Note: This option would take effect in January 2015.
a. If both policies were enacted together, the total effects would be greater than the sum of the effects for each policy because of interactions between the approaches.
For people who have health insurance, including Medicare and other types of coverage, payments for health care fall into two broad categories: premiums and cost sharing. A premium is a fixed, recurring amount paid in advance for an insurance policy (which then limits enrollees’ financial risk by covering some or all of the costs they incur if they use health care services or goods). Cost sharing refers to out-of-pocket payments that enrollees are required to make when they receive health care. In general, premiums spread the cost of medical care across all enrollees, whereas cost sharing concentrates costs on people who use more medical care. To determine the cost-sharing obligations of their enrollees, insurance plans typically vary three basic parameters:
Deductibles and catastrophic caps typically apply on an annual basis. The portion of the cost borne by the enrollee is usually specified as a percentage of the total cost of an item or service (in which case it is referred to as coinsurance) or as a fixed dollar amount for each item or service (in which case it is referred to as a copayment). If other aspects of an insurance plan are the same, lower cost-sharing requirements translate into higher premiums—because insurers must charge more to cover their higher share of medical spending—and higher cost-sharing requirements translate into lower premiums.
Research has shown that people who are not subject to cost sharing use more medical care than do people who are required to pay some or all of the costs of their care out of pocket. The RAND health insurance experiment, which was conducted from 1974 to 1982, examined a nonelderly population and found that health care spending was about 45 percent higher for participants without any cost sharing than for those who effectively faced a high deductible; average spending for people with intermediate levels of cost sharing fell in between those points. A variety of later studies also concluded that higher cost sharing led to lower health care spending—including a 2010 study that found that Medicare beneficiaries responded to increases in their cost sharing by reducing visits to physicians and use of prescription drugs to a degree roughly consistent with the results of the RAND experiment.
Those findings have driven interest in using additional cost sharing as a tool to restrain the growth of health care spending. However, increases in cost sharing expose people to additional financial risk and may deter some enrollees from obtaining valuable care, including preventive care that could limit the need for more expensive care in the future. In the RAND experiment, cost sharing reduced the use of effective care and less effective care (as defined by a team of physicians) by roughly equal amounts. Although the RAND study found that cost sharing had no effect on health in general, among the poorest and sickest participants, those with no cost sharing were healthier by some measures than those who faced some cost sharing. In theory, to address the concern that patients might forgo valuable care, insurance policies could be designed to apply less cost sharing for services that are preventive or unavoidable and more cost sharing for services that are discretionary or that provide limited health benefits. In practice, however, that distinction can be difficult to draw, so trade-offs often occur between providing insurance protection and restraining total spending on health care.
Medicare’s Current Cost Sharing. In the traditional fee-for-service portion of the Medicare program (Parts A and B), the cost sharing that enrollees face varies significantly depending on the type of service provided. Under Part A, which primarily covers the services of hospitals and other facilities, enrollees are liable for a separate deductible for each “spell of illness” or injury for which they are hospitalized; in 2015, that deductible will be $1,240, the Congressional Budget Office estimates. In addition, enrollees are subject to substantial daily copayments for extended stays in hospitals and skilled nursing facilities. Under Part B, which mainly covers outpatient services (such as visits to a doctor), enrollees face an annual deductible that is projected to be $142 in 2015. Once their spending on Part B services has reached that deductible amount, enrollees generally pay 20 percent of allowable costs for most Part B services, although cost sharing is higher for some outpatient hospital care. Certain services that Medicare covers—such as preventive care, hospice services, home health visits, and laboratory tests—require no cost sharing. Because of those variations, enrollees lack consistent incentives to weigh relative costs when choosing among options for their treatment. Moreover, if Medicare patients incur extremely high medical costs, they may be obligated to pay significant amounts because the program does not have a catastrophic cap on cost sharing.
Medicare’s cost sharing differs in two significant ways from that of private plans, which provide health insurance for the majority of people under age 65. First, most private health insurance plans have a single, annual deductible that includes all or most medical costs, rather than the separate deductibles for hospital and outpatient services in fee-for-service Medicare. Second, unlike fee-for-service Medicare, most private health insurance plans include a catastrophic cap on out-of-pocket costs that limits enrollees’ annual spending. Because of those differences, fee-for-service Medicare’s benefit design is more complicated and provides less protection from financial risk than many private insurance plans do. Medicare is not unique, however, in charging different cost sharing for different types of services; many private insurance plans do that as well.
Although proposals to change Medicare’s cost sharing generally focus on the traditional fee-for-service program, roughly a quarter of Medicare enrollees choose private insurance plans (known as Medicare Advantage plans) over the fee-for-service program. Medicare requires Medicare Advantage plans to provide a catastrophic cap on cost sharing but gives insurers some flexibility in structuring other cost-sharing requirements, as long as the overall value of the benefit is at least equal to the benefit that fee-for-service Medicare provides. In general, cost-sharing requirements in Medicare Advantage plans are lower than those in the fee-for-service program and more closely resemble requirements in private insurance plans.
Part D of Medicare, which provides coverage for prescription drugs, is also administered by private insurers, who set each plan’s cost-sharing requirements (subject to certain statutory and regulatory requirements). Once recently enacted changes are fully phased in, the standard Part D benefit will include a deductible, a range of spending over which enrollees face 25 percent coinsurance, and a catastrophic threshold above which enrollees are liable for 5 percent of their drug costs. Beyond those required cost-sharing parameters, Part D insurers have some ability to specify which drugs they cover and what cost sharing enrollees must pay, requiring more cost sharing for expensive, higher-tier brand-name drugs and less cost sharing for lower-tier generic drugs. Because private insurers administering Medicare Advantage and Part D plans have the freedom to specify cost-sharing requirements (within limits) and Medicare enrollees can choose between plans on the basis of cost sharing and other factors, proposals to redesign Medicare’s cost sharing generally do not focus on those parts of the program. Consequently, policies that would affect cost sharing in Medicare Advantage or Part D are not included in this discussion.
Supplemental Insurance for Medicare Enrollees. About 85 percent of people who enroll in fee-for-service Medicare have some form of supplemental insurance coverage that reduces or eliminates their cost-sharing obligations and protects them from high medical costs. (Such supplemental coverage of cost sharing is uncommon outside fee-for-service Medicare and thus is another difference between that program and typical private insurance.) About 15 percent of enrollees in fee-for-service Medicare receive coverage of Medicare’s cost sharing from Medicaid, which is available to Medicare enrollees with low income and assets. About 40 percent of fee-for-service enrollees have supplemental coverage through a current or former employer, which tends to reduce, though not eliminate, their cost-sharing liabilities. About 25 percent of enrollees buy medigap policies—individual insurance policies designed to cover most or all of Medicare’s cost-sharing requirements—and 5 percent of enrollees have various other forms of supplemental coverage.
Federal law requires that medigap plans conform to one of 10 standard plan types. (There are also numerous discontinued plan types; plans of those types may keep their existing enrollees but cannot enroll new members.) The current plan types vary in the extent to which they cover Medicare’s cost sharing, and one type offers only catastrophic coverage (which covers cost sharing only after a deductible of $2,110 has been reached). Even so, 60 percent of people with medigap insurance chose plans that offer “first-dollar” coverage—which pays for all deductibles, copayments, and coinsurance—and most other medigap enrollees chose plans that provide first-dollar coverage for Part A and cover all cost sharing above the deductible for Part B.
According to a recent study done for the Medicare Payment Advisory Commission, Medicare spends 33 percent more per person on enrollees who have medigap coverage, and 17 percent more per person on enrollees who have supplemental coverage from a former employer, than it does on enrollees without supplemental coverage. Those estimates are largely consistent with the results of older studies of the relationship between supplemental coverage and Medicare spending, and they take into account various ways in which medigap policyholders and other Medicare enrollees may differ. The study also concluded that those differences in spending were mainly attributable to higher use of discretionary or preventive services by people with supplemental coverage, particularly those with first-dollar coverage. Another recent study concluded that spending by Medicare enrollees with supplemental coverage was growing at a faster rate than spending by enrollees without supplemental coverage. Neither of those recent studies investigated the effects of supplemental coverage on enrollees’ health.
Raw differences in spending between groups with and without supplemental coverage partly reflect differences in their health status, but studies have generally found that the differences in spending were still large after researchers attempted to account for enrollees’ health status. Even so, people who have medigap policies may differ from other Medicare enrollees in other ways because medigap coverage is not assigned randomly, as it might be in a scientific experiment or trial. The 2010 study of how Medicare beneficiaries respond to increases in their cost sharing makes an important contribution because it more closely resembles such an experiment. That study also found that about 20 percent of the gross savings generated by higher cost sharing for physician visits and prescription drugs—stemming from reduced use of those services—was offset by increases in hospital spending, perhaps because people delayed treatment until their condition worsened.
Collectively, those studies provide considerable evidence that Medicare enrollees who are subject to less cost sharing—because of more generous supplemental insurance—use more medical services than other enrollees do. Enrollees with supplemental coverage are liable for only a portion of the costs of any additional services they use (through any remaining cost sharing and through the effect on their premiums for supplemental coverage); taxpayers (through Medicare) bear most of the cost for the additional services.
Policymakers could alter Medicare’s cost sharing and restrict medigap coverage in various ways to produce savings for the federal government, reduce total health care spending, and create greater uniformity in cost sharing for Medicare enrollees. Those different ways would also alter how health care costs were distributed between healthier and less healthy enrollees.
In particular, four main sets of rules governing Medicare’s cost sharing could be modified: deductibles could be increased, decreased, or combined; coinsurance rates and copayments could be changed; a catastrophic cap could be added; and limits could be imposed on supplemental insurance coverage of Medicare’s cost-sharing obligations. Such changes would interact in important ways: for example, higher deductibles or coinsurance rates would cause enrollees to reach a given catastrophic cap more quickly (and at a lower level of total spending), and limits on supplemental insurance would expose more enrollees to changes in Medicare’s cost-sharing rules and thus increase the impact of those changes on Medicare spending. Policymakers could also “grandfather” current enrollees by maintaining existing rules for them and applying changes only to new enrollees.
Deductibles. In general, raising the Part A and Part B deductibles would generate savings for the federal government in two ways. First, higher deductibles would increase the initial cost borne by enrollees, leading to a corresponding reduction in the cost borne by the government. Second, some enrollees would choose to forgo some care because of its higher cost, decreasing the amount of health care for which the federal government pays. The Part A and Part B deductibles could be increased separately, or they could be combined into a single yearly deductible for all services provided by traditional fee-for-service Medicare. Depending on the dollar value of that combined deductible, federal spending would decrease, increase, or remain the same.
Proposals for a combined deductible generally call for setting it between the levels of the current Part A and Part B deductibles. That approach would tend to increase cost sharing for the roughly 70 percent of enrollees who use only outpatient care in a given year and decrease cost sharing for the roughly 20 percent of enrollees who are hospitalized. (About 10 percent of enrollees use no Part A or Part B services in a given year.) In principle, a combined deductible could also encompass drug spending under Part D, but doing that would be complicated because Part D is administered separately by private insurance plans.
Coinsurance and Copayments. Raising coinsurance rates and copayments would reduce federal spending in the same manner as higher deductibles, shifting some costs from the federal government to Medicare enrollees and causing enrollees to forgo some care because of their higher out-of-pocket costs. Applying higher coinsurance or copayments to types of care that patients are likely to forgo at higher prices, such as elective surgery, would tend to emphasize that effect, decreasing the amount of care provided with little increase in patients’ costs. Conversely, applying higher cost sharing to types of care for which patients are particularly insensitive to price, such as emergency surgery, would tend to increase costs for enrollees with little effect on the amount of care provided. Some proposals envision making wide-ranging changes to Medicare’s cost-sharing rules, whereas other proposals would introduce coinsurance or copayments for specific services that do not currently require cost sharing, such as home health care, laboratory tests, or the first 20 days of a stay in a skilled nursing facility. In general, copayments can give patients more certainty about their costs for treatment than coinsurance does, but copayments can also insulate patients from differences in the total cost of each service.
Catastrophic Caps. Most private insurance plans include a catastrophic cap that limits how much enrollees have to spend out of pocket, but Parts A and B of Medicare have no catastrophic cap on cost sharing. Thus, in the absence of other changes to Medicare’s cost-sharing rules, establishing a catastrophic cap would increase Medicare spending—by requiring the program to pay the entire cost of care above the cap, and possibly by increasing the amount of care sought by enrollees who exceed the cap because they would no longer face any cost for additional care. Generally, a higher cap would produce a smaller increase in federal spending; past proposals have called for caps of more than $5,000 to limit their impact on federal costs.
For enrollees in fee-for-service Medicare who have supplemental coverage, adding a catastrophic cap to Medicare would reduce the costs paid by their supplemental policies, resulting in lower premiums for those policies but little change in enrollees’ financial risk. For enrollees without supplemental coverage, establishing a cap would reduce financial risk and decrease out-of-pocket costs if enrollees’ spending exceeded the cap. Imposing modest cost sharing above the catastrophic cap (as in Part D) could preserve some incentive for enrollees who exceeded the cap to use medical care judiciously (although supplemental coverage of that additional cost sharing would eliminate that incentive).
Supplemental Coverage of Medicare’s Cost Sharing. About 25 percent of enrollees in fee-for-service Medicare purchase medigap policies, and about 40 percent have retiree coverage through a former employer. By reducing or eliminating enrollees’ cost-sharing obligations, those policies can mute the incentives for prudent use of medical care that cost sharing is designed to generate. Lawmakers could impose three types of restrictions on supplemental coverage of Medicare’s cost-sharing obligations:
Grandfathering. Another design choice for policymakers is whether changes to the rules for cost sharing and supplemental insurance would apply to all Medicare enrollees or only to new enrollees—in other words, whether existing enrollees and medigap policyholders would be grandfathered. One rationale for grandfathering medigap policyholders is that changing the terms of medigap policies that have already been purchased could be considered unfair or unduly burdensome. Medicare enrollees who do not buy medigap insurance when they turn 65 may be charged much higher premiums for such insurance if they wait to purchase it until they develop health problems. Thus, many Medicare enrollees pay medigap premiums for years to ensure that they will have access to the financial protection of supplemental insurance if their health deteriorates. In the near term, however, the effects on Medicare spending would be smaller if current enrollees were exempt from changes to cost sharing or restrictions on medigap plans, and operating multiple sets of rules would add to the program’s administrative complexity.
CBO examined three alternative ways to reduce federal spending on Medicare by modifying the cost sharing that enrollees face. The alternatives would apply to all enrollees, with no grandfathering.
If, like the other options, this combined alternative went into effect on January 1, 2015, and the various thresholds were indexed to the growth of per-enrollee Medicare costs thereafter, federal outlays would be $114 billion lower from 2015 through 2023 than they would be under current law, CBO estimates. (Those savings exceed the sum of the savings from the first two alternatives because medigap enrollees would not be entirely insulated by their supplemental coverage from the cost-sharing changes, as they would be in the first alternative, which would reduce their use of care and their cost to the federal government.)
The budgetary effects of changing Medicare’s cost-sharing rules depend significantly on the specific parameters chosen. To illustrate the impact of varying some of those parameters, CBO estimated the effect on federal spending of modestly changing the deductible and catastrophic cap in the third alternative. Raising the 2015 deductible by $100 (to $650), while keeping the catastrophic cap at $5,500, would increase federal savings between 2015 and 2023 by an estimated $22 billion. Raising the catastrophic cap in 2015 by $500 (to $6,000), while keeping the deductible at $550, would add an estimated $31 billion to federal savings through 2023. Making both of those changes together would yield $53 billion in additional savings from 2015 through 2023, compared with the budgetary effects of the third alternative.
Substantial changes to the cost-sharing structure of fee-for-service Medicare and the coverage provided by medigap plans would not only reduce costs to the federal government but also have an impact on Medicare enrollees, on supplemental insurance, and on the administration of the Medicare program.
Effects on Enrollees. The cost-sharing and medigap changes included in this option would affect total health care spending for Medicare enrollees (by changing the amount of health care services they use) and the way in which that spending is divided between the federal government and enrollees and among enrollees themselves. The restrictions on medigap coverage would also affect how much of enrollees’ cost-sharing obligations medigap plans would cover, as well as the premiums that enrollees would pay for those plans.
Under current law, the average fee-for-service enrollee will cost Medicare $10,250 in 2015 and will be obligated to pay $1,700 in cost sharing, CBO estimates. (Cost-sharing obligations may be paid by the enrollee directly out of pocket, by a supplemental insurer, or by some combination of the two.) Those averages mask substantial variation in individuals’ cost-sharing obligations, stemming from differences in health and the use of medical care. For example, CBO estimates that one-quarter of enrollees will have cost-sharing obligations of more than $1,900 in 2015; their obligations will average about $5,250, compared with an average of about $550 for the other three-quarters of fee-for-service enrollees.
Under the full set of changes included in this option (the third alternative), the average fee-for-service enrollee would cost Medicare $10,100 in 2015, CBO estimates, $150 less than under current law. However, under the specific cost-sharing changes and medigap restrictions in that alternative, enrollees’ average cost-sharing obligations would not change—because the higher fraction of total health care costs that enrollees would pay as cost sharing would be offset, on average, by savings from the resulting reduction in their use of health care services. (Different combinations of deductibles, coinsurance rates, catastrophic caps, and medigap restrictions could increase or decrease the average cost-sharing obligations of enrollees.) Even so, that alternative would alter the distribution of cost-sharing obligations among enrollees. One-quarter of enrollees would face cost-sharing obligations of more than $2,300 in 2015; their obligations would average about $4,550, while the other three-quarters of enrollees would have average obligations of about $750. (Roughly 10 percent of enrollees would reach the option’s $5,500 cap on cost-sharing obligations.) Those changes reflect a relatively large average decrease in obligations for enrollees who have serious illnesses that require extended care or hospitalization and a relatively small average increase in obligations for healthier enrollees who use less care.
The medigap restrictions in this option would increase the average amount of cost sharing that a medigap policyholder paid out of pocket and would decrease, to roughly the same extent, the average amount that a medigap plan paid on an enrollee’s behalf. Because medigap insurers must compete for business and are subject to state insurance regulations, they would most likely reduce premiums to reflect that reduction in their costs. Overall, most medigap policyholders would have lower health care expenses under this option because their medigap premiums would decrease more than their out-of-pocket payments would increase (mainly because most of a medigap plan’s liabilities are generated by a small share of policyholders). However, in any given year, some enrollees would face higher combined costs for medigap premiums and out-of-pocket payments under this option.
Beyond altering how and how much Medicare enrollees pay for care, the changes included in this option could have other effects on enrollees. Those changes would give people stronger incentives to use medical services more prudently. However, as noted above, studies have shown that people who are subject to higher cost sharing reduce their use of both effective and ineffective health care. To avoid reductions in effective care, enrollees’ costs could be selectively reduced or eliminated for high-value services—an approach known as “value-based insurance design.” In practice, defining such services can be challenging, and the use of value-based design in private insurance plans has been limited. Furthermore, restricting medigap coverage would prevent Medicare enrollees from buying policies with the low levels of cost sharing that they have shown a preference for in the past. Although most medigap enrollees would have lower overall health care costs under this option, some enrollees would prefer the financial certainty and simplicity of a medigap plan that covered all of their cost-sharing obligations. Those enrollees would object to any legislation or regulation that denied them access to full supplemental coverage for their cost sharing.
Effects on Supplemental Insurance. Altering Medicare’s cost-sharing structure and limiting supplemental coverage could lead to changes in medigap premiums and in enrollees’ demand for medigap policies. If medigap plans were barred from paying the first $550 of an enrollee’s cost-sharing liabilities and then from fully covering all cost-sharing requirements up to a catastrophic cap—as in the second and third alternatives—the costs borne by medigap plans would decrease; as a result, so would premiums for those plans. On the one hand, lower premiums would make medigap policies more appealing. On the other hand, the restrictions on medigap benefits would reduce the value of such policies to enrollees.
A key reason that people buy medigap coverage today is to be protected against high out-of-pocket costs. Adding a catastrophic cap to Medicare would reduce financial risk for enrollees in the traditional fee-for-service program who lack supplemental coverage. Therefore, adding a catastrophic cap to Medicare and restricting the coverage provided by medigap plans could cause some enrollees to not purchase supplemental insurance—especially healthier enrollees, who might expect to consume less health care, and thus spend less on cost sharing, than sicker enrollees. A decrease in medigap enrollment by relatively healthy people would increase average per-enrollee costs for medigap plans, leading to higher policy premiums (if everything else was equal).
Altering the cost-sharing structure of Medicare, as in the first and third alternatives, would also affect costs for employers that provide supplemental coverage for retirees. A unified deductible would tend to increase costs for employers, but the introduction of a catastrophic cap would decrease their costs, particularly for very expensive enrollees. The net effect on an employer’s costs for retiree coverage would depend on the extent of the coverage and the health of the employer’s retirees. Additionally, the creation of a catastrophic cap in Medicare might cause some employers to scale back or discontinue supplemental coverage for current or future retirees, on the theory that their retirees would be sufficiently protected from financial risk by Medicare alone.
The unified deductible and catastrophic cap in the first and third alternatives would have similar effects on federal spending for Medicaid, which provides supplemental coverage for low-income Medicare enrollees. Those dual-eligible beneficiaries have a relatively high prevalence of expensive chronic conditions. Consequently, the introduction of a catastrophic cap would shift some of the cost for those expensive enrollees from Medicaid to Medicare. At the same time, the unified deductible and uniform coinsurance rate would shift some costs from Medicare to Medicaid.
Whether those effects would, on balance, increase or decrease Medicaid’s spending on cost sharing for dual-eligible beneficiaries is unclear. Medicaid avoids paying some cost sharing for those beneficiaries by paying providers on the basis of its own rates, which in many cases are lower than rates paid by Medicare. Specifically, state Medicaid programs often limit the amount they pay for dual-eligible beneficiaries’ cost sharing to the difference (if any) between what Medicare already paid and what Medicaid would pay for the same service—meaning that Medicaid often pays none or only a portion of the cost-sharing obligation. Consequently, a change in cost-sharing obligations for Medicare would not necessarily result in a corresponding change in cost-sharing payments by Medicaid. In addition, Medicare’s payments to providers for bad debt (unpaid cost-sharing obligations) cover much of the cost-sharing obligations that Medicaid avoids, so a fraction of Medicaid’s obligations is ultimately shifted back to the Medicare budget. For those reasons, CBO believes that the estimates shown here include the full federal budgetary effects of this option. (The estimates do not include the option’s effects on states’ Medicaid outlays, however.)
Administrative Issues. Altering the cost-sharing rules for Medicare and medigap plans would raise myriad administrative issues. Health care providers might experience some confusion about how much to collect from a Medicare enrollee during an office visit because it might be difficult to track whether the enrollee’s cost sharing payments had reached the deductible or exceeded the catastrophic cap. Moreover, administering the new cost-sharing structure would require coordination that currently does not exist among the organizations that review and process Medicare claims, insurers who provide supplemental coverage, and Medicare. In addition, changes to Medicare’s cost-sharing structure could affect the total amount of bad debt from unpaid cost-sharing obligations owed to service providers and the distribution of that debt among different types of providers, who are reimbursed by Medicare for bad debt in different ways. At the same time, lower enrollment in supplemental plans and reduced use of medical care by some enrollees with supplemental coverage would decrease the amount of billing paperwork for some supplemental insurers.