Administrative Issues and Effects on Other Programs

Health care proposals that would substantially affect health insurance markets could have a number of important budgetary effects and other implications—for federal and state agencies as well as certain segments of the population—that are not discussed in the preceding chapters. In particular, the federal cost of such proposals and their impact on coverage rates would depend partly on the magnitude of the administrative costs and responsibilities that the proposals entail and partly on their interactions with other government programs.

Large-scale changes to the health insurance system would probably require the federal government to take on new administrative responsibilities. The extent and nature of those responsibilities, and the associated costs, would depend largely on the scope of a proposal—considerations that would also affect its timeline for implementation. In addition, proposals that established or expanded federally funded subsidies for health insurance might either shift costs from state governments and employers to the federal budget directly or create incentives for states and employers to reduce their current payments in ways that boost federal spending. Maintenance-of-effort requirements, under which states or firms must continue activities that they had performed before a new federal program was created, could discourage state governments and employers from cutting back on programs that provide similar health care benefits. Other types of maintenance-of-effort requirements could seek to recapture some or all of the states’ savings resulting from a new federal program, thus reducing the net costs incurred by the federal government. In any case, the impact of such provisions would depend on how effectively they could be monitored and enforced.

The costs of proposals and their effects on coverage rates also depend on how the proposals would treat certain segments of the population and how they would interact with other federal programs. Proposals could affect the receipt of other federal benefits that are not directly related to health care in a number of ways. For example:

For certain segments of the population, such as military service members and veterans, eligibility and payment rules for new subsidies might need to be coordinated with rules under existing government programs. For other segments—in particular, unauthorized immigrants— decisions would need to be made as to whether to deny them federal assistance (as many current programs do). Whether and how proposals sought to include or exclude such populations would affect not only their costs but also their impact on the measured number of uninsured people—both directly and because enforcement provisions could affect enrollment more broadly. Other segments of the population are difficult to reach through existing programs and would present similar challenges to administrators of new subsidy programs or insurance mandates.

Proposals that would significantly change health insurance markets and health care systems could create new administrative responsibilities for the federal government. How long it would take to implement such proposals—and how effectively they would be carried out—depends on their scope and on the funding that is made available for both start-up and operating costs. In assessing the ability of federal agencies to implement large-scale changes, the Congressional Budget Office would take into account the likelihood that adequate funding would be available. Even with sufficient funds, agencies might need several years to make a substantially new system operational and even more time to fully implement it. Various trade-offs would arise if proposals relied on state governments for implementation.

Federal and state governments share responsibilities for the financing and regulation of health insurance and health care (see Table 6-1). Federal agencies perform tasks such as operating the Medicare program, administering tax provisions covering health care and health insurance, regulating the sale of drugs and medical devices, and overseeing employment-based health plans. Although Medicaid and the State Children’s Health Insurance Program are jointly financed by the federal and state governments, state agencies largely administer the programs—determining eligibility and benefit levels and paying providers of health care and insurers. States also license and regulate physicians and health insurers, although many employment-based health plans are exempt from state regulation under terms of the Employee Retirement Income Security Act. (For further discussion of that law, see Box 1-1.)

Error processing SSI file

Source: Congressional Budget Office.

Major proposals affecting health insurance could assign new responsibilities to either an existing or a new agency, and current responsibilities could be shifted between federal and state agencies. Enacting a mandate that required individuals to obtain health insurance or employers to provide it would probably require one or more government agencies to monitor health insurance coverage (as discussed in Chapter 2). Federal and state agencies might also be called on to administer subsidies for health insurance to a broader population than existing programs cover. Proposals to create regional purchasing groups for health insurance might involve interstate collaboration or modify states’ ability to regulate insurance markets.

Trade-offs are implicit in either starting a new agency or relying on existing agencies, and they could affect the costs and impact of implementing a health care proposal. A new agency might benefit from a focused mission and dedicated resources but would face the challenge of "starting from scratch." Creating a new agency could also entail substantial start-up costs. Some of those costs could be reduced by shifting employees from existing agencies to the new agency, but the new agency would then face the additional challenge of aligning the disparate responsibilities and corporate cultures of workers from different government offices. Existing agencies have staff members and other resources already in place but might find it difficult to expand their activities to include new tasks. Moreover, existing agencies might require significant investments in infrastructure if new programs differ substantially from their current responsibilities.

The Internal Revenue Service’s administration of the health coverage tax credit (HCTC) provides an example of the difficulties inherent in implementing a new program. In managing the HCTC program, which assists workers displaced as a result of international trade, the IRS had to take on activities with which it had little previous experience. Typically, taxpayers claim tax benefits on their annual income tax returns and receive those benefits by having their income tax reduced or receiving a larger tax refund. Under the HCTC program, the IRS has to determine throughout the year whether individuals are eligible for benefits and then pay monthly premiums on their behalf to insurers. The program was enacted in 2002 and served about 45,000 people in 2005 (including workers, their spouses, and their dependents). For fiscal years 2003 and 2004, the IRS received a total of $81 million in administrative funds to start the program. As the agency has developed new systems and gained experience, administrative costs have fallen to about $15 million a year.¹

CBO’s analysis of the administrative costs for government agencies to implement a major initiative would reflect current agency budgets and an assessment of the effort required for the new activities. In 2003, the Centers for Medicare and Medicaid Services operated the Medicare Part A and Part B programs with an administrative budget of about $5 billion, or less than 2 percent of the programs’ costs. To put into operation the Medicare drug benefit and carry out the other administrative responsibilities generated by the Medicare Prescription Drug Improvement and Modernization Act of 2003, legislation provided an additional $1.5 billion for CMS and the Social Security Administration in 2004 and 2005. Ongoing administrative costs for the drug benefit are about $900 million per year, or about 2 percent of program costs.

Implementation Timelines

The time required to implement a proposal would vary greatly depending on its nature and scope, but it seems likely that instituting major changes to the health insurance system would take several years. About two years elapsed between enactment of the Medicare drug benefit in December 2003 and the availability of that new benefit to enrollees in January 2006. Approximately half of that time was spent creating a regulatory framework for the program based on the enacted legislation. That framework involved several stages that are common to the regulatory process: First, policies were developed, with input from the public; next, proposed rulings were issued in the Federal Register, followed by a public comment period; and last, final rules were issued, which could take effect up to 60 days after publication in the Federal Register.

Once the regulatory framework was established, CMS solicited and reviewed bids from the private insurance plans that would deliver the benefit, provided information about those plans to enrollees, and processed their choice of plans. Even with that amount of lead time, some problems (regarding assignment of enrollees to plans, filling of prescriptions, and collection of premiums) occurred during the initial months after the drug benefit became available.

The schedule for implementing the Massachusetts health care mandate requiring adults to obtain health insurance coverage also provides insight into the time involved in implementing a major proposal. The legislation was enacted in April 2006, and it took about a year for the new private health insurance options mandated by that law to be delineated and become available. (An accompanying expansion of eligibility for Medicaid was accomplished more quickly.) As part of implementing the new program, the state needed to develop a system to report whether individuals have qualified insurance coverage—an important step because those who do not could be subject to financial penalties. (By design, those penalties are being phased in gradually and will be fully implemented in 2009.)

Implementing a health insurance proposal designed at the federal level could pose additional challenges for the states. Depending on the nature of the proposal, trade-offs could arise between the role given to states and the time it takes to implement the changes involved. In some cases, implementation by each of the 50 states—as well as the District of Columbia and any affected territories—would take longer than action by the federal government. Six state legislatures meet only biennially, limiting their ability to respond quickly to new federal policies if conforming changes in state laws or additional state expenditures are necessary. Moreover, differences among states—in population, geography, or current insurance coverage rates—might generate unanticipated challenges and additional delays. Massachusetts was able to implement its health care mandate quickly because the share of that state’s population that was uninsured was relatively low and because officials were motivated to do so. If other states had to implement a similar policy, they might not be able to move as quickly.

One way to ease implementation of a new federal program would be to build on existing state programs. Initial implementation of SCHIP—a program enacted in August 1997—was able to proceed relatively rapidly because it largely built on the existing infrastructure of Medicaid (a program that already served children in low-income families). As a result, about half of the states were able to have an SCHIP program operating within a year. But a few states took more than two years to begin their programs, and even among the states that acted early, a common approach was to pursue a modest coverage expansion first and then take additional steps in later years. (Another reason for taking a gradual approach may have been concern that the authorizing legislation reduced federal funding by 25 percent after four years.) Consequently, several years elapsed before some states were using the full amount of money available to them under the program.

Delegation Issues

A potential challenge in implementing major changes to the health care or health insurance systems is known in political science as the "agency" problem, which can arise when a principal (through legislation, for example) delegates authority to an agent (such as a government agency). The problem occurs if those agencies implement the legislation in a way that does not reflect the drafters’ intentions.

Another agency problem can arise when states are given responsibility for spending federal dollars; they have less of an incentive to be fiscally prudent because they are not spending their "own" money. The Medicaid program attempts to limit agency problems by tying federal contributions to state spending and other requirements. Specifically, federal matching payments are structured so that each additional dollar in state spending is reimbursed at an average of 57 cents by the federal government. But experience with Medicaid indicates that the program’s funding structure creates opportunities for states to maximize federal spending and minimize state spending. Even federally administered programs such as Medicare are not immune to agency problems—which may arise between federal legislators and program administrators or between federal and local administrators—but federal authorities may have more control over such programs.

Maintenance-of-Effort Requirements

When the federal government creates a new program that could reduce the financial obligations of other payers—particularly states or employers—interest may arise in establishing maintenance-of-effort (MOE) requirements. New federal programs could substitute federal funds for payments now made by states or employers directly, or they could induce states or employers to scale back their activities in ways that would raise federal costs. As applied to states, maintenance-of-effort provisions, which aim to prevent that type of cost shifting, can be structured in two ways:

■	By requiring states to maintain existing programs at historical eligibility or benefit levels, or

■	By requiring them to continue spending funds at certain historical or projected levels or return some of their savings to the federal government.

The two approaches have many similarities and, in both cases, the effects on federal spending would depend heavily on how the requirements were defined and how effectively they could be enforced. Two examples of effective MOE provisions are the ones that were included with the enactment of SCHIP and the Medicare drug benefit. Applying similar MOE requirements to the private sector would be difficult, however, unless proposals specified new enforcement mechanisms and sufficient penalties for violations.

Maintaining Eligibility or Benefit Levels

One goal of provisions that require states to maintain existing programs at historical eligibility or benefit levels may be to prevent states from shifting the responsibility for funding to the federal government for populations or benefits that are already covered and partially subsidized by state funds. Another goal may be to guarantee that people who receive a certain level of services or benefits are not made worse off after the transition to a new system.

When the State Children’s Health Insurance Program was enacted, certain maintenance-of-effort requirements were imposed. SCHIP was designed to cover uninsured children in families whose income is somewhat higher than the maximum allowed for children under Medicaid. SCHIP also provides a higher federal matching rate, covering about 70 percent of the program’s health care costs, on average, compared with 57 percent under Medicaid.² The law froze Medicaid income-eligibility levels for children so that states could not reduce those levels, remove children from the Medicaid rolls, and then reenroll those same children under SCHIP in order to receive a greater federal financial contribution. In a large-scale proposal to change the health insurance system, a similar approach could require states to maintain eligibility levels for low-income people under existing programs.

Maintaining State Funding

Maintenance-of-effort provisions could be implemented that would require states to continue spending funds at certain historical or projected levels. That spending could take the form of reimbursements to the federal government. In that way, MOE provisions could seek to capture some or all of the state savings that might result from a new federal program, in effect offsetting the costs of the new program.

The prescription drug benefit under Medicare provides an example of that approach. The Medicare Modernization Act transferred responsibility for covering outpatient drug costs for people eligible for both Medicare and Medicaid (the "dual eligible" population) from Medicaid to Medicare. By itself, that transfer of responsibility would have substantially reduced states’ costs for Medicaid. But the law also included an MOE provision that effectively requires states to transfer to Medicare most of the estimated funds that they would have spent on drug benefits in the absence of the new benefit. That provision is referred to as a "clawback" mechanism.

In general, requirements for states to maintain funding levels would need to specify several features in order for their effects to be estimated. First, they would have to specify the categories of spending that would be subject to the MOE provisions and establish the initial spending level that would have to be maintained. The MOE requirement for the Medicare drug benefit was a relatively well-defined target, including virtually all Medicaid spending on outpatient prescription drugs for dual-eligible enrollees.³ If a new federal program would cover only a portion of enrollees’ current Medicaid spending, however, disentangling the funding streams might be more difficult.

Another consideration would be whether and how MOE payments would be adjusted over time to approximate what states’ costs would have been in the absence of the new federal program. The initial MOE payment could be indexed in later years to reflect changes over time, such as growth in the population eligible for the program or increases in health care costs. Possible indexes include general price inflation, nominal growth in the economy, or growth in overall health care costs; a faster-growing index would yield larger MOE payments. Even over a decade, the choice of an index could significantly affect the amount of the MOE payments; by CBO’s estimates, general price levels (as measured by the consumer price index) are expected to increase by 21 percent between 2009 and 2018, total health expenditures are expected to grow by about 80 percent in nominal terms over that period, and Medicaid’s costs will roughly double. What states’ costs in the future would actually be in the absence of a new federal program is unknowable, however.

Complying with MOE Requirements

Maintenance-of-effort requirements would be less effective if states are able to take action to minimize their impact. States could minimize such requirements by limiting eligibility or benefit levels for the affected programs or by reducing financing amounts before the requirements could take effect. Lax enforcement of maintenance-of-effort requirements could weaken their effect.

A critical design issue is the date selected to freeze eligibility or benefit levels in existing programs or to establish the initial amount of funding subject to the MOE requirements. If that date fell after the proposal is enacted, states could minimize their costs—and boost federal payments—by scaling back the targeted programs in the interim period. States might even begin to modify programs earlier if they anticipated successful enactment of new subsidies. To limit those possibilities, the SCHIP statute set the date for freezing eligibility levels for Medicaid at a point several months before the law was enacted. Similarly, the Medicare drug benefit program used data from all of 2003 (after its enactment in December of that year) to establish the initial amount of funding subject to the MOE requirements.

Another issue is how the federal government would monitor and enforce the MOE provisions. Monitoring and enforcement would be easier if MOE requirements were applied to existing programs, like SCHIP and Medicaid, with established reporting systems. SCHIP and Medicaid already require federal approval before states can change eligibility levels or program benefits. Similarly, for MOE requirements that recapture state funds, the most effective enforcement mechanisms are those that allow the federal government to recoup any state MOE payments by adjusting downward the amount of funds it would otherwise send to the states to finance the overall program. Under the Medicare drug benefit’s MOE provisions, for example, the federal government can recover any unpaid amounts by reducing its reimbursement to states for the remaining costs of Medicaid—costs that are substantially larger than the MOE payments. The issues of monitoring and enforcement become more complex, however, if the existing programs that are subject to MOE requirements lack established reporting and enforcement mechanisms or if new and untested systems have to be created.

The extent to which states take advantage of opportunities to minimize the effect of such obligations appears to be tempered by their concern about the impact on their residents of cutbacks in services and programs. Although states have frequently found ways to reduce their own expenditures while increasing federal Medicaid payments, the means for shifting responsibility for funding were legal at the time states used them—and states have generally complied with laws intended to prevent future occurrences. Furthermore, states have attempted to avoid reductions in eligibility and benefit levels. More than half of Medicaid spending goes to cover populations and services that are not required by federal law, and states have not taken significant action to reduce eligibility or limit benefits even in difficult economic circumstances.

When analyzing any new subsidy proposals, CBO will evaluate the opportunities available to state governments to minimize the impact of MOE provisions without significantly burdening their residents. States would be expected to fully comply with MOE requirements that are similar in design to those contained in the SCHIP and Medicare drug benefit programs, both of which appear to have been enforced effectively. Depending on the extent to which proposals differ from those provisions, CBO would assess the degree to which states would be likely to shift current expenditures to a new program.

Applying MOE Requirements to Employers

Because new federal subsidies for health insurance could replace payments made by employers—or induce employers to reduce contributions for employees’ health plans or drop plans entirely—proposals might apply MOE requirements to the private sector.⁴ One approach would be to require employers who previously offered health insurance to their employees to maintain prior eligibility rules, benefit amounts, or total contributions. That approach would be similar to the MOE requirements that now apply to state governments. Another approach would be to require employers to increase cash wages by the amounts previously spent on health insurance if their share of those costs declined because of the provision of health insurance from other sources. (In the latter instance, CBO would expect market forces to adjust employees’ compensation, in the aggregate, even in the absence of a requirement on employers.)

For several reasons, however, maintenance-of-effort requirements for employers would be more difficult to implement in the private sector than in the public sector. The federal government does not mandate that employers offer health insurance, nor does it specify the level of benefits that must be provided (with certain limited exceptions) or the contribution that employers offering such coverage must make. Although employers deduct their aggregate contributions toward health insurance as a business expense when calculating their tax liabilities, they do not currently report contributions made on behalf of each employee. As a result, no established reporting or enforcement mechanisms exist that would allow the federal government to closely monitor whether employers were maintaining their effort. Separately identifying and monitoring eligibility, benefit levels, and contributions effectively would be an extremely complex undertaking. Furthermore, imposing a requirement on employers would create an incentive for businesses to reorganize or rename themselves in order to avoid the requirement (an option not available to states). Thus, unless proposals specified a significant new administrative structure and added sufficient penalties for violations, CBO would assume that many employers would not fully comply with MOE provisions regarding coverage or benefits.

Effects on Other Federal Programs

Legislative changes to the health insurance system could affect other federal benefits that are unrelated to health insurance, in a number of ways. For example:

■	A new subsidy for health insurance premiums could be counted as income or assets for means-tested benefit programs (those that require recipients to have limited resources), thereby affecting eligibility and the amount of benefits received under those programs.

■

Modifications to employment-based health insurance could shift compensation from wages to fringe benefits (or vice versa). That change would affect the amount of Social Security benefits as well as benefits under any other government programs for which eligibility or payments are based on a recipient’s earned income.

To the extent possible, CBO would consider those interaction effects in its analysis of proposals.

Inclusion of Subsidy Payments in Income or Assets

A number of federal programs would treat tax credits for health insurance or other subsidies as income or assets under current laws and regulations. The programs examined below do not constitute an exhaustive list but rather illustrate how proposals to change the health insurance system might affect their enrollment or benefits.

Supplemental Security Income. The SSI program is a mandatory program, providing cash payments (based on income) to needy individuals who are aged, blind, or disabled. (In mandatory programs, spending levels are not subject to the annual appropriation process.) To be eligible, a person must have income and assets below certain limits. In most states, SSI recipients automatically qualify for Medicaid.

The SSI program defines income as any cash or in-kind payment that can be used to meet a person’s needs for food or shelter. Needs-based assistance funded by state and local governments and refunds of income taxes are excluded from income. Refundable tax credits—other than the earned income tax credit and the child tax credit—are included in income. Although unspent tax refunds may be included in assets, the portion of income tax refunds that is attributable to the earned income tax credit and the child tax credit is excluded from countable assets for nine months following receipt. In addition, a person’s living arrangements or use of institutional health care can affect SSI receipt. For example, monthly SSI benefits can be reduced if an individual is in a hospital or nursing home for the entire month and Medicaid pays more than half the bill.

A new program that subsidized enrollees’ health insurance through payments to insurers or providers would generally not affect income as defined under SSI eligibility rules because the beneficiary cannot convert those payments (or the medical services received as a result) into food or shelter. But if a new subsidy was structured as a tax credit or other cash payment, it could be considered income or an asset—unless it was specifically excluded. Because SSI recipients have low income, however, they might not benefit from such a tax credit unless it was made refundable.

Supplemental Nutrition Assistance Program. Formerly known as the Food Stamp program, the Supplemental Nutrition Assistance Program provides low-income individuals and households with debit-like cards that can be used to purchase selected groceries. Many recipients of Temporary Assistance for Needy Families or SSI are also eligible for SNAP benefits. Other households must have cash income and assets below specified thresholds to receive those benefits, which decline as income rises. Both cash income and assets are defined broadly for the purpose of determining eligibility. By law, however, nonrecurring payments (such as tax refunds) and certain recurring payments (such as advance payments of the EITC) are excluded from income but are generally counted as assets. EITC payments are also excluded from countable assets for 12 months after receipt.

Whether new subsidies for health insurance would affect eligibility for SNAP benefits would depend, in part, on the subsidies’ design. Programs that provided recipients with in-kind benefits—such as free or low-cost medical services—rather than cash would not directly affect SNAP eligibility or benefits. However, SNAP benefits could decline for disabled or elderly beneficiaries, who can deduct medical expenses from income when determining eligibility for nutrition assistance, if their out-of-pocket medical expenses fell as a consequence of new subsidies. In contrast, recurring cash payments (if not considered reimbursements) would generally be counted as cash income and assets unless explicitly excluded from the program’s eligibility determination.

Temporary Assistance for Needy Families. Temporary Assistance for Needy Families is a mandatory program that primarily assists low-income households. Unlike SSI and SNAP, however, TANF is administered as a block grant provided by the federal government to the states—which then determine eligibility rules and benefit amounts for their residents. Depending on how each state defined income and assets, a new federal subsidy for health insurance could affect the amount of TANF benefits that people received. But those changes would not affect a state’s entitlement to federal funding: Each state would continue to receive the same allotment as specified under federal law. As a result, CBO’s cost estimates for new federal subsidies for health insurance would not reflect offsetting savings in the TANF program.

Changes in Workers’ Earnings and Income

Employers’ payments for health insurance premiums are exempt from income and payroll taxes; similarly, those premium payments are not counted as income to workers when determining eligibility for and the amount of certain transfers (such as Social Security benefits) and tax provisions, including the EITC. Changes to the employment-based health insurance system could affect Social Security benefits or the EITC in two ways. First, proposals that would repeal or limit the current exclusion of employers’ contributions for health insurance would also cause those contributions to be counted as income to workers when determining their Social Security benefits or EITC—unless specifically excluded from income solely for those provisions. Second, any proposal that would induce changes in the allocation of workers’ compensation between wages and health insurance benefits would affect Social Security benefits or the EITC. Offsetting the effects of such changes would present significant challenges.

Social Security. Unlike the programs discussed earlier, Social Security benefits are not means-tested, but the payments that a person receives are a function of his or her past earnings. To the extent that a proposal had an impact on taxable wages—as would happen if the current tax exclusion for employment-based health insurance was limited or eliminated—that outcome would affect not only payroll tax revenues but also future Social Security benefits.

Changes in the overall composition of compensation for workers could have other effects, both immediate and long lasting, on the benefits paid to all people filing new claims for Social Security benefits. The reason is that the calculation of an individual’s initial benefit is based on his or her previous earnings, but all income earned in the years before attaining the age of 60 is indexed to that year according to average wage growth in the economy. As a result, proposals that substantially shifted employers’ health insurance payments to cash compensation could increase measured wage growth sharply during the transition period and thus yield higher benefit payments for all new claimants. Proposals could seek to offset that impact by temporarily adjusting the benefit formula, but they would have to rely on an estimate of the impact because it would be difficult, in practice, to determine precisely what wage growth would have been otherwise.

Earned Income Tax Credit. Low-income people may be eligible for the earned income tax credit, which is refundable (that is, paid to them even if they do not have any income tax liability). To be eligible, individuals must have some wage or self-employment income. The amount of the EITC increases initially as earned income rises until it reaches a maximum credit amount; then, the credit declines gradually as income continues to rise. When income reaches a certain threshold—which varies with marital status and family size—the credit falls to zero.

The current exclusion of health insurance premiums from earned income reduces the credit for workers with very low wages (who are in the credit’s phase-in range) and increases the credit for workers with somewhat higher income (who are in the phaseout range). Correspondingly, proposals that included employers’ contributions for health insurance in earned income or caused workers’ compensation to shift from employer-paid health insurance premiums to wages would increase EITC payments for workers in the phase-in range and reduce the credit for those in the phaseout range.

Proposals that reduced or eliminated the exclusion of employer-paid premiums could hold the EITC "harmless" by specifying that those premiums not affect the computation of the credit. Using a different definition of income to compute the EITC, however, would increase the program’s complexity; workers, for example, would have to compute income two different ways when completing their tax returns—first to compute their income taxes and then to calculate the EITC.

Offsetting the effects of proposals that changed the composition of compensation in a way that held each worker harmless would be all but impossible, for two reasons. The amount spent by employers on health insurance for each worker is not monitored by the IRS, and there is no way to determine with certainty what workers’ earnings otherwise would have been. Even so, an aggregate adjustment could be made to reduce the impact, on average.

Coverage of Certain Populations

Policymakers seeking to expand health care coverage need to decide whether and to what extent to cover certain populations. Those considerations may make the proposals more challenging to administer and may affect their net impact on the federal budget. Some populations, such as members of the military and veterans, receive health insurance or health care from the federal government, so the creation of new subsidy programs or mandates could require additional coordination and could influence the use of those benefits. Other groups, such as prison inmates and unauthorized immigrants, are not eligible for federal assistance for health insurance under current programs. Whether those groups were included or excluded by a proposal would affect its cost and would have an impact on coverage rates. Additional groups, such as the homeless, are difficult to reach through existing programs and would probably present administrative challenges to new efforts to expand coverage.

Military Families and Retirees

Many active-duty members of the military, their families, and military retirees receive health insurance benefits through the Department of Defense. In 2006, about 9 million people were eligible for benefits under DoD’s collection of health plans known as TRICARE.⁵

Some TRICARE beneficiaries have access to other insurance, such as private employment-based coverage offered to a working spouse. If beneficiaries choose to use other insurance plans because they prefer the benefit structure of those plans or because access to TRICARE providers is limited in their area, TRICARE becomes the secondary payer to that other insurance. In 2006, about 12 percent of eligible family members of active-duty service members—400,000 people in total—also had other health insurance. The percentage of eligible military retirees under the age of 65 and their families enrolled in other health insurance plans has declined from 49 percent in 2001 to 33 percent in 2006 (for a total of about 1.1 million retirees under age 65 and family members). If legislation caused private employers or other insurers to provide benefits that were less comprehensive or more expensive, reliance on TRICARE could climb; conversely, more affordable or accessible health coverage options from other sources could cause reliance on TRICARE to fall, decreasing the program’s discretionary spending requirements.

The TRICARE beneficiary population includes more than 200,000 families and military retirees and their families who are either temporarily or permanently living overseas. Most service members are stationed overseas at least once during their career, although not all family members accompany active-duty members who are deployed overseas. In addition, some military retirees choose to move overseas after retirement. Because Medicare does not cover services provided outside the United States, TRICARE is the primary payer for medical expenses incurred by military retirees over age 65 who are living overseas. Thus, legislation that changed TRICARE coverage in some way might affect TRICARE beneficiaries living overseas differently than those residing in the United States.

Veterans

Legislative changes to the health care system might also affect beneficiaries of the Veterans Health Administration, which is part of the Department of Veterans Affairs (VA). Those who previously served in the military can apply to receive health care services from the VA. In fiscal year 2007, there were about 23.5 million veterans, 13.5 million of whom were eligible to enroll in the VA health care system. Of those 13.5 million eligible veterans, 7.8 million were enrolled with the VA (2.8 million of those enrollees did not seek care from the VA) and 5.7 million were not enrolled. Almost 80 percent of enrolled veterans report having some other type of health coverage.⁶ On average, eligible individuals receive no more than half of their total health care through the VA system, even among the groups that rely on VA care most heavily (such as veterans with severe disabilities incurred while serving).

Veterans who apply for enrollment in the VA are assigned to one of eight priority groups on the basis of their service-connected disabilities, service-related exposures, income, assets, and other factors. Veterans with the most severe disabilities and lowest income and assets are placed in the highest-priority groups, and higher-income veterans without service-connected disabilities are placed in the lowest-priority groups. In an effort to reduce the growth in the VA’s medical spending, a freeze on enrollment took effect in January 2003 for veterans assigned to the lowest priority group. About 10 million veterans are ineligible to enroll for the VA’s services because of that freeze.

Legislation that provided alternative coverage for low-income people might draw patients away from the VA system. A substantial portion of the VA’s enrollees have low income and may be going to the VA for medical services because they have limited access to other sources of care or face higher out-of-pocket costs if they seek care from other providers. If alternative coverage allowed them to receive affordable medical services from other providers, they might rely less on the VA’s care, thus opening up additional capacity for other patients, including those in the lowest-priority group. More generally, increases in the affordability of private health insurance might reduce the use of the VA’s services. However, any legislation that required or encouraged veterans to increase their use of the VA’s services would place pressure on existing resources unless it included an accompanying increase in appropriations. (Funding for the Veterans Health Administration is discretionary and thus is subject to annual appropriations.)

Populations Served by the Indian Health Service

The Indian Health Service (IHS) program, and the people it serves, might also be affected by proposals to modify the health care system. The IHS offers a broad range of health care services, including primary care, ancillary services (such as diagnostic laboratory and radiology services), specialty services, and inpatient care to an estimated 1.8 million American Indians and Alaska natives—about 55 percent of all American Indians and Alaska natives eligible to receive care in IHS facilities.⁷ The IHS provides that care through 48 tribal hospitals and more than 600 primary care clinics. Because of staff shortages, limited facilities, and a capped budget, the IHS rarely provides benefits comparable with complete insurance coverage for the eligible population; as a result, estimates of the uninsured population in the United States do not treat the IHS as a source of insurance. As funds permit, the IHS contracts with outside providers to deliver services not available at its facilities.

Under current law, Medicaid, Medicare, and private insurance plans pay the IHS for services furnished to patients who have coverage under those plans. Approximately 17 percent of the program’s total funding is attributable to insurance collections, with payments from Medicaid accounting for about 90 percent of all insurance collections. Proposals that would expand health insurance coverage might increase the funding available to the IHS indirectly through greater insurance recoveries. However, because IHS funding is discretionary, the budgetary effects of those changes would depend on provisions enacted in subsequent appropriation acts.

Children in Foster Care

Under current law, child welfare agencies are responsible for ensuring that children in foster care receive needed health care services. Foster children include those under age 18 living in a foster home, group home, or the home of a relative.

Proposals seeking to expand health care coverage would need to address several issues regarding children in foster care. First, changes in Medicaid’s eligibility rules or benefits could significantly affect those children. Child welfare agencies usually ensure that foster children receive health care services by enrolling them in Medicaid. Of the approximately 800,000 children who received foster care services in 2001, more than 600,000 were enrolled in Medicaid.⁸

Second, the treatment of foster children under proposals mandating that parents or guardians obtain health insurance for children could lead to some confusion. Although child welfare agencies are responsible for ensuring that foster children receive needed health care services, children in foster care can be enrolled in the health plan of a birth parent or a foster family. A mandate requiring parents and guardians to provide health insurance could cause confusion among foster parents, birth parents, and state agencies over which party bears responsibility for ensuring that children are covered.

Third, applying a mandatory enrollment requirement for children of a certain age could affect children in foster care who have reached an age—typically, 18—when the state and their foster families are no longer required to give them assistance. In 2005, approximately 25,000 children were in that category. About half of the states have implemented federal options and waivers to extend Medicaid benefits to that population up to age 21.

Incarcerated Adults and Juveniles

Under current law, correctional authorities, including federal officials, are required to provide medical care to adults and juveniles in their custody. In 2006, an estimated 2.3 million adults and juveniles were in prisons, jails, and residential placement facilities. Proposals to expand coverage for health care could continue to make correctional authorities responsible for such care, or they could extend assistance to incarcerated adults and juveniles.⁹

State, local, and tribal governments do not receive federal matching funds under the Medicaid or IHS programs for care or services they provide to incarcerated individuals. Medicare payments made on behalf of beneficiaries in the custody of law enforcement agencies are also prohibited under federal law. Therefore, spending by state, local, and tribal governments for inmates’ health care—estimated to total more than $6 billion in 2003—is supported predominantly by state and local funds and, to a lesser extent, by money collected from inmates.¹⁰ At the federal level, the Bureau of Prisons estimates that health care costs to treat its approximately 200,000 inmates totaled about $740 million in fiscal year 2007. Those funds are discretionary and thus subject to annual appropriations.

Foreign-Born Residents

In 2006, roughly 37 million foreign-born individuals, representing about 12 percent of the U.S. population, resided in the United States. That category includes naturalized citizens, legal permanent residents, nonimmigrants (noncitizens who are in the country temporarily), refugees, and unauthorized immigrants. Regardless of their citizenship status, foreign-born residents are generally included in the counts of the insured and uninsured in the United States. For that reason, the treatment of foreign-born residents affects the impact that a health care proposal would have on overall rates of coverage. In particular, studies indicate that of the roughly 12 million unauthorized immigrants in this country, about half have health insurance and half are uninsured—so those 6 million uninsured people would account for more than 10 percent of the uninsured population.¹¹

Under current law, naturalized citizens (about one-third of foreign-born residents) are eligible for federal health insurance benefits on the same terms as native-born citizens. However, for many of the remaining two-thirds of foreign-born residents, Medicare and Medicaid rules restrict eligibility.¹² For example, legal permanent residents who entered the country after August 1996 are not eligible to receive most services under Medicaid until they have resided in the United States for five years; similar restrictions apply to Part B of Medicare. Except for emergency care and immediate services for childbirth—which must be provided under federal law—nonimmigrants and unauthorized immigrants are generally prohibited from receiving federally funded services.¹³ In contrast, some state and local governments provide additional services to noncitizens regardless of their immigration status.¹⁴

Depending on how applicants are required to demonstrate citizenship status, proposals that restrict immigrants’ access to health insurance benefits could also affect receipt of benefits by U.S. citizens. Until July 2006, applicants for Medicaid were required to declare, under penalty of perjury, that they met the citizenship or immigration requirements to be eligible for coverage. Under the Deficit Reduction Act of 2005, however, individuals who apply for Medicaid and claim to be U.S. citizens are required to provide certain documents (such as a birth certificate or passport) to prove their citizenship. Following implementation of the new requirements, some states have reported a drop in enrollment, which appears to be concentrated among U.S. citizens who were unable to provide the requisite documentation. CBO has estimated that if states could, instead, verify an individual’s name and Social Security number with the Social Security Administration, enrollment in Medicaid would initially increase by 500,000 and then grow by an additional 200,000 in subsequent years.¹⁵

Residents of U.S. Territories

The unique nature of the laws or economies of the five U.S. territories—Puerto Rico, American Samoa, Guam, the Commonwealth of the Northern Mariana Islands, and the U.S. Virgin Islands—raises several issues under proposals to expand health insurance coverage. Although residents of the U.S. territories are generally not included in counts of the insured and uninsured, the manner in which proposals treat them would have implications for federal spending and overall coverage rates.

One important distinction is that the roughly 4 million residents of the five territories generally do not pay federal income taxes, even though they are U.S. citizens. Instead, each territory has its own individual income tax. Therefore, proposals that sought to encourage participation in a health plan by imposing a federal tax penalty for noncompliance with a mandate or by offering a federal tax credit to offset the costs of insurance premiums would have an impact on territorial residents only to the extent that the territory’s tax code "mirrored" the federal tax code. The U.S. territories with such mirrored provisions in their tax codes are Guam, the Commonwealth of the Northern Mariana Islands, and the U.S. Virgin Islands, which, in total, represent about 10 percent of the population in the five territories.

Individuals in the U.S. territories also are less likely to be insured by their employer than the population in the 50 states. In addition, individuals living in the territories face challenges in accessing health care services that are similar to those faced by residents of rural areas within the United States, including limited access to health care facilities and a shortage of primary care practitioners and other health care providers. Proposals that provided a subsidy for health insurance might therefore have a limited impact in the territories if the infrastructure and staffing of health care services remained at current levels.

Proposals that modified or expanded the Medicaid and SCHIP programs also would have a limited impact on the U.S. territories unless the proposals were specifically structured to apply to the territories. Although the territories participate in those programs, they operate them under rules different from those that are applied to the 50 states and the District of Columbia. For example, territories are not required to cover the same eligibility groups, and they are permitted to use different financial standards (regarding income and assets) in determining eligibility. The U.S. territories are also subject to overall spending caps for both programs and must provide matching funds up to the cap and pay all costs above the cap.

Individuals in Households Headed by an Unmarried Couple

In the United States, approximately 5.5 million households are headed by unmarried couples (including both opposite-sex and same-sex couples). More than 2.2 million of those households have children less than 18 years of age living with them. Many unmarried couples purchase multiple health insurance policies to cover household members, possibly in response to current tax laws. Under current law, an individual whose employer provides benefits for domestic partners generally would not be entitled to the same tax exclusion as his or her married coworker. Thus, any premiums paid by an employer on behalf of a domestic partner would be subject to income and payroll taxes.¹⁶ Those considerations could affect the design of a proposal to expand health insurance coverage.

Homeless People

People who are homeless are more likely to suffer from mental illness, substance abuse, and other illnesses and disabilities. The homeless population has been estimated at approximately 670,000 on a given night and 1.6 million over the course of a year (in 2007). Most homeless people are uninsured and face financial and other barriers to obtaining health care services.¹⁷ As many as 30 percent of the homeless receive Medicaid; although the remainder are poor, many do not qualify because they are not aged, disabled, or the caregivers of minor children. Others may be eligible but do not enroll in Medicaid—in some cases, because they are unaware of the program, lack documentation to confirm eligibility, or submit incomplete applications (some applications by the homeless may leave the address or phone number blank, even though Medicaid agencies allow applicants to designate a third party as a mailing address).

Homeless people who are not enrolled in Medicaid generally obtain care from "safety net" providers, such as federally funded health centers, local health departments, and public hospitals. The Health Care for the Homeless program, which is part of the federal health centers program, is specifically designed to deliver primary care and other services to the homeless. In 2007, 200 organizations received federal grants under the Health Care for the Homeless program and served approximately 700,000 patients.¹⁸ No national estimates are available on the number of homeless people who use other types of health care providers or the services they receive.

Proposals that extended health insurance coverage to the homeless population could increase the funding available to safety net providers (through the greater insurance payments they would receive) and could also give homeless people the means to enroll in a private plan. Enrolling homeless people in an insurance plan would present administrative challenges, however, because many of them would probably lack the necessary documentation to verify their citizenship or their eligibility for subsidies.

Individuals Refusing Coverage for Religious Reasons

A segment of the population refuses medical treatment because of their religious faith. For example, the Church of Christ, Scientist, teaches that medical problems should be treated through prayer rather than through traditional medical care. Isolated cases have been reported in the media in which members of other churches, as well as people who are not affiliated with an organized religion, rely on prayer rather than medical treatment or do not consider it appropriate to purchase health insurance. A proposal that sought to achieve universal coverage would need to specify whether people who refuse medical treatment or insurance coverage for religious reasons would be excluded from the system (and whether that exclusion would apply only to adults or whether parents would have the right to exclude their children).

The program’s relatively high administrative costs—roughly $300 per enrollee each year—partly reflect the fact that certain ongoing expenses are spread across a relatively small number of participants.

Federal matching rates under both programs vary across states and cover a higher percentage of medical costs in states with lower income per capita. The minimum matching rate is 50 percent for Medicaid and 65 percent for SCHIP. Different rates may apply for administrative costs.

States’ Medicaid programs may continue to cover and receive federal matching funds for a few classes of drugs that were excluded from the Medicare drug benefit.

As noted in Chapter 1, economists generally agree that cash wages and other forms of compensation would increase as payments for health insurance declined, but it is less clear how that adjustment would apply to individual workers within a firm.

The TRICARE program is available to current military service members, their families, and those who qualify for military retirement by serving at least 20 years in uniform or by receiving a disability retirement. Service members who separate before achieving military retirement do not receive TRICARE coverage but instead can apply for enrollment with the Department of Veterans Affairs.

In 2007, most had some type of non-VA coverage, including Medicare Part A (39 percent), Medicare Part B (37 percent), Medicare Part D (19 percent), medigap (22 percent), Medicaid (8 percent), TRICARE (17 percent), and private insurance (51 percent).

Eligibility for IHS services is limited to individuals who are members of a federally recognized tribe and who live in the service area of an IHS facility. Thus, not all of the estimated 3.3 million American Indians and Alaska natives living in the United States are eligible to receive health care services through the IHS.

All children for whom federal foster care payments are made—approximately 50 percent of children in foster care—are eligible for Medicaid. The remaining 50 percent are generally eligible for Medicaid under guidelines established by state Medicaid programs.

Incarcerated adults and juveniles are not included in most counts of the insured and uninsured. Those counts generally exclude the institutionalized population.

That estimate is based on data from the Census Bureau on local government expenditures and the 2002–2003 State Health Expenditure Report copublished by the Milbank Memorial Fund, the National Association of State Budget Officers, and the Reforming States Group in June 2005.

Karina Fortuny, Randy Capps, and Jeffrey S. Passel, The Characteristics of Unauthorized Immigrants in California, Los Angeles County, and the United States (Washington, D.C.: Urban Institute, March 2007); and Jeffrey S. Passel, Unauthorized Migrants: Numbers and Characteristics (Washington, D.C.: Pew Hispanic Center, June 14, 2005).

Refugees and individuals who have been granted asylum are eligible to receive most federal benefits for the first seven years they are in the United States.

Unauthorized immigrants may receive care through Medicaid under the following circumstances: if they meet certain income requirements and are pregnant; if they are under the age of 19 or at least 65 years old; if they are disabled; or if they are the caregiver of a child under the age of 18. However, such "emergency" Medicaid coverage pays for only those services that are necessary to stabilize a patient.

Section 1011 of the Medicare Modernization Act made $1 billion available to states between fiscal years 2005 and 2008 for the health care costs of their unauthorized immigrants.

Congressional Budget Office, letter to the Honorable Nancy Pelosi regarding the budgetary impact of section 211 of H.R. 3963 (October 25, 2007).

If the worker could claim his or her domestic partner as a dependent, then the tax exclusion might apply. However, an individual must have low income and meet other support and residency tests to be claimed as a dependent.

Department of Housing and Urban Development, Office of Community Planning and Development, The 2007 Annual Homeless Assessment Report (July 2008); and Patricia A. Post, Casualties of Complexity: Why Eligible Homeless People Are Not Enrolled in Medicaid (Nashville, Tenn.: National Health Care for the Homeless Council, May 2001).

Department of Health and Human Services, Health Resources and Services Administration, "Health Care for the Homeless Program Marks 20th Anniversary" (July 2007).