people who need help the most--those with a limited ability to pay for their prescription drugs, those with high drug costs, those without insurance coverage.  Combining a discount card, a cash subsidy in a personal account, and catastrophic insurance would provide some of the elements for a sensible benefit for millions of seniors.

It would be a mistake to create even a limited Medicare drug benefit that repeats the mistakes of the past.  Attempts to limit federal cost by overall spending caps, price controls, and restrictions on beneficiaries and providers lead to worse health outcomes, and at best have only a temporary ability to hold down spending.  Competition among drug plans, with flexibility to design their benefits and negotiate their best prices, will lead to more effective cost control.  Proposals that give beneficiaries more purchasing power and more choice will result in better value and provide the basis for future improvements in the Medicare program.

Joseph Antos is the Wilson H. Taylor Scholar in Health Care and Retirement Policy at AEI.