After Social Security, Medicare and Medicaid are the largest federal entitlement programs. Together with the State Children's Health Insurance Program, they provide federally funded or subsidized health insurance coverage to millions of low-income, disabled, or aged beneficiaries. This fiscal year, Medicaid will spend about $115 billion on health care for 43 million low-income people and on subsidies to hospitals that serve low-income populations. SCHIP--which gives federal grants to states to provide health insurance for uninsured children who do not qualify for Medicaid--will spend $2 billion. And Medicare will pay for the health care of some 39 million elderly and disabled people at a gross cost of about $221 billion. (Premiums paid by participants cover some of those costs.) Together, those three programs will account for about 18 percent of federal outlays in 2000.

The President's budget for 2001 includes a variety of initiatives to expand health insurance coverage of low-income people through Medicaid and SCHIP and coverage of disabled workers and certain people ages 55 and older through Medicare. The President also proposes adding a prescription drug benefit to Medicare and encouraging health plans to compete on the basis of price by enabling Medicare beneficiaries who choose low-cost plans to pay lower Medicare premiums. In addition, the budget includes various proposals to reduce the growth rate of spending for Medicaid and Medicare.

The President also proposes transferring revenues from the general fund of the Treasury to Medicare's Hospital Insurance Trust Fund. Although that transfer would delay the date on which the trust fund is exhausted, it would not provide any new resources to address the budgetary pressures that will result from the projected rapid growth of Medicare spending over the coming decades.

Taken together, the health care initiatives in the President's budget would raise federal spending by $159 billion over the 2001-2010 period, the Congressional Budget Office estimates (see Table 2-1). Of that total, $26 billion would be spending for Medicaid, $64 billion for SCHIP, and $69 billion for Medicare. In addition, the President's proposals to expand eligibility for Medicare would reduce tax revenues by $8.4 billion and increase Social Security spending (which is off-budget) by $1.4 billion over the same period.
 

Because of the magnitude of federal health insurance programs and the scale of the President's proposals to alter them, an understanding of how those programs spend money is important in evaluating the proposals. Consequently, this chapter reviews trends in spending and enrollment for Medicaid and SCHIP before examining the changes to those programs proposed in the President's budget. It then does the same for Medicare. (The President's trust fund proposals are discussed in Chapter 3.)
 

Spending and Enrollment Trends in Medicaid and the State Children's Health Insurance Program

CBO projects that spending for Medicaid and SCHIP will more than double over the next 10 years. Federal outlays for Medicaid (a joint federal/state program) will rise from $115 billion in 2000 to $264 billion in 2010--an average increase of 8.7 percent a year. That projected growth rate is higher than the rates experienced in recent years but well below the double-digit increases seen in the early 1990s. Federal outlays for SCHIP, which was created by the Balanced Budget Act of 1997, are projected to rise from $2 billion in 2000 to nearly $5 billion in 2010 (see Table 2-2).
 

Medicaid

Medicaid covers a range of acute and long-term care services for eligible beneficiaries. Under current law, spending on those benefits is projected to grow from $100 billion this year to $240 billion in 2010. The program also makes payments (known as DSH payments) to "disproportionate share" hospitals that serve large numbers of Medicaid recipients and other low-income people. CBO expects that statutory limits will keep those payments relatively flat over the next decade; they are estimated to rise only from $9 billion in 2000 to $10 billion in 2010. Administrative expenses are projected to grow from $6 billion this year to more than $14 billion at the end of the decade. Medicaid also provides funding of about $0.5 billion a year to the Vaccines for Children program operated by the Centers for Disease Control and Prevention (CDC).

CBO estimates that total Medicaid enrollment will rise from 43 million in 2000 to almost 51 million by 2010. Enrollees can be divided into four general eligibility categories: aged, blind or disabled, children, and adults. Although the first two categories represent less than 30 percent of total enrollment, they account for more than 70 percent of total spending on benefits. Low-income children and adults make up the majority of enrollees, but they account for a smaller portion of spending because they have much lower health care costs per capita.

Because states share the costs of the Medicaid program with the federal government, states' decisions about eligibility and the scope of benefits affect federal spending. The federal government pays for a percentage of each state's Medicaid outlays according to a formula based on the state's per capita income. That federal matching rate varies from state to state, ranging from 50 percent to 83 percent. On average, the federal government pays for 57 percent of total Medicaid spending; states are responsible for the other 43 percent. The state share of Medicaid spending is also projected to increase dramatically over the next decade--from $88 billion to $202 billion. Combining federal and state shares, total Medicaid spending will rise from $203 billion to $466 billion during that period.

Past and Future Growth of Federal Medicaid Outlays. After a period of rapid growth in the early 1990s, federal spending for Medicaid rose by only 3.3 percent in 1996 and 3.9 percent in 1997. CBO attributes the slowdown in growth to a number of factors: a decline in caseloads due to a combination of strong economic growth and welfare reform, one-time savings as states enrolled Medicaid beneficiaries in managed care, actions by states to limit reimbursements to health care providers, and statutory changes that limited the amount of DSH payments a hospital could receive to 100 percent of its uncompensated care.

That slow growth has not lasted. Medicaid spending increased by 5.9 percent in 1998 and 6.4 percent in 1999. CBO expects higher growth rates in the future: 7 percent in 2000, rising to 9 percent a year after 2003. Although growth of 9 percent represents a sharp increase from the low rates of the mid-1990s, it remains well below growth rates of a decade ago (see Figure 2-1).
 

Figure 2-1. Annual Growth of Federal Medicaid Outlays, 1978-2010 (By fiscal year)

SOURCES: Congressional Budget Office projections and historical data from the Office of Management and Budget.

NOTE: The growth rate for 1999 is estimated.

CBO anticipates that a number of factors will boost spending growth over the next several years. Some of those factors reflect states' decisions about the desired scope of their Medicaid programs; others are a function of the rising cost of providing health care benefits.

First, many of the forces that slowed growth in the mid-1990s have proved temporary. Medicaid caseloads are rising again as states reenroll children and adults who dropped out of the program after welfare reform legislation was enacted in 1996. (Higher enrollment of children has occurred as states identify children who are eligible for Medicaid while they seek out children to enroll in their SCHIP programs.) Health care providers are beginning to negotiate higher payment rates. And a modest slowdown in the growth of enrollment in Medicaid's managed care plans and pressure by those plans to raise payment rates are likely to reduce the savings from moving people into managed care.

Second, many states are expanding Medicaid eligibility under the program's existing authority and under waivers granted by the Health Care Financing Administration (HCFA). Those expansions will probably be modest in scope, focusing on pregnant women and other adults, or be waiver programs that specifically limit enrollment. However, states may choose to expand Medicaid eligibility further using revenues from their multibillion-dollar settlement with the tobacco industry and from the continuing strong economy. States may also expand enrollment of disabled people who receive long-term care services at home or in the community because of concerns about complying with the Americans with Disabilities Act.

Third, administrative costs are expected to rise rapidly as states spend more on enrolling people who left the rolls under welfare reform and on other aspects of program management. States will also continue their long-standing efforts to receive additional federal matching payments by converting programs that they now fund themselves into Medicaid programs.

Fourth, Medicaid costs are expected to increase because of recipients' higher utilization of services such as prescription drugs and long-term care provided in community settings.

Trends in Spending on Benefits. Health care benefits for enrollees account for the vast majority of Medicaid spending. More than 60 percent of spending on benefits goes for acute care services--a category that includes payments to fee-for-service providers, capitated (per capita) payments to managed care plans, and Medicare expenses for certain beneficiaries. The rest of benefit spending goes for institutional and noninstitutional long-term care services.

Acute Care Services. Payments made to health care providers on a fee-for-service basis are projected to increase from $43 billion in 2000 to $102 billion in 2010--an average of about 9 percent a year (see Table 2-2). That projection reflects the movement of beneficiaries out of fee-for-service arrangements and into managed care plans. The fastest growing component of fee-for-service acute care is spending for prescription drugs, which is projected to rise by about 12 percent annually because of both higher utilization and higher drug prices. Spending for some other acute care services--such as outpatient hospital, clinic, diagnostic, and screening services--is also expected to grow faster than the average, rising by about 10 percent a year.

Outlays for Medicaid's managed care plans, which receive a fixed payment per enrollee, are projected to rise from $15 billion in 2000 to $42 billion in 2010. Those figures include payments for fully and partially capitated plans, primary care case-management fees, and other payments for uncompensated care that are included in some states' waiver programs. CBO projects that the annual growth of managed care payments will slow from 15 percent in 2000 to 10 percent in 2004. That slower growth is expected to occur for two reasons: some plans will leave the market because of insufficient capitation rates, and enrollment will increase more slowly as states face difficulties moving chronically ill aged and disabled beneficiaries into managed care.

Besides its regular benefits, Medicaid also pays the premiums and out-of-pocket expenses incurred by some low-income Medicare beneficiaries. For so-called qualified Medicare beneficiaries (QMBs), who have income below the federal poverty level, Medicaid pays all of Medicare's premiums and cost-sharing amounts. (About three-quarters of QMBs are dually eligible for full benefits under both programs.) For specified low-income Medicare beneficiaries (SLMBs), who have income between 100 percent and 120 percent of the poverty level, Medicaid pays the premiums for Part B of Medicare (Supplementary Medical Insurance). In addition, certain qualified individuals who have income between 120 percent and 175 percent of the poverty level are eligible to have Medicaid pay some or all of their Part B premiums through 2002. (Although state Medicaid programs initially make the Part B premium payments for the latter group, Medicare reimburses states for 100 percent of the costs.) All together, CBO estimates that Medicaid spending on Medicare premiums for those three categories of low-income beneficiaries will more than double in the next decade, from $2.5 billion to $6.1 billion.

Long-Term Care Services. CBO projects that federal Medicaid spending on long-term care services will rise from $39 billion this year to $90 billion in 2010. In that projection, spending on long-term care provided in institutions such as nursing homes is expected to grow less rapidly (8 percent a year) than spending on noninstitutional care (12 percent a year). Noninstitutional care includes a range of medical and support services, such as home health care, personal care, and other services offered under special waivers for home- and community-based care. Because of its faster growth rate, noninstitutional care will account for an increasing share of total spending on long-term care, rising from 28 percent in 2000 to 37 percent in 2010.

Trends in DSH Payments. The Balanced Budget Act set specific state allotments for disproportionate share hospital payments for each year of the 1998-2002 period. After that, state allotments will rise at the same rate as the consumer price index but will be capped at 12 percent of a state's total Medicaid spending. The act also limited state DSH payments to mental health facilities. CBO estimates that those limits, along with statutory restrictions on the amount that states may pay individual hospitals, will prevent some states from spending their entire DSH allotments. As a result, federal DSH payments are expected to decrease from $8.6 billion this year to $8.1 billion in 2002. After that, they are projected to grow by about 2.4 percent a year, reaching $9.8 billion in 2010.

Trends in Administrative Costs. CBO estimates that Medicaid's spending on administration will grow at an average annual rate of 9 percent through 2010, rising from $6 billion to $14 billion. States are expected to drive that growth by increasing their efforts to enroll people who left the rolls under welfare reform and continuing to shift the administrative expenses of other welfare programs into Medicaid to maximize federal matching payments. Also contributing to growth is higher spending for improvements to computer systems, administration related to the provision of school-based health services, and mental health case management.

Trends in Enrollment. CBO estimates that over the 2000-2010 period, Medicaid enrollment will rise at an average annual rate of 1.6 percent. The number of blind or disabled enrollees will increase most quickly--by 3 percent a year, on average. That growth is tied largely to continued increases in the number of disabled recipients in the Supplemental Security Income (SSI) program, which usually confers eligibility for Medicaid as well. CBO also anticipates some growth in disabled enrollees because of states' efforts to comply with the Americans with Disabilities Act.

Enrollment of children is expected to increase by 3.1 percent next year before moderating in later years. As noted earlier, enrollment will jump in the near term as states work to enroll children in SCHIP and, in the process, identify children who are eligible for Medicaid. (Such children cannot participate in SCHIP.) In addition, children who dropped off the rolls after welfare reform have begun to return to the Medicaid program. Enrollment of adults is also likely to increase modestly in the near term as states expand coverage of parents, as permitted by the 1996 welfare reform law.

State Children's Health Insurance Program

The Balanced Budget Act established SCHIP as title XXI of the Social Security Act. The program gives states matching funds to provide health insurance to children who are uninsured but do not qualify for Medicaid. States can use their SCHIP funds to enroll children in their existing Medicaid program or establish a separate program. To date, about 70 percent of states have chosen to establish separate programs. To encourage states to participate in SCHIP, the federal government matches their SCHIP spending at a higher rate than Medicaid spending (70 percent versus 57 percent, on average).

Unlike Medicaid, which is an open-ended entitlement program, federal funding for SCHIP is limited. The federal government allocates money to the states each year for SCHIP on the basis of how many low-income and uninsured children they have. Total SCHIP allotments will be $4.2 billion in 2000 and 2001, drop to $3.1 billion in 2002, and then rise again, reaching $5.0 billion by 2007. States have three years to spend their SCHIP allotments; after that, any unspent funds are reallocated to states that have exhausted their allotments and are made available for another year.

If a state runs out of SCHIP money, the consequences for children enrolled in the program depend on whether the state operates a separate program or one that enrolls children in Medicaid. Children in states that expanded their Medicaid programs with SCHIP funds are entitled to Medicaid benefits. If such a state uses up its SCHIP allotment, children can continue to receive services through Medicaid (although states will receive the lower Medicaid matching rate for those services). Children in states with separate programs, by contrast, are not entitled to Medicaid benefits; they may have to wait until the following year's funding becomes available to continue receiving services.

Trends in Spending. CBO projects that federal outlays for SCHIP will rise from $2.0 billion this year to $4.8 billion in 2010 (see Table 2-2). Outlay growth will be particularly strong in the next few years as states build up their programs. In 1998 and 1999 combined, federal spending was only about $1 billion. That amount was much lower than the total available allotments for several reasons: the lengthy lead time necessary to develop large-scale health insurance programs, the constraints of states' legislative calendars, and program rules that limit administrative and outreach spending to 10 percent of total spending. States may also have been exercising caution as they planned the size of their programs out of concern about sharp drops in their allotments beginning in 2002.

Some states have already planned programs with large enrollment goals that will require as much or more funding than will be provided in their allotment amounts. Other states will need a few more years to develop larger programs. Still others may have programs that never use available allotment amounts because of the states' financial constraints or the preferences of state governments. In most years, states with large programs are expected to rely in part on the reallocation of unused funds from other states to support their programs. CBO estimates that in 2001, $1.8 billion of the total 1998 SCHIP allotment will be reallocated, and $0.5 billion of that amount will never be spent. In later years, a smaller proportion of the total allotment is likely to be reallocated as states develop greater capacity to spend their allocations within three years. Given states' preferences and constraints, SCHIP programs are expected to be fully developed by 2003.

Trends in Enrollment. Assessing enrollment levels in SCHIP is difficult for two reasons. First, to date, the program's experience suggests that children enroll in SCHIP for short spells, moving on and off the rolls as their family's income changes. Variation in earnings is common among families with children eligible for low-income health programs. SCHIP is designed to continue children's health coverage when their family's income rises above eligibility levels for Medicaid; conversely, those children can return to Medicaid from SCHIP when their family's income falls. In fact, many children who leave SCHIP reportedly reenroll in Medicaid.

Second, as states build up their SCHIP programs, reported levels of enrollment often account for only a few months of coverage during any given year. Consequently, data on the number of children enrolled in the program at a particular point in time, or at any time since the program's inception, overstate the number of children covered on a full-year-equivalent basis.

CBO estimates that enrollment in SCHIP could reach 2.1 million this year, with about 1.6 million children enrolled on a full-year-equivalent basis. After 2002, enrollment could reach a steady state of 2.5 million on an average annual basis, or closer to 2 million on a full-year-equivalent basis.
 

The President's Proposals for Medicaid and SCHIP

The President's budget includes numerous proposals that would affect the Medicaid and SCHIP programs. CBO estimates that those proposals would increase federal outlays for Medicaid by $26 billion through 2010 (see Table 2-3). They would raise federal spending for SCHIP by $64 billion over the same period.
 

The main Medicaid and SCHIP proposals focus on expanding eligibility. The most far-reaching of those initiatives, the FamilyCare proposal, would provide health insurance coverage to parents of children enrolled in Medicaid or SCHIP. The President's budget would also expand Medicaid and SCHIP eligibility to 19- and 20-year-olds, some aged or disabled people with income below 300 percent of the SSI benefit standard, and certain women with breast or cervical cancer. Other proposals would restore full Medicaid eligibility to immigrants who lost coverage under welfare reform, allow states to use SCHIP funds to cover immigrant children, and extend the Medicaid coverage of people who lose eligibility because of an increase in their earnings.

Other proposals in the budget would expand states' outreach efforts to children who are eligible for but not enrolled in Medicaid and SCHIP, reduce Medicaid spending for prescription drugs, and decrease federal reimbursement for some administrative costs related to enrolling welfare recipients in Medicaid. Proposals by the President that would affect prescription drug coverage, the Part B premium, and cost sharing in the Medicare program (discussed later in this chapter) would also increase spending for Medicaid. Finally, the President's budget includes proposals to combat fraud in Medicaid, improve the program's operations, and improve public health.

The FamilyCare Proposal

Under current law, Medicaid coverage in most states is far more expansive for children than for parents. Because of a combination of federal requirements and optional eligibility expansions by states, Medicaid typically covers pregnant women and children at income levels at or above the federal poverty level. For most parents, by contrast, eligibility standards for Medicaid remain in line with standards for welfare programs, which are substantially below the poverty level. Although current law allows states to expand coverage of those parents, states have been cautious in their approach.⁽¹⁾ Also, states cannot use their SCHIP funds to cover the parents of children eligible for that program, although several states have expressed interest in doing so.

The President's proposal would add $50 billion in budget authority between 2002 and 2010 to existing funds for SCHIP and rename that program FamilyCare. The FamilyCare proposal would expand eligibility for parents of Medicaid and SCHIP children--using an enhanced matching rate initially to motivate states to participate and a new federal requirement later to compel states to take part. The federal government would match FamilyCare spending at the same enhanced rate that applies in SCHIP (70 percent, on average). The proposal would also make eligibility standards for Medicaid more uniform among states. It would require that by 2006, all parents with income below the poverty level would be eligible for Medicaid and all children with family income below 200 percent of the poverty level would be eligible for either Medicaid or SCHIP. The program would provide enhanced matching funds to states that already cover parents above the poverty level and children with family income above federally mandated eligibility requirements.

The FamilyCare funding in the President's budget, however, would be insufficient to cover the full costs of the program after 2006, CBO estimates. Once annual FamilyCare funds were exhausted, additional costs for many of the program's beneficiaries would shift to Medicaid. Those additional Medicaid outlays would be matched at the lower Medicaid rate (57 percent, on average). Overall, CBO estimates, net federal spending on FamilyCare and Medicaid under this proposal would total $56 billion through 2010 (see Table 2-3). If spending were not constrained by the insufficient funds, the net costs of the proposal would increase by $13 billion over the 2001-2010 period.

How the Program Would Operate. The FamilyCare proposal would be implemented in two distinct phases: an optional phase between 2001 and 2005 and a mandatory phase starting in 2006. The ways in which states could use their FamilyCare funds would differ tremendously between the two phases. During the mandatory phase, states would be required to pay for coverage of some FamilyCare beneficiaries through Medicaid (and some Medicaid beneficiaries through FamilyCare). States would receive an enhanced matching rate for some beneficiaries even though those people would be covered by Medicaid. The FamilyCare proposal is complex, and some key details are not specified in the President's budget.

Phase I: Voluntary Expansion, 2001-2005. To participate in the new program, states would first be required to expand SCHIP eligibility to children with family income up to 200 percent of the poverty level and eliminate waiting lists. The President's proposal would make FamilyCare funds available for five years (rather than the current three years under SCHIP) to give states more time to spend their money. CBO assumes that states could use their new FamilyCare funds to meet those initial requirements if their existing SCHIP allotments had run out.

After meeting those requirements, states could voluntarily extend coverage to parents of children in Medicaid or SCHIP. States would be required to enroll parents in the same program as their children and to enroll the parents of Medicaid children before the parents of SCHIP children. Starting in 2002, funding for those coverage expansions would come from the new FamilyCare allotments. (CBO assumes that any coverage expansions made in 2001 would be paid for out of current SCHIP allotments.) The federal government would match states' spending for FamilyCare at the same rate that now applies to SCHIP (although states would not be able to receive that rate for parents who have income below eligibility limits that were established before January 1, 2000).

Phase II: Mandatory Expansion, 2006 and After. Starting in 2006, all states would be required to expand eligibility to parents of Medicaid-eligible children up to the poverty level. That mandate has a complicated twist. States would have to cover all of those parents through Medicaid, including parents in that income category who were previously covered by FamilyCare. Therefore, some of the parents would shift from FamilyCare to Medicaid, but others would be newly enrolled in Medicaid. Expanded coverage for parents with income above the poverty level would still come from the FamilyCare allotments. Either way, states would receive the enhanced matching rate for all parents enrolled under the proposal. Only states that had expanded eligibility to parents under the poverty level before January 1, 2000, would fail to get an enhanced matching rate.

However, states that had covered parents with income above the poverty level before January 1, 2000, or that covered children above federally mandated eligibility requirements would be rewarded. In the first phase of the proposal, they would continue to receive the regular Medicaid match for those beneficiaries, even as other states that expanded their programs after the proposal's enactment received an enhanced match through FamilyCare. In phase II, however, the early-acting states would be able to transfer financing of those parents and children to FamilyCare (with its higher matching rate), even though those beneficiaries would remain entitled to Medicaid benefits.

CBO's Estimate. The President's budget would pay for the FamilyCare program with specified levels of budget authority that would fall short of full funding. To assess the extent of that shortfall, CBO estimated the implications of the proposal for the federal budget in two ways--without and with the constraints of the available budget authority (see Table 2-4). The first approach treats FamilyCare as an open-ended entitlement program and illustrates the full scope of the proposed expansion. For the second approach, CBO had to make assumptions about the amount of FamilyCare funding that would be available each year because the President's budget does not specify annual amounts. CBO then assessed whether that funding would be enough to cover the program's projected spending each year. The two approaches yield very different results.
 

Phase I: Voluntary Expansion, 2001-2005. CBO assumed that all states would participate in the FamilyCare program to receive higher matching funds for expanding eligibility. Nonetheless, some states might choose not to take part during the voluntary phase. Poorer states already have higher-than-average matching rates in Medicaid and would receive only a modest enhancement under the proposal. In addition, states that now have more restrictive eligibility standards for adults and children and are not spending all of their SCHIP allotments might see little reason to participate voluntarily. States might also tailor the timing and magnitude of their participation in FamilyCare with an eye to the shortfall occurring in phase II.

CBO estimates that the first phase of FamilyCare would increase spending by $26.6 billion through 2010. About 20 percent of that spending would be for states to meet the conditions of participating in FamilyCare. Approximately 20 states now have eligibility limits for SCHIP below 200 percent of the poverty level. Expanding those limits to qualify for FamilyCare would lead some of the states to exhaust their SCHIP allotments; CBO assumed that they would be able to use funds from their FamilyCare allotments to cover any shortfall. Eliminating waiting lists for SCHIP would have a similar effect on states whose SCHIP programs were near their allotment limits.

Making SCHIP funds available for five rather than three years, as the President proposes, would reduce SCHIP outlays by an estimated $0.9 billion in 2001. Because SCHIP funds that remain unspent after three years are reallocated to states that have used up their allotments, giving states two more years to spend their allotments would essentially shift SCHIP funds to states that are less likely to spend them. States that had been planning to use reallocated SCHIP money in 2001 would have to compensate for the change by cutting their program, using state funds to make up the difference, or (if their SCHIP program was implemented as a Medicaid expansion) covering the affected beneficiaries under the state's regular Medicaid program. In later years, those states could also respond by using a portion of their FamilyCare allotments.

During the 2001-2005 period, CBO estimates, states would spend $10.9 billion of their FamilyCare allotments. About 85 percent of that would cover parents of Medicaid children, with the rest covering parents of SCHIP children. Total SCHIP/FamilyCare allotment amounts over the period would be sufficient to pay for that spending, particularly because many states have significant unspent SCHIP funds. CBO assumes that states would expand their coverage of parents gradually, enrolling 50 percent of eligible parents with income below the poverty level, 25 percent of parents of Medicaid children with family income above the poverty level, and 12.5 percent of parents of SCHIP children by 2005.

The FamilyCare program would affect Medicaid in several ways during that period. First, Medicaid spending would increase as states that expanded their SCHIP programs identified children who were eligible for Medicaid. Second, Medicaid would pick up the costs for some of the children who lost SCHIP coverage because of the extension of funding availability from three to five years. Third, FamilyCare would induce some parents of children who were eligible for Medicaid but not enrolled to enroll their children. In the other direction, several states that would expand coverage to parents of Medicaid children anyway (after January 1, 2000) under current law would probably implement those expansions through FamilyCare, with its enhanced federal match, instead of through Medicaid. That switch would reduce Medicaid outlays, CBO estimates. On balance, the FamilyCare program would increase federal spending for Medicaid by $0.3 billion over the 2001-2005 period (see Figure 2-2).
 

Figure 2-2. Budget Authority and Outlays Under the President's FamilyCare Proposal, 2001-2010 (By fiscal year)

SOURCE: Congressional Budget Office.

NOTE: The numbers for FamilyCare include budget authority and outlays projected under current law for the State Children's Health Insurance Program, which FamilyCare would replace. They assume the spending constraints in the President's budget.

Phase II: Mandatory Expansion, 2006 and After. In 2006, as states were required to make all parents of Medicaid children with family income below the poverty level eligible for Medicaid (at an enhanced matching rate), FamilyCare spending on those parents would drop. At the same time, federal Medicaid spending would surge by more than $4 billion a year to cover those parents and others made newly eligible under the requirement.

From a budgetary standpoint, a more significant change would occur in 2006: states that already covered parents of Medicaid children with family income above the poverty level as well as children beyond federally mandated levels would begin paying for those beneficiaries through FamilyCare instead of Medicaid (at an enhanced matching rate). Ignoring any constraints imposed by available funding, CBO estimates that the shift would cost FamilyCare $14.6 billion in 2006 and a total of $85.4 billion over the 2006-2010 period. Medicaid outlays would plummet by $11.9 billion in 2006 and a total of $69.5 billion through 2010 as a result of that shift. The net effect on federal outlays--a $15.8 billion increase over the 2006-2010 period--reflects the higher cost of covering those beneficiaries under FamilyCare's enhanced matching rate. About 75 percent of the additional outlays would be for children.

Insufficient Funding. If the FamilyCare proposal was adequately funded, program spending would total $112 billion over the 2001-2010 period. (Including outlays for SCHIP, that amount would be $150 billion.) But in CBO's view, the amounts requested for FamilyCare in the President's budget would not be enough, even when combined with SCHIP allotments. Problems would begin in 2007, when states would have only $17 billion available to cover an estimated $23 billion in spending (see Table 2-4).

If states ran out of FamilyCare funds, about 90 percent of FamilyCare beneficiaries would retain Medicaid coverage. Those beneficiaries include parents of Medicaid children with family income above the poverty level and Medicaid children with family income above the federally mandated minimums. States would continue to cover those people under their Medicaid programs but would receive the lower Medicaid matching rate.

Prospects for SCHIP children and their parents would be murkier. In states that had enrolled them through an expanded Medicaid program, those enrollees would still be entitled to Medicaid after FamilyCare funds ran out. But enrollees would have no such guarantee in states that had created separate programs. Those states would face three choices when FamilyCare funds were exhausted: scale back their programs, fund any additional costs entirely from state coffers, or make their Medicaid eligibility standards less restrictive (which would allow them to receive additional federal matching funds, albeit at the lower Medicaid rate).

The constraints imposed by the President's budget request for FamilyCare would reduce the program's outlays substantially--to $64 billion over the 2001-2010 period, CBO estimates, compared with $112 billion. Likewise, Medicaid savings would be only $7 billion over 10 years, compared with $43 billion (see Table 2-4).

Comparison with the Administration's Estimate. CBO's estimate of net federal outlays under the FamilyCare proposal is substantially lower than the Administration's estimate. CBO projects that those outlays, including savings in Medicaid, would total $56.2 billion through 2010 (with funding constraints). By contrast, the Administration estimates the net cost of the proposal at $76.0 billion.

Other Proposals to Expand Eligibility

Expanded Coverage of 19- and 20-Year-Olds. Under current law, states are limited in their ability to cover 19- and 20-year-olds in Medicaid. People in that age group can qualify for Medicaid if they are pregnant, disabled, or have children of their own enrolled in the program and meet welfare-related standards. Some people who were formerly in foster care are also eligible. In all, about 1 million 19- and 20-year-olds, mostly with income below 200 percent of the poverty level, receive Medicaid through one of those categories. States are not allowed to cover 19- and 20-year-olds in their SCHIP programs.

The President's budget would give states the option of covering 19- and 20-year-olds in both programs. CBO anticipates that states that together contain 25 percent of the eligible people in that age group would take up the option for Medicaid and that states with 5 percent of the eligible population would do so for SCHIP. As a result, CBO estimates, the proposal would increase enrollment of 19- and 20-year-olds in Medicaid by about 10 percent, raising spending by $1.7 billion over the next decade. Spending for SCHIP would increase by $0.4 billion (see Table 2-3).

Expanded Eligibility for Long-Term Care Services. The President's budget would also allow states to extend Medicaid eligibility for noninstitutional long-term care services to certain aged or disabled people with income up to 300 percent of the SSI standard (which is now $512 per month). Qualified individuals would require the same high level of medical and social support needed by people in institutions and those receiving home- or community-based services.

Current law lets states expand eligibility for all aged or disabled people with income up to the poverty level. Many states set substantially higher income limits for people requiring institutional long-term care; that limit is commonly set at 300 percent of the SSI standard (which corresponds to about 225 percent of the poverty level). In addition, states can offer non-institutional long-term care services to qualified people at similar income levels through special waivers for home- and community-based services.

Under the President's proposal, states could expand eligibility more broadly than under a waiver, which frequently limits enrollment to a specific, narrow set of people. However, CBO assumes that relatively few states--representing 5 percent of total Medicaid enrollment--would take advantage of the option, because of concerns about potentially large jumps in enrollment. (Many people in the expanded income-eligibility range who do not receive long-term care services under current law could meet the qualifications for institutional or waiver care.) CBO estimates that this proposal would increase Medicaid spending by a total of $1.3 billion over the 2001-2010 period.

Coverage for Certain Women with Breast or Cervical Cancer. The President's budget would give states the option of providing Medicaid coverage to women who have been screened under the CDC's Breast and Cervical Cancer Early Detection Program and found to have breast or cervical cancer. Currently, screening services under that program are available only to women with income up to 250 percent of the poverty level (most states set their eligibility criteria at about 200 percent of the poverty level). States that exercised this option would receive the regular Medicaid matching rate from the federal government. CBO estimates that states that account for 15 percent of total Medicaid costs would expand eligibility in 2001 under this proposal. By 2005, that share would rise to around 30 percent. Under those assumptions, the proposal would increase federal Medicaid spending by $0.6 billion over 10 years, CBO estimates.

Eligibility for Some Legal Immigrants. The President's budget contains three proposals that would soften the restrictions on Medicaid eligibility for legal immigrants that were enacted as part of welfare reform. Under that reform, most legal immigrants who entered the country after August 22, 1996, are not eligible for nonemergency Medicaid services until they accumulate 10 years of Social Security-covered employment or become naturalized citizens. Eligibility was also tightened for other means-tested programs, such as SSI and Food Stamps.

Under the President's budget, immigrants who had been in the country for five years and who became disabled after entry would be allowed to receive SSI disability benefits. That proposal would also affect Medicaid because most SSI recipients are automatically eligible for Medicaid. In addition, the budget would give states the option of expanding their Medicaid programs to cover immigrant children and pregnant women. (The option for immigrant children would also make them eligible for SCHIP.)

CBO estimates that those proposals would increase federal outlays for Medicaid by $10.2 billion and for SCHIP by $0.2 billion through 2010. Most of the additional outlays--$8.8 billion--would be for disabled immigrants receiving SSI. Those estimates are based on historical data about immigrants' participation in SSI and Medicaid; they also account for the fact that immigrants already receive limited Medicaid benefits if they require emergency medical treatment. On the basis of discussions with state officials, CBO assumed that states that together contain 90 percent of total Medicaid enrollees would ultimately choose to cover immigrant children and pregnant women.

Extension of Transitional Medicaid. Under current law, states must provide "transitional" Medicaid coverage to beneficiaries who would normally lose their Medicaid coverage when their earnings increase. States are required to cover those people for at least six months without regard to their earnings and for up to a year subject to certain income limits. Some states have extended that coverage to 18 or 24 months. The requirement will expire at the end of fiscal year 2001. The President's budget would permanently extend it and simplify income-reporting requirements for people enrolled in transitional Medicaid. CBO estimates that those changes would increase spending by $4.8 billion through 2010.

Proposals to Increase Outreach

Some 4 million to 5 million children meet the eligibility standards for Medicaid but are not enrolled in the program. The President's budget includes three initiatives that aim to get more of those children enrolled by increasing outreach efforts. The first would let states expand the kinds of entities that can conduct presumptive eligibility determinations (which enroll poor, uninsured children in Medicaid on a temporary basis until a full determination of their eligibility can be made) to include schools, SCHIP workers, and homeless shelters. The second initiative would give states access to income information from the National School Lunch Program, which would give them another tool to identify eligible but unenrolled children. The third proposal would require states to make their enrollment requirements for Medicaid as simple as those for their SCHIP programs.

CBO estimates that those outreach proposals would bring less than 5 percent of the eligible but unenrolled children into Medicaid. As a result, the proposals would increase federal Medicaid spending by $1.2 billion over the next decade. They would also increase SCHIP spending by $0.1 billion through 2010.

Proposals Affecting Medicaid's Prescription Drug Benefit

Unlike Medicare, Medicaid covers prescription drugs for many beneficiaries. To lower Medicaid's spending on prescription drugs, HCFA negotiates rebates with drug manufacturers. The President's budget contains three proposals that would affect the program's prescription drug benefit. They would provide data about drug prices to state Medicaid programs, require an additional rebate from manufacturers of generic drugs, and extend coverage to certain drugs intended to help people stop smoking.

Providing Average Manufacturer Price Data to State Medicaid Programs. The President proposes allowing the Secretary of Health and Human Services (HHS) to give state Medicaid programs information about the average manufacturer price (AMP) of each drug that Medicaid covers. The AMP is the price that wholesalers pay, including discounts and price reductions, for drugs distributed to retail pharmacies. Drug manufacturers are required to provide that information on a confidential basis to HCFA in order to participate in Medicaid. HCFA uses the AMP information to calculate the amounts that manufacturers owe under Medicaid's rebate program (which is explained below). Most states set their drug reimbursement rates on the basis of estimates of pharmacies' acquisition costs plus a dispensing fee. States generally arrive at those estimates by applying a discount to the published list price. Certain drugs are also subject to specific upper payment limits.

Having AMP data would let states reevaluate their current payment rates. But the degree to which they would incorporate the new information is highly uncertain. CBO assumes that states would use the AMP data to reduce their reimbursement rates, lowering Medicaid outlays by $1.3 billion through 2010.

Requiring an Additional Rebate from Manufacturers of Generic Drugs. Under current law, pharmaceutical manufacturers must sign a rebate agreement with the Secretary of HHS before Medicaid will cover their products. Rebates are based specifically on the AMP and on the number of units of the drug purchased by Medicaid. Manufacturers of both brand-name and generic drugs pay a basic rebate to Medicaid, but those rebates are calculated using different formulas. For makers of generic and other over-the-counter drugs, the basic rebate equals 11 percent of the AMP. For brand-name drugs, manufacturers must pay an additional rebate if they increase the prices of those drugs at a higher rate than general inflation. The additional rebate discourages manufacturers from circumventing the effect of the basic rebate by raising prices. The President's budget would extend that additional rebate to generic drugs.

Given the competitive nature of the market for generic drugs, extending the additional rebate to cover those drugs would bring in significantly less money than the additional rebates on brand-name drugs do. Nevertheless, recent evidence suggests that price increases for some generic products have exceeded general inflation. CBO estimates that this proposal would save $0.8 billion over 10 years.

Mandating Coverage of Smoking-Cessation Drugs. Medicaid allows states to exclude coverage of certain types of drugs, including drugs for fertility, cosmetic purposes, and smoking cessation. However, an increasing number of states have recently expanded their coverage of smoking-cessation drugs, although the scope of that coverage varies from state to state. The President proposes requiring all states to cover prescription drugs, and at least one over-the-counter product, for smoking cessation. CBO estimates that this proposal would increase Medicaid outlays by $0.2 billion over the 2001-2010 period.

Proposal to Reduce Medicaid's Administrative Costs

Before the 1996 welfare reform law, the three major public assistance programs--Aid to Families with Dependent Children (AFDC), Food Stamps, and Medicaid--all reimbursed states for 50 percent of most administrative costs. States usually charged the common administrative costs of those programs to AFDC. When the welfare reform law replaced AFDC with Temporary Assistance for Needy Families (TANF), an amount equal to historical administrative costs (including the common costs of administering Medicaid and Food Stamps) was included in the states' block grants. Under current law, states must charge part of the common costs of Medicaid and TANF to Medicaid, even if those costs are already included in their TANF block grants.

The President's budget proposes to reduce federal reimbursement for Medicaid administrative costs to reflect costs that are estimated to be covered by the TANF block grant. The reduction would amount to about one-third of the common costs of administering those programs. However, the proposal would allow states to use TANF funds to pay those costs. CBO estimates that the proposal would reduce federal Medicaid outlays by $3.6 billion over 10 years. But outlays for TANF would rise by $0.8 billion over that period as states offset some of the drop in Medicaid reimbursements.

Interactions with Medicare Proposals

Because Medicaid pays premiums and cost-sharing amounts for many Medicare beneficiaries with low income, policies that affect Medicare spending often have an impact on the Medicaid program. Several of the President's proposals for Medicare (which are described later in this chapter) would affect Medicaid spending for those beneficiaries. CBO estimates that the net effect of the proposals (excluding the proposal to add a prescription drug benefit to Medicare) would be to increase federal Medicaid outlays by $0.8 billion through 2010.

The President's prescription drug proposal for Medicare would have a substantially greater effect on Medicaid spending for low-income Medicare beneficiaries. That effect is described in greater detail below, but its net cost to Medicaid would be $12.4 billion over the next decade. Moreover, Medicaid would incur further costs because the new drug benefit would induce more low-income Medicare beneficiaries to enroll in Medicaid. CBO estimates those additional costs at $6.3 billion through 2010.

Other Proposals

The President's budget contains several other proposals that would have a small net effect--less than $50 million a year--on Medicaid spending. They include initiatives that would save money by combating fraud and improving program operations, such as providing the Secretary of HHS with stronger enforcement tools, allowing civil penalties to be pursued against nursing home chains, and strengthening enforcement of orders for parents to pay medical child support. The budget also includes proposals that would increase Medicaid spending, such as launching an asthma-management initiative, exempting caregivers when Medicaid costs are recovered from estates, changing the car allowance rules in the Food Stamp program, and making certain housing funds contingent on the use of Medicaid waivers for home- and community-based services.
 

Spending and Enrollment Trends in Medicare

The growth rate of Medicare spending has slowed dramatically in recent years. After increasing by an average of 11 percent per year from 1990 through 1995, spending rose by 8 percent in 1996 and just 1.5 percent in 1998. In 1999, Medicare spending did not grow at all but instead declined by 0.7 percent.⁽²⁾

The decline in 1999 stems from several factors. First, it reflects a continuation of two trends that began in the mid-1990s: a slowing of growth in enrollment, and the effect of antifraud initiatives on compliance with Medicare's rules for payment. Second, the drop reflects changes in payment rates and other program rules required by the Balanced Budget Act (BBA).

The lull in the growth of Medicare spending is likely to prove short--in part because the Medicare, Medicaid, and SCHIP Balanced Budget Refinement Act of 1999 (BBRA) will increase Medicare's payments to health care providers beginning in 2000. CBO projects that spending growth will resume in the next few years and will average about 7 percent annually through both 2005 and 2010 (see Table 2-5). Total Medicare outlays (mandatory and discretionary) are projected to nearly double by the end of the decade, to $438 billion. Much of that increase will reflect rising costs per beneficiary. Enrollment will expand only modestly, as the last of the relatively small cohorts born in the late 1930s and early 1940s reach age 65.
 

In the decades after 2010, Medicare spending will grow more rapidly, as the baby boomers begin to turn 65. Between 2010 and 2030, the elderly population will increase at a rate three times faster than between 2000 and 2010. Medicare costs are likely to keep growing considerably faster than program enrollment, however, because of advances in medical technology that are expected to raise health care costs and because of a continued increase in beneficiaries' use of services.

Medicare+Choice

The Balanced Budget Act established the Medicare+ Choice program to expand the range of health plans available to beneficiaries and to lay the foundation for a more competitive Medicare system. Building on Medicare's previous risk-based sector, in which all of the plans were health maintenance organizations, Medicare+Choice allows a wider variety of health plans (including preferred provider organizations, private fee-for-service plans, and provider-sponsored organizations) to participate in Medicare. Whereas traditional Medicare pays health care providers on a fee-for-service basis, Medicare+Choice plans receive a fixed amount per enrollee for providing services.

CBO projects that payments for Medicare+Choice and other group plans will soar from $41 billion this year to $133 billion in 2010, as enrollment in those plans continues to expand (see Table 2-6). That spending increase also reflects projected growth in costs per enrollee that (under current law) will roughly mirror growth in the fee-for-service sector. Within that sharply upward trend, however, annual changes in Medicare+Choice spending will vary considerably. Those fluctuations result from technical aspects of Medicare's reimbursement policy rather than sudden changes in underlying spending patterns.⁽³⁾
 

CBO projects that enrollment in Medicare's risk-based plans will grow by about 5 percent this year, to 6.4 million.⁽⁴⁾ That projection is lower than CBO had previously predicted. The reason is that Medicare beneficiaries have become less likely to switch from fee-for-service to Medicare+Choice plans since those plans announced higher premiums and reduced benefits for 2000 and withdrew from some localities. Over the longer run, however, CBO expects Medicare+ Choice plans to continue offering more generous benefit packages than fee-for-service Medicare. Consequently, enrollment will continue to grow over the next decade--from 16 percent of Medicare enrollees this year to 31 percent in 2010. Because per-enrollee payments to Medicare+Choice plans are tied to fee-for-service expenditures, higher enrollment in those plans does not necessarily slow the growth rate of Medicare spending.

Fee-for-Service Medicare

CBO projects that spending in Medicare's fee-for-service sector will increase from $176 billion in 2000 to $299 billion in 2010. That growth will occur despite shrinkage in fee-for-service enrollment--which will decline by 500,000 over the next decade--and cuts in the growth of payment rates for many services. Spending growth for different types of services will vary considerably over that period.

Postacute Care Services. Growth in payments for skilled nursing facility (SNF) and home health services--the fastest-growing areas of fee-for-service spending in the decade before the Balanced Budget Act--slowed significantly beginning in 1998. Spending for home health care fell by 14.9 percent in 1998 and by an even more dramatic 34.9 percent in 1999. SNF expenditures, by contrast, rose by 8.8 percent in 1998, but that was less than half the rate of the previous year. Spending for SNF services then dropped by 12.5 percent in 1999. Growth in payments for hospice services slowed to 1 percent in 1998, down from 5.7 percent the previous year. In 1999, however, growth in hospice payments soared to 18.6 percent.

The slowdown in spending for SNF and home health services mostly resulted from the new prospective payment systems enacted in the BBA. Increases in the time to process claims also played a role, particularly for SNF services. The delay between provision of services and payment by Medicare accounted for 1.5 percentage points of the drop in home health payments, on average, in 1998 and 1999. The payment lag accounted for 2.2 percentage points of the drop in SNF payments in 1998 and 5.5 percentage points in 1999. CBO expects claims for postacute care services to be processed more quickly in later years, eliminating the drag on spending.

The BBRA temporarily increased the payment rates for SNF services (from April 2000 through September 2002) and postponed by a year the 15 percent cut in payments for home health services that had been scheduled for October 2000. The transition to prospective payment systems and the implementation of those BBRA provisions are expected to restore the growth of spending for postacute services. That spending is projected to increase through 2010 at an average annual rate of 7 percent for SNF services and 12 percent for home health services. Growth in payments for hospice services is expected to decrease to its long-term trend of around 5 percent per year by 2005.

Inpatient Hospital Services. Medicare payments for inpatient hospital services fell by 1.3 percent in 1999, to $86 billion. The factors contributing to that drop included a decline in the volume of services provided as well as provisions in the BBA that froze payment rates for most operating costs, reduced capital-related payment rates by 17.8 percent, and cut subsidies for medical education. In addition, the case-mix index (a measure of the relative costliness of the cases treated in hospitals paid under the prospective payment system) fell by 0.5 percent in 1999 after falling by the same amount in 1998. The drop in that index may be attributable to the widespread adoption by hospitals of less aggressive billing practices following antifraud initiatives that focused on those practices.

For most hospitals, the Balanced Budget Act limits cumulative increases in payment rates for operating costs to about 6 percentage points below the total rate of inflation over the 1999-2002 period. Although the BBRA eased some of those limits for certain hospitals, continuing limits on rate increases will result in only a 0.6 percent rise in total payments for inpatient hospital services in 2000, CBO projects. After the BBA limits expire, however, annual growth rates are expected to accelerate again, averaging 4.2 percent from 2001 through 2010.

Physicians' Services. Medicare payments for physicians' services rose by 5.1 percent in 1999, to $33 billion. Payments are projected to increase to $36 billion this year and to grow at an average annual rate of 2.9 percent over the next decade, reaching $48 billion in 2010. That growth rate results from payment formulas enacted in the BBA that tie the growth of per-enrollee spending for physicians' services to the growth of gross domestic product per capita.

Outpatient Services. Payments to outpatient facilities--such as hospitals' outpatient departments, therapy providers, dialysis facilities, and rural health clinics--fell by 8.2 percent in 1999, in part because the BBA's caps on therapy services were implemented. Those payments are projected to rebound this year and to grow by 15.2 percent in 2001 because of increases in prospective payment rates enacted in the BBRA. Annual growth rates are then expected to level off at 7 percent to 8 percent for the rest of the decade.

In 1999, Medicare spent almost $13 billion on nonphysician professional services and other outpatient ancillary services--including Medicare-covered prescription drugs, durable medical equipment, ambulance services, and chiropractic care. Those outpatient costs are projected to grow, on average, by roughly 7 percent a year for the rest of the decade. That growth results in large part from rising payments for the limited category of drugs covered under Part B of Medicare.
 

The President's Proposals for Medicare

The President's budget request for 2001 includes provisions to expand Medicare eligibility to new populations, extend Medicare coverage to services such as prescription drugs, and reduce the growth of program spending for services covered under current law. Populations newly eligible for Medicare would include certain people between the ages of 55 and 64, who would be allowed to buy in to the program. The cost of those expansions would be offset by savings in Medicare's fee-for-service sector, which would have spillover effects on Medicare+Choice spending and also result in lower Part B premiums. The net effect of the President's Medicare proposals would be to increase Medicare spending by a total of $69 billion through 2010 (see Table 2-7).
 

The budget also includes a $750 million demonstration project to let Medicare beneficiaries participate in clinical trials. That program would be paid for through the Treasury's general fund rather than the Medicare trust funds.

Proposals to Modify Traditional Medicare

The President proposes a variety of policy changes that would affect beneficiaries, providers, and health plans participating in Medicare, including:

• Reductions in payments for certain services;
  
  
• Modernization initiatives that would add innovations in health care financing used in the private sector to fee-for-service Medicare;
  
  
• Adjustments to beneficiaries' cost-sharing requirements; and
  
  
• New requirements to improve compliance with Medicare's payment rules.

Those proposals would reduce projected fee-for-service spending by $40.4 billion between 2001 and 2010. Because the growth of spending in Medicare+ Choice plans is linked to the growth of spending in the fee-for-service sector, those reductions would also lower payments to Medicare+Choice plans (by $14.0 billion over 10 years). The President also proposes to accelerate implementation of methods of adjusting payments to Medicare+Choice plans to reflect health risks more accurately. That provision would reduce spending during the next decade by $0.5 billion.

One-quarter of the gross savings in spending for Medicare Part B would be returned to beneficiaries in the form of lower premiums. Thus, beneficiaries would save $6.3 billion through 2010 (see Table 2-7).

Reductions in Payments. Under the President's proposals, payments to certain providers and suppliers would fall significantly. In principle, those rates are updated each year to reflect changes in the costs of inputs (such as wages, medical equipment, drugs and other supplies, and so on). For many services, the Balanced Budget Act holds the increases in payment rates below the annual rate of inflation through 2002, with full adjustment for inflation resuming in 2003. The President's proposals would continue holding those payment increases below inflation through 2005.

The largest savings would come from extending the BBA's reductions in annual payment updates for inpatient hospital services. The President proposes to reduce the annual updates for hospitals paid under the prospective payment system by 0.8 percentage points for urban hospitals and 0.4 percentage points for rural hospitals between 2003 and 2005. In addition, the BBA's provision to lower prospective capital payments to hospitals by 2.1 percent would be extended through 2005. Those provisions would save $14.0 billion through 2010 (see Table 2-8). Additional savings would come from extending the BBA's update reductions through 2005 for hospitals paid on the reasonable-cost basis established by the Tax Equity and Fiscal Responsibility Act of 1982; for suppliers of durable medical equipment, prosthetics and orthotics, and parenteral and enteral nutrition; for clinical laboratory services; and for ambulance services. Those provisions would lower spending by another $5.0 billion through 2010.
 

The BBA reduced Medicare's payments for the bad debts that hospitals incur. The President's budget would further reduce those payments and would extend the reduction in payments for bad debts to other providers. Those providers include SNFs, providers of outpatient physical therapy, comprehensive outpatient rehabilitation facilities, community mental health clinics, federally qualified health centers, and rural health clinics. Total savings from reducing bad-debt payments would be $5.3 billion through 2010.

Other services for which payments would be reduced include outpatient drugs, tests performed by clinical laboratories, and prosthetic and orthotic devices. The Administration also proposes to lower Medicare's payments for erythropoietin, a drug used by patients with end-stage renal disease who are receiving dialysis. In addition, the President's budget would eliminate Health Professional Shortage Area bonus payments for nonprimary care physicians practicing in urban areas. If enacted, those changes would save $6.2 billion through 2010.

Fee-for-Service Modernization. The Balanced Budget Act took important steps toward improving the efficiency of Medicare's fee-for-service sector by establishing prospective payment systems for several services. The President's budget would seek further efficiencies by extending and making permanent a "centers of excellence" program that lets Medicare contract with certain hospitals to treat particular disorders. Those hospitals would be chosen on a competitive basis.

Under the proposal, the Secretary of HHS would be authorized to pay selected hospitals a single, bundled rate for all services associated with an acute hospital admission. The first contracts incorporating such payments would be established in 2002 for cardiac surgery and knee and hip replacements. Contracts for other procedures and medical conditions could be established in the future. CBO estimates that the proposal would save $0.9 billion through 2010.

The President's budget would also authorize the Secretary of HHS to negotiate discounted payment rates for Medicare services with physicians and hospitals organized as preferred provider organizations. Those providers could make up for the loss in revenue from lower Medicare payments by attracting more patients, who would have lower cost sharing than under fee-for-service Medicare. CBO expects that the negotiated discounts would be small because the majority of Medicare enrollees (about 85 percent) have supplemental coverage that insulates them from cost-sharing requirements and because beneficiaries tend to stay with their current providers. Nevertheless, providers in competitive markets might feel that being designated a "preferred Medicare provider" would be necessary to maintain their patient base or attract new Medicare patients. Those arrangements would save $3.5 billion through 2010, CBO estimates.

The President also proposes to give the Secretary 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 choosing 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. CBO estimates that allowing more competitive acquisition would save $0.8 billion through 2010.

In addition, the President would allow both insurance companies and other entities that are experienced in processing claims to compete for Medicare business. The expanded competition would result in more accurate processing of claims, which CBO estimates would save $0.8 billion over 10 years.

The President would also provide disease-management and primary care case-management services to certain Medicare beneficiaries in the fee-for-service sector. Based on the experience of the Medicare Case Management Demonstration projects conducted between 1993 and 1995--which found that case management failed to reduce overall use or cost of Medicare-covered services--CBO estimates that the proposal would have a negligible effect on spending for Medicare benefits over the 2001-2010 period.

Adjustments to Beneficiaries' Cost Sharing. Other provisions of the President's budget would require fee-for-service enrollees to pay more for Medicare services by indexing the Part B deductible to inflation and requiring coinsurance payments for clinical laboratory services. At the same time, coinsurance for certain preventive services would be eliminated. The net effect of those changes would be to reduce Medicare outlays by an estimated $2.2 billion through 2010.

Under Part B, beneficiaries must pay for the first $100 of covered services each year before Medicare begins paying. That deductible amount has remained the same since 1991. Under the President's proposal, it would increase by the percentage change in the consumer price index beginning in 2003. In that year, CBO estimates, the deductible would be $103, rising to $122 in 2010.

Clinical laboratory services are an exception to the deductible rules; Medicare pays for 100 percent of those. The President's budget proposal would impose the standard Part B deductible and 20 percent coinsurance requirement on clinical laboratory services (other than preventive services) beginning in 2002. For certain preventive services, however, the proposal would eliminate both the deductible and the 20 percent coinsurance requirement. That change would substantially increase the use of those services and also increase demand for other services.

Requirements to Improve Compliance with Medicare's Payment Rules. The President's budget includes several initiatives to improve compliance with Medicare's payment rules and reduce fraud and abuse. In particular:

• Group health plans would be required to notify Medicare of beneficiaries for whom they provide primary coverage. HCFA would then know immediately whether Medicare or a private insurer had primary responsibility to pay for a beneficiary's health services.
  
  
• New rules would restrict the provision of partial hospitalization services--outpatient services furnished to patients who might otherwise be hospitalized for the treatment of mental health conditions. In addition, the Secretary of HHS would receive more authority to screen out unqualified providers of that benefit and could impose civil monetary penalties on physicians who falsely certified that patients needed those services.

Together, those provisions would save an estimated $1.9 billion over 10 years.

Other Proposals. The budget also includes a proposal to lengthen coverage of immunosuppressive drugs for certain beneficiaries who receive organ transplants paid for by Medicare. Transplant recipients who have not exhausted their current drug coverage would qualify for a total of 48 months of continuous coverage for immunosuppressive drugs after their transplant. All new transplant recipients would also be eligible for 48 months of drug coverage. That policy would amend temporary coverage extensions enacted in the BBRA. CBO estimated that the BBRA allowed for eight additional months of coverage beyond the former 36-month limitation for people eligible in 2000 and 11 additional months of coverage for people eligible in 2001 through 2004. Lengthening drug coverage would generate savings by averting costs associated with kidney rejection, such as rehospitalization, dialysis, and retransplantation. After accounting for those offsetting savings, CBO estimates that permanently extending the period of drug coverage to 48 months would cost $0.2 billion over the 2001-2010 period.

The President's budget proposes making various changes to the Medicare+Choice program. The most sweeping proposal--to replace Medicare+Choice in 2003 with a new system for making payments to private health plans, called the competitive defined benefit system--is discussed in a separate section below. Before that system is put in place, the budget proposes to repeal the BBRA provision that slowed the implementation of improved methods of adjusting rates paid to Medicare+Choice plans to reflect differences in risk. The President would reestablish the original implementation schedule. Speeding the phase-in of improved risk adjustment would save $0.5 billion between 2001 and 2010. In addition, the budget would shift the payments owed to Medicare+Choice plans in October 2002 to the end of September 2002. That shift would increase Medicare outlays for fiscal year 2002 and reduce outlays for 2003 but would have no cumulative effect.

Proposals to Expand Medicare Eligibility

The President's proposals to let people under age 65 buy in to the Medicare program are similar to proposals in last year's budget, with one exception. Under the current proposal, participants would be able to claim up to 25 percent of their buy-in premiums as an income tax credit. Two groups would be eligible to participate: people ages 62 to 64 who do not have private health insurance, Medicaid, or other public coverage; and certain workers ages 55 to 61 who lose their health insurance because of a job loss. The terms of participation would differ for the two groups. Because of the tax credit, CBO's estimate of participation in the buy-in option is higher than the estimate for last year's version, and its estimate of adverse selection among participants is significantly less.

A third proposal to expand Medicare coverage, which would apply to disabled workers, is part of a broader initiative to allow the disabled to return to work and maintain their health insurance coverage.

Buy-In for People Ages 62 to 64. The Administration proposes allowing people ages 62 to 64 who do not have employment-based health insurance, Medicaid, or coverage through another government program to enroll voluntarily in Medicare, provided they do so as soon as they are eligible. Events that would make people eligible include turning 62 or losing employment-based health insurance between ages 62 and 64 under certain circumstances.

Medicare premiums under the buy-in would be paid in two parts, both of which would be adjusted for geographic variations in Medicare's costs and updated annually:

• Before age 65, enrollees would pay premiums that reflected the average expected cost of benefits if everyone ages 62 to 64 participated in the buy-in. The national average monthly premium would be about $326 in 2002, the first year of the program. Up to 25 percent of those premiums could be claimed as an income tax credit.
  
  
• At age 65 and thereafter, buy-in participants would pay a premium surcharge (in addition to their regular Medicare premium) to recapture for the government the extra costs that Medicare would pay as a result of adverse selection in the buy-in program. The surcharge would reflect the difference between the premium paid before age 65 and the higher average costs of people who chose to participate in the program. CBO estimates that a person who enrolled in the buy-in program in 2002 at age 62 and stayed until age 65 would pay a surcharge of about $4 a month in 2005. No part of those premiums could be claimed as an income tax credit.

The buy-in for people ages 62 to 64 would raise outlays for Medicare benefits by $46.2 billion between 2002, when the program would begin, and 2010, CBO estimates (see Table 2-9).⁽⁵⁾ Premiums would total slightly more than that, resulting in net Medicare savings of $0.1 billion. Tax revenues would be reduced by about $7.7 billion because of the tax credit, which CBO assumed about three-quarters of participants would claim. About 650,000 people would participate in the program in 2002, rising to about 1.3 million by 2010. In addition, Social Security benefits would increase by about $1.4 billion through 2010, under the assumption that approximately 1 percent of people ages 62 to 64 would retire if Medicare coverage was available to them.
 

Buy-In for Displaced Workers Ages 55 to 61. The Administration also proposes to allow certain workers ages 55 to 61 who lose health insurance because of a job loss to buy in to Medicare. (Their spouses would be eligible for coverage as well.) The program would be available only to people who met several eligibility requirements:

• Having health insurance coverage for at least 12 months immediately before enrolling in the program;
  
  
• Participating in their employer's plan immediately before losing their job;
  
  
• Being eligible for unemployment insurance benefits; and
  
  
• Being ineligible for any other employment-based or federal health insurance coverage. (That requirement means that workers would first have to exhaust the 18 months of continued coverage from their former employer available under the Consolidated Omnibus Budget Reconciliation Act of 1985, or COBRA.)

Monthly premiums for the buy-in would be about $460 per person in 2002, but participants could claim a tax credit for up to 25 percent of their payments. Premiums would be updated annually and adjusted for geographic differences in costs. Those premiums would not quite cover the costs of the program, however, because the program would attract enrollees who expected to have high medical costs. As a result, CBO projects that the program would increase net Medicare outlays by about $0.2 billion between 2002 (when it began) and 2010, reflecting outlays for benefits of $4.1 billion and premiums of $3.9 billion (see Table 2-9). The bulk of the program's costs would come from forgone tax revenue due to the tax credit, amounting to about $0.7 billion through 2010. The proposal would also encourage a small number of additional workers to seek unemployment insurance, raising federal outlays for unemployment compensation by an estimated $0.1 billion over 10 years.

Participation in the program would be limited because of the stringent eligibility requirements and the significant premiums that enrollees would pay, although the tax credits for both the buy-in premiums and COBRA premiums would result in substantially higher participation than otherwise. By 2010, CBO estimates, about 90,000 people would be enrolled in the program at any one time.

Medicare Coverage for the Working Disabled. The Work Incentives Improvement Act of 1999 extended coverage under Medicare's Part A (Hospital Insurance) by four and a half years for certain disabled people who return to work. The President proposes to make that extension permanent. CBO estimates that the change would increase net Medicare outlays by $0.1 billion over the 2001-2010 period.

Proposal to Add a Prescription Drug Benefit to Medicare

By far the President's costliest proposal for Medicare is to create a voluntary outpatient prescription drug benefit under a new Part D of the program. That benefit would begin in 2003 and be fully phased in by 2009. The benefit would pay half of the cost of prescription drugs, up to a specified cap. It would be financed half from premium payments by enrollees and half from general government 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. (The President's budget also provides $35 billion from 2006 through 2010 for a possible catastrophic benefit, but no policy is specified. Consequently, that amount is not included in the estimates discussed below.)

How the Benefit Would Work. In 2003, all Medicare enrollees would have a one-time chance 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 be given one opportunity to enroll after their 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 pharmacy 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 with drug manufacturers and pharmacies. The benefit would have no deductible and would generally pay 50 percent of an enrollee's prescription drug costs, up to a limit of $1,000 in 2003. That cap would gradually rise to $2,500 in 2009. Thus, in 2009, a beneficiary who spent $5,000 on prescription drugs would receive the maximum reimbursement of $2,500. That beneficiary would also pay $578.40 in Part D premiums that year. After 2009, 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 cap would shrink, thus eroding the benefit.

Certain low-income beneficiaries would receive help with drug-related costs through the Medicaid program. Medicaid would pay both the premiums and cost-sharing expenses under the drug benefit, at the usual federal/state matching rate, for participants who were also fully eligible for Medicaid (so-called dual-eligibles, who now receive full drug coverage through Medicaid) 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 below 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 2-10). Eligibility for those subsidies would be determined by state Medicaid agencies. Neither the federal nor state governments would be liable for covering any drug expenses above the Part D cap for low-income beneficiaries who were not fully eligible for Medicaid. For dual-eligibles, though, Medicaid would pay all drug costs not paid by Medicare, including expenses above the cap.
 

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 the employers' 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 of the program. Consequently, their costs would not be included in the cap on Part D benefits.

CBO's Estimate. The new Part D provisions would add a total of $149 billion to federal costs through 2010, CBO estimates. (By comparison, the Administration's estimate is about $160 billion.) Of CBO's total, almost $131 billion would represent outlays for Medicare (net of premium receipts), and nearly $19 billion would represent federal outlays for Medicaid (see Table 2-11). States would also face additional Medicaid costs--totaling some $4 billion through 2010. CBO estimates that the premium for Part D would start at $24.10 a month in 2003 and rise to $50.90 in 2010 (see Table 2-12).
 

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 employer-sponsored drug coverage for retirees would keep that coverage rather than opt for the new benefit. In addition, CBO assumes that people who are eligible for benefits under Part B but do not actually enroll would also not enroll in Part D. Under those assumptions, nearly 36 million people would sign up for Part D in 2003, representing approximately 88 percent of total Medicare enrollment.

In 2003, about 31 percent of participants would have drug expenses exceeding the $1,000 cap on Part D benefits. By 2010, when the cap would be $2,562, about 21 percent of participants would have expenditures exceeding it. The Part D benefits paid per participant would average $568 in 2003, rising to $1,240 in 2010.

CBO estimates higher Medicare costs for the prescription drug benefit than the Administration does but lower Medicaid costs. As a result, its estimate of net federal costs is about 7 percent lower than the Administration's. The two base their estimates of future drug spending on patterns reported in Medicare's Current Beneficiary Survey. However, CBO and the Administration differ in the adjustments they make to those data to account for underreporting in the survey and growth since the survey year. In particular, CBO uses a larger adjustment factor (1.33) to account for underreporting by noninstitutionalized respondents than the Administration does (1.15).⁽⁶⁾ Further, CBO assumes somewhat higher rates of growth in drug spending over the next few years than the Administration. Both CBO and the Administration assume that the new drug subsidies for low-income people will induce more participation in Medicaid, but the Administration's estimate of that effect is larger.

Compared with its estimate of a similar proposal by the President last year, CBO made only two significant changes in the assumptions underlying its estimate of the Part D benefit (aside from reflecting the delayed start of the benefit from 2002 to 2003).⁽⁷⁾ One change affects Medicare costs and the other Medicaid costs. For Medicare, CBO's current assumption (based on the results of an employer survey conducted by Hewitt Associates) is that only 25 percent of employers would accept the premium subsidy and keep their current drug coverage for Medicare-eligible retirees.⁽⁸⁾ (In last year's estimate, CBO assumed that 75 percent of employers would accept the subsidy.) For Medicaid, CBO expects less of an increase in Medicaid participation because of low-income subsidies for the new drug benefit than it did last year (70 percent versus 80 percent). That change was made because of better information about the proportion of income-eligible people who would also meet the asset requirements for Medicaid eligibility.

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. With the proposed prescription drug benefit, uncertainties exist about the nature and value of the benefit, the effectiveness of PBMs in controlling drug costs, participation in Part D by Medicare beneficiaries who now have drug coverage, and the impact of the new benefit on Medicaid spending.

The 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 unclear. A number of innovative drugs are likely to be approved 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. Thus, projections of the rate of growth in drug use and prices are highly uncertain even without changes in insurance coverage. For this estimate, CBO assumes that recent growth trends will continue for several years and then moderate.

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 in the President's plan. For the other half, CBO estimates that the new Part D coverage would increase drug utilization by up to 25 percent.

Part D is designed to ensure that most enrollees would receive some benefit. However, because of the annual cap, it would not protect enrollees with chronic conditions who are dependent on prescription drugs 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, a program that did not provide first-dollar coverage but limited an enrollee's out-of-pocket costs to some annual maximum (or stop-loss) would be less likely to cause a large increase in drug utilization and would better protect enrollees from catastrophic expenses. Under such a program, however, fewer enrollees would be likely to benefit. Further, a catastrophic benefit might result in higher prices for some drugs with no close substitutes because enrollees whose expenses exceeded the stop-loss amount would no longer be price-conscious.

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

PBMs save money for private-sector 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 because 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 may be more likely to 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.

Whether the PBMs chosen to administer the Part D benefit would have as much freedom to use those cost-saving techniques as they have in aggressive private insurance plans is unclear. For example, the President's 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 varying 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 very 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 though nominal discount and rebate rates were unchanged, if the prices from which discounts and rebates were calculated rose as a result of the new benefit.

Program Participation. CBO's estimate assumes that everyone who participates in Part B of Medicare would also participate in Part D, with one exception: a quarter of beneficiaries who have drug coverage through health plans for retirees would retain that coverage. 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 kept (or instituted) prescription drug coverage at least as good as the Part D benefit. That subsidy payment--together with the tax exclusion of their health plan costs--would induce some 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.) CBO assumes that about three-quarters of Medicare enrollees who now have drug coverage through a retiree health plan would enroll in Part D because their employers would either eliminate their drug coverage altogether or make it secondary to the Medicare benefit.

Because of the 50 percent coinsurance rate and the benefit cap, the benefits provided under Part D would be limited. Moreover, through their premiums, enrollees would pay for 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 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. CBO also expects states to enroll their dual-eligibles because that would shift some of the states' costs for drug coverage to Medicare. Other low-income people eligible for Medicaid assistance under the new drug benefit would also enroll in Part D to gain drug coverage.

Effects on Medicaid Costs. As Table 2-11 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 who are fully eligible for both programs. 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.

Certain low-income Medicare beneficiaries who are not eligible for full Medicaid benefits would also become eligible for assistance to pay for their Part D premiums and cost sharing. As noted previously, 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 part 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.

The President's proposal would also increase Medicaid spending for services not related to the new drug benefit. As noted above, many low-income Medicare beneficiaries who are ineligible for full Medicaid benefits are eligible to have their Medicare Part A and B premiums paid by Medicaid--and in some cases, their cost sharing as well. A sizable number of them do not enroll in Medicaid, however. CBO estimates that about 1.5 million Medicare beneficiaries with income below the poverty level are eligible for partial or full Medicaid assistance but do not participate in the program. A further 1.0 million beneficiaries with income between 100 percent and 120 percent of the poverty level who are eligible to have their Part B premiums paid by Medicaid do 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 benefits that Medicaid pays for as well as the prescription drug benefit. Participation in Medicaid by beneficiaries who are eligible for full Medicaid benefits might also increase, although their participation is already greater than that of other groups.

For this estimate, CBO assumed 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's costs would be lower. Conversely, if Medicare paid more for drugs, Medicaid's costs would be higher.

Proposal to Create a Competitive Defined Benefit Program

The President is proposing to give Medicare's managed care plans various incentives to compete on the basis of price as well as quality through what the budget calls a competitive defined benefit program. CBO estimates that the program would save Medicare $13.7 billion through 2010, although that estimate is subject to great uncertainty (see Table 2-7).

How the Program Would Operate. Beginning in 2003, the current system of paying for Medicare+ Choice plans would be replaced with a bidding system. Under that approach, the premium that Medicare beneficiaries paid would depend on the plan they chose. Beneficiaries who stayed in the fee-for-service sector would pay the regular Part B premium, and those who chose the proposed prescription drug benefit would also pay the regular Part D premium. Competing plans would be free to charge a different premium than the fee-for-service program. However, beneficiaries who chose cheaper plans would generally pay lower premiums, and those who opted for more costly plans would pay the extra costs of that choice.⁽⁹⁾ Managed care plans would submit a bid price for the standard Medicare benefit package (including the drug benefit for those who chose it), enabling beneficiaries to make price comparisons among plans.

The actual amount that beneficiaries paid would depend on the difference between the bid price of their plan and a county-specific reference price, which would be the larger of two amounts: the payment rate established by the BBA for Medicare+Choice plans or 96 percent of average Medicare spending per enrollee in their county (that average is adjusted for the difference between the average health status of the county's Medicare enrollees and all Medicare enrollees).⁽¹⁰⁾ If beneficiaries enrolled in a plan with a bid price below the reference price, their Medicare premiums would be reduced by 75 percent of the difference (but not below zero). If they chose a plan with a bid price above the reference price, they would pay the full difference.

Managed care plans would receive their full bid price for the defined benefit package regardless of whether that price was above or below the reference price. But given the price structure that beneficiaries would face, plans would have a strong incentive to keep their bid price below the reference price; otherwise, they would have trouble competing against the 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.⁽¹¹⁾

The government would adjust the payments to health plans to reflect differences in expected risk based on health status. Plans enrolling beneficiaries with greater-than-average health risks would receive higher federal payments than other plans. Risk adjustment has been considered a perennial problem for the Medicare program, however, and full implementation of Medicare's new risk-adjustment system is not expected until after 2003.

The amount the program would save (or lose) on people enrolling in plans would depend on the difference between the plan's bid price and the reference price, the health risk of the enrollee, and the difference between the reference price and spending per enrollee in the absence of the competitive defined benefit. (The following example makes the simplifying assumption that spending in the absence of the competitive defined benefit is equal to spending in the fee-for-service sector.)

Suppose, for example, that average costs in the fee-for-service sector were $7,500, the annual premium for beneficiaries enrolled in that sector was $1,200 ($100 a month), and the reference price was $7,200.⁽¹²⁾ Beneficiaries choosing a less expensive plan with a bid price of, say, $6,500 would have their annual premium reduced by 75 percent of the difference between the bid price and the reference price, or $525 (75 percent of $700). As a result, their annual premium would be $675, or $56.25 a month.

For a beneficiary in good health, costs in the fee-for-service sector might be expected to be only half the average, or $3,750. Medicare would pay that person's plan half of the bid price, or $3,250. The competitive defined benefit program would save $500 compared with the fee-for-service sector but would give $525 to the beneficiary through lower premiums. Thus, the program would lose $25 on that person.

By contrast, the expected cost in the fee-for-service sector for a beneficiary in poor health might be twice the average, or $15,000, and the payment to his or her plan would be twice the bid price, or $13,000. The new program would save $2,000 compared with the fee-for-service sector but would return $525 to the beneficiary in lower premiums. Thus, Medicare's net savings for that person would total $1,475. (In those examples, Medicare's net savings would be lower if the reference price was higher and greater if the reference price was lower.)

CBO's Estimate. Gross savings from the competitive defined benefit program would total $85.2 billion through 2010, CBO estimates, of which $71.5 billion would be returned to beneficiaries in the form of lower Medicare premiums (see Table 2-7). Thus, the program would reduce net Medicare outlays over 10 years by $13.7 billion. (The Administration's estimate is $11.9 billion.)

CBO analyzed data from Medicare's risk-based plans to estimate the bid prices such plans would offer under the proposal's bidding process. Those plans prepare adjusted community rate (ACR) proposals that indicate whether Medicare's payment is excessive and how much they would return to beneficiaries in the form of additional benefits or waived premiums. CBO's analysis relied on ACR data for 1997, the latest year for which per-enrollee spending in the fee-for-service sector is available at the county level.⁽¹³⁾ The analysis used the difference between the payment rate and the amount returned to beneficiaries in the form of additional benefits or waived premiums as the measure of the bid price that plans would have submitted in 1997 if competition had been based on price.

CBO assumed that bid prices would increase at an average rate of about 5.5 percent per year, slightly lower than the growth rate of Medicare+Choice payments under current law. Plans would lose market share if they bid above the reference price. To maintain enrollment levels, such plans could subsequently reduce their bid prices, presumably by reducing their cost of providing services. CBO assumed that plans would not offer a Medicare+Choice product in counties where they could not operate profitably.

CBO projects that more counties would have managed care plans under the proposal than under current Medicare rules. The reference price would equal or exceed current-law payment rates in areas where the BBA rules will increase those rates--areas that often have few or no Medicare+Choice plans. Moreover, the reference price would be higher than current payment rates in areas where the BBA rules will reduce those rates (compared with pre-BBA rules). CBO's analysis assumed that the number of Medicare enrollees living in areas with access to a managed care plan would increase by 6 million from 2002 through 2010--equal to the projected growth in Medicare enrollment during that period.

Plans that offered beneficiaries substantial reductions in Medicare premiums would tend to gain market share at the expense of both the fee-for-service sector and plans that offered smaller reductions in Medicare premiums. However, plans that lowered their premiums might also reduce the benefits they offered. Beneficiaries would take into account both the savings from lower premiums and the benefits they would have to give up in deciding whether to move to a less expensive plan. The size of any gains in a plan's market share from one year to the next was assumed to be positively related to its market share in the preceding year. CBO also assumed that a substantial number of Medicare beneficiaries who would initially choose not to switch to a lower-cost plan would reconsider that choice and switch in later years. The initial change in market share in response to lower premiums would account for only one-third of the ultimate change.

Under current law, CBO projects, enrollment in Medicare+Choice plans will increase from 7.2 million in 2002 to 14.1 million in 2010. Many beneficiaries enroll in Medicare+Choice plans to obtain prescription drug benefits that are not available in the fee-for-service sector. CBO assumes that the President's proposal to offer a drug benefit to enrollees in the fee-for-service sector would dramatically slow the growth of enrollment in Medicare+Choice. CBO estimates that with that benefit in place, enrollment in competitive defined benefit plans would be 11.6 million in 2010.

By that year, two-thirds of enrollees in those plans would not pay any Medicare premiums. They would live in areas where the reference price was substantially higher than the cost to an efficient plan of providing the defined package of Medicare benefits. Many beneficiaries, however, live in areas where the reference price would be more in line with the actual cost of providing services. They would not have access to plans that offered a substantial reduction in Medicare premiums.

CBO assumes that beneficiaries who chose to enroll in competitive defined benefit plans would tend to have better-than-average health and that risk adjustment would accurately reflect that. Based on the projected geographic distribution of enrollment and the health status of enrollees, CBO estimates that almost 85 percent of the savings achieved through the bidding process would be returned to beneficiaries in the form of lower premiums.

Other Issues. Promoting greater price competition in Medicare could broaden the options available to beneficiaries and slow the growth of spending. Those outcomes are by no means assured, 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 the 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 Medicare's risk-based program to date suggests that competition is more likely to occur in large, high-cost urban markets, although the nature of the reference-price mechanism 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 allowed 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 would go only partway, however, toward establishing a competitive model for Medicare. The fee-for-service sector--in which the large majority of Medicare beneficiaries are still enrolled--would not be required to compete fully on the basis of price with the private plans participating in Medicare. The special status of the fee-for-service sector could result in lower savings for the Medicare program than other competitive strategies might yield.

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, beneficiaries have been attracted away from fee-for-service Medicare to managed care plans by 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, slowing the growth of enrollment in managed care. How far lower premiums might offset that effect is unknown.

The mechanics for bidding and setting prices in the President's proposal are not clear, which adds to the difficulty of predicting the effects of the proposal on plans' behavior. For example, efficient plans in areas with a high reference price 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 would be able to compete against the fee-for-service sector and each other on the basis of both price and benefits. Such competition would be less possible in markets with lower reference prices. Thus, although the proposal aims to reduce the current disparities in benefits among Medicare+Choice plans across the country, it might not end up doing so.

Even if beneficiaries proved highly responsive to reductions in Medicare premiums and plans chose to compete on that basis, the effects of the proposal on the growth of Medicare spending are speculative. Would there be one-time savings--possibly stretched out over several years--as beneficiaries in fee-for-service shifted to managed care plans? Or would competitive forces be strong enough to foster efficiencies throughout the system, slowing the growth of costs in the future? Debate over those questions has been going on in the private sector since the mid-1990s, when many people with employer-sponsored health insurance plans began to shift from fee-for-service to more tightly managed plans. That debate has yet to be resolved.

1. States can expand Medicaid coverage of parents either through authority granted under section 1115 of the Social Security Act or through permissive state income and resource requirements allowed under section 1931 of the act.

2. Spending for Medicare benefits dropped by 0.9 percent in 1999, but that decrease was partially offset by an increase in spending for program administration.

3. Medicare generally pays Medicare+Choice plans on the first day of the month. When the first day falls on a weekend or holiday, payments are shifted to the last business day of the preceding month. In addition, the Balanced Budget Act alters some payment dates for group plans. For those reasons, the number of payments varies each fiscal year from 11 to 13. The growth of Medicare spending for group plans surges in years with 13 payments and slows in years with 11 payments.

4. Another 0.5 million beneficiaries are enrolled in group plans participating in a demonstration project or paid on a cost basis. CBO estimates that enrollment in those plans will decline to 0.1 million in 2010.

5. The basis for that estimate is similar to the approach that CBO used in estimating previous versions of the proposal. See Congressional Budget Office, An Analysis of the President's Budgetary Proposals for Fiscal Year 1999 (March 1998), pp. 37-42.

6. Last year, the Administration used a 1.30 factor to adjust for underreporting.

7. For last year's estimate, see the statement of Dan L. Crippen, Director, Congressional Budget Office, before the Senate Committee on Finance, July 22, 1999.

8. Hewitt Associates, Retiree Health Coverage: Recent Trends and Employer Perspectives on Future Benefits (Menlo Park, Calif.: Henry J. Kaiser Family Foundation, October 1999).

9. Because most beneficiaries' Medicare premiums are withheld from their Social Security checks, beneficiaries would see the effect of their plan choice as a change in the amount of their Social Security checks.

10. This factor is conceptually similar to the pre-BBA payment rate for risk-based plans, which was 95 percent of average Medicare spending per enrollee.

11. Plans could also offer additional benefits--beyond the basic Medicare benefit and drug benefit--for a separate premium, which would give them another way to compete against the fee-for-service sector and other managed care plans.

12. The reference price in this example is 96 percent of the average fee-for-service cost. The reference price would be higher than the average fee-for-service cost in areas where the BBA's payment rate for Medicare+Choice plans was higher than that cost. The reference price would be below 96 percent of the average fee-for-service cost in areas where the payment rate was below 96 percent of that average, because the reference price would be the greater of the BBA payment rate and the weighted average of per-enrollee spending in the fee-for-service and capitated sectors.

13. ACR proposals provide notoriously unreliable data on the cost to plans of providing basic Medicare benefits. The share of Medicare payments that plans report they are willing to give back to beneficiaries in the form of additional benefits and waived premiums may be a less unreliable measure of the discount they would offer if price played a role in competition.

---

Please select a format in which to retrieve a document.

(Additional information about our document formats is available.)

---