Order Code RL32545
CRS Report for Congress
Received through the CRS Web
Health Care Spending: Context and Policy
August 26, 2004
Jennifer Jenson
Specialist in Health Economics
Domestic Social Policy Division
Congressional Research Service ˜ The Library of Congress

---

Health Care Spending: Context and Policy
Summary
The United States spends a large and growing share of national income on
health care. In 2004, health spending is expected to approach $1.8 trillion, more than
15% of gross domestic product (GDP). We spend substantially more than other
developed countries, both per capita and as a share of GDP. However, given our
wealth, such spending is not necessarily a problem. On the one hand, depending on
our preference for health care compared with other things, we may wish to spend
even more. On the other hand, regardless of the preferred level for national spending,
our nation might use available resources more efficiently and equitably.
Health care costs put significant pressure on the federal budget — both directly,
through spending on Medicare, Medicaid, and other federal benefits; and indirectly,
through tax expenditures for health insurance and expenses. The Congressional
Budget Office projects spending for Medicare, Medicaid, and the State Children’s
Health Insurance Program will total $476 billion and account for about 21% of
federal spending in 2004. Federal tax expenditures for health benefits; health
coverage for military personnel, veterans, and federal employees; and spending by
Public Health Service agencies are expected to add $235 billion in costs. Given
competing constituent interests and the complex interdependence of public and
private benefits and actors, policymakers face difficult challenges in helping to
ensure access to health care and health insurance without exacerbating federal budget
pressures or contributing to market-wide inflation.
Three broad policy directions have both promise and limitations for addressing
health spending and costs: (1) changing health care, (2) changing federal programs,
and (3) using tax policy to make health care more affordable for individuals and
families. The first, changing health care, focuses on what government might do to
help improve the production and delivery of health services. This direction focuses
on innovations that could improve quality and potentially reduce costs throughout the
health care system. A key limitation comes from uncertainty about whether any
particular innovation will reduce or increase health spending.
The second direction, changing federal programs, focuses more narrowly on
federal spending for federal benefits. To influence federal spending, policymakers
can set budgets for programs, services, or beneficiaries. They can change eligibility
rules or program benefits. And they can change other program features, including
payment methods and amounts, and how beneficiaries obtain coverage. In this
category, the primary challenge is balancing explicit tradeoffs between competing
goals regarding access and spending.
The final direction, using tax policy to make health care more affordable for
individuals and families, focuses on helping consumers pay for health insurance and
health care. Tax exclusions, credits, and deductions, and tax-advantaged accounts
are examples of subsidies in this category. The promise of these tools relates to
flexibility: In general, they help consumers buy the health insurance and health care
they prefer. An important limitation is that tax subsidies drive up consumer demand,
health care prices, and both public and private spending. This report will be updated.

---

Contents
Health Spending: The Big Picture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
International Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Valuing Spending on Health Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Economics and Valuing Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Distribution Matters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Key Issue for the Congress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Federal Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Constituents and Complexity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Three Policy Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Changing Health Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Changing Federal Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Program Budgets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Eligibility and Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Other Program Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Using Tax Policy to Make Health Care More Affordable . . . . . . . . . . . . . . 11
Insurance Subsidies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Other Subsidies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
List of Tables
Table 1. National Health Expenditures and Gross Domestic Product . . . . . . . . . 1
Table 2. U.S. Spending on Health Care and Other Items . . . . . . . . . . . . . . . . . . . 4

---

Health Care Spending: Context and Policy
Health care costs and spending are persistent concerns for the Congress. On the
one hand, policymakers worry about access to care and the burden of health costs on
household and employer budgets. On the other hand, rising costs put growing
pressure on the federal budget from Medicare, Medicaid, and tax expenditures for
private health insurance. This report seeks to put health spending in context. How
much does this nation spend, and is it too much? Why is policy action so difficult?
And what types of policies can the Congress pursue in seeking to balance concerns
regarding spending and access?
Given the breadth of the topic, this report is not intended to be comprehensive.
Instead, it introduces selected issues, using varied policy examples to make ideas
more concrete.
Health Spending: The Big Picture

Health spending in the United States is expected to approach $1.8 trillion in
2004, an estimated $6,167 per capita, according to the Centers for Medicare and
Medicaid Services (CMS). As Table 1 shows, although growth in spending appears
to have been slowing recently, the rate continues to outpace change in gross domestic
product (GDP) by a healthy margin.
Table 1. National Health Expenditures
and Gross Domestic Product
2001
2002
2003a
2004a
National Health Expenditures (NHE, in
billions)b
$1,421
$1,553
$1,674
$1,794
NHE per capitab
$5,021
$5,440
$5,808
$6,167
NHE growth from prior year
8.5%
9.3%
7.8%
$7.2%
GDP growth from prior year
2.6%
3.6%
4.8%
5.8%
NHE as percent of GDP
14.1%
14.9%
15.3%
15.5%
Sources: Stephen Heffler et al., “Health Spending Projections Through 2013,” Health Affairs — Web
Exclusive, Feb. 11, 2004, at [http://content.healthaffairs.org/webexclusives/index.dtl?year=2004], pp.
W4-80-W4-81 (PDF version). Katharine Levit et al., “Health Spending Rebound Continues in 2002,”
Health Affairs, vol. 23, no. 1 (Jan./Feb. 2004), pp. 147-159.
a. Projected.
b. Amounts include spending for health services and supplies, and investment (research and
construction).

---

CRS-2
Data on health care spending per privately insured American tell the same story:
growth is slowing, but still rapid compared with changes in personal income. In
2003, spending on health care services — including hospital inpatient and outpatient
services, physician services, and prescription drugs — rose by 7.4% per capita,
compared with 9.5% in 2002 and 10% in 2001.1 These rates for health services
compare with an average annual increase in personal income of 2.0% over the 2001-
2003 period.2
Is the U.S. spending level a problem? What about the rate of growth?3
International Perspective
Compared with other developed countries, the United States spends both more
per capita and a greater share of its national income on health care. According to
data from the Organization for Economic Cooperation and Development (OECD),
in 2001, per capita health spending in the United States was more than double the
OECD median.4
Also based on OECD data, U.S. health spending consumed 13.9% of GDP,
compared with median spending of 8.1% of GDP for OECD countries.5 After
America, countries spending the highest shares of GDP were: Switzerland (11.1%),
Germany (10.7%), Canada (9.7%), France (9.5%), and Greece (9.4%). The U.S.
spending level is not necessarily too high. About 90% of the variation in health
1 Center for Studying Health System Change, Tracking Health Care Costs: Trends Turn
Downward in 2003, Data Bulletin no. 27, June 2004. Spending on inpatient and outpatient
hospital care together accounted for more than half the increase in spending in 2003, in part
because hospital spending accounts for a substantial fraction of total spending, and in part
because spending on outpatient services grew at a rapid 11.0%. Growth in spending on
prescription drugs is slowing but still rapid, falling from 13.8% in 2001, to 13.2% in 2002
and 9.1% in 2003.
2 Bureau of Economic Analysis, “National Income and Product Accounts, Table 7.1 —
Selected Per Capita Product and Income Series in Current and Chained Dollars,” last revised
June 25, 2004, at [http://www.bea.gov/bea/dn/nipaweb/TableView.asp#Mid]. Annual
growth rates for personal income were: 2.3% in 2003, 1.2% in 2002, and 2.3% in 2001.
3 For additional information on health spending, see CRS Report RL31374, Health
Expenditures in 2002, and CRS Report RL31094, Health Care Spending: Past Trends and
Projections, both by Paulette C. Morgan.
4 Uwe E. Reinhardt, Peter S. Hussey, and Gerard F. Anderson, “U.S. Health Care Spending
in an International Context,” Health Affairs, vol. 23, no. 3 (May/June 2004), pp. 10-25. U.S.
spending in purchasing-power-parity international dollars was $4,887, compared with
median spending in OECD countries of $2,161. OECD countries include Australia, Austria,
Belgium, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary,
Iceland, Ireland, Italy, Japan, Korea, Luxembourg, Mexico, Netherlands, New Zealand,
Norway, Poland, Portugal, Slovak Republic, Spain, Sweden, Switzerland, Turkey, and the
United Kingdom.
5 OECD and CMS report slightly different estimates of health spending as a share of GDP
in 2001 (13.9% vs. 14.1%). Given uncertainty in estimating both health spending and GDP,
this difference is not meaningful.

---

CRS-3
spending across OECD countries in 2001 can be explained by differences in GDP per
capita, suggesting that countries with higher national income are able and willing to
spend this income on more health care.6
Valuing Spending on Health Care
Criticism of U.S. spending levels generally boils down to the argument that
Americans benefit little from the additional money they spend on health care.
Despite paying more than twice as much per capita as other OECD countries, basic
health statistics for the United States are about average.7 In 2001, the U.S. infant
mortality rate was 6.8 deaths/1000 live births, essentially the same as the mean (6.9
deaths/1000 live births) for all OECD countries. In the same year, U.S. life
expectancy at birth was just below OECD averages. U.S. females were expected to
live 79.8 years, compared with 80.4 years for females in all OECD countries; for
males, the U.S. and OECD numbers were about equal at 74.4 and 74.5 years,
respectively.
A related argument regarding the uncertain value of health spending points to
variation within the United States itself that cannot be fully explained by differences
in health status or prices, and that is not correlated with better outcomes or
satisfaction with care.8 For example, according to the Medicare Payment Advisory
Commission (MedPAC), in 2000, Medicare spending per beneficiary varied from
about $3,500 in Santa Fe, New Mexico, to almost $9,200 in Miami, Florida.9
Unfortunately, variation in measures of health on the one hand and spending on
the other are difficult to interpret. In the former case, many things besides health care
affect mortality and life expectancy, including nutrition, sanitation and hygiene,
housing, and the prevention and control of infectious disease. In the latter case,
although more spending is not necessarily better, it also is not necessarily worse:
prices may appropriately differ across markets with different input costs; and quality
6 Reinhardt, et al., “U.S. Health Care Spending in an International Context.” Said another
way, as income increases, spending on health care increases both absolutely and as a
proportion of income. This characteristic implies, in economic jargon, that health care is
a luxury good.
7 Organization for Economic Cooperation and Development, “OECD Health Data 2004 —
Frequently Requested Data, Table 1 — Life expectancy (in years)” and “Table 2 — Infant
Mortality, Deaths per 1000 Live Births,” data released June 3, 2004
[http://www.oecd.org/document/16/0,2340,en_2649_37407_2085200_1_1_1_37407,00.h
tml], visited July 12, 2004.
8 See, for example, Fisher et al., “The Implications of Regional Variations in Medicare
Spending. Part 2: Health Outcomes and Satisfaction with Care,” Annals of Internal
Medicine, vol. 138, no. 4 (Feb. 18, 2003), pp. 288-298.
9 Medicare Payment Advisory Commission, “Geographic Variation in Per Beneficiary
Medicare Expenditures,” Report to the Congress: Variation and Innovation in Medicare,
(Washington: MedPAC, June 2003), pp. 3-15.

---

CRS-4
problems take various forms, including not only overuse, but also underuse and
misuse of services.10
Economics and Valuing Spending
Economics offers additional concepts for thinking about whether U.S. spending
levels are desirable or affordable. Despite high spending, we may conclude as a
society that it is worthwhile to devote the same, or even more, resources to health
care. This conclusion depends on preferences for health care, relative to other things.
If we value health care more than what we would otherwise produce with the same
resources, diverting resources to health care from other uses will increase social
welfare.
We also may conclude that spending levels are affordable based on the
observation that it is possible, in a growing economy, to spend more both on health
care and on other goods and services. As Table 2 shows, over the 1960-1999 period,
increasing national income was sufficient to support both rapid growth in per capita
spending for health care and growth in spending for items other than health care.
Whether our economy will be able to support a similar trend in the future depends
on the extent to which increases in health spending continue to outpace change in
GDP.11
Table 2. U.S. Spending on Health Care and Other Items
(in 1996 dollars)
1960
1970
1980
1990
1999
Per capita GDP (sum of
spending on health care and
items other than health care)
$12,764
$17,022
$21,271
$26,388
$31,962
Per capita spending on health
care
646
1,197
1,870
3,165
4,192
Per capita spending on items
other than health care
12,118
15,825
19,401
23,223
27,770
Source: Chernew et al., “Increased Spending on Health Care: How Much Can the United States
Afford?” Health Affairs, vol. 22, no. 4 (July/Aug. 2003), p. 9.
10 Mark R. Chassin, “Assessing Strategies for Quality Improvement,” Health Affairs, vol.
16, no. 3 (May/June 1997), pp. 151-161.
11 Michael E. Chernew, Richard A. Hirth, and David M. Cutler, “Increased Spending on
Health Care: How Much Can the United States Afford?” Health Affairs, vol. 22, no. 4
(July/Aug. 2003), pp. 15-25. Based on simulation analysis, the authors conclude health
spending will continue to be affordable through 2075 if real per capita growth in health care
costs exceeds real growth in GDP by 1%. If the gap is instead 2%, spending would be
affordable only through 2039.

---

CRS-5
Distribution Matters
Even if health spending is generally affordable for society, the cost of health
insurance and health care may be too much for certain individuals and families. For
example, in 2002 about 17.2% of Americans under age 65 went without health
insurance for the entire year. Low income individuals were more likely to be
uninsured: about one-third of those earning less than 150% of the poverty level, and
just over one-quarter of those with income between 150 and 199%, were uninsured,
compared with about one in ten people earning at least 200% of poverty.12
Given the cost of health insurance, these rates are not surprising. In 2002, the
average annual premium for family coverage under an employer-sponsored plan was
$7954, with the workers’ share of this amount averaging $2084.13 In the same year,
the poverty threshold for a family of four was $18,392.14
Even having insurance may not guarantee ready access to health care. For
example, according to MedPAC, although Medicare beneficiaries enjoy good access
overall, population subgroups report delaying care because of cost. Even after
controlling for income, health status, and other demographic variables, beneficiaries
with only Medicare are more likely to delay care than those with Medicare and
supplemental coverage of some sort. This finding is statistically significant for all
reported sources of supplemental coverage, including Medicaid, Medigap, employer-
sponsored, and health maintenance organization (HMO).15
Key Issue for the Congress
Regardless of whether America can afford to spend more of its national income
on health care, health spending is a key issue for the Congress both because it
12 CRS Report 96-891 EPW, Health Insurance Coverage: Characteristics of the Insured and
Uninsured Populations in 2002, by Chris L. Peterson. Based on data from the Mar.
Supplement to the Current Population Survey, 33.6% of those earning less than 100% of the
poverty level were uninsured in 2002. Rates for other income groups were: 32.2% (100-
149% of poverty), 26.8% (150-199%), and 11.3% (200% or more).
13 The Kaiser Family Foundation and Health Research and Educational Trust, Employer
Health Benefits, 2002 Annual Survey, Henry J. Kaiser Family Foundation, 2002, pp. 26 and
77.
14 CRS Report 95-1041 EPW, Poverty in the United States: 2002, by Thomas Gabe.
15 Medicare Payment Advisory Commission, “Access to Care in the Medicare Program,”
Report to the Congress: Medicare Payment Policy, (Washington: MedPAC, Mar. 2003), p.
167. For significance testing, MedPAC calculated adjusted odds ratios using pooled data
(1996-1999) from the Medicare Current Beneficiary Survey. The Commission also reported
the unadjusted proportion of beneficiaries delaying care because of cost, noted here for a
sense of magnitude. Of beneficiaries with only Medicare, 16.1% reported delaying care
because of cost, compared with 7.9% of beneficiaries with both Medicare and Medicaid.
Rates of delay for those with other sources of supplemental coverage were: 3.8% (Medicare
and Medigap), 2.9% (Medicare and employer-sponsored insurance), and 2.7% (HMO).

---

CRS-6
constitutes a substantial share of federal spending and because it affects all
constituents in one way or another.
Federal Spending
Medicare and Medicaid generally top the list of concerns about federal health
spending. According to the Congressional Budget Office, Medicare spending is
projected to be $297 billion in 2004, and the federal shares of spending for Medicaid
and the State Children’s Health Insurance Program are expected to be $174 billion
and $5 billion, respectively.16 The sum of these amounts, $476 billion, represents
about 21% of estimated federal outlays ($2.3 trillion) for 2004. Costs for Medicare
and Medicaid are expected to grow significantly as the population ages.
Federal tax expenditures for health benefits are also substantial. Although
difficult to measure, recent estimates by the Joint Committee on Taxation suggest
personal income tax expenditures for health benefits will exceed $100 billion in
2004.17 Most of this amount represents forgone revenue because employer-provided
health benefits are excluded from federal income and employment taxes. Other tax
expenditures include the itemized deduction for unreimbursed medical and dental
expenses above 7.5% of adjusted gross income, the deduction for health insurance
for the self-employed, and the deduction and exclusion for health savings accounts.
Federal spending on health benefits for military personnel, veterans, and federal
employees is expected to total $84 billion in 2004. This amount comprises outlays
of $28.5 billion for defense health benefits, $26.6 billion for veterans medical care,
and $28.7 billion for Federal Employees Health Benefits.18
In addition to the health and tax benefits noted already, program budgets for
Public Health Service agencies sum to $51 billion in 2004. This amount includes
$28.0 billion for the National Institutes of Health, $7.3 billion for the Health
Resources and Services Administration, $6.2 billion for the Centers for Disease
Control and Prevention, $3.6 billion for the Indian Health Service, $3.4 billion for
the Substance Abuse and Mental Health Services Administration, $1.7 billion for the
16 U.S. Congressional Budget Office, “CBO’s Current Budget Projections,” Mar. 8, 2004,
at [http://www.cbo.gov/showdoc.cfm?index=1944&sequence=0#table1], visited July 7,
2004.
17 Joint Committee on Taxation (JCT), Estimates of Federal Tax Expenditures for Fiscal
Years 2004-2008, Joint Committee Print #JCS-8-03, Dec. 22, 2003, p. 27. John Sheils and
Randall Haught of the Lewin Group also have estimated tax expenditures for health benefits.
See: “The Cost of Tax-Exempt Health Benefits in 2004,” Health Affairs — Web Exclusive,
Feb. 25, 2004, at [http://content.healthaffairs.org/webexclusives/index.dtl? year=2004], pp.
W4-106-W4-112 (PDF version). Estimates by JCT and Lewin are not precisely comparable;
nonetheless, estimates of 2004 expenditures under similar tax provisions total $111 billion
for JCT and $122 billion for Lewin.
18 Executive Office of the President, Office of Management and Budget, Historical Tables,
Budget of the United States Government, Fiscal Year 2005 (Washington: U.S. Government
Printing Office, 2004), p. 294.

---

CRS-7
Food and Drug Administration, and $0.3 billion for the Agency for Healthcare
Research and Quality.19
Constituents and Complexity
Influencing health spending is complicated. Broadly, the Congress must balance
fiscal constraints against the desire to help constituents. Beyond this general
challenge, the details can be mind-numbing: constituent groups often have
competing objectives, or at least different priorities; public and private actions are
highly interdependent; and policy actions inevitably have both intended and
unintended consequences.
For example, health spending and cost trends affect:
! Taxpayers, who pay for public benefits and tax subsidies;
! Individuals and families, who may receive coverage through public
programs, benefit from tax subsidies for health insurance, or find
themselves uninsured or underinsured because of the high cost of
health care;
! Employers, who must balance providing an attractive compensation
package, including health insurance, for employees against the need
to keep labor costs under control;
! States, who share responsibility with the federal government to
provide coverage for certain vulnerable populations;
! Insurers and health plans, who must balance offering attractive
products at reasonable prices against profit goals and the risk of
financial loss; and
! Health care providers, whose income depends on insurance
coverage and a functioning market for health care.
Together, these actors make up a complex market in which it is hard to discern
the beginning or end of public and private influences. Public programs depend on
private providers to deliver health care services; and they depend on private entities
to administer benefits, whether by processing claims or by providing private health
plan options for beneficiaries. The private insurance market in turn depends on
substantial tax subsidies to increase demand for coverage and make the price of
insurance more affordable for purchasers. Public subsidies — such as Medicare and
Medicaid payment add-ons for hospitals that train physicians or treat low-income
people — help ensure access to care not only for beneficiaries of public programs,
but also for uninsured and privately insured individuals. Ultimately, all policies
affecting public benefits influence the private market, and vice versa.
Given the complicated interdependence of actors, unintended consequences are
inevitable. For example, although Medicare and Medicaid have provided both
19 For this accounting, program budgets include both agency appropriations and funding
from other sources, including user fees, transfers between Public Health Service agencies,
and transfers both from the Department-level budget for Health and Human Services, and
from other federal Departments.

---

CRS-8
financial protection and access to care for millions of beneficiaries, the programs also
contribute to health care inflation because insured consumers are less price sensitive.
Similarly, expanding public benefits, as the Congress has done in enacting drug
coverage under Medicare and creating the State Children’s Health Insurance
Program, inevitably crowds out private spending, regardless of efforts by
policymakers to prevent this substitution. Maximizing the benefits and minimizing
the costs of any policy action is a difficult challenge.
Three Policy Directions
Rapid growth in health spending makes clear thought about policy strategy both
more difficult, and more important. Accordingly, the following pages introduce three
broad approaches for using policy to influence health spending and costs: changing
health care, changing federal programs, and using tax policy to make health care
more affordable.
These broad directions are neither mutually exclusive nor exhaustive. In
addition, controlling spending — whether national spending or federal spending —
is not assumed to be their only objective. As discussed above, devoting a high share
of national income to health care is not necessarily a problem. Nevertheless,
policymakers generally are concerned about whether health services are worth their
cost, as well as about how benefits and subsidies are distributed.
Changing Health Care
This broad direction — changing health care to increase its value and reduce its
underlying cost — focuses on the health system. The basic idea is that policy may
be able to help improve efficiency in the production and delivery of health care, and
in so doing lower the cost of health services. Benefits from lower costs would affect
both public and private health spending.
If health care were a car, policy goals might be building a better car for the same
price, building the same car at a lower price, or getting the car from factory to
customer more efficiently. For cars, the “invisible hand” of the market helps make
these things happen; but health care is different, its market complicated by insurance
coverage, poor information, and other problems.
Two hot policy topics: evidence-based medicine20 and health information
technology, are examples of promising tools for helping produce better health care
for the same price, the same health care at a lower price, and a more efficient delivery
system. Both offer the potential to improve quality by providing information,
including general information about effective treatment strategies, and specific
20 According to the Oxford Centre for Evidence-Based Medicine, “evidence-based medicine
is the conscientious, explicit and judicious use of current best evidence in making decisions
about the care of individual patients. The practice of evidence-based medicine means
integrating individual clinical expertise with the best available external clinical evidence
from systematic research.”

---

CRS-9
information about patients. Both also may help reduce costs by encouraging cost-
effective care and more efficient administrative systems.
However, given America’s fragmented and competitive health care system,
private entities generally have been unwilling or unable to invest sufficient capital
to realize the potential of these tools. Government might help by subsidizing
research and the development of systems, by disseminating information, and by
establishing standards to facilitate coordination.
This policy direction offers both promise and risk. Investment in the tools
described above and in other types of information, technology, and systems might
help reduce costs; or it might not. Although technological change tends to lower the
cost of most products, new discoveries in health care tend to have the opposite effect.
Still, inefficiency and waste in the U.S. health care system, like that implied by
relatively poor health statistics and regional variation in health spending, suggest
opportunities for improvement relative to multiple objectives, including reducing
spending, enhancing value, and possibly also rationalizing the distribution of
resources.
Changing Federal Programs
Whereas strategies to change health care focus on the health system generally,
a second broad direction — changing federal programs — focuses on federal
spending for federal benefits.
Given limited resources and growing costs, many policymakers note the need
to control spending on federal programs. But other goals, such as improving program
benefits and ensuring adequate payments for health care providers, are also
important. Medicare illustrates the tension from competing objectives. Over the past
decade, repeated legislative efforts alternately have emphasized limiting spending or
increasing spending, with most bills including provisions for doing both.
Whether the Congress seeks to reduce spending or not, policymakers have
different types of options for changing federal programs, including specifying
budgets, changing eligibility and/or benefits, and changing features that define how
programs work.
Program Budgets. In a way, the simplest tool for influencing federal
spending on health care is to set a budget. For example, the Congress limits outlays
on health benefits for veterans by specifying a budget through the appropriations
process. By changing the appropriation, Congress can reduce or increase spending
for this population.
The appropriation for veterans’ health care is an example of a program-level
budget, but other possibilities include budgets for certain services or beneficiaries.
For example, the Congress has attempted to control spending for physician services
under Medicare through the sustainable growth rate system, which in effect sets a
budget for spending on care by physicians; and it could limit federal spending for
individuals in entitlement programs through capitation payments. Of course the

---

CRS-10
details matter. Setting a budget can restrict spending; but if set too high, it also can
lead to higher spending than would occur otherwise.
In addition, a mismatch between funding, demand, and supply can lead to access
problems. In the veterans case, some argue that queuing for services is the result of
appropriations that have failed to grow in tandem with health care costs and
congressionally mandated changes in benefits. In the Medicare case, some
physicians threatened to stop seeing beneficiaries when the sustainable growth rate
system recently led to payment cuts.
Eligibility and Benefits. Another tool for influencing federal spending is
changing eligibility and benefits under entitlement programs. The Congress can use
this tool to reduce spending, but usually it has done the opposite. For example, over
the years, policymakers have expanded eligibility for Medicare and Medicaid,
notably in the former case to certain disabled persons and individuals with end-stage
renal disease, and in the latter case to successive subgroups of pregnant women and
children.
The Congress also has expanded benefits. Examples in Medicare include
coverage for hospice services and, more recently, for various clinical preventive
services and outpatient prescription drugs. In Medicaid, most new benefits have been
optional for states. Mandatory additions have included limited coverage for
professional services by non-physician providers (dentists, nurse midwives, and nurse
practitioners), coverage for care provided in rural health clinics and federally
qualified health centers, and coverage for family planning and pregnancy-related
services.
Changing cost sharing is another way policymakers can modify benefits. For
example, under the Balanced Budget Act of 1997 (BBA), policymakers increased
coverage for hospital outpatient services under Medicare by reducing beneficiaries’
liability for coinsurance. In contrast, the Congress increased beneficiaries’ financial
liability when it required future increases in the Part B deductible under the Medicare
Prescription Drug, Improvement, and Modernization Act of 2003.
Changes to eligibility and benefits have fairly straightforward tradeoffs. In
general, expansions increase access, but also spending. Restrictions reduce spending,
but may limit access. Distribution and incentives matter. While a policy change to
reduce covered services or increase cost sharing requirements might seriously limit
access for some beneficiaries, the same cutback likely would encourage others to be
appropriately prudent in seeking health services.
Other Program Features. The Congress can influence spending under
entitlement programs by changing program features other than eligibility or benefits.
Key tools include changing payment methods and amounts, and changing how
beneficiaries obtain coverage.
For example, over the years policymakers have changed payment methods for
most Medicare services. Beginning in 1983 and accelerating with the BBA, cost-
based payment has been abandoned in favor of prospectively determined rates for
hospital, physician, skilled nursing facility, home health, and other services. Under

---

CRS-11
prospective payment, providers have a greater incentive to be efficient because they
are at risk for costs above payments amounts, and can profit if costs are below
payment amounts.
In addition to encouraging efficiency through payment methods, policymakers
can influence spending by changing payment updates. For example, under the BBA,
Congress restricted payment updates for most Medicare services to control rapid
growth in spending. Since the BBA, Congress has increased updates on multiple
occasions to ensure adequate payments for providers.
Increasingly, policymakers have looked to private health plans to provide
benefits under public programs, including Medicare, Medicaid, and the State
Children’s Health Insurance Program. Some people emphasize the inherent value of
offering different coverage options for beneficiaries. Others argue that greater
reliance on private plans will reduce program spending because the plans can provide
benefits more efficiently. That outcome depends, among other things, on how much
private plans are paid.
Like changes in eligibility and benefits, changes in other program features must
balance competing goals regarding spending and access to care. Payment amounts,
whether for particular services or for all services under a health plan, must cover the
cost of efficiently caring for beneficiaries; and payment methods should encourage
the provision of adequate, but not wasteful, care.
Using Tax Policy to Make Health Care More Affordable
A third broad direction — using tax policy to make health care more affordable
for individuals and families — focuses more on consumers, compared with strategies
to change health care or federal programs. In this area, limiting tax expenditures
generally has not been a priority; instead, tax policies emphasize helping consumers
pay for health insurance and health care.
Insurance Subsidies. The subsidy for employer-provided health benefits
is by far the largest tax expenditure for private insurance. Payments for health
insurance are excluded from the income and employment tax base, effectively
lowering the price of insurance for those obtaining coverage under employer-
sponsored plans.21 With the annual cost of such plans averaging $3,383 for
individuals and $9,068 for families in 2003,22 and with most households facing
marginal income tax rates of 15 or 25 percent, savings can be substantial. On the one
hand, these savings make insurance more affordable.
On the other hand, the subsidy encourages people to purchase more insurance
than they would otherwise. Having more insurance drives up demand for health care,
21 For more information on this tax subsidy and others, see CRS Issue Brief IB98037, Tax
Benefits for Health Insurance: Current Legislation, by Bob Lyke.
22 The Kaiser Family Foundation and Health Research and Educational Trust, Employer
Health Benefits, 2003 Annual Survey, Henry J. Kaiser Family Foundation, 2003, p. 1.
Average premium costs include both employer and worker contributions.

---

CRS-12
which in turn drives up health care prices and spending. In addition, the exclusion
for employer-provided benefits disproportionately helps those who least need
assistance: high-income workers who face high marginal tax rates.
Increasingly, policymakers have shown interest in another type of subsidy for
purchasing insurance — tax credits. Depending on the details, tax credits arguably
offer potential advantages compared with the exclusion for employer-provided
benefits. First, credits are less regressive because subsidies are not a function of
marginal income tax rates. Second, because tax credits need not be tied to
employment, they provide a tool for expanding insurance coverage that can reach a
larger population.
Key disadvantages relate to access and complexity. Even credits of $1000 or
more may not be enough to make insurance affordable for the unemployed or lower-
income workers, especially if they must purchase insurance in the individual market,
which is generally more expensive. In addition, unless tax credits are carefully
targeted, subsidies may crowd out private spending without expanding coverage.
Finally, ensuring that low-income individuals receive subsidies when premiums are
due (as opposed to as much as a year later, when they file their taxes) is complicated.
Tax credits can be paid in advance, but doing so requires both more administrative
resources and sufficient knowledge and sophistication among potential beneficiaries.
Other Subsidies. In addition to helping consumers purchase insurance, tax
subsidies — including the itemized deduction for unreimbursed medical and dental
expenses and several tax-advantaged accounts — help consumers pay for health
expenses not covered by insurance. Like insurance subsidies, these subsidies reduce
the apparent cost of health care, and have the same unintended impact: increasing
demand, prices, and spending. They also provide larger benefits to taxpayers in the
highest brackets.
The deduction for unreimbursed medical and dental expenses is less regressive
than the subsidy for employer-provided insurance because eligibility is related to
income (taxpayers who itemize deductions can deduct expenses exceeding 7.5% of
adjusted gross income). In addition, because the deduction covers catastrophic costs,
some might regard it a higher-priority use of limited public dollars.
The policy trend favors tax-advantaged accounts to help consumers pay for
unreimbursed expenses. These accounts — health care Flexible Spending Accounts
(FSAs), Health Reimbursement Accounts (HRAs), Archer Medical Savings Accounts
(Archer MSAs), and Health Savings Accounts (HSAs) — differ on various
dimensions.23 But they are more similar than different, offering account holders
significant flexibility in using balances to cover health care expenses.
Two of the accounts, HSAs and Archer MSAs, were crafted with an eye to
limiting the impact of insurance on demand. Because the accounts must be used in
23 For more information, see CRS Report RS21573, Tax-Advantaged Accounts for Health
Care Expenses: Side-by-Side Comparison, by Bob Lyke and Chris L. Peterson.

---

CRS-13
conjunction with high-deductible health insurance plans, some believe the
combination will encourage consumers to be more prudent in seeking health services.
The underlying assumption — that consumers with savings accounts and high-
deductible plans will think twice before seeking discretionary health services — is
worth evaluation, although incentives under this arrangement are not completely
obvious. If consumers accrue large account balances over time, will they continue
to be price sensitive, or will they instead act as if they have first-dollar coverage?
And will high-deductible plans affect spending at all among consumers with health
care expenses that easily exceed even high deductibles?
Few consumers established MSAs, but interest in HSAs appears to be greater.
Regardless of how HSAs play out, the idea of influencing consumer demand by
favoring certain types of insurance coverage merits further analysis. Next steps
might include evaluating consumer behavior under different types of arrangements,
and developing strategies that target subsidies according to need.
Conclusion
The good news is that policymakers have a full toolbox for pursuing goals
regarding health care costs and spending. They can use government resources and
leadership to help improve health care. They can change federal programs to
influence both access to care and federal spending. And they can use tax policy to
support and shape the market for health insurance and health care.
The bad news is that both problems and solutions are complicated. Does the
United States spend too much on health care, or not? How should society allocate
its resources among members? And how should policymakers set priorities among
competing goals and interests? Even assuming agreement on these questions, the
Congress faces difficult challenges in choosing the best combination of policy tools
for achieving whatever objective is adopted.

---