Of Sausage-Making and Medicare
By Joseph Antos, John E. Calfee
Posted: Monday, January 26, 2004

HEALTH POLICY OUTLOOK
AEI Online (Washington)
Publication Date: January 1, 2004


The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 created a new prescription-drug benefit and opened Medicare to broader competition among health plans. While these major reforms promote market principles, they do so with heavy regulation: private plans will be given more opportunity to offer services to seniors, including subsidized drug plans, but with restrictions that could blunt the effectiveness of competition in restraining cost growth. The new law also tweaked the Medicare system in a number of ways, continuing a long tradition of creating new regulatory complexities in a vain attempt to ameliorate the distortions caused by old rules.


Bismarck famously compared the mysteries of democratic law making to the messiness of sausage manufacturing. Those mysteries are most keenly felt when the result is both massive and manufactured in haste. By that standard, the Medicare Prescription Drug, Improvement, and Modernization Act of 2003, signed into law in early December 2003 and dubbed MMA for the "Medicare Modernization Act," merits a close and skeptical examination of its ingredients and likely consequences.

The new Medicare law was enacted following two contentious votes in the U.S. House of Representatives, each yielding a razor-thin margin in the wee hours of the morning amid complaints that members were voting on a bill they had not read and did not understand. Outside the Washington beltway, most people are only dimly aware of what has changed in Medicare, and many seniors do not even know that a law was passed. During the week after the president signed the Medicare bill, four out of ten seniors surveyed by the Kaiser Family Foundation either did not know the bill had passed or thought that the bill had failed in Congress.[1]

What most people do know is that a drug benefit has been added to Medicare. But legislation that fills over four hundred pages of text plus hundreds of pages of explanation is loaded with provisions that in large and small ways could change the landscape of Medicare and reverberate throughout the health system.[2] Although the drug benefit has received most of the headlines, an equally important matter is whether the legislation does anything to retard or reverse the long march toward a single-payer health care system in which virtually all citizens pay for each other's health care through government revenues and expenditures.

The journey to a single payer has been largely completed in Canada (where only pharmaceuticals occupy a vestige of private markets), Western Europe, Japan, Australia, and New Zealand. Following decades of expansion of Medicare and Medicaid and the creation of smaller government health programs, the addition of a Medicare drug benefit could have propelled the United States along that route.

The drug benefit is indeed a major expansion of the government health entitlement, but the impulse toward a single-payer system has been partly offset by the requirement that the new benefit operate through competing private plans. Still, the long-run effectiveness of competition in the drug benefit is far from certain. The system could very easily careen toward ineffectual cost-plus operations that lead relentlessly toward uncontrolled expenditure growth followed by the extension of Medicare price controls to the pharmaceutical sector.

MMA includes measures to strengthen competition in the rest of Medicare by revamping the role of private health plans and pointing the way to other, more modest changes in the traditional Medicare program. Those provisions are important, but their success in improving the functioning of the Medicare market will be limited by the climate of over-regulation that has historically characterized the program.

Perhaps the most promising component of the new law has nothing to do with Medicare. Health Savings Accounts (HSAs) create new incentives for people under age sixty-five to become more careful health care consumers and gives them opportunities to save for their own health needs. The potential significance of this provision for improving private health insurance markets is only now becoming clear. A later Outlook will more fully analyze HSAs.

The New Drug Benefit

The Medicare drug benefit rolls out in 2006. Beginning in May 2004, however, Medicare beneficiaries will be able to purchase a government-approved drug discount card. The cards provide no insurance, but low-income beneficiaries with no other source of prescription-drug coverage will also receive an annual $600 subsidy toward their purchase of pharmaceuticals. The discount card program ends when the full drug benefit begins operation in January 2006.

The federal Centers for Medicare and Medicaid Services (CMS) expects the cards to provide modest discounts on the order of 10 to 15 percent. In addition to negotiating volume-based, across-the-board discounts, card sellers could obtain selective discounts by constructing formularies and negotiating steep discounts for favored brands. Average prices, however, may be no better than those offered by Wal-Mart and other mass retailers. Moreover, the selective discounts predicted by CMS are no greater than those already available from cards offered by pharmacy benefit managers (PBMs) and others.[3] This may be why CMS expects only about 7 million Medicare beneficiaries to enroll in the discount card program, with two-thirds of them attracted by the $600 cash subsidy.

The Drug Benefit for 2006 and Later. The Medicare drug benefit itself will begin operation January 1, 2006, with the addition of a new Part D to Medicare. The Part D drug plans will be voluntary and will be privately run. The plans will come in two forms. New drug-only insurance plans will be designed for beneficiaries in traditional fee-for-service Medicare. Comprehensive health plans, such as preferred provider organizations (PPOs) and health maintenance organizations (HMOs) in the new Medicare Advantage program, will be able to offer drug coverage to their enrollees. In regions with fewer than two competing plans, beneficiaries would be served by a single government-sponsored cost-plus plan (the "federal fallback"). Medigap insurance will be prohibited from covering drugs (although beneficiaries who are currently enrolled in such plans will be permitted to retain that coverage).

The law describes a "standard" benefit plan in which enrollees will be responsible for a deductible of $250 and coinsurance of 25 percent for the next $2,000 of expenses. If drug expenses go beyond that level, enrollees pay 100 percent of the next $2850 in costs, the notorious "donut hole" between $2,250 and $5,100 of drug spending. After that, the plans cover 95 percent of costs. Insurance payments from other sources, however, would not be counted toward satisfying the $3,600 total out-of-pocket payout before catastrophic coverage would begin. All these cutoffs will increase annually in proportion to Medicare's rapidly increasing outpatient drug spending; the Congressional Budget Office (CBO) estimates that in 2013 the donut hole will range between $4,000 and $9,000. Premiums for the drug benefit will be set by competition, not by regulation or statute. CBO expects premiums to average about $35 per month in 2006 and increase to perhaps $58 by 2013.

No insurer, public or private, offers health insurance of any sort with a donut hole in the benefit, which is both confusing and unpopular. MMA also permits "alternative" benefits that can depart to a highly uncertain extent from the standard. However, such benefits must be at least actuarially equivalent to the standard benefit and must respect the basic parameters, including the deductible and the donut hole.

The Part D plans will be heavily subsidized, with about 75 percent of the cost of the benefit covered by the federal taxpayer. (Low-income beneficiaries will be given generous additional subsidies; these are described below.) The subsidy is intended to yield premiums far below the actuarial value of the drug benefit, inducing broad participation among Medicare beneficiaries. Broad participation avoids adverse selection--a situation in which only people with above-average pharmaceutical needs participate, driving up costs and premiums, and discouraging participation of those with lesser drug use until little of the market is left. CBO predicts success, estimating that almost three-quarters of Medicare beneficiaries will have drug coverage through Part D plans, with employer-sponsored retiree plans or other programs (such as the Veterans Health Administration) covering the remaining beneficiaries.

The law is designed to keep existing retiree drug plans in operation. MMA provides a 28-percent subsidy to employers for drug spending by eligible beneficiaries in the range between $250 and $5,000, but only if the retiree plan matches or exceeds the new Medicare drug benefit. The subsidy is paid even if firms offload some (but not all) of their costs onto retirees through increased deductibles and copayments.[4] Although firms are likely to reduce the value of their retiree drug coverage somewhat, fewer firms will drop that coverage completely. CBO projects that the subsidy will cost taxpayers $81 billion in the first decade, but that some 2.7 million Medicare beneficiaries will nonetheless lose their retiree drug coverage as a consequence of MMA.

What Will Part D Do to Pharmaceutical Markets? Although the basic contours of the benefit are set forth in the law itself, the way the benefit will play out is extremely unclear. All enrollees in the new entitlement will receive their benefits from competing private plans, so prices and expenditures will be determined primarily by competition rather than a system of administered prices. There seems little reason to expect an immediate escalation in expenditures well beyond predicted levels, but spending trajectories after the first two years or so are conjectural and may prove highly unstable. The competition that is expected to constrain prices and spending may be severely limited by the nature of MMA itself. Medicare spending for prescription drugs could begin to grow very rapidly, as so often happens in the wake of new entitlements. If that happens, MMA could prove to be a path toward pharmaceutical price controls, with the unintended consequence of slowing the development of new and more effective drugs.

The New Cost-Sharing Arrangement. The new Part D constitutes the largest new entitlement since the creation of Medicare itself in 1965. All Medicare beneficiaries will be eligible to participate in the new program, but, as noted above, those with low incomes will receive a far more generous benefit. Means testing is a striking departure from Medicare's tradition of uniform benefits regardless of ability to pay.

Beneficiaries with incomes below 135 percent of the poverty level ($16,362 for couples in 2003) will receive the largest subsidy. Those individuals will face no deductible or donut hole, and copayments will be minimal. CBO estimates that 12 million beneficiaries will be eligible for the maximum low-income subsidy. Another 2 million beneficiaries with incomes between 135 and 150 percent of poverty will be eligible for lower--but still generous--subsidies. Those beneficiaries will be obligated to pay a $50 deductible, 15 percent coinsurance (but no donut hole) up to the catastrophic coverage level, and nominal copayments above that. Consequently, more than a third of all Medicare beneficiaries will be subject to minimal cost-sharing requirements.

In contrast, Part D enrollees who do not receive special subsidies will face substantial out-of-pocket costs. Taking into account cost sharing (deductibles and coinsurance) and insurance premiums, the first $2,250 of prescription drugs will cost an enrollee over $1,000 and the first $5,100 of drugs will cost over $3,600. These large out-of-pocket payments are one reason why total pharmaceutical usage by Medicare beneficiaries may initially increase by only a few percentage points even though the new benefit increases federal spending significantly. This modest effect also reflects the fact that for many beneficiaries, the new federal benefit substitutes for private coverage and other sources of funding. Another reason to expect only a modest immediate rise in demand is the simple fact that even though roughly one-quarter of beneficiaries lack drug insurance at present, more than 97 percent of them typically fill all of their prescriptions.[5] The current wave of new generics will ease the financial burden on low-income beneficiaries who are not eligible for the maximum subsidy. So will the discounts negotiated for all Part D purchases, even those in the donut hole.

The Dynamics of Competition. Competing plans will have full authority to negotiate pharmaceutical prices, and MMA bars the government from interfering. Plans could employ formularies with tiered copayments involving lower prices for favored brands (the only proven way to obtain substantial manufacturer discounts). Each plan, however, would have to offer drugs from every therapeutic category as defined by the (nongovernmental) U.S. Pharmacopoeia. The category definitions will have a strong influence on competition generally and on the depth of discounting in particular--just one example of the many crucial details yet to be determined. Plans would also have to withstand CMS scrutiny of the reasonableness of their costs.

Subject to these constraints, each plan can work out its own combination of formularies, negotiated discounts, rebates and other cost-sensitive arrangements, prescription and usage monitoring, disease management contracts, and marketing. Plans would only have to maintain "actuarial equivalence" with the standard 25 percent coinsurance plans while respecting the specified deductible and donut hole boundaries.

Notwithstanding these freedoms, the limits to innovation are severe. Plans could not, for example, trade larger cost sharing below the donut hole in return for partial elimination of the hole. Even the limited flexibility permitted in theory may prove elusive when CMS addresses the difficult task of determining the actuarial characteristics of diverse plans that bear no more than a passing resemblance to anything ever seen in the marketplace. An additional factor is that regulations (not yet written, of course) may make it easy for physicians and patients to ignore the restrictions imposed by formularies, which in turn would undermine the ability of alternative plans to negotiate discounts.

Regulation could easily suppress useful competition in other ways. The ability to exploit generic drugs is an example. Roughly 10 percent of the pharmaceutical market is turning generic each year as a result of the current wave of blockbuster patent expirations. Some plans could spend more money on devising ways to accelerate and monitor generic use, especially in the all-important donut hole--where a 20-percent cut in prices would save many beneficiaries far more than it would in the 25 percent coinsurance region. The extent to which CMS would permit such administrative costs as part of the allowable base could prove contentious and hard to predict. Again, some of the most creative and innovative activities may be curtailed. This could be part of a general pressure to cause plans to strongly resemble one another, limiting the scope and vigor of competition.

Of special concern is the bidding process wherein plans gain permission to go to market. This promises to be quite tricky. Premiums will be the difference between the CMS subsidy and projected costs. Because that subsidy would equal 75 percent of the projected costs of the weighted average of all accepted plans, individual firms would not know the size of the subsidy when they submit their bids. Nor could they be sure of their own cost structure. Bidders will not have reliable estimates of either the size or composition of their enrollee population until after the bidding process is complete. They will find it difficult to predict such essential cost elements as negotiated discounts and rebates, pharmacy networks and associated contracts, and innumerable details involving generic drugs, copayment tiers, and the like.

One particular problem is that in order to negotiate selective brand discounts, plans have to offer the possibility of shifting large numbers of enrollees to the favored brands. But the size of each plan's enrolled population will depend strongly on the attractiveness of drug prices that have yet to be determined--something the manufacturers will understand very well as they negotiate with competing bidders. Many of these imponderables are actually familiar to Congress and the U.S. Office of Personnel Management, as the Medicare bidding process resembles the annual negotiations for federal employee health plans. But the new drug plans will be very different from traditional health plans, partly because they will not be simple extensions of ongoing operations. It is not clear how this conundrum can be solved. The danger is that we could end up with a bidding process that generated rather small selective discounts and tends toward a relatively uniform price structure across plans. This would inhibit the vigorous and adventurous drug benefit competition that may be necessary to maintain efficient pharmaceutical use at a reasonable cost.

Finally, it is worth noting that drug plans will have to market aggressively to educate Medicare beneficiaries and attract enrollment. Plans with low premiums may be inferior to plans charging higher premiums but offering more comprehensive lists of preferred drugs. Plans with higher copayments may offer lower drug prices, which translate into savings within the donut hole where enrollees pay the full negotiated price. The informational requirements for making good consumer choices will be substantial. To some extent, intermediaries (perhaps organizations like AARP) will fill the informational gap, but advertising and other promotional tools will also be essential. The extent to which CMS will regulate or limit promotional activities remains unclear, however.

Stand-Alone Drug Plans versus Comprehensive Health Plans. The drug benefit can be provided either as stand-alone drug insurance or as part of the new Medicare Advantage plans, which are explained below. MMA specifies that Medicare Advantage drug benefits must meet essentially the same standards as stand-alone plans. There is some danger, however, that this requirement could prove mischievous.

Drug benefits embedded in a comprehensive health insurance plan are inherently superior to stand-alone plans. Comprehensive health plans avoid the notorious "silo effect" in which medical costs are treated as separate categories, impeding exploitation of the large and mounting synergies between medical technology and health care. MMA's requirement that drug benefits meet the standards of stand-alone plans, including rough actuarial equivalence, risks recreating the silo effect by limiting how the drug benefit may be structured. If comprehensive plans were allowed more freedom to pursue their natural advantages, they might induce more efficient pharmaceutical usage and ultimately pose stronger competition with traditional fee-for-service Medicare and its myriad inefficiencies.

Risk-Bearing versus Cost-Plus Operations. The drug benefit established by MMA carries a substantial risk that the individual plans will rapidly evolve into cost-plus operations with little incentive to restrain costs. One reason is that a large fraction of spending, perhaps on the order of half, will not involve significant enrollee cost-sharing, which is the most effective tool (other than price discounts) for controlling costs.

A second problem is that drug plans may end up bearing relatively little risk for unrestrained costs. A well-designed drug benefit would reward firms for saving money (for themselves and their customers) and would penalize them for spending too much. MMA removes most of the rewards and penalties. If a plan's actual experience (excluding administrative costs) departs from expected experience by more than a "risk corridor" (equal to plus or minus 2.5 percent of their initial projection of cost for the benefit), firms must refund 80 percent of any savings or will be reimbursed 80 percent of any overrun. These percentages gradually decline in later years, but this mechanism is nonetheless a force for making all plans into average plans and making the average firm act like a cost-plus enterprise.

Even that limited level of risk bearing may be absent in many regions. MMA permits CMS to accept plans that assume only "limited" risk, which is presumably even less risk than the usual arrangement with its risk corridors and refunds. Finally, if CMS does not think at least two bids for a specific region are adequate, it can accept bids for plans that are explicitly cost-plus arrangements.

A Path toward Pharmaceutical Price Controls? Medicare price controls for hospital, physician, and other services were born of the attempt to rein in costs that were escalating far beyond predicted levels. Pressure to limit the growth of prescription-drug spending under the new benefit could once again lead Congress to impose price controls, but with potentially more serious consequences.

Two factors suggest that the Medicare drug benefit will prove far more expensive than the current $400 billion estimate for the first ten years. First, the plan may be expanded before it even begins. Many people, especially Democrats and their traditional ally, AARP, want very much to sweeten the benefit. If Congress were to narrow or eliminate the donut hole, the most criticized feature of the bill, the expected costs of the benefit would escalate. The same would be true if Congress were to expand generous low-income subsidies to enrollees with incomes above 135 percent of poverty. AARP has already announced plans to lobby for a smaller donut hole and less stringent means-testing. It has also announced support for direct price negotiations by CMS if prices do not drop in the new system.

Second, even without new legislation, costs might climb far higher than projected. The myriad details of the enabling legislation, combined with regulations to be drafted, seem likely to inhibit the kinds of competition with the greatest potential to restrain costs. Manufacturers' discounts may therefore prove modest. Large regions may be served by a single government-sponsored plan, paid on a cost-plus basis with little incentive to constrain input costs. Even the standard plans, competitive in theory but possibly restrained in practice, may evolve rapidly toward cost-plus arrangements. Numerous experiments with so-called no-fault auto insurance have demonstrated that cost-plus insurance is a route to extravagant costs.

Oddly enough, the least controversial and least discussed part of the new drug benefit--catastrophic coverage for 95 percent of annual costs over $5,100--could turn out to be the most expensive and difficult to control. A number of powerful drugs costing more than $10,000 per year have been introduced in the past few years. These include the new generation of targeted cancer therapies such as Gleevec, Herceptin, Rituxin, and Zevalin ($28,000 per dose); new AIDS drugs including Fuseon; Avonex for multiple sclerosis; Xolair for asthma; and various hepatitis C drugs. We can expect more of these, along with creative new uses of expensive old drugs, paralleling what has happened with medical procedures-such as bone marrow transplants, at well over $100,000 per procedure--and medical devices--such as left ventricular assist devices costing $60,000 each.

Such developments would add to the already sizeable financial obligations imposed on Medicare by MMA. Coupled with rising demands from other parts of the federal budget, Congress may soon face immense pressure to curtail the cost of the drug benefit without reducing its generosity. Such pressures are endemic to Medicare as they are to all state-run health care systems. Beginning in the 1980s, Medicare responded with price controls that rapidly became hideously comprehensive and complicated. Since price controls can become too strict, causing shortages and hindering improvements in care, policymakers have adopted a "watchful waiting" strategy: watch for shortages and gross distortions in the market, and then adjust the controls accordingly. This has happened repeatedly with Medicare price controls over hospitals and physicians. In fact, MMA itself contains numerous adjustments to correct actual or perceived pricing problems (discussed below).

With pharmaceuticals, the situation will be fundamentally different. Drugs typically sell at large margins over manufacturing costs, a necessary result when a small number of successful products recoup research and development (R&D) expenses for a much larger number of unavoidable failures. If CMS pushes pharmaceutical prices well below market levels, shortages will not occur in the supply of existing drugs. Instead, fewer new drugs will reach the Medicare population as private funding for R&D dries up. Investment in high-risk, high-return research is the only promising path toward solutions to the great unsolved medical problems of Alzheimer's, diabetes, congestive heart failure, and many other diseases that plague Medicare patients.

Reforming or Destroying Medicare?

Opponents of the House version of the Medicare bill argued vehemently against the competition provisions, focusing their attacks on the "premium support" program that was to begin in 2010. Under that provision, private health plans would have competed directly with traditional Medicare in local markets, and the premium paid by enrollees in traditional Medicare would increase if private plans could provide services more cheaply. Some critics feared that seniors remaining in the traditional program would see sharply rising premiums as healthier beneficiaries moved to lower-cost private plans.

Politics and pragmatism have reduced the premium-support provision to a six-city demonstration project for what is now called "comparative cost adjustment." That demonstration will probably not be implemented, based on past experience with controversial projects of this sort.[6] That does not satisfy some critics, who continue to fret about undermining traditional Medicare.

MMA is not the revolution that some on the political left fear and some on the right hail. The legislation is simply the latest (but not the last) attempt by Congress to fix problems it created in the Balanced Budget Act (BBA) of 1997--excessive regulation and unrealistic payment rates that did not reflect the costs actually faced by individual health plans.[7] MMA makes some necessary changes in the rules for private health plans in Medicare, but it fails to give seniors the kinds of realistic plan options that are available under the Federal Employees Health Benefits Program (FEHBP).

Prior to the BBA, private health plans (primarily HMOs) were a small but growing alternative to traditional Medi-care. Those plans attracted enrollment by offering additional benefits, mainly prescription-drug coverage. The BBA attempted to expand the types of plans available to Medicare beneficiaries under the newly named Medicare+Choice (M+C) program, but that expansion never occurred and the number of plans participating in M+C plummeted. The proliferation of complex and frequently changing regulatory requirements, formula-driven payment rates that did not keep up with the rising cost of health services, and changes in the broader business climate facing the plans account for that failure.[8]

The Medicare Advantage (MA) program has replaced M+C, and more than just the name has been changed. PPOs, which have become the dominant type of health plan for those under age sixty-five, are expected to enter MA. Plans will be able to offer drug coverage integrated with other health benefits, and all MA plans will operate special programs to help those with chronic illness. Federal payments to plans will be based on a higher benchmark than under M+C, which will attract greater interest in the program. The government will pay health plans based on bids that reflect their cost of services, with adjustments upward or downward for enrollees who are sicker or healthier than average. Lower-cost plans are allowed to share those cost-savings with their enrollees in the form of premium rebates, which could be substantial. Seniors in traditional Medicare will see no change in their premiums because competition is permitted only among the private plans.

In other ways, however, the MA program does not break with mistakes of the past. MA remains heavily prescriptive, specifying in detail how health plans are to operate. To assure a choice of health plans for every Medicare beneficiary, Congress expects some plans to operate in large multi-state regions even though such regions will not coincide with the markets that plans have already established. MMA attempts to force private plans into a government mold, as M+C tried to do unsuccessfully over the past five years. The result is likely to be fewer plan choices available to beneficiaries, less competition among plans, and less downward pressure on costs than would have been possible in a less regulatory environment.

The new law may have strengthened Medicare (including the traditional program), and it has taken steps to revive health plan options for beneficiaries. Competitive bidding in place of arbitrary payment formulas used in M+C will promote cost containment without disrupting the supply of services. Failing to involve all plans (including traditional Medicare) in the bid process will limit the scope of competition, but health plan provisions in MMA clearly intend to foster rather than suppress a competitive market in Medicare.

Tweaking Medicare

Armies of Washington consultants and government bureaucrats devote their careers to tweaking the Medicare program, and the new legislation does little to divert their energies to more wholesome pursuits. Indeed, tweaks are the bread and butter of Medicare legislation. A small change here or there can swing millions of dollars toward (or away from) a specific group of providers. Legislative adjustments, and the regulations that inevitably follow, can change the way providers deliver health care. That can have a significant long-run impact (for good or ill) on the broader health system.

Fully half of the new legislation is devoted to tweaking the current Medicare program without making fundamental changes. Buried in the new law are adjustments in the level and structure of payments to individual health care providers, changes in the rules determining how health care is delivered, modifications in the administration of Medicare, and adjustments in what beneficiaries pay for and how much they pay.

The sheer multitude of payment adjustments and other changes make it difficult to assess how individual groups of providers fared under MMA. Hospitals are a case in point. Medicare payments for inpatient services will increase by the full amount of the government's index of hospital cost inflation in 2004, an increase higher than hospitals typically receive. Between 1991 and 2001, for example, the hospital payment update was less than the inflation index in all but one year.[9] In addition to the general payment increase for hospitals, however, there are numerous payment adjustments that affect various types of hospitals or even specific institutions. The list of affected groups is long: rural hospitals, teaching hospitals, hospitals that treat a large share of low-income patients, "critical access" hospitals, hospitals in states with large populations of illegal aliens, and hospitals that disproportionately use new treatment technologies, among others. Rural counties in seven states have been reclassified, allowing twenty hospitals to receive higher payments typically paid in larger metropolitan areas. A rural community hospital demonstration program will pay higher rates to fifteen small rural hospitals. A hospital in Saginaw, Michigan, that would have no longer qualified for certain Medicare payments is now allowed to retain them.[10]

That list barely scratches the surface of payment adjustments made by MMA, but it gives some idea of the amount of effort that is expended every year to fine-tune Medicare's complex price-fixing schemes. Political fine-tuning moves resources around the health system and changes provider and consumer behavior in ways that can be unintended and unwanted. Worse still, political gridlock often delays actions that might correct those unforeseen consequences.

Physician payment is an example. Congress is gradually undoing a provision from 1997 that keeps Medicare spending for physician services under control--too tightly. Automatic across-the-board cuts in doctors' fees have created a political uproar. An increasing number of physicians refuse to see new Medicare patients because their fees have been cut and they cannot increase their charges to patients. Despite these disruptions, the 1997 provision has not successfully slowed the growth of Medicare spending; outlays for physician services in 2002 grew by $3 billion even with a fee reduction of 5.4 percent. MMA reversed the automatic fee cuts in 2004 and 2005, but did not otherwise seek to reform a clearly unmanageable price-fixing system.

One of the most startling tweaks in MMA effectively halts the development of new specialty hospitals for the next eighteen months. Responding to Medicare's generous reimbursement policies, certain physician specialists (such as orthopedic or cardiac surgeons) have begun to move their practices from general hospitals to specialized hospitals in which they have a financial interest. That has seriously cut into the revenues of some general hospitals, which have broader responsibilities to provide emergency services and act as a health safety net in their communities. Congress has temporarily halted this reaction to the existing Medicare price incentive by imposing more regulation, this time directly on the supply of services. Whatever the merits of the case, the political solution of piling one regulation on another cannot resolve the underlying misallocation of resources caused by government controls.

The Shape of Things to Come

Congress and the president redefined the boundaries of politically acceptable Medicare policy with the Medicare Prescription Drug, Improvement, and Modernization Act. As with all compromises, some boundaries have been stretched while others remain rigidly in place.

Every Medicare beneficiary will at last be able to buy prescription-drug coverage, and most will. Beneficiaries will have new choices of health plans, which in turn will be attracted to the Medicare Advantage model by the higher payment rates and the ability to bid competitively. Some of the problems with provider payment in traditional Medicare have been resolved, albeit temporarily. Quality improvement initiatives have been advanced, including new ways of managing the care of high-cost and chronically ill patients.

At the same time, the drug benefit threatens to displace the substantial private coverage already available in the market. MMA does not fund the new benefit, placing a tremendous additional financial burden on younger generations. Medicare Advantage will not reduce program spending, at least in the short run, and regulatory burdens that contributed to the collapse of Medicare+Choice could resurface in the new program. Congress remains wedded to complex price-setting schemes and has signaled a new interest in directly controlling the allocation of health system resources. The drug benefit could easily evolve into yet another comprehensive system of price controls.

We cannot know at this point how the new law will turn out. For all of its complexity and length, MMA is an incomplete outline to guide the development of Medicare in the coming years. The legislation leaves substantial discretion to CMS, and undoubtedly there are undiscovered omissions and contradictions in the law that will have to be resolved.

Perhaps the deepest mystery is how MMA will shape the American health care system. The new law attempts to balance a step toward comprehensive single-payer health care with promising steps in the opposite direction. These include the competitive aspects of the drug benefit, the resurrection of PPOs and managed care, and (looking beyond Medicare) the potentially very substantial role of Health Savings Accounts--which may yet stake out a path away from the low-deductible, low-copayment, employer-based model for prepaid health care and instead more toward true health insurance.

Notes

1. Kaiser Family Foundation, "Knowledge about News: Medicare Prescription Drug Bill," available at http://headlines.kff.org/healthpollreport/templates/detail.php?page=2&feature=hni (accessed on January 21, 2004).

2. U.S. House of Representatives, Medicare Prescription Drug, Improvement, and Modernization Act of 2003: Conference Report to Accompany H.R. 1 (Washington: Government Printing Office, 2003).

3. U.S. General Accounting Office, Prescription Drug Discount Cards: Savings Depend on Pharmacy and Type of Card Used (Washington: GAO, September 2003), p. 4.

4. Ellen E. Schultz and Theo Francis, "U.S. Drug Subsidy Benefits Employers," Wall Street Journal, January 8, 2004; letter from Chairman Bill Thomas, Ways and Means Committee, to Secretary Tommy Thompson, HHS, January 16, 2004.

5. Benjamin M. Craig, David H. Kreling, and David A. Mott (2003) "Do Seniors Get the Medicines Prescribed for Them? Evidence from the 1996-1999 Medicare Current Beneficiary Survey, Health Affairs, v. 22, n. 3, p. 175-182 (May-June).

6. Len M. Nichols and Robert D. Reischauer, "Who Really Wants Price Competition in Medicare Managed Care?," Health Affairs, vol. 19, no. 5 (September-October 2000), pp. 30-43; Sarah Lueck and David Rogers, "Medicare Test May Prove Moot," Wall Street Journal, November 21, 2003; and Milt Freudenheim, "Medicare Plan for Competition Faces Hurdles," New York Times, November 28, 2003.

7. Previous legislative fixes were attempted in the Medicare Balanced Budget Refinement Act of 1999 and the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000.

8. Bruce M. Fried and Janice Ziegler, The Medicare+Choice Program: Is it Code Blue?, Washington: ShawPittman, June 8, 2000; and Marsha Gold, "Medicare+Choice: An Interim Report Card," Health Affairs, vol. 20, no. 4 (July-August 2001), pp. 120-138.

9. Medicare Payment Advisory Commission, Report to Congress: Medicare Payment Policy (Washington: MedPAC, March 2003), Table D-1, p. 266.

10. Christopher Lee, "Medicare Bill Partly a Special Interest Care Package," Washington Post, November 23, 2003.

Joseph Antos is the Wilson H. Taylor Scholar in Health Care and Retirement Policy at AEI. John E. Calfee is a resident scholar at AEI.