Skip Navigation

Mr. Chairman and Members of the Committee, I am pleased to be here today to discuss the President's recommended changes to the Medicare program. Those recommendations build on several of the major Medicare provisions in the President's budget proposal for fiscal year 2000. They also reflect some of the ideas generated by the Bipartisan Commission on the Future of Medicare, which completed its work in March. In addition, the President's proposal takes into account the growing concerns that some groups of health care providers have about the effects of the Balanced Budget Act of 1997 on Medicare payments.

Key features of the President's proposal include adding a prescription drug benefit to Medicare, making broad changes to the traditional fee-for-service program, converting the Medicare+Choice program into a competitive defined benefit program, and transferring revenues from the general fund to Medicare. The proposal lacks specificity in several important areas, however. That vagueness limits the Congressional Budget Office's (CBO's) ability to estimate the costs of some parts of the proposal and makes the estimates that CBO has been able to produce more uncertain.

My testimony today describes the major provisions of the President's proposal as outlined in the July 2, 1999, report from the Domestic Policy Council. It then discusses CBO's analysis of those provisions and provides cost estimates where feasible.
 

OVERVIEW OF THE ESTIMATE

CBO estimates that the President's proposal would increase outlays for Medicare and Medicaid by $111.1 billion over the 2000-2009 period (see Table 1). By comparison, the Administration estimates the 10-year cost of the proposal at $45.7 billion. In CBO's view, outlays for the prescription drug benefit would be $168.2 billion, offset in part by $57.1 billion in savings from fee-for-service changes and from greater price competition among managed care plans (see Table 2). More than one-quarter of the net increase in federal spending would occur in the Medicaid program, including new spending for prescription drugs that would be paid for entirely by the federal government.
 

PRESCRIPTION DRUG BENEFIT

The President's proposal would create a voluntary outpatient prescription drug benefit under a new Part D of Medicare. The benefit would begin in 2002 and would be fully phased in by 2008. The benefit would pay half of the cost of prescription drugs (up to a specified cap) and would be financed by premium payments from enrollees and general revenues. Taking cost sharing and premiums into account, the average enrollee would pay about 75 percent of the cost of covered drugs up to the cap.

Description of the Proposal

In 2002, all Medicare enrollees would have a one-time opportunity to purchase the new benefit. In later years, enrollees would be permitted to choose the Part D option only when they first became eligible for Medicare, with two exceptions: beneficiaries whose primary coverage was employer sponsored would have a one-time opportunity to enroll after retirement (or after the retirement or death of the working spouse), and beneficiaries with employer-sponsored retiree health plans would have a one-time option to enroll if their former employer dropped prescription drug coverage for all retirees.

The new drug benefit would be administered by a pharmaceutical benefit management company (PBM) in each geographic area, selected through competitive bidding. All Part D enrollees would gain from the below-retail prices that PBMs can typically negotiate. The benefit would include no deductible and would generally pay 50 percent of an enrollee's prescription drug costs, up to an annual cap per enrollee. That cap would be set at $1,000 in 2002 and would gradually rise to $2,500 in 2008. Thus, in 2008, a beneficiary who purchased $5,000 in prescription drugs would receive the maximum reimbursement of $2,500. That beneficiary would also pay $634.80 in Part D premiums that year. After 2008, the cap would be indexed to annual changes in the consumer price index (CPI). Assuming that the cost of prescription drugs continued to rise more rapidly than the CPI, the real value of the benefit cap would shrink, thereby eroding the benefit.

Low-income participants would receive subsidies through the Medicaid program. Medicaid would pay both the premiums and the cost-sharing expenses, at the usual federal/state matching rate, for participants who were also fully eligible for Medicaid (so-called dual-eligibles) or who had income below the poverty line. The federal government would pay all of the premiums and cost-sharing expenses for other Part D enrollees with income less than 135 percent of the poverty line and part of the premiums for Part D enrollees with income between 135 percent and 150 percent of the poverty line (see Table 3).
 

Eligibility for those subsidies would be determined by state Medicaid agencies. Neither the federal nor the state governments would be liable for covering any drug expenses above the Part D benefit cap for low-income beneficiaries who were not fully eligible for Medicaid.

The President's proposal also includes an incentive that is intended to retain employer-sponsored drug coverage for retirees. Medicare would pay employers 67 percent of the premium-subsidy costs it would have incurred if their retirees had enrolled in Part D instead. In addition, enrollees in Medicare's managed care plans would receive their prescription drug coverage through those plans, which for the first time would be paid directly for providing such coverage (for enrollees who opted for the Part D benefit).

Medicare now pays for a limited list of drugs provided on an outpatient basis. Those drugs would continue to be covered under Part B. Consequently, their costs would not be included in the cap on Part D benefits.

CBO's Estimate

CBO estimates that the new Part D provisions would add a total of $168 billion to federal costs through 2009. (By comparison, the Administration's estimate of Part D costs is about $119 billion.) CBO estimates that Medicare outlays (net of premium receipts) would be $136 billion, and federal outlays for Medicaid would be $32 billion (see Table 4). States would also face additional Medicaid costs--totaling some $12 billion through 2009. CBO estimates that the premium for Part D would start at $25.20 a month in 2002 and rise to $52.90 in 2008 when the program was fully phased in (see Table 5).
 

CBO's cost estimate assumes that most people who are enrolled in Part B of Medicare would also enroll in Part D. But some of those who have employee-sponsored drug coverage for retirees would keep that coverage rather than enroll in the new program. CBO assumes that such people account for about 20 percent of Part B enrollees. In addition, about 7 percent of those eligible for benefits under Part B do not actually enroll. CBO assumes that they would also not enroll in Part D. Under those assumptions, about 31 million people would enroll in Part D in 2002, representing approximately 80 percent of total Medicare enrollment.

In 2002, about 36 percent of participants would have drug expenses exceeding the $1,000 cap on Part D benefits. By 2008, when the benefit cap would be $2,500, about 25 percent of participants would have expenditures exceeding the cap. Part D benefits paid per participant would average about $600 in 2002, rising to around $1,280 in 2008.

CBO is estimating higher costs for the Part D benefit than the Administration. Both CBO and the Administration base their estimates of future drug spending on patterns reported in Medicare's Current Beneficiary Survey, and both adjust the amounts reported by noninstitutionalized people by approximately the same factor to account for underreporting. However, CBO's estimate also attempts to account for spending on prescription drugs by residents of nursing homes. The estimates also differ in their assumptions about the rate of growth in enrollees' spending on prescription drugs. The latest projections of national health expenditures indicate that the recent rapid rates of growth in drug spending will slow sharply over the next few years. CBO, however, assumes that the slowdown will not occur as rapidly as those projections suggest.

Other Issues

Estimating the cost of a service not now covered by Medicare is inherently more difficult than estimating the cost of a change in the way a current service is paid for. The cost of the President's proposal for covering prescription drugs is uncertain because many design aspects of the new benefit have not yet been fully specified.

Nature and Value of the Benefit. Per capita spending for prescription drugs has been growing at double-digit rates in recent years--faster than other components of health care spending. Whether that rapid growth will continue, accelerate, or moderate is uncertain. A number of innovative drugs are likely to be cleared for marketing in the near future, which would tend to increase both the use and the average price of prescription drugs. However, a number of heavily used brand-name drugs are about to lose their patent protection (allowing entry of generic substitutes), which would tend to reduce prices. Hence, projections of the rate of growth in drug use and prices are highly uncertain even in the absence of changes in insurance coverage. For this estimate, CBO assumes that recent growth trends will continue for several years and then moderate somewhat.

Another area of uncertainty is the extent to which the coverage provided under the President's proposal would increase drug utilization by enrollees. Half of Medicare enrollees already have coverage for prescription drugs (typically through a retiree health plan or Medicaid) that is at least as generous as the coverage offered under the President's plan. For the other half, the new Part D coverage would increase drug utilization by up to 25 percent, CBO estimates.

Part D is designed to ensure that most enrollees would receive some benefit. However, because of the cap on benefits, it would not protect enrollees with drug-dependent chronic conditions from very large out-of-pocket expenses. Although the benefit cap would reduce Medicare's exposure to increases in prescription drug costs, it would also limit the value of the benefit to people who are especially vulnerable to those costs. Alternatively, insurance that provided no first-dollar coverage but limited an enrollee's out-of-pocket costs would be less likely to cause increased utilization and more likely to protect enrollees from catastrophic expenses. Under such an alternative, however, fewer enrollees would expect to benefit.

Effectiveness of the PBMs. The President proposes to administer the drug benefit through private-sector PBMs, which private health plans commonly use to negotiate price discounts and control utilization. A single PBM, selected through competitive bidding, would administer the benefit in each geographic area. CBO's cost estimate assumes that those PBMs would reduce costs below the level that an uninsured retail purchaser would face by about 12.5 percent--savings that are smaller than PBMs now generate for large, tightly managed health plans. That estimate could change, however, as details of the proposal's design emerge.

PBMs produce savings for private health plans in four main ways. First, they negotiate discounts with pharmacies that agree to participate in their networks. Second, they obtain rebates from manufacturers of brand-name drugs in exchange for preferred status on the health plan's formulary. (A formulary is a list of drugs preferred by the plan's sponsor, in part on the basis of their lower prices.) Third, PBMs use mail-order pharmacies, which are often better able than retail pharmacies to save money. Mail-order pharmacies are likely to have lower average operating costs, and they can substitute generic or other lower-cost drugs for the ones prescribed. Finally, PBMs establish differential copayment requirements that encourage beneficiaries to select lower-priced options such as generic, preferred formulary, or mail-order drugs. Some PBMs also use management techniques such as on-line utilization review and prior approval to evaluate care and encourage the most cost-effective treatment practices.

It is uncertain whether the PBMs chosen to administer the Part D benefit under the President's proposal would have as much freedom to use those cost-saving techniques as they have in aggressive private insurance plans. For example, the proposal specifies that PBMs would have to set dispensing fees high enough to ensure participation by most retail pharmacies, which could reduce their ability to negotiate substantial discounts from pharmacies. The proposal also specifies that beneficiaries would be guaranteed access to off-formulary drugs when medically necessary, reducing PBMs' ability to negotiate rebates from manufacturers. Further, the proposal would limit their ability to encourage beneficiaries to choose lower-cost drugs through differential copayments. Although PBMs would not be prohibited from charging differential copayments, those copayments could not exceed 50 percent. Some private drug plans require enrollees to pay the full difference between the cost of a brand-name drug and its generic equivalent (if one exists) unless the prescribing physician specifically states that the brand-name drug is medically necessary. Such an approach would apparently not be permitted in the Part D program.

Indeed, how much incentive PBMs would have to generate savings under the program is uncertain. The President's proposal envisions competitive bidding to select the PBM for each geographic area, but it is unclear what financial risks, if any, the winning PBM would bear beyond the costs of processing claims. The proposal indicates that contractual incentives (such as performance bonuses) might be used to encourage PBMs to focus more aggressively on generating savings, but those mechanisms have not yet been specified. Nor is it clear how savings would be measured. Actual savings could disappear, even while nominal discount and rebate rates were unchanged, if the prices against which discounts and rebates were calculated rose as a consequence of the new benefit.

Program Participation. CBO's estimate assumes that everyone who participates in the Part B program would also participate in Part D, with one exception: most people who have drug coverage through retiree health plans would remain with those plans. Those assumptions are quite speculative, however, and participation rates might well be lower or higher.

As noted above, employers would receive federal payments equal to 67 percent of the Part D premium subsidy for eligible retirees if they retained (or instituted) prescription drug coverage at least as good as the new Part D benefit. That subsidy payment, together with the tax deductibility of their health plan costs, would help induce employers to keep full drug coverage in their retiree health plans rather than eliminate it or wrap their plans' benefits around the new Part D package. (Employers with a wraparound plan would require Medicare to be the primary payer for prescription drugs, with the employer's plan serving as a supplement.) For their part, most retirees in employer-sponsored plans would probably prefer to continue with those plans rather than Medicare Part D, for two reasons. First, they would generally pay a lower premium for equivalent drug coverage in a retiree health plan than in Part D because employers typically pay more than 50 percent of the benefit costs. Second, retiree health plans usually provide much more generous drug coverage than Part D would, and getting all drug benefits through the retiree plan would avoid the problems associated with coordinating benefits. Nevertheless, CBO assumes that about one-quarter of Medicare enrollees who now have drug coverage through a retiree health plan would enroll in Part D because some employers would eliminate their drug coverage altogether.

The benefits provided under Part D would be very limited because of the 50 percent coinsurance rate and the benefit cap. Moreover, through their premium payments, enrollees would pay half of whatever benefits were paid out. Consequently, the federal subsidy under Part D would amount to less than one-quarter of enrollees' drug costs, on average. Despite those limitations, Part D would offer a more generous drug benefit package than standard medigap plans do, and at a lower premium. As a result, the three medigap plans that now offer drug coverage would no longer be competitive and might ultimately be replaced by a plan that supplemented the coverage offered under Part D.

Because of the one-time option to enroll and the 50 percent subsidy of premium costs, CBO expects that all Part B enrollees with medigap coverage or with no supplementary coverage would choose to enroll in Part D. People receiving Medicaid benefits under the proposal would also enroll in Part D because states would be required to cover their drug costs if they applied.

Effects on Medicaid Costs. As Table 4 showed, the President's proposal would increase Medicaid's costs for drugs and other benefits--substantially in the case of federal costs and less sharply in the case of state costs. Although Medicaid would no longer have to pay all drug costs for Medicare beneficiaries who now receive full Medicaid benefits, those savings would be more than offset by additional Medicaid spending on behalf of other Medicare beneficiaries.

Part D would pay for a portion of the drug costs that Medicaid now pays for Medicare enrollees at all income levels who are also fully eligible for Medicaid. That expansion of Medicare's role would lower both federal and state Medicaid costs by shifting them to Medicare. But the savings would be partly offset by the Part D premiums that Medicaid would have to pay for those dual-eligibles.

Low-income Medicare beneficiaries who are ineligible for full Medicaid benefits would also become eligible for assistance to pay for their Part D premiums and cost sharing. As noted above, the federal and state governments would share those costs for people with income below the poverty level. But the federal government alone would pay the premiums and cost sharing for beneficiaries with income between 100 percent and 135 percent of the poverty level, without any financial participation by the states. It would also pay a portion of the Part D premium costs for beneficiaries with income between 135 percent and 150 percent of the poverty level. To receive those benefits, however, eligible Medicare beneficiaries would have to enroll in the Medicaid program, and not all of them would choose to do so.

Medicaid spending would rise by more than the cost of the new prescription drug benefit. Many low-income Medicare beneficiaries who are ineligible for full Medicaid benefits are eligible to have their Medicare premiums paid by Medicaid--and in some cases, their cost sharing as well. A sizable number of them do not enroll in Medicaid, however. In 1998, an estimated 1.3 million Medicare beneficiaries with income below the poverty level were eligible for partial or full Medicaid assistance but did not participate in the program.⁽¹⁾ A further 1.3 million beneficiaries with income between 100 percent and 120 percent of the poverty level who were eligible to have their Part B premiums paid by Medicaid did not participate. The availability of a free drug benefit, made possible by enrollment in Medicaid, would attract more Medicare beneficiaries into the Medicaid program, boosting spending for other Medicaid benefits as well as for prescription drugs. Participation in Medicaid by beneficiaries who are eligible for full Medicaid benefits might also increase somewhat, although their participation is already greater than that of other groups.

For this estimate, CBO assumes that the price of drugs under the proposed Medicare benefit for Medicaid beneficiaries would be similar to the price that Medicaid obtains under current law (including Medicaid rebates). If Medicare received deeper discounts and rebates, Medicaid costs would be lower. Conversely, if Medicare paid more for drugs, Medicaid costs would be higher.
 

FEE-FOR-SERVICE CHANGES

The President is proposing a host of policy changes for the traditional fee-for-service sector of Medicare. Those changes include modifying the pricing rules that govern payments to providers, changing beneficiaries' cost-sharing requirements, and permitting the Secretary of Health and Human Services (HHS) to supplement certain administered pricing systems with new options for paying providers. Together, those fee-for-service policies would reduce federal spending by an estimated $48 billion through 2009. (The Administration's estimate of fee-for-service savings is $64 billion.)

Adjustments to Providers' Payments

The proposal would increase payments to certain providers beginning in 2000, redirect some payments to hospitals that serve a large number of low-income patients, and reduce the growth in payment rates for many services after 2002. The net effect of those provisions would be to lower payments to fee-for-service providers by an estimated $28 billion through 2009

To relieve some of the financial pressures that the Balanced Budget Act of 1997 imposed on providers, the President proposes changing how certain provisions of that act are put into effect. Those changes can be made administratively and do not require legislative action. They include allowing more rural hospitals to be reclassified as urban hospitals to receive higher payment rates; delaying collection of past overpayments from home health agencies; increasing payments to certain hospitals for outpatient services; and delaying the expansion of the "transfer policy," which would have reduced some hospital payments. CBO does not "score" those changes in administrative policy because they do not involve a change in law, even though they would increase baseline spending. CBO will take the policy changes that the Administration implements into account in its next baseline projection of Medicare spending under current law.

The President is also proposing to establish a "quality assurance fund" to pay for future legislative changes that would increase payments to certain providers beginning in 2002. But his proposal does not specify policies to accomplish that increase in spending. Thus, CBO's estimate of the net impact of policies to adjust provider payments includes the Administration's figure of $7.4 billion, although that amount could change depending on specific legislative proposals.

Another proposed change is designed to help hospitals with large caseloads of indigent patients. The portion of payment rates for Medicare's managed care plans that reflects disproportionate share hospital (DSH) payments would be eliminated. (DSH payments are additional payments that Medicare makes when beneficiaries receive inpatient care from hospitals that serve a large number of low-income patients.) Instead, Medicare would make DSH payments directly to those hospitals when they provide inpatient care to patients enrolled in managed care plans. CBO estimates that redirecting DSH payments in that way would have a negligible effect on Medicare spending.

The President's proposal would also significantly reduce payments to certain providers in the longer term by continuing payment reductions imposed by the Balanced Budget Act beyond 2002. For many services, the act holds the increases in payment rates below the rate of inflation through 2002, with full adjustment for inflation resuming in 2003. The proposal would hold those increases below inflation through 2009 for hospital inpatient care, ambulance services, prosthetics and orthotics, hospice care, ambulatory surgical center care, durable medical equipment, clinical laboratory services, and parenteral and enteral nutrition. In addition, the proposal would extend a 2.1 percent reduction in payment rates to hospitals for capital-related costs through 2009.

Adjustments to Beneficiaries' Cost Sharing

Other provisions of the President's proposal would require fee-for-service enrollees to pay more for Medicare services by indexing the Part B deductible to inflation and instituting coinsurance for clinical laboratory services. At the same time, the proposal would eliminate coinsurance for certain preventive services. The net effect of those changes would be to reduce Medicare outlays by an estimated $5 billion through 2009.

The deductible for Part B has been $100 since 1991. Under the proposal, it would increase by the percentage change in the consumer price index beginning in 2002.

Medicare currently pays 100 percent of the approved fee for clinical laboratory services. Except for preventive services, the proposal would impose the standard Part B deductible and 20 percent coinsurance requirement on clinical laboratory services beginning in 2002.

By contrast, the President's proposal would waive both the deductible and the 20 percent coinsurance requirement for certain preventive services. That change would substantially increase the use of those services and would also increase demand for other services--particularly those furnished by physicians. However, much of the increase in spending for physicians' services would be offset by other policies that would reduce updates to the physician fee schedule.

New Payment Options

Under current law, Medicare has limited authority to contract selectively, establish payment rates through competition or negotiation, or use many of the other techniques that private plans employ to manage spending and quality of care. The President's proposal would give the Secretary of HHS authority to adopt some of those techniques, including contracting with preferred provider organizations (PPOs), negotiating discounted rates for specific services, and developing systems to manage the care (in a fee-for-service setting) of certain diseases or beneficiaries.

The potential savings from those changes are substantial. The Administration estimates that granting the Secretary additional flexibility to manage pricing and utilization would save $25 billion over the next decade. However, major impediments stand in the way of realizing those savings. Thus, CBO estimates that the provisions would reduce payments to fee-for-service providers by less than $4 billion.

Providers often contract at a discount with private plans in the expectation of treating more patients. In turn, plans often require patients to pay substantially higher prices when they use providers who have not granted price concessions. As currently structured, Medicare's fee-for-service program does not have the tools that private plans use to extract such price concessions. About 85 percent of Medicare enrollees are indifferent to changes in cost-sharing requirements because they are insulated from those requirements by supplemental coverage--through employer-sponsored insurance, medigap insurance, a Medicare managed care plan, or Medicaid. Moreover, the 15 percent of enrollees without supplemental coverage might have little incentive to switch to providers granting discounts. Under current law, Medicare's coinsurance mechanism for Part B services would limit their savings to no more than 20 percent of the discount. Consequently, it is not clear that the proposal for Medicare to contract with existing PPOs is feasible. Given the limited potential for increasing their market share, PPOs would probably not be willing to offer substantial discounts to Medicare.

Other contracting options proposed by the President might yield more savings to the extent that they promoted the efficient delivery of health services by high-quality providers. Those options include the Centers of Excellence proposal (which bundles payments for facilities and physicians for certain inpatient services, including treatment of heart conditions and joint surgeries); the global payment proposal (which bundles payments for facilities, professionals, and suppliers for all care provided at a specific site); and the proposal to coordinate care for certain high-cost conditions. Those proposals account for about two-thirds of CBO's estimate of savings from granting the Secretary additional flexibility.

The President also proposes that the Secretary be given authority to contract selectively for some Part B services other than those furnished by physicians. That proposal would expand on a demonstration project in Polk County, Florida, in which Medicare is selecting suppliers through a competitive bidding process for five types of products: oxygen equipment and supplies, hospital beds and accessories, enteral nutrition products and supplies, urological supplies, and surgical dressings. The demonstration, which is still in the development stage, has produced bids between 13 percent and 31 percent lower than Medicare's existing fee schedule for those supplies. However, negotiations with bidders--including some who were unsuccessful in the first round--are continuing, and CBO anticipates that some of those potential savings will erode over time.

Moreover, the Secretary faces substantial challenges in expanding competitive bidding to other areas and other services. In recent years, providers and elected representatives have voiced significant opposition in communities in which the Secretary has tried to reduce spending through competitive bidding and selective contracting. CBO assumes that such opposition will continue to be a substantial impediment to expanding the competitive bidding model and realizing the potential savings from selective contracting.
 

COMPETITIVE DEFINED BENEFIT PROGRAM

The President proposes to give Medicare's managed care plans various incentives to compete on the basis of price as well as quality. This "competitive defined benefit" proposal is extremely complex, and many of its details are unclear. CBO has not yet estimated the costs of the proposal and, for the present, is using the Administration's estimate as a placeholder. That estimate indicates that Medicare would save $8.9 billion through 2009.

Description of the Proposal

Beginning in 2003, the premium that Medicare beneficiaries paid would depend on the plan they chose. Beneficiaries who stayed in the traditional fee-for-service sector would pay the regular Part B premium. But those who chose cheaper plans would generally pay a lower premium, and those who opted for more costly plans would pay the extra costs of that choice. Managed care plans would submit a premium offer for the standard Medicare benefit package, enabling beneficiaries to make price comparisons among plans.

The actual amount that beneficiaries paid would depend on the difference between the premium of the plan they chose and a reference price, which would be 96 percent of the average costs in the fee-for-service sector. If they enrolled in a plan with a premium below the reference price, their Part B premium would be reduced by 75 percent of the difference (with the remaining 25 percent accruing to the government). What they would pay if they chose a plan with a premium above the reference price is less clear. But the proposal indicates that the federal payment would be capped at the amount the government would pay a plan whose premium was equal to the reference price. Consequently, beneficiaries would apparently pay the full difference between the cost of the plan and the reference price, which is more than the difference between the cost of the plan and the average fee-for-service cost. That requirement would mean that enrollees in plans with a premium just below the average fee-for-service cost--say, at 98 percent of that cost--would have to pay more than the Part B premium. More generally, beneficiaries choosing plans with premiums above the reference price could face hefty additional premium payments.

Suppose, for example, that average costs in the fee-for-service sector were $7,000 and the annual Part B premium for beneficiaries enrolled in that sector was $840, or $70 a month. The reference price would be 96 percent of $7,000, or $6,720. Beneficiaries choosing a less expensive plan with a premium, say, of $6,300 would have their Part B premium reduced by 75 percent of the difference ($420), or $315. So their annual premium would be $525, or $43.75 a month. The government would capture 25 percent of $420, or $105, and would pay a total of $5,775, which is the difference between the plan's premium and the beneficiary's payment.

In this example, if beneficiaries enrolled in plans with premiums at or below 80 percent of average fee-for-service costs, or $5,600, their contributions would be reduced to zero and the government would pay the full premium. By contrast, if they chose a plan with a premium at 110 percent of fee-for-service costs, or $7,700, their Part B premium would be $1,820 (about $152 a month)--more than double the fee-for-service premium. The government's contribution would be capped at $5,880, the difference between the reference price and the fee-for-service premium. That premium structure would give beneficiaries strong incentives to choose lower-cost plans if any were available in their market.

Managed care plans would receive their full premiums for the defined benefit package regardless of whether those premiums were above or below the reference price. But given the price structure that beneficiaries would face, plans would have a strong incentive to keep their premium offers below the reference price; otherwise, they would have difficulty competing against the traditional fee-for-service program. In markets with multiple plans, they would also have an incentive to compete against other managed care plans on the basis of price.

The government would adjust the payments to health plans to reflect differences in risk and geographic differences in cost. Plans enrolling beneficiaries with greater-than-average health risks and plans in high-cost areas would receive higher federal payments than other plans. Payments by beneficiaries would not be adjusted for those factors, however. Rather, beneficiaries would face premiums calculated as if all plans had average risk selection and were in average-cost areas.

Risk adjustment has been considered a perennial problem for the Medicare program, and full implementation of Medicare's new risk-adjustment system is not expected until after 2003. Geographic adjustments have also been problematic. Under this proposal, the government would increase payments to managed care plans in high-cost areas to reflect "full local costs." Payments in low-cost areas would not be reduced, however, below the levels mandated by the Balanced Budget Act.

Although the basic benefit would nominally be standardized, plans would be given the flexibility to reduce or eliminate Medicare's cost sharing as long as the value of cost-sharing reductions did not exceed 10 percent of the value of the benefit package. Plans could offer additional benefits for a separate premium. Both of those options would give them other means to compete against the fee-for-service sector and other managed care plans.

Other Issues

Promoting greater price competition in the Medicare program could broaden the options available to beneficiaries and slow the rate of growth of Medicare spending. Those outcomes are by no means guaranteed, however. Much would depend on the details of the proposal, many of which are unclear, and on the responses of beneficiaries and health plans to new incentives, which are uncertain. Moreover, the potential for effective price competition among health plans varies from market to market across the country. Experience with the Medicare risk program to date suggests that competition is more likely to occur in large, high-cost urban markets, although the nature of the geographic payment adjustment could modify that conclusion.

Under current law, there is effectively no price competition among Medicare+Choice plans. Medicare uses an administered pricing system to set its payments to plans, and plans are not permitted to offer cash rebates or other financial incentives to encourage enrollment. Instead, they have incentives to increase optional benefits rather than to reduce costs. Consequently, even though beneficiaries gain if they enroll in managed care plans that are more efficient than the fee-for-service sector, Medicare does not. Moreover, beneficiaries who might prefer less generous benefits for a lower price do not have that option. The President's proposal would remove that bias and allow both beneficiaries and the Medicare program to benefit from less costly choices.

The proposal goes only part way, however, toward establishing a competitive model for Medicare. The traditional fee-for-service sector--in which the large majority of Medicare beneficiaries are still enrolled--would not be required to compete fully on price with the private plans participating in Medicare. The special status of the fee-for-service sector could result in lower savings than other competitive strategies might yield.

Unlike a competitive model in which the reference premium was based on some average premium in the market, beneficiaries would not have to make payments in addition to the Medicare premium to remain in the fee-for-service sector. Moreover, the presence of low-cost plans would not affect the savings that other plans could offer beneficiaries, because the reference premium would be unaffected. Nonetheless, because the Medicare premium would be based on fee-for-service costs, if those costs rose faster than the costs of managed care plans, those plans might be able to offer beneficiaries significant premium discounts relative to the fee-for-service sector.

How plans would structure their offerings in this new type of competitive environment is very uncertain. It would depend on how responsive beneficiaries proved to be to changes in premiums. To date, what has attracted beneficiaries to switch from fee-for-service Medicare to managed care plans has been the lower cost-sharing requirements and additional benefits (especially coverage of prescription drugs) that those plans offer. With prescription drug coverage available in the fee-for-service sector under the President's proposal, managed care plans would lose one of their major comparative advantages, potentially slowing the growth of enrollment in managed care. How far reduced premiums might offset those effects is unknown. But if medigap premiums continue to rise as rapidly as they have in recent years and employers continue to limit their retirees' health benefits, plans with lower premiums that also offered reduced cost sharing would become increasingly attractive.

The mechanics for bidding and setting prices in the President's proposal are unclear, which adds to the difficulty of predicting the effects of the proposal on plans' behavior. With regard to the hold-harmless provision, for example, the proposal states that the increases in payments to low-cost areas included in the Balanced Budget Act would be maintained, but it does not provide details. The nature of the geographic adjustments for high-cost areas is also unclear. The effects on payments to plans would vary considerably if those adjustments reflected only price differences or if they also included differences in utilization patterns.

In particular, if the geographic adjustment took both price and utilization effects into account, efficient plans in high-cost areas might be able to use high payment rates to subsidize packages of supplemental benefits as well as offer the basic Medicare package for a low or zero premium. (Although plans would be required to charge a separate premium for supplemental benefits, there is no indication that such a premium would have to be anything more than nominal.) Under those circumstances, plans in high-cost markets would be able to compete against the fee-for-service sector and each other on both price and covered benefits. Such competition would be less possible in low-cost markets. Thus, although the proposal intends to reduce the current disparities in benefits among Medicare+Choice plans across the country, that outcome would be quite uncertain.

Another novel factor affecting plans' behavior is the new prescription drug option. The proposal would require plans to offer Part D benefits to beneficiaries who chose to participate in the program. Plans would receive a premium payment from Medicare for those beneficiaries, and they could also offer a separate prescription drug benefit for an additional premium. The premium offers that plans would make would apparently cover both Part B and Part D benefits for those choosing to enroll in Part D. Plans might compete by offering Part D coverage at a low rate or offering additional drug coverage for only a modest extra premium.

Given all of the uncertainties about how the proposal would be implemented and how plans and enrollees might respond, predicting future enrollment trends in Medicare's managed care plans is hazardous. In the short term, the growth of managed care enrollment might slow or even reverse if beneficiaries saw less need to switch from the fee-for-service sector once a prescription drug benefit was available. Even if beneficiaries proved to be highly responsive to reductions in the Part B premium and plans chose to compete on that basis, the effects of the proposal on the growth of Medicare spending are quite speculative. Would there be one-time savings--possibly stretched out over several years--as beneficiaries in fee for service shifted to managed care plans, essentially accelerating the current enrollment trend? Or would competitive forces be strong enough to foster efficiencies throughout the system, slowing the growth of costs in the future? That debate has been going on in the private sector since the mid-1990s, when many enrollees in employer-sponsored plans began to shift from fee-for-service to more tightly managed plans. It has yet to be resolved.
 

TRANSFERS FROM THE GENERAL FUND

The President is proposing to augment Medicare's financing by making transfers from the general fund of the U.S. Treasury to the program's trust funds. Consistent with the policy outlined in the President's budget for fiscal year 2000, CBO estimates that $288 billion would be transferred from the general fund to the Hospital Insurance (HI) Trust Fund over the next decade. That transfer would delay by several years the projected date on which the HI trust fund will become insolvent by committing future general revenues to the program. It would do nothing to address the underlying rapid growth in spending for Medicare that will eventually outrun the revenues dedicated to the program.

An additional $40 billion would be transferred from the general fund to the Supplementary Medical Insurance (SMI) Trust Fund to finance part of the cost of the new prescription drug benefit. (For administrative purposes, Medicare's spending for prescription drugs and beneficiaries' premiums for that benefit would be accounted for in the trust fund.) The transfer would not materially alter the financial status of the trust fund. SMI benefits are funded by premiums, which cover 25 percent of costs, and general revenue, which covers the rest. The statutory formula allows SMI to maintain a small reserve to cover unforeseen contingencies, but the trust fund does not build up substantial reserves. Thus, the additional transfer associated with the prescription drug benefit simply means that the government's costs will be paid for out of general revenues.
 

OTHER INITIATIVES

The President's proposal includes provisions outlined in his last two budgets to allow people under age 65 to buy into Medicare. Although the buy-in provisions have not changed significantly, other facets of the President's proposal might alter the estimates that CBO made earlier this year of participation in the buy-in program and associated costs. The proposal also calls on the National Association of Insurance Commissioners (NAIC) and the Secretary of HHS to develop new supplemental insurance options to protect beneficiaries from catastrophic costs. Such options could fundamentally alter the market for private medigap plans, which supplement Medicare.

The buy-in would be open to two groups: people ages 62 to 64 who do not have access to employment-based health insurance, Medicaid, or other public coverage; and displaced workers age 55 or older who have lost their health coverage because of a job loss. The Administration's description of the provisions, including the estimates of the premiums that participants would pay, is essentially unchanged from the description in the President's budget. But the Medicare program itself would change significantly as a result of the other reforms that the Administration is proposing, especially the addition of a prescription drug benefit. The proposal does not address how the buy-in provisions would be modified by those changes or whether participants would be able to purchase prescription drug coverage. If prescription drugs were included in the benefit package, the buy-in premiums would probably be significantly higher than the Administration is suggesting, and problems of adverse selection in the buy-in program would be exacerbated.

The President's medigap provisions partly address a significant limitation in Medicare benefits--the absence of stop-loss coverage that would protect beneficiaries from catastrophic health expenses. Those provisions would mandate several actions, short of restructuring Medicare benefits.

First, the NAIC would be asked to develop a new medigap option that would limit out-of-pocket expenses and reduce, but not eliminate, beneficiaries' payments for deductibles and coinsurance. (The President's proposal assumes that prescription drug costs would not be covered by the new option.) Such a plan could provide important financial protection while maintaining some cost sharing, which would discourage unnecessary use of covered services. The medigap plans that are now available cover most of Medicare's cost-sharing requirements, and Medicare must bear the cost of the additional use of services induced by such coverage. If people who buy medigap insurance switched to the lower-cost, more basic coverage option, Medicare might reap significant savings.

Second, the Secretary of HHS would be authorized to review the standard medigap packages to determine whether changes should be made to their content or number. The Secretary would also report to the Congress on policy options for improving supplemental coverage for Medicare beneficiaries, including the possibility of having Medicare offer additional, optional coverage to limit out-of-pocket spending. A Medicare-sponsored supplemental plan would probably be extremely popular with beneficiaries, who might view it as more valuable than private insurance because it would be backed by the federal government. Such an insurance policy would severely limit the market for the slimmed-down medigap option that the NAIC is being asked to develop.
 

CONCLUSION

The President's proposal provides a framework for making significant changes to the Medicare program. It is intended to modernize Medicare's benefits, enable the federal government to become a more prudent purchaser of health services, and encourage price competition among health plans to slow the growth of Medicare spending in the longer term. CBO estimates that the President's Medicare reform plan would increase federal outlays by $111 billion over the next decade.

The President proposes a new prescription drug benefit that would provide first-dollar coverage, with an annual limit of $2,500 in 2008, when the benefit was fully phased in. Although most Medicare enrollees would receive some benefit, the proposal would not substantially protect those in poor health who incurred very large out-of-pocket expenses for prescription drugs.

The President proposes to pay for the federal share of the prescription drug benefit through transfers from the general fund. Those transfers are simply promises to pay future benefits with future tax dollars. How burdensome that commitment might become depends on both the growth of future spending for prescription drugs and the growth of the economy over the coming decades.

The President proposes to extend some provisions of the Balanced Budget Act that limit payment updates beyond their 2002 expiration date. The President would also provide a small amount of additional funds to reduce the impact of the act's payment reductions through as-yet-unspecified legislation. On balance, payments to providers would be reduced from baseline levels, although those reductions would accrue only after 2002.

Reducing payment rates for fee-for-service providers would yield Medicare savings without contributing to the program's efficiency. But improving the efficiency of the fee-for-service sector is key to achieving short-term cost savings and longer-term reform. Fee for service is likely to remain the plan of choice for most Medicare enrollees over at least the next decade, even under the most favorable assumptions about the growth of enrollment in managed care plans. Successful adoption of the contracting and payment methods that private health plans use to manage their costs could establish the basis for a competitive fee-for-service sector. But recent efforts to test such methods have not found much acceptance among providers, and the President's proposal treads lightly on that issue.

The President's provisions for rationalizing cost-sharing requirements would modestly increase some of those requirements and lower others, without reducing their complexity. A more thorough reform might subject all Medicare-covered services to a single deductible and uniform coinsurance rates, at the same time placing an annual limit on the amount that enrollees paid in cost sharing for all covered services (including drugs if that benefit was added to the program).

The proposed competitive defined benefit would provide new opportunities for Medicare's managed care plans to compete on the basis of price as well as the generosity of benefits and the quality of service. Although the President's proposal would introduce elements of competition among health plans that could help slow the growth of Medicare spending in the longer term, it would fall short of a fully competitive program. By establishing the fee-for-service sector as the benchmark for defining Medicare benefits and setting premiums for health plans, it would blunt the incentives for efficiency. For that reason, CBO has reservations about the magnitude of savings that could be expected from the competitive defined benefit. CBO has not completed an estimate of that part of the proposal, but the modest savings predicted by the Administration may be reasonable.

The overall effect of the President's proposal is to increase Medicare spending, largely funded with general revenues. Although it would move toward a more competitive system, the proposal would do little to reform the traditional fee-for-service sector.

1. Ellen O'Brien, Diane Rowland, and Patricia Keenan, Medicare and Medicaid for the Elderly and Disabled Poor (Washington, D.C.: Kaiser Commission on Medicaid and the Uninsured, May 1999), p. 9.