Mr. Chairman and Members of the Committee: Thank you for inviting me to appear before you.  I am Joseph Antos, the Wilson H. Taylor Scholar in Health Care and Retirement Policy at the American Enterprise Institute.  I am also adjunct professor in the School of Public Health at the University of North Carolina at Chapel Hill.  I have previously served as the assistant director for health and human resources at the Congressional Budget Office, and earlier held several research and management positions in the Health Care Financing Administration, the precursor to the Centers for Medicare and Medicaid Services (CMS).  The views I present today are my own, and do not represent the position of the institutions with which I am associated.

 The Medicare program faces unprecedented challenges.  Medicare’s benefit structure is outmoded and inadequate, failing to cover outpatient prescription drugs or to provide adequate financial protection for millions of enrollees.  Physicians, hospitals, and other providers have developed new and better ways to diagnose, treat, and cure diseases—causing Medicare spending per enrollee to grow rapidly.  The impending retirement of 78 million baby-boomers, beginning in just 8 years, will rapidly escalate demands on Medicare’s finances.

 These challenges cannot be avoided.  Steps must be taken to improve and strengthen Medicare so that it can meet the changing needs of seniors and the disabled, now and in the future.  There is a model for such a reform in the Federal Employees Health Benefits Program (FEHBP), which has served the health insurance needs of federal workers, their dependents, and retirees for over 40 years.

 Any proposal to modify a program as popular as Medicare raises concerns among policy experts and the general public.  Much controversy has developed over the role private competition should play in reforming Medicare.  Although the current and future problems of the program are widely acknowledged, there is concern that the cure might be worse than the disease.  The following arguments have been raised against reform proposals that would enhance the role of private plans in Medicare:

* Private plans would be less successful than Medicare in controlling spending.
* Private plans would limit access to health care providers.
* Private plans would systematically exclude older, sicker Medicare beneficiaries.
* Private plans are unreliable and are likely to drop out of the program with little warning.

 My testimony today will address these serious charges.  It is essential that we craft a reform that assures a better Medicare program that is more responsive to the needs of beneficiaries, and that places the program on a sustainable financial footing for the long term.  I will argue that private competition is an essential element in achieving those goals efficiently and equitably.  But the way in which private competition is introduced matters greatly.  A poorly designed reform proposal could make matters worse.

Does Medicare Control Costs Better than the Private Sector?

A recent analysis by Boccuti and Moon [1] argued that Medicare has been more successful than the private sector in constraining spending growth over the long term.  They ascribe this success in part to the system of government price setting and regulation that has been the hallmark of the Medicare program.  But a closer look at the evidence suggests that the private sector has performed as well or better than Medicare in constraining cost growth.

 Between 1970 and 1999, private insurance spending grew 18 percent faster than Medicare spending.  That figure accounts for differences in enrollment patterns and is restricted to comparable benefits, in this case hospital and physician services only (see the Appendix for details of the calculation).

 But private insurance became more generous over that time period, covering a growing proportion of the total cost of health services.  In 1970, private insurance paid for about 60 percent of the total private cost of hospital and physician services.  By 1999, that had grown to 85 percent.  This expansion in generosity is not the result of adding new types of benefits (such as a prescription drug benefit), since the analysis is confined to hospital and physician services.  Instead, it reflects a shift away from out-of-pocket payments toward more first-dollar private insurance coverage—a shift promoted by federal tax preferences for employment-based insurance.

 Accounting for the increasing generosity of private coverage appears to reverse the Boccuti/Moon conclusion.  Although private insurance spending has risen faster than Medicare spending over the past thirty years, the unit cost of private coverage grew more slowly. [2]  That suggests that Medicare does not have an advantage over the private sector in limiting the growth of health care spending.

 This is like comparing two boxes of cereal in the supermarket.  The large box is more expensive than the regular size.  In other words, if you buy more cereal, it costs more.  But the cost per ounce of cereal is lower with the large box.

 We should be careful not to make too much of these calculations.  It is difficult to make all the adjustments that would be necessary for a completely accurate comparison.  In particular, adjustments should be made to reflect any changes in Medicare’s generosity, parallel to the estimate for private insurance.  That is not possible with the National Health Accounts data used here and in the earlier study.   However, any error is not likely to be great since Medicare’s benefit package has remained largely unchanged for decades.

 Other comparisons confirm that Medicare does not have an advantage over the private sector in controlling spending.  The Federal Employees Health Benefits Program and the California Public Employees’ Retirement System (CalPERS) are large public programs that purchase health insurance on the private market.  Over the past decade, FEHBP’s average spending growth was slightly higher than Medicare’s while CalPERS’ spending growth was somewhat lower. [3]

 Two glaring differences between Medicare and the other public programs should be noted.  First, Medicare does not have a prescription drug benefit.  Consequently, it did not have to deal with rapid spending increases in drug spending that occurred over the last few years.  Second, Medicare has experienced serious disruptions in the delivery of care not shared by the other programs.  Flawed government pricing formulas have helped drive HMOs from the Medicare program and threatened access to physician services in some markets.

 Recent experience with physician payment reveals the difficulties of government price controls.  (HMO payment issues are discussed in a later section.)  In 2002, Medicare reduced physician fees by 5.4 percent following the congressionally-mandated “sustainable growth rate” formula.  This unprecedented across-the-board reduction was intended to recoup excess spending for physician services that occurred in earlier years.

 Reports from some parts of the country indicated that certain physicians were closing their practices to new Medicare beneficiaries.  This did not show up as a decline in the overall participation rate of physicians, but it nonetheless was a disruption of service to some patients. 

 Ironically, the rate cut did not save money.  Although some physicians were paring back their Medicare business, others reacted to the fee cut by increasing the volume of services paid by the program.  Medicare physician payments increased by nearly $3 billion even though the average fee for each service was reduced.  That was the result of an abnormally large increase in service volume—a 7.9 percent increase in 2002 compared to 3.5 percent increases in 2000 and 2001, when fees increased.[4]  It would be difficult to argue that such a sharp increase in volume last year was justified solely on clinical grounds.

 This illustrates a problem common to price setting methods in Medicare’s fee-for-service system.  If the administered price is reduced below the market price, providers find ways to make part of the reduction up by expanding services.  Tighter controls that also restrict the use of services could prevent that, but such restrictions would have adverse consequences for the health of beneficiaries.

Would Medicare Beneficiaries Have to Change Doctors?

The fear of having to find a new doctor because of changes in Medicare policy is palpable.  Medicare beneficiaries often develop close ties with their physicians, and there can be adverse health consequences if a patient shifts to another physician during the course of treatment.  The traditional Medicare fee-for-service program places no direct limits on which providers a patient may see.  This offers flexibility and an important sense of security to many beneficiaries, but it comes at a significant cost.

 The structure of traditional Medicare reflects the insurance practices that existed in 1965 at the inception of the program.  Medicare does not cover important services, including outpatient prescription drugs and many preventive services.  It imposes a confusing, inequitable, and inefficient cost-sharing structure on beneficiaries.  And the program fails to provide financial protection for those with the highest health costs.

 Most beneficiaries supplement their Medicare benefits to overcome many of those gaps.  Such additional coverage is available through private Medigap insurance, employer-sponsored retiree plans, Medicaid, and Medicare managed care plans.  About 91 percent of beneficiaries have supplemental coverage. [5]

 The cost to beneficiaries of the traditional Medicare program depends to a large extent on whether they are eligible for or can afford to purchase supplemental coverage.  The 4 million people who qualify for Medicaid benefits have relatively low out-of-pocket spending, roughly $750 in 2000. [6]  About 10 million people buy Medigap policies, spending on average about $3,600 out-of-pocket in 2000.  Supplemental insurance premiums accounted for about $1,700 of that total.

 The conclusion is clear: Although traditional Medicare allows a wide choice of providers, nearly everyone is saying they want something different than what the government allows.  The flexibility of the traditional program comes at an additional cost of thousands of dollars to millions of beneficiaries.

 A reform modeled after FEHBP would offer beneficiaries a broad set of options, not just plans that restrict access to a specific panel of doctors.  Health plans that rely on closed panels of providers and intrusive management of care have given way to more flexible plans that are the norm in the private market today.  About 70 percent of workers with insurance from an employer are in preferred provider organizations (PPOs) or point-of-service plans (POSs), which allow beneficiaries to go to the provider of their choice. [7]  Such options, as well as HMOs and the traditional fee-for-service plan, should be made available to Medicare beneficiaries.  A beneficiary who wants to remain in the traditional program should be allowed to do so.

 Should we give beneficiaries in traditional fee-for-service a prescription drug benefit that is similar to the benefit that would be offered by competing plans?  Pharmaceuticals have become one of the essential tools of modern health care.  If we did not add a drug benefit to traditional Medicare, we would not be able to take advantage of new therapeutic approaches and methods for controlling cost.  Disease management programs, which manage the care of high-cost patients and promise both cost savings and better health outcomes, cannot function unless prescription drugs are part of the benefit.  But health plans should be allowed to structure their benefits to foster efficiency, and they should be able to pass savings back to their enrollees (perhaps through lower premiums).  A one-size-fits-all approach will not work.

 A concern also has been raised that PPOs typically charge more when patients go out of network for treatment.  The Blue Cross Standard Option in FEHBP, for example, requires an enrollee to pay $15 for a standard office visit to an in-network doctor or 25 percent of the charge for a doctor who is out of the network.  Such differences in beneficiary cost sharing can amount to substantial sums for someone who chooses to receive most of her care out of the plan’s network.  That places a responsibility on beneficiaries to make their plan choices carefully, assuring that they can stay within the network for most services and keep their out-of-pocket spending low.

 Beneficiaries need good information to make sound choices.  The Office of Personnel Management, which operates FEHBP, provides comparative information on health plans (including lists of preferred providers).  Private groups also publish information to guide beneficiaries in selecting a plan. [8]  A similar information strategy for a competitive Medicare reform would assure beneficiaries that the plan they chose included their personal physicians.

Would Private Plans Avoid Sicker Beneficiaries?

A market-based reform of Medicare would offer beneficiaries a choice of several competing health plans.  Under premium support models, premiums would be partly subsidized.  Beneficiaries would face higher premiums for more expensive plan options.  Plans would have an incentive to keep premium costs down to remain competitive.  They might accomplish that through improved efficiency.  But they might also keep cost down by skimping on care and enrolling people with below-average health costs (the “good risks”). [9]  This is known as risk selection.

 Medicare has dealt with risk selection primarily by improving the accuracy of payments to health plans.  Methods have been developed to adjust federal payments to health plans based on the likely cost of providing services to enrollees.  If plans received larger compensation for sicker people (whose expected health costs are above average), the plans would have a greater incentive to enroll them.

 Medicare’s current risk adjusters are based mostly on demographic factors (including the age, sex, and Medicaid status of the enrollee), with only limited information on the actual use of services.  New methods have been developed that should be more effective in predicting differences in health spending among beneficiaries.  CMS recently announced that it would begin to phase in a “selected significant condition” model in January.  That new model is expected to make more accurate payments for the sickest patients, rewarding plans that can generate real cost savings through disease management.  That could become an important tool under a more competitive Medicare program.

 Risk selection is probably not a major problem for beneficiaries in the current Medicare program since most are enrolled in traditional fee-for-service.  But how much of a problem would risk selection pose under a market-based reform?  Evidence from FEHBP suggests that the problem could be negligible, depending on the design of Medicare’s reform.

 FEHBP provides a generous premium subsidy to enrollees, equal to 75 percent of the health plan’s premium subject to a dollar cap.  Payment risk adjusters are not used, and employees can freely choose among plans in their area during the annual open enrollment season.  One might expect to find substantial risk selection across plans under those circumstances, with higher-cost plans attracting an older and sicker group of enrollees.  But a recent study by Florence and Thorpe [10] found very small differences in the average age of enrollees in low- and high-cost plans because FEHBP pays such a large portion of the premium.

 The potential for risk selection in FEHBP (and other choice-based systems) is also limited by the inertia most people exhibit when given the opportunity to change health plans.  Francis [11] found that many FEHBP enrollees who could save money by changing plans during open season fail to do so.  Inertia in personal decision making helps to stabilize plan enrollment, reducing the chances that more generous plans would be driven out of the market by the cost of attracting sicker enrollees.  But that also implies that beneficiaries would pay more than they would have if they were willing to change plans.

 If most people don’t change plans, how is a choice system supposed to work?  Indeed, how do markets work if most consumers remain loyal to their favorite brands?  A competitive market will generate pressures on plans to lower costs or improve their product so long as some consumers are prepared to shift to a competitor.  The credible threat of a loss of market share—or the prospect of gaining market share—is sufficient incentive in a competitive market.

 These arguments do not imply that we can be complacent about risk selection under a reformed Medicare program.  Health markets are far from the competitive ideal, and the consequences of a poorly functioning choice system could be quite serious for an older population.

 Beneficiaries and their families will need reliable user-friendly information about plan options.  Techniques including improved payment risk adjusters will help, particularly in a program that provides large premium subsidies.  The Medicare agency must be vigilant in its oversight, ready to take corrective action when necessary.  But prudent oversight does not have to mean a large increase in administrative costs.  The Office of Personnel Management (OPM) successfully manages FEHBP with a handful of staff and minimal regulatory requirements.

Would Private Plans Stay in the Program?

Recent experience with private health plans in Medicare+Choice has been sobering.  In the mid- to late-1990s, Medicare HMOs were highly competitive with the traditional fee-for-service program, offering better benefits at a reasonable price.  Congress created Medicare+Choice in 1997, intending to expand the number of private plan choices available to beneficiaries.  Participation in Medicare+Choice peaked in 1998 at 346 plans, and that number has dropped ever since.  This year only 153 contractors remain in the program.

 Millions of Medicare beneficiaries have had to change health plans over the past five years as the Medicare+Choice program shrank.  Although most beneficiaries affected by plan dropouts could turn to other managed care plans, and all beneficiaries could enroll in the traditional fee-for-service program, this kind of disruption could have adverse health consequences for some people.  These developments have been taken by some observers as evidence that competition cannot work in Medicare.

 That’s half right.  Plans will not be able to compete in Medicare unless we also change the government’s approach to managing the program.  Medicare+Choice did not break away from Medicare’s history of top-down price setting and complex regulations.  Administrative inflexibility, unrealistically low payments, and an inability to adjust to changing market conditions were key factors leading to the decline of Medicare+Choice.

 In broad terms, Medicare’s experience with managed care parallels that of the private sector.  After initial enthusiasm that HMOs would be able to control health costs, consumers and providers began to reject the constraints imposed by such plans.  The goal of cost containment became less important in a booming economy and a tight labor market.  Consumers shifted to less restrictive plan options, including PPOs and POSs.  Providers demanded and got larger fee increases, driving premiums up.

 Those changes were led by consumers, and the market responded by offering new options.  Traditional managed care plans lost market share relative to the new plans, which offered a more attractive package of benefits, premiums, and cost containment practices.

 In contrast, Medicare+Choice was locked into a set of requirements by the 1997 legislation and did not have the authority (or experience) to make changes necessary under the new circumstances.  For example, most Medicare+Choice plans could expect only a 2 percent annual increase in federal payments even though their cost of providing care was growing at 10 percent a year or more.  It is not surprising that plans left the program.

 Medicare cannot be immunized from the pressures of the wider health care market.  It can, however, become a smarter purchaser of health care, and it can become more responsive to the demands of its beneficiaries.  Medicare should not be afraid to drive a hard bargain with health plans and expect them to provide excellent care.  But Medicare must also be allowed to respond to changing circumstances before problems get out of hand that can result, as we have seen, in the exodus of plans from the program.

 Elements of a successful Medicare choice program are contained in FEHBP.  Rather than setting into law detailed pricing formulas and other requirements that often do not stand the test of time, OPM has broad discretion to negotiate rates and conditions of plan participation.  It has the ability to accept new health plan options (other than new fee-for-service carriers) without new legislative authority, but it remains accountable to Congress.  Such principles should be incorporated into a market-based reform of Medicare.

Conclusion

Congress has an opportunity to help Medicare fulfill its promise to millions of seniors and people with disabilities.  We can build on Medicare’s successes, but we need not repeat its mistakes.  The program must expand to cover prescription drugs, preventive benefits, and protection against uncapped medical costs.  The program must be made financially sustainable if the taxpayers of today are to receive their Medicare benefits tomorrow.

 The Federal Employees’ Health Benefits Program is a good place to start.  A Medicare reform modeled after FEHBP would provide both the incentive and the opportunity for seniors to choose health plans that best meet their needs.  Beneficiaries would be able to select from competing plans, including a modernized fee-for-service Medicare that offers a sensible set of benefits.  Such a reform would also create incentives for health care providers to produce high quality care at lower cost.  Medicare would remain a government-run program under an FEHBP approach, assuring appropriate oversight and protection for the most vulnerable.

 There are certainly risks in attempting to reform a program as important as Medicare.  But there are also risks from failing to take the prudent actions necessary to make Medicare an effective and sustainable program.  Adapting the FEHBP model to the specific circumstances facing the Medicare population can be a successful strategy for using effective competition to make lasting and meaningful improvements in an essential program.
 
 Appendix

Data from the National Health Accounts confirm that private insurance spending grew more rapidly than Medicare spending over the past three decades, even when differences in enrollment and differences in the types of benefits covered by the different insurance programs are taken into account.  Between 1970 and 1999, spending by private insurance for hospital and physician services grew 18.1 percent faster than comparable Medicare spending, both measured on a per-enrollee basis (see Chart 1).

 But private insurance became more generous over that time period, covering a growing proportion of the total cost of health services (see Chart 2).  In 1970, private insurance paid for about 60 percent of the total private cost of hospital and physician services.  By 1999, that had grown to 85 percent.  This expansion in generosity is not the result of adding new types of benefits (such as a prescription drug benefit), since the analysis is confined to hospital and physician services.  Instead, it reflects a shift away from out-of-pocket payments toward more first-dollar insurance coverage in the employer market.
 
 Accounting for the increasing generosity of private coverage, the unit cost of private coverage grew more slowly than Medicare (see Chart 3).  The “unit” in this case refers to the percent of spending covered by insurance.  An insurance policy that is more generous covers more of a person’s total health spending, and thus has more “units” than a less generous policy.

 We were unable to adjust Medicare spending for possible increases in that program’s generosity because of data limitations of the National Health Accounts.  However, we know that Medicare benefits did not appreciably increase over the three decades.  Some preventive benefits were added, for example, but all the major health services have been covered by Medicare since its inception.

 In an attempt to verify this observation, we examined other sources of information.  Data on spending for all health services is available from the National Health Care Expenditures Study (NHCES) for 1977 and the Medical Expenditure Panel Survey (MEPS) for 1996.  We calculated the percent of total spending paid by private insurance for people under age 65, and the corresponding percent of total spending paid by Medicare for people 65 and older.  This estimate is not limited to hospital and physician spending, and is consequently not directly comparable to the earlier analyses.

 The value of Medicare has not kept pace with private insurance (see Chart 4).  For persons under age 65, the generosity of private health insurance grew by 41.5 percent between 1977 and 1996.  For persons 65 and older, the generosity of Medicare grew by only 22.2 percent over the same period.

 Comparisons using National Health Accounts data cannot prove the superiority of one model of cost containment over the other.  Private insurance spending includes spending on behalf of Medicare beneficiaries, many of whom have private retiree policies or private Medigap insurance that supplements their Medicare coverage.  The spending data cannot account for differences in the age, health status, or other characteristics of the beneficiary population, which clearly affect the use of health services.  The direction of any bias caused by inadequate data cannot be determined.

Source: National Health Accounts Note: Private health insurance data not adjusted for the increasing generosity of private health insurance during 1970-99

Source: National Health Accounts

Source: National Health Accounts Note: Private health insurance data adjusted for the increasing generosity of private insurance during 1970-99; this adjustment was performed by accounting for displacement of private out-of-pocket spending by insurance

Sources: National Health Care Expenditure Survey (1977), Medical Expenditure Panel Survey (1996)

[1] Cristina Boccuti and Marilyn Moon, "Comparing Medicare and Private Insurers: Growth Rates in Spending Over Three Decades," Health Affairs (March/April 2003), pp. 230-237.

[2] "Unit" refers to the percent of spending covered by insurance. An insurance policy that is more generous covers more of a person’s total health spending, and thus has more "units" than a less generous policy.

[3] Medicare Payment Advisory Commission, Report to Congress: Medicare Payment Policy, March 2003, pp. 11-12.

[4] The increase in the volume of services are estimated as the increase in total allowed charges less the fee update and the increase in fee-for-service enrollment. Unpublished data are available from the Office of the Actuary, CMS.

[5] Medicare Payment Advisory Commission, Report to Congress: Assessing Medicare Benefits, June 2002, p. 17.

[6] Glenn M. Hackbarth, Testimony before the Subcommittee on Health, Committee on Ways and Means, U.S. House of Representatives, May 1, 2003.

[7] The Kaiser Family Foundation and Health Research and Educational Trust, Employer Health Benefits: 2002 Annual Survey (Menlo Park and Chicago: KFF and HRET, 2002), Exhibit 5.1, p. 69.

[8] See, for example, Checkbook’s Guide to Health Plans for Federal Employees, produced annually by the Center for the Study of Services, Washington, DC.

[9] The term "risk" is often misused in discussions of health insurance. Risk denotes uncertainty. But insurers usually consider people who have a high expected cost of care to be "poor risks." For example, people with AIDS are virtually certain to incur high health care costs, but they are called poor risks. The text applies the term as it is commonly (mis)understood.

[10] Curtis S. Florence and Kenneth E. Thorpe, "How Does The Employer Contribution For The Federal Employees Health Benefits Program Influence Plan Selection?," Health Affairs (March/April 2003), pp. 211-218.

[11] Walton Francis, Testimony before the Subcommittee on the Civil Service, Census and Agency Reorganization, Committee on Government Reform, U.S. House of Representatives, December 11, 2002.

Joseph Antos is the Wilson H. Taylor Scholar in Health Care and Retirement Policy at AEI.