This is the accessible text file for GAO report number GAO-03-650T 
entitled 'Medicare: Observations on Program Sustainability and 
Strategies to Control Spending on Any Proposed Drug Benefit' which was 
released on April 09, 2003.

This text file was formatted by the U.S. General Accounting Office 

(GAO) to be accessible to users with visual impairments, as part of a 

longer term project to improve GAO products’ accessibility. Every 

attempt has been made to maintain the structural and data integrity of 

the original printed product. Accessibility features, such as text 

descriptions of tables, consecutively numbered footnotes placed at the 

end of the file, and the text of agency comment letters, are provided 

but may not exactly duplicate the presentation or format of the printed 

version. The portable document format (PDF) file is an exact electronic 

replica of the printed version. We welcome your feedback. Please E-mail 

your comments regarding the contents or accessibility features of this 

document to


Before the Committee on Ways and Means, House of Representatives:

United States General Accounting Office:


For Release on Delivery Expected at 11:00 a.m.

Wednesday, April 9, 2003:


Observations on Program Sustainability and Strategies to Control 

Spending on Any Proposed Drug Benefit:

Statement of David M. Walker:

Comptroller General of the United States:


GAO Highlights:

Highlights of GAO-03-650T, a testimony before the Committee on Ways and 

Means, House of Representatives 

Why GAO Did This Study:

The House Committee on Ways and Means is holding a hearing on 

modernizing Medicare and integrating prescription drugs into the 


There are growing concerns about gaps in the Medicare program, most 

notably the lack of outpatient prescription drug coverage, which may 

leave Medicare’s most vulnerable beneficiaries with high out-of-pocket 


At the same time, Medicare already faces a huge projected financial 

imbalance that has worsened significantly in the past year. 

This statement discusses the challenges of adding a drug benefit to 

Medicare in the context of the program’s current and projected 

financial condition.  It also examines program design issues to be 

considered with respect to administering any proposed drug benefit.  

Specifically, it discusses how private sector health plans have used 

entities called pharmacy benefit managers (PBM) to control drug benefit 


What GAO Found:

The recent publication of the 2003 Medicare Trustees’ annual report 

reminds us that Medicare in its current condition—without a 

prescription drug benefit—is not sustainable.  At the same time there 

are growing concerns about gaps in the Medicare program, most notably 

the lack of outpatient prescription drug coverage, that may leave 

Medicare’s most vulnerable beneficiaries with high out-of-pocket 


The Hospital Insurance (HI) portion of Medicare faces a huge projected 

financial imbalance that has worsened significantly in the past year.  

Under the Trustees’ 2003 intermediate estimates, the present value of 

HI’s actuarial deficit is $6.2 trillion—a 20 percent increase over the 

prior year.  Beginning in 2013, HI’s program outlays are expected to 

begin to exceed program tax revenues, putting increased pressure on the 

federal budget to raise the resources necessary to meet program costs.  

In addition, Supplementary Medical Insurance is projected to place an 

increasing burden on taxpayers and beneficiaries.


GAO’s long-term budget simulations show that, absent meaningful 

entitlement reforms, demographic trends and rising health care spending 

will drive escalating federal deficits and debt.  Neither slowing the 

growth of discretionary spending nor allowing the 2001 tax reductions 

to sunset will eliminate the imbalance.  While additional economic 

growth will help ease our burden, the potential fiscal gap is too great 

to grow our way out of the problem.

The application of basic health insurance principles to any proposed 

benefit could help moderate the cost for both beneficiaries and 

taxpayers. These include beneficiary protections against the risk of 

catastrophic medical expenses and premium contributions and cost-

sharing arrangements that encourage beneficiaries to be cost conscious.

The private sector’s use of PBMs to control drug expenditures may be 

instructive for Medicare, but the program’s unique role and nature may 

moderate how such entities would be used and the potential efficiency 

gains afforded in attempting to transfer PBM-like strategies to 


To view the full statement, click on the link above.  For more 

information, contact William J. Scanlon at 202 512-7114.

[End of section]

Mr. Chairman and Members of the Committee:

I am pleased to be here today as you discuss issues related to an 

outpatient prescription drug benefit for Medicare beneficiaries. There 

are growing concerns about gaps in the Medicare program, most notably 

the lack of outpatient prescription drug coverage, which may leave 

Medicare’s most vulnerable beneficiaries with high out-of-pocket costs. 

Recent estimates suggest that, at any point in time, about a third of 

Medicare beneficiaries lack prescription drug coverage. The rest have 

at least some drug coverage through various sources--most commonly 

employer-sponsored health plans--although recent evidence indicates 

that this coverage is beginning to erode.

At the same time, however, the recent publication of the 2003 Trustees’ 

annual report reminds us that Medicare in its current condition--with 

no prescription drug benefit--already faces a huge projected financial 

imbalance that has worsened significantly in the past year. 

Furthermore, as the Medicare Trustees made clear over 10 years ago, the 

current Medicare program is not fiscally sustainable in its present 


In 10 years, Hospital Insurance (HI) Trust Fund outlays will begin to 

exceed tax receipts, and by 2026 the HI trust fund will be exhausted. 

However, trust fund insolvency does not mean the program will cease to 

exist; program tax revenues will continue to cover a portion of 

projected annual expenditures.[Footnote 1]

The huge fiscal pressures created by the retirement of the baby boom 

generation and rising health care costs are on our 10-year budget 

horizon. Between now and 2035, the number of people age 65 and older 

will double. Federal health and retirement spending are expected to 

surge as people live longer and spend more time in retirement. In 

addition, advances in medical technology are likely to keep pushing up 

the cost of providing health care. Moreover, the baby boomers will have 

fewer workers to support them in retirement.

We must also remember that Medicare has grown substantially as a 

percent of the federal budget since its enactment in 1965. In addition, 

it is expected to represent an increasing percentage of the federal 

budget in the years ahead. After a brief slowdown in the late 1990s, 

Medicare spending growth has recently accelerated. In fiscal year 2001, 

growth in program spending reached nearly 9 percent, with spending on 

certain services increasing much more rapidly. For example, spending 

for home health services grew about 30 percent and spending for skilled 

nursing facility care grew slightly over 20 percent. For the first 5 

months of fiscal year 2003, Medicare spending has been growing at 7.6 

percent.[Footnote 2]

A significant problem that hobbles Medicare’s ability to achieve a 

desirable degree of efficiency is that the program too often pays 

overly generous rates for certain services and products. For example, 

for certain services, our recent work has shown substantially higher 

Medicare payments relative to providers’ costs--as much as 35 percent 

higher for home health care and 19 percent higher for skilled nursing 

facility care.[Footnote 3] Similarly, Medicare has overpaid for various 

medical products. In 2001, we reported that Medicare paid over $1 

billion more than other purchasers in 2000 for certain outpatient drugs 

that the program covers. Excessive payments hurt not only the taxpayers 

but also the program’s beneficiaries or their supplemental insurers, as 

beneficiaries are generally liable for copayments equal to 20 percent 

of Medicare’s approved fee. For certain outpatient drugs, Medicare’s 

payments to providers were so high that the beneficiaries’ copayments 

exceeded the price at which providers could buy the drugs. The Centers 

for Medicare & Medicaid Services (CMS) has not acted on our 

recommendation that Medicare establish payment levels for drugs more 

closely related to actual market transaction costs, using information 

available to other public programs that pay at lower rates.[Footnote 4]

In the face of these short-term and long-term cost pressures, I 

continue to maintain that substantive financing and programmatic 

reforms are necessary to put Medicare on a sustainable footing for the 

future. These fundamental reforms are vital to reducing the program’s 

growth, which threatens to absorb ever-increasing shares of the 

nation’s budgetary and economic resources. Thus, any proposals to help 

seniors with the costs of prescription drugs would need to be carefully 

crafted to avoid further erosion of the projected financial condition 

of the Medicare program. Stated differently, it will be prudent to 

adopt a modified Hippocratic oath for Medicare reform--namely, any such 

reform proposals should “do no further harm” to Medicare’s already 

serious long-range financial imbalance.

As you deliberate on ways to modernize Medicare’s benefit package while 

striving for program sustainability, I would like to highlight several 

key considerations:

* The traditional measure of HI Trust Fund solvency is a misleading 

gauge of Medicare’s financial health. Long before the HI Trust Fund is 

projected to be insolvent, pressures on the rest of the federal budget 

will grow as HI’s projected cash flow turns negative and the gap 

between program tax revenues and expenditures escalates. Moreover, a 

focus on the financial status of HI ignores the increasing burden 

Supplemental Medical Insurance (SMI)--Medicare part B--will place on 

taxpayers and beneficiaries.

* GAO’s most recent long-term budget simulations continue to show that, 

absent meaningful entitlement reforms, demographic trends and rising 

health care spending will drive escalating federal deficits and debt. 

To obtain budget balance, massive spending cuts, tax increases, or some 

combination of the two would be necessary. Neither slowing the growth 

of discretionary spending nor allowing the 2001 tax reductions to 

sunset will eliminate the imbalance. In addition, while additional 

economic growth will help ease our burden, the potential fiscal gap is 

too great to grow our way out of the problem.

* Under the huge budgetary pressures that we are sure to face in the 

coming years, we must set priorities so that any benefit expansions are 

in line with available resources. In this regard, the application of 

basic health insurance principles to any proposed benefit could help 

moderate the cost for both beneficiaries and taxpayers. Under these 

principles, beneficiaries receive protections against the risk of 

catastrophic medical expenses while remaining conscious of the cost of 

care through their premium contributions and cost-sharing arrangements. 

Given our already huge Medicare financial imbalance, it is also 

important that benefit expansion proposals include targeting mechanisms 

to ensure that federal support is directed at the beneficiaries with 

the greatest financial risk.

* The private sector’s use of entities called pharmacy benefit managers 

for controlling drug expenditures may be instructive for Medicare, but 

the program’s unique role and nature may moderate how these strategies 

will be used and the potential efficiency gains afforded in attempting 

to transfer these strategies to Medicare.

Outlook Worsening for Medicare’s Long-Term Sustainability:

Today the Medicare program faces a long-range and fundamental financing 

problem driven by known demographic trends and projected escalation of 

health care spending beyond general inflation. The lack of an immediate 

crisis in Medicare financing affects the nature of the challenge, but 

it does not eliminate the need for change. Within the next 10 years, 

the first baby boomers will begin to retire, putting increasing 

pressure on the federal budget. From the perspectives of the program, 

the federal budget, and the economy, Medicare in its present form is 

not sustainable. Acting sooner rather than later would allow changes to 

be phased in so that the individuals who are most likely to be 

affected, namely younger and future workers, will have time to adjust 

their retirement planning while helping to avoid related “expectation 

gaps.” Since there is considerable confusion about Medicare’s current 

financing arrangements, I would like to begin by describing the nature, 

timing, and extent of the financing problem.

Demographic Trends and Expected Rise in Health Care Costs Drive 

Medicare’s Long-Term Financing Problem:

As you know, Medicare consists of two parts--HI and SMI. HI, which pays 

for inpatient hospital stays, skilled nursing care, hospice, and 

certain home health services, is financed by a payroll tax. Like Social 

Security, HI has always been largely a pay-as-you-go system. SMI, which 

pays for physician and outpatient hospital services, diagnostic tests, 

and certain other medical services, is financed by a combination of 

general revenues and beneficiary premiums. Beneficiary premiums pay for 

about one-fourth of SMI benefits, with the remainder financed by 

general revenues. These complex financing arrangements mean that 

current workers’ taxes primarily pay for current retirees’ benefits 

except for those financed by SMI premiums.[Footnote 5]

As a result, the relative numbers of workers and beneficiaries have a 

major impact on Medicare’s financing. The ratio, however, is changing. 

In the future, relatively fewer workers will be available to shoulder 

Medicare’s financial burden. In 2002 there were 4.9 working-age persons 

(18 to 64 years) per elderly person, but by 2030, this ratio is 

projected to decline to 2.8. For the HI portion of Medicare, in 2002 

there were nearly 4 covered workers per HI beneficiary. Under their 

intermediate 2003 estimates, the Medicare Trustees project that by 2030 

there will be only 2.4 covered workers per HI beneficiary. (See fig. 


Figure 1: Ratio of HI-Covered Workers to Beneficiaries:

[See PDF for image]

Note: Projections based on the intermediate assumptions of The 2003 

Annual Report of the Boards of Trustees of the Federal Hospital 

Insurance and Federal Supplementary Medical Insurance Trust Funds.

[End of figure]

The demographic challenge facing the system has several causes. People 

are retiring early and living longer. As the baby boom generation ages, 

the share of the population age 65 and over will escalate rapidly. A 

falling fertility rate is the other principal factor underlying the 

growth in the elderly’s share of the population. In the 1960s, the 

fertility rate was an average of 3 children per woman. Today it is a 

little over 2, and by 2030 it is expected to fall to 1.95--a rate that 

is below replacement. The combination of the aging of the baby boom 

generation, increased longevity, and a lower fertility rate will drive 

the elderly as a share of total population from today’s 12 percent to 

almost 20 percent in 2030.

Taken together, these trends threaten both the financial solvency and 

fiscal sustainability of this important program. Labor force growth 

will continue to decline and by 2025 is expected to be less than a 

third of what it is today. (See fig. 2.) Relatively fewer workers will 

be available to produce the goods and services that all will consume. 

Without a major increase in productivity, low labor force growth will 

lead to slower growth in the economy and slower growth of federal 

revenues. This in turn will only accentuate the overall pressure on the 

federal budget. This slowing labor force growth is not always 

recognized as part of the Medicare debate, but it is expected to affect 

the ability of the federal budget and the economy to sustain Medicare’s 

projected spending in the coming years.

Figure 2: Labor Force Growth:

[See PDF for image]

Note: GAO analysis based on the intermediate assumptions of The 2003 

Annual Report of the Board of Trustees of the Federal Old-Age and 

Survivors Insurance and the Federal Disability Insurance Trust Funds. 

Percentage change is calculated as a centered 5-year moving average.

[End of figure]

The demographic trends I have described will affect both Medicare and 

Social Security, but Medicare presents a much greater, more complex, 

and more urgent challenge. Unlike Social Security, Medicare spending 

growth rates reflect not only a burgeoning beneficiary population, but 

also the escalation of health care costs at rates well exceeding 

general rates of inflation. The growth of medical technology has 

contributed to increases in the number and quality of health care 

services. Moreover, the actual costs of health care consumption are not 

transparent. Third-party payers largely insulate covered consumers from 

the cost of health care decisions. These factors and others contribute 

to making Medicare a greater and more complex fiscal challenge than 

Social Security.

HI’s Trust Fund Faces Cash Flow Problems Long before the HI Trust Fund 

Is Projected to Be Insolvent:

Current projections of future HI income and outlays illustrate the 

timing and severity of Medicare’s fiscal challenge. Today, the HI Trust 

Fund takes in more in taxes than it spends. Largely because of the 

known demographic trends I have described, this situation will change. 

Under the Trustees’ 2003 intermediate assumptions, program outlays are 

expected to begin to exceed program tax revenues in 2013. (See fig. 3.) 

To finance these cash deficits, HI will need to draw on the special-

issue Treasury securities acquired during the years of cash surpluses. 

For HI to “redeem” its securities, the government will need to obtain 

cash through some combination of increased taxes, spending cuts, and/or 

increased borrowing from the public (or, if the unified budget is in 

surplus, less debt reduction than would otherwise have been the case). 

Neither the decline in the cash surpluses nor the cash deficits will 

affect the payment of benefits, but the negative cash flow will place 

increased pressure on the federal budget to raise the resources 

necessary to meet the program’s ongoing costs. This pressure will only 

increase when Social Security also experiences negative cash flow and 

joins HI as a net claimant on the rest of the budget.[Footnote 6]

Figure 3: Medicare’s HI Trust Fund Faces Cash Deficits as Baby Boomers 


[See PDF for image]

Note: GAO analysis based on the intermediate assumptions of The 2003 

Annual Report of the Boards of Trustees of the Federal Hospital 

Insurance and Federal Supplementary Medical Insurance Trust Funds.

[End of figure]

The gap between HI income and costs shows the severity of HI’s 

financing problem over the longer term. This gap can also be expressed 

relative to taxable payroll (the HI Trust Fund’s funding base) over a 

75-year period. This year, under the Trustees’ 2003 intermediate 

estimates, the 75-year actuarial deficit is projected to be 2.40 

percent of taxable payroll--a significant increase from last year’s 

projected deficit of 2.02 percent. This means that to bring the HI 

Trust Fund into balance over the 75-year period, either program outlays 

would have to be immediately reduced by 42 percent or program income 

immediately increased by 71 percent, or some combination of the two. 

These estimates of what it would take to achieve 75-year trust fund 

solvency understate the extent of the problem because the program’s 

financial imbalance gets worse in the 76th and subsequent years. As 

each year passes, we drop a positive year and add a much bigger deficit 


The projected exhaustion date of the HI Trust Fund is a commonly used 

indicator of HI’s financial condition. Under the Trustees’ 2003 

intermediate estimates, the HI Trust Fund is projected to exhaust its 

assets in 2026. This solvency indicator provides information about HI’s 

financial condition, but it is not an adequate measure of Medicare’s 

sustainability for several reasons. In fact, the solvency measure can 

be misleading and can serve to give a false sense of security as to 

Medicare’s true financial condition. Specifically, HI Trust Fund 

balances do not provide meaningful information on the government’s 

fiscal capacity to pay benefits when program cash inflows fall below 

program outlays. As I have described, the government would need to come 

up with cash from other sources to pay for benefits once outlays 

exceeded program tax income.

In addition, the HI Trust Fund measure provides no information on SMI. 

SMI’s expenditures, which currently account for about 43 percent of 

total Medicare spending, are projected to grow even faster than those 

of HI in the near future. Moreover, Medicare’s complex structure and 

financing arrangements mean that a shift of expenditures from HI to SMI 

can extend the solvency of the HI Trust Fund, creating the appearance 

of an improvement in the program’s financial condition. For example, 

the Balanced Budget Act of 1997 modified the home health benefit, which 

resulted in shifting a portion of home health spending from the HI 

Trust Fund to SMI. Although this shift extended HI Trust Fund solvency, 

it increased the draw on general revenues and beneficiary SMI premiums 

while generating little net savings.

Ultimately, the critical question is not how much a trust fund has in 

assets, but whether the government as a whole and the economy can 

afford the promised benefits now and in the future and at what cost to 

other claims on available resources. To better monitor and communicate 

changes in future total program spending, new measures of Medicare’s 

sustainability are needed. As program changes are made, a continued 

need will exist for measures of program sustainability that can signal 

potential future fiscal imbalance. Such measures might include the 

percentage of program funding provided by general revenues, the 

percentage of total federal revenues or gross domestic product (GDP) 

devoted to Medicare, or program spending per enrollee. As such measures 

are developed, questions would need to be asked about actions to be 

taken if projections showed that program expenditures would exceed the 

chosen level.

Absent Reform of Medicare and Other Entitlements for the Elderly, 

Budgetary Flexibility Will Disappear:

Taken together, Medicare’s HI and SMI expenditures are expected to 

increase dramatically, rising from about 12 percent of federal revenues 

in 2002 to more than one-quarter by midcentury. The budgetary challenge 

posed by the growth in Medicare becomes even more significant in 

combination with the expected growth in Medicaid and Social Security 

spending. As shown in figure 4, Medicare, Medicaid, and Social Security 

have already grown from 13 percent of federal spending in 1962 before 

Medicare and Medicaid were created to 42 percent in 2002.

Figure 4: Composition of Federal Spending by Budget Function, 1962, 

1982, and 2002:

[See PDF for image]

[End of figure]

This growth in spending on federal entitlements for retirees will 

become increasingly unsustainable over the longer term, compounding an 

ongoing decline in budgetary flexibility. Over the past few decades, 

spending on mandatory programs has consumed an ever-increasing share of 

the federal budget.[Footnote 7] In 1962, prior to the creation of the 

Medicare and Medicaid programs, spending for mandatory programs plus 

net interest accounted for about 32 percent of total federal spending. 

By 2002, this share had almost doubled to approximately 63 percent of 

the budget. (See fig. 5.):

Figure 5: Federal Spending for Mandatory and Discretionary Programs, 

Fiscal Years 1962, 1982, and 2002:

[See PDF for image]

[End of figure]

In much of the past decade, reductions in defense spending helped 

accommodate the growth in these entitlement programs. However, even 

before the terrorist attacks of September 11, 2001, this ceased to be a 

viable option. Indeed, spending on defense and homeland security will 

grow as we seek to combat new threats to our nation’s security.

GAO prepares long-term budget simulations that seek to illustrate the 

likely fiscal consequences of the coming demographic tidal wave and 

rising health care costs. These simulations continue to show that to 

move into the future with no changes in federal retirement and health 

programs is to envision a very different role for the federal 

government. Assuming, for example, that the tax reductions enacted in 

2001 do not sunset and discretionary spending keeps pace with the 

economy, by midcentury federal revenues may not even be adequate to pay 

Social Security and interest on the federal debt. Spending for the 

current Medicare program--without any additional new benefits--is 

projected to account for more than one-quarter of all federal revenues. 

To obtain budget balance, massive spending cuts, tax increases, or some 

combination of the two would be necessary. (See fig. 6.) Neither 

slowing the growth of discretionary spending nor allowing the tax 

reductions to sunset eliminates the imbalance. In addition, while 

additional economic growth would help ease our burden, the projected 

fiscal gap is too great for us to grow our way out of the problem.

Figure 6: Composition of Spending as a Share of GDP Assuming 

Discretionary Spending Grows with GDP after 2003 and the 2001 Tax Cuts 

Do Not Sunset:

[See PDF for image]

Note: Assumes currently scheduled Social Security benefits are paid in 

full throughout the simulation period. Social Security and Medicare 

projections are based on the Trustees’ 2003 intermediate assumptions.

[End of figure]

Indeed, long-term budgetary flexibility is about more than Social 

Security and Medicare. While these programs dominate the long-term 

outlook, they are not the only federal programs or activities that bind 

the future. The federal government undertakes a wide range of programs, 

responsibilities, and activities that obligate it to future spending or 

create an expectation for spending. A recent GAO report describes the 

range and measurement of such fiscal exposures--from explicit 

liabilities such as environmental cleanup requirements to the more 

implicit obligations presented by life-cycle costs of capital 

acquisition or disaster assistance.[Footnote 8] Making government fit 

the challenges of the future will require not only dealing with the 

drivers--such as entitlements for the elderly--but also looking at the 

range of other federal activities. A fundamental review of what the 

federal government does and how it does it will be needed. This 

involves looking at the base of all major spending and tax policies to 

assess their appropriateness, priority, affordability, and 

sustainability in the years ahead.

Medicare Is Projected to Absorb Ever-Increasing Shares of the Economy:

At the same time, it is important to look beyond the federal budget to 

the economy as a whole. Figure 7 shows the total future draw on the 

economy represented by Medicare, Medicaid, and Social Security. Under 

the 2003 Trustees’ intermediate estimates and the Congressional Budget 

Office’s (CBO) most recent long-term Medicaid estimates, spending for 

these entitlement programs combined will grow to 14 percent of GDP in 

2030 from today’s 8.4 percent. Taken together, Social Security, 

Medicare, and Medicaid represent an unsustainable burden on future 


Figure 7: Social Security, Medicare, and Medicaid Spending as a 

Percentage of GDP:

[See PDF for image]

Note: Projections based on the intermediate assumptions of the 2003 

Trustees’ Reports, CBO’s March 2003 short-term Medicaid estimates, and 

CBO’s June 2002 Medicaid long-term projections under midrange 


[End of figure]

Although real incomes are projected to continue to rise, they are 

expected to grow more slowly than has historically been the case. At 

the same time, the demographic trends and projected rates of growth in 

health care spending I have described will mean rapid growth in 

entitlement spending. Taken together, these projections raise serious 

questions about the capacity of the relatively smaller number of future 

workers to absorb the rapidly escalating costs of these programs.

As HI trust fund assets are redeemed to pay Medicare benefits and SMI 

expenditures continue to grow, the program will constitute a claim on 

real resources in the future. As a result, taking action now to 

increase the future pool of resources is important. To echo Federal 

Reserve Chairman Alan Greenspan, the crucial issue of saving in our 

economy relates to our ability to build an adequate capital stock to 

produce enough goods and services in the future to accommodate both 

retirees and workers in the future.[Footnote 9] The most direct way the 

federal government can raise national saving is by increasing 

government saving; that is, as the economy returns to a higher growth 

path, a balanced fiscal policy that recognizes our long-term challenges 

can help provide a strong foundation for economic growth and can 

enhance our future budgetary flexibility. It is my hope that we will 

think about the unprecedented challenge facing future generations in 

our aging society. Putting Medicare on a sustainable path for the 

future would help fulfill this generation’s stewardship responsibility 

to succeeding generations. It would also help to preserve some capacity 

for future generations to make their own choices for what role they 

want the federal government to play.

As with Social Security, both sustainability and solvency 

considerations drive us to address Medicare’s fiscal challenges sooner 

rather than later. HI Trust Fund exhaustion may be more than 20 years 

away, but the squeeze on the federal budget will begin as the baby boom 

generation begins to retire. This will begin as early as 2008, when the 

leading edge of the baby boom generation becomes eligible for early 

retirement.[Footnote 10] CBO’s current 10-year budget and economic 

outlook reflects this. CBO projects that economic growth will slow from 

an average of 3.2 percent a year from 2005 through 2008 to 2.7 percent 

from 2009 through 2013, reflecting slower labor force growth. At the 

same time, annual rates of growth in entitlement spending will begin to 

rise. Annual growth in Social Security outlays is projected to 

accelerate from 5.2 percent in 2007 to 6.6 percent in 2013. Annual 

growth in Medicare enrollees is expected to accelerate from 1.1 percent 

today to 2.9 percent in 2013. Acting sooner rather than later is 

essential to ease future fiscal pressures and also provide a more 

reasonable planning horizon for future retirees. We are now at a 

critical juncture. In less than a decade, the profound demographic 

shift that is a certainty will have begun.

As Bleak Fiscal Future Looms, Efforts to Address Medicare Coverage Gaps 

Are Being Considered:

Despite a common awareness of Medicare’s current and future fiscal 

plight, pressure has been building to address recognized gaps in 

Medicare coverage, especially the lack of a prescription drug benefit 

and protection against financially devastating medical costs. Filling 

these gaps could add significant expenses to an already fiscally 

overburdened program. Under the Trustees’ 2003 intermediate 

assumptions, the present value of HI’s actuarial deficit is $6.2 

trillion, a 20-percent increase from the prior year.[Footnote 11] This 

difficult situation argues for tackling the greatest needs first and 

for making any benefit additions part of a larger structural reform 


The Medicare benefit package, largely designed in 1965, provides 

virtually no outpatient drug coverage. Beneficiaries may fill this 

coverage gap in various ways. According to the Medicare Current 

Beneficiary Survey, nearly two-thirds of Medicare beneficiaries had 

some form of drug coverage from a supplemental insurance policy, health 

plan, or public program at some point during 1999. All beneficiaries 

have the option to purchase supplemental policies--Medigap--when they 

first become eligible for Medicare at age 65. Those policies that 

include drug coverage tend to be expensive and provide only limited 

benefits. Some beneficiaries have access to coverage through employer-

sponsored policies or private health plans that contract to serve 

Medicare beneficiaries. In recent years, coverage through these sources 

has become more expensive and less widely available. Beneficiaries 

whose incomes fall below certain thresholds may qualify for Medicaid or 

other public programs. More than one-third may lack drug coverage 


In recent years, prescription drug expenditures have grown 

substantially, both in total and as a share of all heath care outlays. 

Prescription drug spending grew an average of 15.9 percent per year 

from 1996 to 2001, more than double the 6.5 percent average growth rate 

for health care expenditures overall. (See table 1.) As a result, 

prescription drugs account for a growing share of health care spending, 

rising from 6.5 percent in 1996 to 9.9 percent in 2001. By 2012, 

prescription drug expenditures are expected to account for almost 15 

percent of total health expenditures.

Table 1: National Expenditures for Prescription Drugs and Health Care, 

1996 to 2001:

Year: 2001; Prescription drug expenditures (in billions): $140.6; 

Annual growth in prescription drug expenditures from previous year 

(percent): 15.4; Annual growth in health care expenditures from 

previous year (percent): 8.7.

Year: 2000; Prescription drug expenditures (in billions): 121.8; Annual 

growth in prescription drug expenditures from previous year (percent): 

17.3; Annual growth in health care expenditures from previous year 

(percent): 6.9.

Year: 1999; Prescription drug expenditures (in billions): 103.9; Annual 

growth in prescription drug expenditures from previous year (percent): 

19.2; Annual growth in health care expenditures from previous year 

(percent): 5.7.

Year: 1998; Prescription drug expenditures (in billions): 87.2; Annual 

growth in prescription drug expenditures from previous year (percent): 

15.1; Annual growth in health care expenditures from previous year 

(percent): 5.4.

Year: 1997; Prescription drug expenditures (in billions): 75.7; Annual 

growth in prescription drug expenditures from previous year (percent): 

12.8; Annual growth in health care expenditures from previous year 

(percent): 4.9.

Year: 1996; Prescription drug expenditures (in billions): 67.2; Annual 

growth in prescription drug expenditures from previous year (percent): 

10.5; Annual growth in health care expenditures from previous year 

(percent): 5.0.

Year: Average annual growth from 1996 through 2001; Prescription drug 

expenditures (in billions): [Empty]; Annual growth in prescription drug 

expenditures from previous year (percent): 15.9; Annual growth in 

health care expenditures from previous year (percent): 6.5.

Source: CMS, Office of the Actuary.

[End of table]

In 2002, CBO projected that the average Medicare beneficiary would use 

$2,440 worth of prescription drugs in 2003. This is a substantial 

amount considering that some beneficiaries lack any drug coverage and 

others may have less coverage than in previous years. Moreover, 

significant numbers of beneficiaries have drug expenses much higher 

than those of the average beneficiary. CBO also estimated that, in 

2005, 12 percent of Medicare beneficiaries would have expenditures 

above $6,000.

In focusing on the need for prescription drug coverage, we should not 

forget that Medicare does not provide complete protection from 

catastrophic losses. Under Medicare, beneficiaries have no limit on 

their out-of-pocket costs attributable to cost sharing. The average 

beneficiary who obtained services had a total liability for Medicare-

covered services of $1,700, consisting of $1,154 in Medicare copayments 

and deductibles in addition to the $546 in annual part B premiums in 

1999, the most recent year for which data are available on the 

distribution of these costs. For beneficiaries with extensive health 

care needs, the burden can be much higher. In 1999, about 1 million 

beneficiaries were liable for more than $5,000, and about 260,000 were 

liable for more than $10,000 for covered services. In contrast, 

employer-sponsored health plans for active workers typically limited 

maximum annual out-of-pocket costs for covered services to less than 

$2,000 per year for single coverage.[Footnote 12]

Recently, several proposals have been made to add a prescription drug 

benefit to the Medicare program. While different in scope and detail, 

the proposals have certain features in common--including use of a 

third-party entity to administer the new drug benefit. The remainder of 

my remarks will focus on the lessons learned from our work regarding 

the private sector’s use of such an entity to manage the drug benefits 

of insurers’ policyholders and health plans’ enrollees.

Private Sector Strategies for Controlling Drug Expenditures May Be 

Instructive for Medicare:

Some proposals to add a Medicare outpatient prescription drug benefit 

look to private sector strategies as a means to administer a drug 

benefit and control costs. Most employer-sponsored health plans 

contract with private entities, known as pharmacy benefit managers 

(PBM), to administer their prescription drug benefits, and those that 

do not contract with PBMs may have units in their organizations that 

serve the same administrative purpose. Typically, on behalf of the 

health plans, PBMs negotiate drug prices with pharmacies, negotiate 

rebates with drug manufacturers, process drug claims, operate mail-

order pharmacies, and employ various cost-control techniques, such as 

formulary management and drug utilization reviews. In 2001, nearly 200 

million Americans had their prescription drug benefits administered 

through PBMs. This year, we reported on the use of PBMs by health plans 

in the Federal Employees’ Health Benefits Program (FEHBP).[Footnote 13] 

In considering the application of these findings to Medicare, we are 

reminded that Medicare’s unique role and nature may temper how the 

strategies and potential efficiency gains afforded by private sector 

PBMs may be transferred to benefit the program.

Private Sector Uses PBMs to Leverage Price Negotiations through Volume 


PBMs use purchasing volume to leverage their negotiations with 

pharmacies and drug manufacturers in seeking favorable prices in the 

form of discounts, rebates, or other advantages. Through negotiations, 

PBMs create networks of participating retail pharmacies, promising the 

pharmacies a greater volume of customers in exchange for discounted 

prices. PBMs may be able to secure larger discounts by limiting the 

number of network pharmacies. However, smaller networks provide 

beneficiaries fewer choices of retailers, thereby limiting convenient 

access. These are trade-offs health plans must consider in deciding how 

extensive a pharmacy network they want their PBMs to offer 

beneficiaries. The health plans we reviewed in our FEHBP study 

generally provided broad retail pharmacy networks. The average 

discounted prices PBMs obtained for drugs from retail pharmacies were 

about 18 percent below the average prices cash-paying customers without 

drug coverage would have paid for 14 selected widely used brand-name 

drugs. For 4 selected generic drugs, the PBM-negotiated retail pharmacy 

prices were 47 percent below the price paid by cash-paying customers.

PBMs also use their leverage to negotiate with drug manufacturers for 

rebates. Rebates generally depend on the volume of a manufacturer’s 

products purchased. Health plans and PBMs can add to that volume by 

concentrating beneficiaries’ purchases for particular types of drugs 

with certain manufacturers. Health plans can steer their beneficiaries’ 

purchases to specific drugs through the use of a formulary--that is, a 

list of prescription drugs that health plans encourage physicians to 

prescribe and beneficiaries to use. Determining whether a drug should 

be on the formulary involves clinical evaluations based on a drug’s 

safety and effectiveness, and decisions on whether several drugs are 

therapeutically equivalent.[Footnote 14] Restricting the formulary to 

fewer drugs within a therapeutic class can provide the PBMs with 

greater leverage in negotiating higher rebates because they can help 

increase the manufacturer’s market share for certain drugs. However, a 

restricted formulary provides beneficiaries with fewer preferred drug 

alternatives and makes the policies governing coverage of nonformulary 

drugs or the cost sharing for them critical to beneficiaries.[Footnote 


The FEHBP plans and PBMs we reviewed provided enrollees with generally 

nonrestrictive drug formularies across a broad range of drugs and 

therapeutic categories.[Footnote 16] The manufacturer rebates that the 

PBMs passed through to the FEHBP plans effectively reduced plans’ 

annual spending on prescription drugs by a range of 3 percent to 9 

percent. The share of rebates PBMs passed through to the FEHBP plans 

varied subject to contractual agreements negotiated between the plans 

and the PBMs.

PBMs also assisted the FEHBP plans by providing a less expensive mail-

order drug option. Mail-order prices for the FEHBP plans we reviewed 

averaged about 27 percent lower than cash-paying customers would pay 

for the same quantity at retail pharmacies for 14 brand-name drugs and 

53 percent lower for 4 generic drugs. The FEHBP plans generally had 

lower cost-sharing requirements for drugs purchased through mail order, 

particularly for more expensive brand-name drugs or maintenance 

medications for chronic conditions.

The claims and information processing capabilities PBMs offered also 

helped the FEHBP plans to manage drug costs and monitor quality of 

care. PBMs maintain a centralized database on each enrollee’s drug 

history that can be used to review for potential adverse drug 

interactions or potentially less expensive alternative medications. 

They also use claims data to monitor patterns of patient use, physician 

prescribing practices, and pharmacy dispensing practices. Their systems 

provide “real-time” claims adjudication capabilities that allow a 

customer’s claim for a drug purchase to be approved or denied at the 

time the pharmacist begins the process of filling a prescription. Two 

plans in our FEHBP study reported savings ranging from 6 to 9 percent 

of the plan’s annual drug spending; the savings were associated 

primarily with real-time claims denials preventing early drug refills 

and safety advisories cautioning pharmacists about potential adverse 

interactions or therapy duplications.

Use of Private-Sector Strategies in Medicare Would Represent Departure 

from Traditional Policies and Practices:

While Medicare’s sheer size would provide it with significant leverage 

in negotiating with pharmacies and drug manufacturers, doing so would 

represent a departure from traditional Medicare. Medicare beneficiaries 

represent less than 15 percent of the population but a 

disproportionately higher share--about 40 percent--of prescription 

drug spending. However, because of Medicare’s design and obligations as 

a public program, its current purchasing strategies vary considerably 

from those of the private sector.

* Any willing provider. In contrast with private payers’ reliance on 

selective contracting with providers and suppliers, the traditional 

Medicare program has generally allowed any hospital, physician, or 

other provider willing to accept Medicare’s reimbursements and 

requirements to participate in the program. With respect to drug 

purchasing in particular, private plans determine the extent of their 

enrollees’ access by the choices they make about the size of their 

participating pharmacy network and breadth of their drug formulary. 

Allowing any pharmacy willing to meet Medicare’s terms to participate 

or allowing all therapeutically equivalent drugs equal coverage on a 

formulary would restrict the program’s ability to secure advantageous 

prices. Moreover, health plans and PBMs currently make formulary 

determinations privately. In contrast, Medicare’s policies have 

historically been open to public comment.

* Administrative rate-setting. Whereas private health plans typically 

rely on price negotiations to establish payment rates, Medicare 

generally establishes payment rates administratively. As discussed 

earlier, Medicare’s rates often exceed market prices and this is the 

case for some of the few outpatient prescription drugs covered by 

Medicare.[Footnote 17] The program’s method of paying for these drugs 

is prescribed in statute: In essence, Medicare pays 95 percent of a 

drug’s “average wholesale price” (AWP). Despite its name, however, AWP 

is not necessarily a price that wholesalers charge and is not based on 

the price of any actual sale of drugs by a manufacturer. AWPs are 

published by manufacturers in drug price compendia, and Medicare bases 

providers’ payments on these published AWPs. Other public and private 

purchasers typically use the leverage of volume and competition to 

secure better prices. By statute, Medicaid, the nation’s health 

insurance program for certain low-income Americans, is guaranteed 

manufacturers’ rebates based on prices charged other 

purchasers.[Footnote 18] Certain other public payers can pay at rates 

set in the federal supply schedule, which uses verifiable confidential 

information on the prices drug manufacturers charge their “most 

favored” private customers. Manufacturers agree to these prices, in 

part, in exchange for the right to sell drugs to the more than 40 

million Medicaid beneficiaries.

* Low-budget program administration. Duplicating the type of controls 

PBMs have exercised over private-sector drug benefits would likely 

involve devoting a larger share of total expenditures to administration 

than is spent by Medicare currently. Medicare’s administrative costs 

historically have been extremely low, averaging about 2 percent of the 

cost of the services themselves.[Footnote 19] This level of expenditure 

may not be consistent with the level needed to review the volumes of 

claims data associated with prescription drugs for the elderly or 

acquire and maintain the on-line systems and databases PBMs use to 

employ such utilization controls as real-time claims adjudication. The 

number of prescriptions for Medicare beneficiaries could easily exceed 

the current number of claims for all other services combined, or over 1 

billion annually.

Decisions about the Extent of Latitude and Competition Allowed Are 

Critical to Administering a Medicare Drug Benefit:

Medicare would undoubtedly need assistance from external entities to 

administer a drug benefit, just as it has used insurers to process 

claims in the traditional program and Medicare+Choice plans to go 

further by also managing services and assuming risk. Decisions about 

the roles assigned an entity or entities and the latitude allowed them 

in carrying out those roles would be critical. These decisions would 

undoubtedly affect the benefit’s value to beneficiaries and the 

efficiencies and savings secured for both beneficiaries and taxpayers. 

Some of these decisions parallel those made by FEHBP plans that I 

discussed--trade-offs about beneficiaries’ interests in broad pharmacy 

networks and formularies versus potential savings. Others stem from the 

uniqueness of Medicare, its likely disproportionate share of the drug 

market, and its position as a public program requiring transparency and 


Insurers and PBMs have been successful in securing some savings on drug 

purchases by leveraging their volume to move market share from one 

product to another. Medicare’s leverage, given that purchases by the 

elderly constitute about 40 percent of the drug market, could be 

considerable. Yet the large market share may also be likely to attract 

considerable attention. The administration of a Medicare drug benefit 

could then be subject to the same intensity of external pressures from 

interested parties regarding program prices and rules that can often 

inhibit the program from operating efficiently today. The potential for 

micromanagement could compromise trying to use the very flexibility 

PBMs have employed in negotiating prices and selecting preferred 

providers in order to generate savings. An alternative would be to 

sacrifice some of the program’s leverage and grant flexibility to 

multiple PBMs or similar entities so that any one entity would be 

responsible for administering only a share of the market.

Contracting with multiple PBMs or similar entities, however, would pose 

other challenges. If each had exclusive responsibility for a geographic 

area, beneficiaries who wanted certain drugs could be advantaged or 

disadvantaged merely because they lived in a particular area. To 

minimize inequities, Medicare could, like some private sector 

purchasers, specify core benefit characteristics or maintain clinical 

control over formulary decisions instead of delegating those decisions 

to its contractors.

If multiple PBMs or similar entities operated in a designated area, 

beneficiaries could choose among them to administer their drug 

benefits. These organizations would compete for consumers directly on 

the basis of differences in their drug benefit offerings and 

administration. This contrasts with the private sector where drug 

benefits are typically part of an overall insurance plan, and PBMs 

typically compete for contracts with insurers or other purchasers. 

Competition could be favorable to beneficiaries if they were adequately 

informed about differences among competing entities offering drug 

benefits and shared in the savings. However, adequate oversight would 

need to be in place to ensure that fair and effective competition was 

maintained. For example, a means to ensure that beneficiaries received 

comprehensive user-friendly information about policy and benefit 

differences among competing entities would be necessary. Monitoring 

marketing and customer recruitment strategies and holding entities 

accountable for complying with federal requirements would require 

adequate investment. The contracting entities could need protections as 

well. Some mechanism would be needed to risk adjust payments for 

differences in beneficiaries’ health status so that those entities 

enrolling a disproportionate share of high-use beneficiaries would not 

be disadvantaged.

Concluding Observations:

Medicare’s financial challenge is very real and growing. The 21st 

century has arrived and our demographic tidal wave is on the horizon. 

Within 5 years, individuals in the vanguard of the baby boom generation 

will be eligible for Social Security and 3 years after that they will 

be eligible for Medicare. The future costs of serving the baby boomers 

are already becoming a factor in CBO’s short-term cost projections.

Frankly, we know that incorporating a prescription drug benefit into 

the existing Medicare program will add hundreds of billions of dollars 

to program spending over just the next 10 years. For this reason, I 

cannot overstate the importance of adopting meaningful reforms to 

ensure that Medicare remains viable for future generations. Adding a 

drug benefit to Medicare requires serious consideration of how that 

benefit will affect overall program spending. If competing private 

entities are to be used to administer a drug benefit, it is important 

to understand how these entities can be used in the Medicare context to 

provide a benefit that balances beneficiary needs and cost containment.

Medicare reform would be done best with considerable lead time to phase 

in changes and before the changes that are needed become dramatic and 

disruptive. Given the size of Medicare’s financial challenge, it is 

only realistic to expect that reforms intended to bring down future 

costs will have to proceed incrementally. We should begin this now, 

when retirees are still a far smaller proportion of the population than 

they will be in the future. The sooner we get started, the less 

difficult the task will be.

We must also be mindful that health care costs compete with other 

legitimate priorities in the federal budget, and their projected growth 

threatens to crowd out future generations’ flexibility to decide which 

competing priorities will be met. In making important fiscal decisions 

for our nation, policymakers need to consider the fundamental 

differences among wants, needs, and what both individuals and our 

nation can afford. This concept applies to all major aspects of 

government, from major weapons system acquisitions to issues affecting 

domestic programs. It also points to the fiduciary and stewardship 

responsibility that we all share to ensure the sustainability of 

Medicare for current and future generations within a broader context of 

providing for other important national needs and economic growth.

The public sector can play an important role in educating the nation 

about the limits of public support. Currently, there is a wide gap 

between what patients and providers expect and what public programs are 

able to deliver. Moreover, there is insufficient understanding about 

the terms and conditions under which health care coverage is actually 

provided by the nation’s public and private payers. In this regard, GAO 

is preparing a health care framework that includes a set of principles 

to help policymakers in their efforts to assess various health 

financing reform options. This framework will examine health care 

issues systemwide and identify the interconnections between public 

programs that finance health care and the private insurance market. The 

framework can serve as a tool for defining policy goals and ensuring 

the use of consistent criteria for evaluating changes. By facilitating 

debate, the framework can encourage acceptance of changes necessary to 

put us on a path to fiscal sustainability. I fear that if we do not 

make such changes and adopt meaningful reforms, future generations will 

enjoy little flexibility to fund discretionary programs or make other 

valuable policy choices.

Mr. Chairman, this concludes my prepared statement. I will be happy to 

answer any questions you or other committee members may have.

Contacts and Acknowledgments:

For future contacts regarding this testimony, please call William J. 

Scanlon, Director, Health Care Issues, at (202) 512-7114. Other 

individuals who made key contributions include Rashmi Agarwal, Linda 

Baker, John Dicken, Hannah Fein, Kathryn Linehan, James McTigue, 

Jennifer Rellick, and Melissa Wolf.


[1] Under the Trustees 2003 intermediate assumptions, revenues from the 

HI payroll tax and the taxation of certain Social Security benefits are 

initially projected to cover about three-fourths of projected 

expenditures once the trust fund is exhausted. This ratio, however, is 

projected to decline rapidly.

[2] Congressional Budget Office, Monthly Budget Review (Washington, 

D.C.: Mar. 10, 2003).

[3] See U.S. General Accounting Office, Medicare Home Health Care: 

to Home Health Agencies Are Considerably Higher than Costs, GAO-02-663 

(Washington, D.C: May 6, 2002) and Skilled Nursing Facilities: Medicare 

Payments Exceed Costs for Most but Not All Facilities, GAO-03-183 

(Washington, D.C: Dec. 31, 2002).

[4] U.S. General Accounting Office, Medicare: Payments for Covered 

Outpatient Drugs Exceed Providers’ Costs, GAO-01-1118 (Washington, 

D.C.: Sept. 21, 2001). 

[5] Another small source of funding derives from the tax treatment of 

Social Security benefits. Under certain circumstances, up to 85 percent 

of an individual’s or couple’s Social Security benefits are subject to 

income taxes. Under present law, the Old-Age and Survivors Insurance 

(OASI) and Disability Insurance (DI) Trust Funds are credited with the 

income taxes attributable to the taxation of the first 50 percent of 

OASDI benefit payments. The remainder of the income taxes attributable 

to the taxation of up to 85 percent of OASDI benefit payments is 

credited to the HI Trust Fund. Any other income taxes paid by retirees 

would also help finance the general revenue contribution to SMI.

[6] Under the Trustees’ intermediate 2003 projections, this will occur 

for Social Security (OASDI) in 2018.

[7] “Mandatory spending” refers to outlays for entitlement programs 

such as food stamps, Medicare, and veterans’ pensions; payment of 

interest on the public debt; and outlays for certain nonentitlement 

programs such as payments to states from Forest Service receipts. In 

2002 Social Security, Medicare, and Medicaid accounted for over 71 

percent of mandatory spending.

[8] U.S. General Accounting Office, Fiscal Exposures: Improving the 

Budgetary Focus on Long-Term Costs and Uncertainties, GAO-03-213 

(Washington, D.C.: Jan. 24, 2003).

[9] Testimony before the Senate Committee on Banking, Housing, and 

Urban Affairs, July 24, 2001.

[10] In 2008, the first baby boomers will reach age 62 and become 

eligible for Social Security benefits; in 2011, they will reach age 65 

and become eligible for Medicare benefits.

[11] This estimate represents the present value of HI’s future 

expenditures less future tax income, taking into account the amount of 

HI trust fund assets at hand at the beginning of the projection period 

and adjusting for the ending target trust fund balance. Excluding the 

ending target trust fund balance, HI’s unfunded obligation is estimated 

to be $5.9 trillion over the 75-year period under the Trustees’ 2003 

intermediate assumptions. 

[12] The Kaiser Family Foundation and Health Research and Education 

Trust, Employer Health Benefits: 2000 Annual Survey (Menlo Park, Calif. 

and Chicago: 2000).

[13] U.S. General Accounting Office, Federal Employees’ Health 

Benefits: Effects of Using Pharmacy Benefit Managers on Health Plans, 

Enrollees, and Pharmacies, GAO-03-196 (Washington, D.C.: Jan. 10, 

2003). FEHBP covered about 8.3 million federal employees, retirees, and 

their dependents as of July 2002, and the three FEHBP plans we reviewed 

accounted for about 55 percent of FEHBP enrollment. The FEHBP plans and 

PBMs we reviewed were Blue Cross and Blue Shield, which contracted with 

AdvancePCS for retail pharmacy services and Medco Health Solutions for 

mail-order services; Government Employees Hospital Association, which 

contracted with Medco Health Solutions; and PacifiCare of California, 

which contracted with Prescription Solutions, another subsidiary of 

PacifiCare Health Systems. 

[14] A pharmacy and therapeutics committee within the health plan or a 

PBM typically makes decisions about whether to include particular 

brand-name or generic drugs on the plan’s formulary. 

[15] Plans generally encourage the use of formulary drugs by having 

lower cost sharing or requiring special approval of a nonformulary 

drug. For example, health plans have increasingly adopted three-tiered 

cost-sharing strategies whereby enrollees incur the lowest out-of-

pocket costs for using generic drugs, higher costs for brand-name drugs 

on the formulary, and the highest costs for brand-name drugs not 

included on the formulary. 

[16] Our report compared the FEHBP plans’ formularies to the Department 

of Veterans Affairs (VA) National Formulary, considered by the 

Institute of Medicine to be not overly restrictive. Each FEHBP plan we 

reviewed included over 90 percent of the drugs listed on the VA 

formulary or therapeutically equivalent alternatives, and included at 

least one drug in 93 percent to 98 percent of the therapeutic classes 

covered by VA.

[17] GAO-01-1118.

[18] Since the enactment of the Omnibus Budget Reconciliation Act of 

1990, drug manufacturers are required to provide rebates to state 

Medicaid programs on outpatient drugs based on the “lowest” or “best” 

prices they charged other purchasers or a minimum of 15.1 percent of 

the average manufacturers’ price (AMP) for brand-name drugs. Rebates 

must be at least 11 percent of AMP for generic drugs.

[19] U.S. General Accounting Office, Medicare: HCFA Faces Challenges to 

Control Improper Payments, GAO/T-HEHS-00-74, (Washington, D.C.: Mar. 9,