Order Code RS22008
Updated January 25, 2007
Federal Spending for Older Americans
Analyst in Social Legislation
Domestic Social Policy Division
William Joseph Klunk
Domestic Social Policy Division
The federal government currently spends more than one-third of its budget on
benefits and programs for older Americans. For people age 65 and over, estimated
FY2007 spending for Social Security, Medicare, and Medicaid alone is $833 billion.
Spending for retired federal employees and military personnel and for veterans is also
significant, and smaller programs add to the total as well. Tax expenditures might also
be taken into account to obtain a more complete budget picture.
If current trends and policies continue, federal spending for older Americans is
expected to grow substantially. Contributing factors include a growing share of the U.S.
population being of retirement age, real increases in Social Security benefits due to wage
growth, and rising health care costs. Whatever the magnitude, this spending growth will
confront Congress with difficult choices. The share of the federal budget devoted to
older Americans is growing, leaving less available for other spending priorities and
leading many observers to warn that current policies are unsustainable because they
would require markedly higher taxation or debt levels that could adversely affect the
economy. Although Social Security spending can be altered via changes to the benefit
formula, options for reducing long-term growth in Medicare and Medicaid may be less
Background. The federal government provides income and other assistance to
older Americans through a number of programs. Among people age 65 or older, the most
significant support comes from three entitlement programs: Social Security, which pays
monthly cash benefits to about 34 million; Medicare, which provides health coverage to
about 36 million; and Medicaid, which provides health and long-term care coverage to
about 5 million.1 Other assistance for older Americans includes pension benefits and
Social Security reflects December 2005, Medicare reflects July 2005, and Medicaid reflects
health insurance for federal and military retirees, and similar benefits for certain veterans.
Additional assistance is provided to low-income individuals through the Supplemental
Security Income program and through funding for housing, nutrition, social services,
energy assistance, and other needs.
Relative to younger populations, federal spending for older Americans is high. The
Congressional Budget Office (CBO) has estimated that FY2000 federal spending on the
elderly was 3½ times the amount spent for families with children under 18. Historically,
state and local governments have taken the lead on spending for children through
education and other programs. Although public revenues for elementary and secondary
education are significant (representing about 4% of U.S. gross domestic product (GDP)
in the 2001-2002 school year), less than 10% come from federal sources.2
Measuring how much the federal government spends on older Americans depends
on how that population is defined. In this report, the term “older American” generally
refers to people age 65 and older; however, other ages could be used. The minimum age
could be as low as 62 (when Social Security retirement benefits are first available), 59½
(when the early withdrawal penalty no longer applies to qualified retirement plans), 55
(when federal employees may retire if they have sufficient years of service), or as high as
67 (when future retirees will be eligible for full Social Security benefits). While the age
chosen would affect dollar estimates, the general relationship between spending for older
Americans and the federal budget or U.S. economy would not change.
Another measurement issue is that support and benefits for older Americans can take
many forms. Cash payments are direct benefits, whereas tax expenditures — not all of
which may be captured in available data sources — are indirect. Even when federal
support is measurable and clearly benefits a particular age group, other people may
indirectly gain (for example, income and medical benefits for the elderly may save their
children from these costs). Although a more complete picture would consider all of these
issues, this report focuses on direct cash outlays for older Americans in a given year.
Current Outlook. A significant — and growing — portion of the federal budget
and the U.S. economy is devoted to spending on older Americans. In FY1971, $46 billion
was spent on people age 65 or older, which was 22% of the federal budget and 4% of
GDP. In FY2007, the same spending figure is expected to reach $935 billion, or nearly
35% of the federal budget and 7% of GDP. (Dollar figures are nominal, not adjusted for
inflation.) Table 1 shows how these figures have changed over time. CBO projects that
unless policy changes are made, Social Security and Medicare alone will account for
about 15% of GDP in 2050.3
average monthly in 2005. Although Social Security and Medicare also cover people under age
65 (primarily those who are disabled), older Americans comprise a majority of enrollees in both
programs. In contrast, about 90% of Medicaid enrollees are under age 65.
U.S. Department of Education, National Center for Education Statistics, The Condition of
Education 2005, 2005 and Revenues and Expenditures for Public Elementary and Secondary
Education: School Year 2002-03, Oct. 2005.
CBO, The Long-Term Budget Outlook, Dec. 2005. Represents scheduled (rather than payable)
Table 1. Estimated Federal Spending on People Age 65 or Older,
Billions of dollars
Total federal spending on age 65+
Total as a % of federal budget
Total as a % of GDP
Average annual growth between periods
Total federal spending on age 65+
Source: CBO, Federal Spending on the Elderly and Children, July 2000 and CRS estimate for FY2007.
Note: Dollar figures are nominal (not adjusted for inflation). All other includes Federal Civil Retirement,
Military Retirement, Veterans’ Compensation, Supplemental Security Income, and several programs that
contribute less than 1% each to overall spending. Figures do not include the state share of total Medicaid
spending, which averages about 43%. FY2000-FY2007 Medicaid growth is low in part due to the shifting
of drug costs for seniors to Medicare Part D in 2006; in addition, based on an examination of actual data
that were not available to CBO in July 2000, it appears that the FY2000 Medicaid figure is a slight
overestimate. After FY2000, the “all other” category is assumed to grow at the rate of inflation as measured
by the Consumer Price Index-All Urban Consumers. Figures may not sum due to rounding.
Between FY1971 and FY2007, Social Security, Medicare, and Medicaid accounted
for 80% to 90% of federal spending on older Americans. Figure 1 shows that as
percentage of total federal spending on older Americans, Social Security and “all other”
spending is decreasing, while Medicare spending is increasing. Medicaid’s share has
remained fairly level. Figure 2 shows the dollar growth in each program.
The American population is aging, and a larger portion will receive Social Security,
Medicare, and other age-related federal benefits in the future. At the same time, a smaller
portion of the population will be of working age. Because Social Security and part of the
Medicare program are financed through payroll taxes paid by current workers, the
financial health of these programs is sensitive to the ratio of beneficiaries to workers,
sometimes called the dependency ratio.4 In addition to payroll taxes, growth in other
outlays for Social Security and intermediate spending assumptions for Medicare.
CRS Report RL32981, Age Dependency Ratios and Social Security Solvency, by Laura B.
revenue sources (e.g., income taxes) may slow as well. Under current policies, there is
concern that a combination of upward pressure on spending and downward pressure on
revenues could lead to large federal deficits, rising debt, and a weakened economy.5
Figure 1. Estimated Federal Spending on People Age 65 or Older
by Program Share, FY1971-FY2007
Source: See Table 1.
Figure 2. Estimated Federal Spending on People Age 65 or Older,
Source: See Table 1.
CBO, The ABCs of Long-Term Budget Challenges, remarks by Donald B. Marron, Dec. 8, 2006.
Drivers of Spending Growth. Increases in federal spending for older Americans
are being driven by two factors: growth in program enrollment and growth in spending
per beneficiary.6 The number of Social Security, Medicare, and Medicaid beneficiaries
will increase as the U.S. population ages. At the same time, spending per beneficiary will
increase because of price inflation and “real” growth in benefit costs (that is, growth that
exceeds price inflation). While general price inflation grew at an average annual rate of
3.0% between 1985 and 2005, Social Security spending per beneficiary grew 0.8
percentage points faster (at a rate of 3.8%) and Medicare spending per beneficiary grew
3.5 percentage points faster (at a rate of 6.5%).7
Social Security benefits are adjusted annually to reflect price inflation, allowing
current beneficiaries to maintain purchasing power from year to year. Social Security
benefits also grow in real terms as beneficiaries are added to the program, because initial
benefit levels for newly eligible individuals are adjusted for wage growth in the economy,
which generally exceeds price inflation. (Once a person’s initial benefit level is
determined, the annual adjustment for price inflation applies.)
As with health care costs in the private sector, real growth in Medicare and Medicaid
benefit costs are driven by increases in utilization and the complexity or “intensity” of
services provided. In turn, both utilization and intensity are driven by advances in
medical technology, changes in medical practice, and changes in the beneficiary
population. For example, Medicare beneficiaries — whose average age is increasing —
may visit doctors more frequently. In turn, doctors may order a greater number of
diagnostic tests or offer treatments — such as drug-coated cardiac stents — that were not
available 10 years ago. Legislative changes, such as the establishment of a Medicare Part
D drug benefit, also affect benefit costs.
CBO estimates that over the next 10 years, only about one-third of the projected
increase in Social Security spending and about 30% of the projected increase in Medicare
and Medicaid spending can be attributed to increasing enrollment.8 The remainder can
be attributed to increases in spending per beneficiary. Under intermediate assumptions
for 75-year projections by the Social Security and Medicare trustees, less than one-third
Note that figures in the remainder of this report reflect total spending and enrollment for Social
Security and Medicare and do not attempt to separate out beneficiaries under age 65 (primarily
disabled individuals) who account for less than a third of current spending on each program.
Based on the Consumer Price Index-All Urban Consumers from the Bureau of Labor Statistics
and data from the Social Security and Medicare trustees. Benefit growth that exceeds GDP
growth (rather than price inflation) is also an important concept used in Medicare and Medicaid
cost projections, which are highly uncertain (Social Security cost projections face less uncertainty
because benefit levels are formula-driven). Over the past several decades, both Medicare and
Medicaid spending per beneficiary have outpaced per capita GDP growth by one or more
percentage points — a phenomenon referred to as “excess cost growth.” Although it has slowed
in recent years, any level of excess cost growth means that Medicare and Medicaid will continue
to consume a growing share of U.S. economic output (in excess of what is expected due to
growth in the share of the U.S. population enrolled in these programs). Under intermediate
assumptions that annual growth in Medicare spending per beneficiary will outpace per capita
GDP by one percentage point, CBO projects that the share of GDP devoted to Medicare will rise
from 3.0% in 2006 to 8.6% in 2050. See CBO, The Long-Term Budget Outlook, Dec. 2005.
CBO, The Budget and Economic Outlook: Fiscal Years 2008 to 2017, Jan. 2007.
of total spending growth in these two programs can be attributed to increasing
Implications. The share of the federal budget devoted to older Americans is
growing, leaving less available for other spending priorities and leading many observers
to warn that current policies are unsustainable because they would require markedly
higher taxation or debt levels that could adversely affect the economy. Although the share
of the U.S. economy devoted to federal spending for older Americans is also growing, the
extent to which this is a problem depends on one’s view of how the country’s economic
resources should be allocated.
It is likely that some combination of benefit reductions (for example, through
changes in eligibility, payment levels, covered health care services, etc.) and tax increases
— which could have a direct impact on individual Americans — will be necessary to
bring federal spending to a sustainable level. As policy options are proposed, Congress
may consider the potential effects on individuals from a number of perspectives. For
example, if future Social Security benefits are reduced relative to current law, but many
people still enjoy higher total income in retirement than today’s beneficiaries due to rising
wages and overall economic growth, are they worse off? One way to think about this
question might be in terms of poverty rates. Another might be in terms of how the income
of older Americans compares to that of working age people, since disparities would affect
the standard of living that different generations are able to afford. Yet another way might
be in terms of whether current and soon-to-be retirees are spared from policy changes,
with any necessary tax increases or benefit cuts falling to their children instead.
Viewed from any perspective, growth in federal spending for older Americans
presents issues that call for congressional action, which is likely to be program-specific
in nature. Social Security spending can be altered via changes to the benefit formula, but
options for reducing long-term growth in Medicare and Medicaid may be less
straightforward. As discussed earlier, spending for these two programs is driven by a
variety of factors, many of which are difficult for policymakers to control. Although no
single path to slower health care cost growth presents itself, a few of the potential options
include curtailing enrollment, making beneficiaries pay more for their care, limiting
coverage of new items and services, and changing provider payments. Because many of
these options could simply shift costs to other public and private payers, Congress may
consider whether the goal of any particular reform is to reduce federal health care
spending, to reduce economy-wide health care spending, or both.10
Authors’ calculations based on 2006 projections.
See discussion in CRS Report RL32747, Social Security and Medicare: The Economic
Implications of Current Policy, by Marc Labonte.