Health care services account for almost 90 percent of spending in function 550. Health-related research and training programs consume a little over 10 percent, and about 1 percent of spending goes to consumer and occupational health and safety. On average, spending for health care services and health research and training has grown by 7 percent annually since 2002; spending for consumer and occupational health and safety programs has grown by about 5 percent per year over the same period.

The largest component of mandatory spending for health care services in function 550 is Medicaid, which funds health services for low-income women, children, and elderly people and for people with disabilities. (Medicare, in budget function 570, is the largest federal health care program.) The federal government shares the cost of Medicaid with the states, and, since 2002, federal Medicaid spending has grown at an average annual rate that is slightly above 5 percent. The Congressional Budget Office projects that federal Medicaid spending will total $192 billion in 2007 and will grow by roughly 8 percent per year from 2007 through 2017.

Other mandatory programs in function 550 pay for health care services for children in some low-income families and for federal civilian or military retirees. Most of the discretionary spending for health care is disbursed by the Centers for Disease Control and Prevention, the Health Resources and Services Administration (HRSA), the Indian Health Service, and the Substance Abuse and Mental Health Services Administration.

Spending for health research and training mainly funds programs of the National Institutes of Health (NIH) and HRSA that provide grants or loans to health professionals. NIH funding grew by a total of 22 percent from 2002 to 2006.

The federal government pays a portion of the costs that states incur to administer their Medicaid programs. The basic federal matching rate is 50 percent for most administrative activities. In some cases, however, the federal subsidy is higher. For example, the federal government pays 75 percent of the cost of employing skilled medical professionals for Medicaid administration, 75 percent of the cost of utilization review (the process of determining the appropriateness and medical necessity of various health care services), 90 percent of the cost of developing systems to manage claims and information, and 75 percent of the cost of operating such systems.

This option would set the federal matching rate for all Medicaid administrative costs at 50 percent. That change would save $1.2 billion in 2008 and $8.4 billion over five years. The President has included this proposal in his 2008 budget.

Enhanced matching rates were designed to encourage states to develop and support particular administrative activities that the federal government considers important for the Medicaid program. Once those administrative systems are operational, however, there may be less reason to continue the higher subsidy. Moreover, because states pay, on average, about 43 percent of the cost of health care for Medicaid beneficiaries, they have a substantial incentive to maintain efficient information systems and employ skilled professionals.

A potential drawback of this option is that a reduced federal subsidy might cause states to cut back on some beneficial activities, with adverse consequences for program management. For example, states might hire fewer nurses to conduct utilization reviews and oversee care in nursing homes, or they might make fewer improvements to their information-management systems.

The federal government's three major public assistance programs—Temporary Assistance for Needy Families (TANF), Food Stamps, and Medicaid—have certain administrative tasks in common. For instance, during the enrollment process, each program requires that potential recipients provide information about their family's income, assets, and demographic characteristics. Before the 1996 welfare reform law, which replaced Aid to Families with Dependent Children (AFDC) and some related programs with the TANF block-grant program, all three programs reimbursed states for 50 percent of most administrative costs. As a matter of convenience, states usually charged the full amount of those common administrative costs to AFDC.

The TANF block grants are calculated on the basis of past federal welfare spending, including what the states received as reimbursement for administrative costs. Because states had previously paid the common administrative costs of their AFDC, Medicaid, and Food Stamp programs from AFDC funds, those amounts are now included in their TANF block grants. However, the Department of Health and Human Services now requires each state to charge Medicaid's share of common administrative costs to the federal Medicaid program, even if that amount is already implicitly included in the state's TANF block grant. As a result, many states are in effect being paid twice for at least a portion of Medicaid's share of common administrative costs.

For any state that receives such a double payment, this option would limit the federal reimbursement for administrative costs for Medicaid to the amount not included in the state's TANF block grant. Federal outlays would decline by $280 million in 2008 and by almost $1.8 billion through 2012. Overall, the reduction in Medicaid funding would equal about one-third of the common costs of administering the Medicaid, AFDC, and Food Stamp programs that were charged to AFDC in 1996—the base period used to determine the amount of the TANF block grant. (A similar adjustment has already been made in the amount that the federal government pays the states to administer the Food Stamp program.) The President's 2008 budget includes this proposal.

A rationale for this option is that it would eliminate the current implicit double payment to states. A potential drawback is that reducing federal reimbursement could hamper states' efforts to enroll additional eligible children in Medicaid and the State Children's Health Insurance Program. Such action could also prompt states to restrict eligibility or services for those two programs.

The federal government reimburses states for about 50 percent of the cost of managing their Medicaid programs. Under this option, the federal government would reduce its spending for Medicaid's administrative costs by capping the per-enrollee amount that it pays each state for Medicaid administration. The cap would grow by 5 percent annually—a rate slower than that at which administrative costs have grown in the past—from a base-year amount that represents the per-enrollee administrative costs for which each state claimed matching payments in 2006.

A rationale for this option is that such a change would result in savings totaling $460 million in 2008 and $4.0 billion through 2012. (Limiting federal payments for administrative costs to a 5 percent growth rate would produce savings because the actual growth rate of those costs is projected to be about 9 percent in 2007, about 8 percent in 2008 and 2009, and then about 7 percent in ensuing years.) Another rationale for implementing the option is that it would give states a stronger incentive to improve the efficiency with which they manage their Medicaid programs.

An argument against this option is that, faced with fewer administrative resources, states might cut back on some activities that could improve the functioning of their Medicaid programs. For example, they might reduce funding for efforts to combat waste, fraud, and abuse.

Spending by the Medicaid program for prescription drugs increased at an average real (inflation-adjusted) rate of 13 percent annually between 2000 and 2005, with the federal component of that spending reaching $17.9 billion in 2005. With the introduction in January 2006 of the Medicare drug benefit, Medicaid spending for prescription drugs fell substantially, to $10.2 billion, largely because coverage for so-called dual eligibles—people who are covered under both Medicare and Medicaid—is now provided by Medicare. The lower level of spending for Medicaid, however, is still subject to upward pressures similar to those affecting overall prescription drug spending.

The amount that Medicaid pays for a particular drug depends on two prices: the average wholesale price (AWP), a list price published by the manufacturer; and the average manufacturer's price (AMP), which is the average price that the manufacturer actually receives for drugs distributed to retail pharmacies and mail-order establishments. For brand-name drugs, state Medicaid agencies typically pay the AWP minus a percentage (ranging from 10 percent to 15 percent, depending on the state) plus a dispensing fee. A portion of that spending is recouped by both the federal and state governments through a rebate paid by the manufacturer to Medicaid.

For brand-name drugs, the basic rebate is equal to the maximum of a fixed, or flat, percentage of the AMP—15.1 percent currently—and the difference between the AMP and the "best price" at which the manufacturer sells the drug to any private purchaser. An additional rebate applies if the AMP grows faster than inflation. (Makers of generic drugs must rebate 11 percent of the AMP to the state Medicaid agency.) Overall, Medicaid receives an average rebate from manufacturers of slightly more than 20 percent under the current pricing system (not including the additional rebate tied to price inflation).

This option would boost the flat rebate from 15.1 percent to 20 percent. That change would increase the average Medicaid rebate (relative to the AMP) to about 24 percent, reducing mandatory federal spending by $130 million in 2008 and by $1.4 billion through 2012.

Although many manufacturers offer large discounts to private purchasers, the best-price provision discourages them from offering discounts beyond the flat rebate because any such discount automatically triggers a greater Medicaid rebate. A higher flat rebate percentage, however, would allow manufacturers to offer slightly greater discounts without triggering the best-price provision. Thus, beyond reducing Medicaid spending for prescription drugs, this option might in certain cases enable some private purchasers to buy certain drugs at lower prices. The interaction of the higher basic rebate with the additional inflation-adjusted rebate, however, makes the ultimate effect of this option on prices paid by private purchasers difficult to predict.

A potential drawback of this option is that pharmaceutical firms, faced with reduced revenues from Medicaid, might invest less money in research and development in certain drug classes whose use is heavily concentrated in the Medicaid population.

The Medicaid program funds coverage for two broadly different types of health care: acute care (including services such as inpatient hospital stays and visits to physicians' offices, and products such as prescription drugs); and long-term care (services such as nursing home care and home- and community-based assistance). The program is financed jointly by the states and the federal government, with the federal government's share determined as a percentage of overall Medicaid spending. That percentage, referred to as the federal matching rate, can range from 50 percent to 83 percent, depending on a state's per capita income. (The matching rate averages 57 percent nationwide.) Although the federal match helps states provide health coverage to disadvantaged populations, it may also encourage higher spending by subsidizing each additional dollar spent on Medicaid by states. The federal share of Medicaid's outlays in 2007 is estimated to be $113 billion for acute care and $59 billion for long-term care.

This option would convert the federal share of Medicaid's payments for acute care services into a block grant, as 1996 legislation did with funding for welfare programs. (Long-term care would continue to be financed as under current law.) Each state's block grant would equal its 2006 federal Medicaid payment for acute care, indexed to the increase in input prices faced by providers of medical care. (An "input" is a factor used in the production of medical care, such as professional labor, office space, and so on.) That change in financing would reduce federal outlays by $7.0 billion in 2008 and by $83.2 billion over five years. The change generates savings because federal Medicaid payments are projected under current law to grow faster than input prices. (Alternatively, block grants could be indexed both to increases in input prices and to the change in each state's population. In that case, savings would be $5.6 billion in 2008 and would grow at a slower rate thereafter, totaling $65.7 billion over five years.) In exchange for slower growth in payments, states would be given more flexibility in how they could use the funds to meet the needs of their low-income and uninsured populations.

A rationale for this option is that a block grant rather than federal matching payments would eliminate the federal subsidy for each additional dollar spent by states on acute care. Block-grant proposals in the past have typically coupled a change in financing with increased discretion for states to design and administer their programs. For example, states, if given increased discretion, could modify their benefit packages and make corresponding adjustments in the number of people covered. In addition, block grants would eliminate states' ability to use funding strategies designed to maximize federal assistance.

An argument against this option is that converting acute care payments into a block grant would reduce the total amount of federal support for Medicaid and also shift the cost burden to the states. As a result, ending federal matching payments for acute care services could provide an incentive for states to scale back their Medicaid spending. Unless states were willing to pay more themselves or were able to find ways to provide more cost-effective care, access to health services for lower-income people might be reduced. Another argument against the option is that distinguishing between acute and long-term care for the purposes of financing could be difficult administratively. For example, in order to facilitate their recovery, former hospital patients often require services after an inpatient stay that resemble long-term care. Finally, greater state discretion creates the potential for increased disparity across states in eligibility requirements and benefit packages.

Hospitals that serve a disproportionately large share of low-income patients may receive higher payments from Medicaid—if the hospitals meet certain federal criteria—than other hospitals do. States have some discretion in determining not only which hospitals receive those so-called disproportionate share hospital (DSH) payments but also the size of those payments. During the late 1980s and early 1990s, many states engaged in funding transfers using the DSH program to obtain increased federal Medicaid funding without raising their net spending on DSH hospitals—effectively boosting the federal matching rate above that specified in law.

To rein in that practice, lawmakers enacted a series of restrictions on Medicaid DSH payments during the 1990s that included setting fixed ceilings on DSH payments to each state. The Medicare Modernization Act of 2003 raised those ceilings by $1.2 billion in 2004 and by smaller amounts in later years. The Congressional Budget Office projects that under current law, federal outlays for Medicaid DSH payments, which totaled an estimated $8.8 billion in 2006, will rise to $10.2 billion in 2012.

This option would convert the Medicaid DSH program into a block grant to the states. The grant could be reduced below levels under current law; or its future growth could be limited to a slower rate than that at which Medicaid DSH payments would increase under current law; or both approaches could be implemented. In exchange for less funding, states could be given greater flexibility to use the funds to meet the needs of their low-income and uninsured populations in more cost-effective ways.

As an illustration of how this option could be structured, the block grant for each state in 2008 could equal 90 percent of the state's Medicaid DSH allotment for 2007. In subsequent years, the block grant could be indexed to the increase in the consumer price index for all urban consumers minus 1 percentage point. In that case, outlay savings from this option would total $600 million through 2012. (The option would increase costs at first because states do not currently spend all of their allotted DSH money as a result of the criteria and conditions that must be met—conditions that would be removed under this option.)

A rationale for converting to a block grant is that, in addition to the budgetary savings that would eventually result under this option, the increased latitude provided to the states could result in DSH funds being better targeted to facilities and providers that serve low-income populations. For example, states would have greater flexibility to use those funds to support outpatient clinics and other nonhospital providers that treat Medicaid beneficiaries and low-income patients.

State governments, however, might not increase their contributions to make up for the reduction in federal subsidies. As a result, hospitals (and health care providers in general) could receive less in combined federal and state Medicaid subsidies and, consequently, they might not be able to serve as many low-income patients. Another potential effect is that giving states more flexibility to allocate DSH payments could alter the distribution and amount of assistance among hospitals, possibly resulting in some hospitals' receiving less public funding than they do now. Finally, states may already have enough flexibility under current rules to allocate DSH payments to achieve the maximum benefit.

Medicaid is a joint federal/state program that pays for health care services for a variety of low-income individuals. The states operate the program and receive financial assistance from the federal government in the form of matching payments: The states pay for services for Medicaid beneficiaries; submit evidence of payments for medical claims to the federal government; and receive matching federal funds that may range from 50 percent to 83 percent of those states' payments, depending on the per capita income of the state. That payment mechanism can, in some instances, create an opportunity for states to inflate their payments for medical claims in order to maximize the federal assistance they receive.

Many states finance part of their share of Medicaid spending by imposing taxes on health care providers. States typically impose taxes on a particular type of provider and use the revenues to increase payment rates to those same providers. In the process, states collect federal Medicaid funds to cover a portion of those higher payments. In a simple example, a state pays a provider $100 for services provided to Medicaid beneficiaries and receives a federal matching payment of 50 percent of that amount. In the absence of a provider tax, the state and federal governments would each pay $50 for the services. But suppose the state assesses a tax on the provider of 6 percent of gross revenue—the maximum allowed in 2007—and pays the provider $106 for Medicaid services (amounts are rounded to the nearest dollar). The provider=s net payment from the state for the Medicaid services is still $100 ($106 - $6). The federal government reimburses the state for 50 percent of the payment—$53 ($106 * 0.5), leaving the state paying only $47 ($53 - $6). The effective federal matching payment thus increases from 50 percent to 53 percent in that example.

The 109th Congress reduced the rate of allowable taxes levied on Medicaid providers from 6 percent to 5.5 percent for the period beginning January 1, 2008, and ending September 30, 2011 (see Public Law 109-432); this option would gradually reduce that rate further, to 3.0 percent by 2010, and make the limit permanent. This option would reduce federal spending by $130 million in 2008 and by $4.5 billion over five years.

The primary rationale for this option is that it would reduce federal Medicaid expenditures. By lowering the ceiling on allowable taxes, this option would limit the extent to which states could use such taxes to effectively inflate their federal matching rate. An argument against this option is that the lower payments to states that would result could lead to fewer people receiving Medicaid assistance, less assistance being provided per covered Medicaid beneficiary, or reduced spending by states on other activities.

In low-income families, children are much more likely than adults to qualify for public health insurance. As a result of the Medicaid expansions that began in the late 1980s and enactment of the State Children's Health Insurance Program (SCHIP) in 1997, the great majority of children in families with income below 200 percent of the federal poverty level are now eligible for either Medicaid or SCHIP. For parents, however, states generally limit Medicaid eligibility to those with income substantially below the federal poverty level ($17,170 for a family of three in 2007). Several states have expanded eligibility for public coverage to parents at higher income levels.

Under this option, states would be required to expand Medicaid eligibility to all parents with income below the federal poverty level. That new requirement, which would provide coverage to 2.8 million low-income adults and children by 2012, would increase federal outlays by about $900 million in 2008 and by about $14.9 billion over five years.

The main rationale for this option is that it would expand health insurance coverage to low-income parents and their children. In 2005, roughly 40 percent of low-income parents were uninsured. Among parents who would be newly eligible under this option, participation rates would probably be similar to rates among their children who are currently eligible for Medicaid or SCHIP. Coverage for newly eligible parents may boost participation for children currently eligible for Medicaid or SCHIP but not enrolled, since parents and their children would be covered under the same insurance.

A potential drawback of this option is that expanded eligibility could result in some parents with private insurance dropping that coverage to obtain public insurance. Moreover, employers with disproportionate numbers of lower-income workers might be less inclined to offer health insurance to their workforce as a whole because the perceived demand would be lessened by the availability of the new alternative coverage. Also, the increased amounts that states would be required to spend under this option could lead some to cut back on optional health care services that they would otherwise have provided.

As a result of expansions to the Medicaid program and enactment of the State Children's Health Insurance Program (SCHIP) in 1997, the majority of children in families with income below 200 percent of the federal poverty level are now eligible for either Medicaid or SCHIP. That coverage based on family income typically ends for young adults when they turn 19, however. And most low-income young adults cannot qualify for public coverage on their own because eligibility is generally limited to parents, disabled adults, and pregnant women; furthermore, income-eligibility requirements for adults are generally more restrictive than they are for children.

This option would require states to expand their Medicaid eligibility to young adults ages 19 to 23 with income below the federal poverty level. That new requirement, which would provide coverage to an estimated 600,000 low-income young adults by 2012, would increase federal outlays by $300 million in 2008 and by $6.2 billion over five years.

A rationale for this option is that it would increase the currently low rates of health insurance coverage for low-income young adults. (In 2005, almost half of those adults were uninsured.) Low-income young adults generally have less access to employer-sponsored health insurance than do other adults because they often work part time or for employers that do not offer coverage. Expanding Medicaid eligibility to that group would increase their access to preventive care and lower the risk of high medical expenditures in the case of unforeseen illness.

An argument against this option is that expanding Medicaid would increase the program's expenditures at a time when it already faces budgetary pressures. Another potential argument against this option is that many young adults are uninsured by choice. (Because low-income young adults tend to be healthier than the overall population, their perceived low risk of having health problems makes them less likely to buy health insurance.) Finally, the increased amounts that states would be required to spend under this option could lead some of them to cut back on optional health care services that they would otherwise provide to their existing Medicaid-covered populations.

Enacted as part of the Balanced Budget Act of 1997, the State Children's Health Insurance Program (SCHIP) provides health care coverage for certain uninsured children from low-income families. States administer SCHIP through their Medicaid programs, as a separate program, or a combination of both. The program, which began operation in 1998, is authorized through 2007. For the purpose of its baseline budget projections and consistent with statutory guidelines, the Congressional Budget Office assumes that funding for the program in later years will continue at its 2007 level of $5.0 billion. Such funding for SCHIP would gradually cover a progressively smaller proportion of the nation's children because of the rising cost of medical care and the increasing size of the population of children nationwide.

This option would index SCHIP funding after 2007 to the rates of growth in per capita health expenditures—using the projections of national health expenditures (NHE) from the Centers for Medicare and Medicaid Services—and in the number of children. According to the most recent NHE projections, per capita health expenditures will grow by about 6 percent annually after 2007. Those changes would increase SCHIP funding to about $6.8 billion by 2012, raising outlays by $70 million in 2008 and by a total of $1.9 billion through 2012.

This option would reduce, but not eliminate, the shortfalls in funding that many states will face if SCHIP funding continues at its current annual level of $5.0 billion. Under their current allocation of SCHIP funds, a number of states are not receiving sufficient federal money to finance their programs as currently operated. (CBO estimates that 14 states will face a total shortfall of $735 million this year.) In addition, the number of children who are eligible for SCHIP will probably grow more quickly than the overall population of children as the share of people who have private health care coverage continues to gradually decline.

An argument for this option is that without such a funding increase, many states will be unable to maintain their current level of benefits and coverage beyond 2007. (Even with this increase, some states would still face shortfalls in the future.) Many states are already experiencing shortfalls under the current funding levels that cannot be fully addressed by the redistribution of SCHIP funds from states with surpluses. Therefore, to stay within their budget, states will have to reduce the level of benefits they provide to recipients, restrict the number of children deemed eligible for aid, or implement some combination of the two.

An argument against this option is that current funding levels reflect the Congress's intent to establish SCHIP as a program with limited funding. According to that argument, states should design programs with those limits in mind and pay for any additional spending entirely with their own funds. In general, states have flexibility in setting eligibility levels and benefit packages provided under their programs, and they may alter those criteria to reflect the availability of federal and state funds. Some states have even used unspent SCHIP funds for demonstration projects to expand coverage to low-income adults. (The Deficit Reduction Act of 2005 prohibits new SCHIP waivers for nonpregnant, childless adults, however.) It can be argued that SCHIP was intended to cover children and that additional funding need not be provided if some of the program's resources are being used to cover adults.