Order Code RL34424
Trends in Discretionary Spending
March 26, 2008
D. Andrew Austin
Analyst in Economic Policy
Government and Finance Division

---

Trends in Discretionary Spending
Summary
Discretionary spending by the federal government is defined as non-entitlement
spending that is authorized through appropriations acts. Discretionary outlays
increased, on average, 5% a year in real terms from FY1999 to FY2008. The share
of discretionary spending as a proportion of total federal outlays rose from 33.6% in
FY1999 to an estimated 38.2% in FY2008. Higher defense outlays accounted for
75.6% of the increase in discretionary spending in the past decade.
From FY1962, when discretionary spending accounted for 47.2% of total
outlays, to FY2007, discretionary spending’s share of federal outlays has fallen by
almost a fifth. Rapid growth of entitlement outlays and slower growth in defense
spending relative to other federal spending are two major causes of the long-term fall
in discretionary spending’s share of federal spending. FY2007 federal spending
accounted for a fifth (20.0%) of the U.S. economy, nearly equal to its average share
of gross domestic product (GDP) since FY1962. FY2008 discretionary spending is
estimated to account for 7.9% of GDP.
Discretionary spending is often divided into defense, domestic discretionary,
and international outlays. Defense and domestic discretionary spending comprise
nearly all of discretionary spending. In FY1962, when discretionary spending was
12.3% of GDP, defense spending was 9.0% of GDP. In FY2007, defense spending
was 4.0% of GDP, because defense spending has grown more slowly than GDP.
In the last six years, military spending has increased sharply. On average, from
FY1999 to FY2008, defense outlays grew 8% per year in real terms, while non-
defense discretionary outlays grew 2.5% per year in real terms.
Projected future growth in entitlement program outlays may put severe pressure
on discretionary spending unless policies change or federal revenues are increased.

This report will be updated annually.

---

Contents
What Does Discretionary Spending Include? . . . . . . . . . . . . . . . . . . . . . . . . 1
Trends in Discretionary Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Discretionary Defense Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Discretionary Non-defense Spending . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Discretionary Spending in the President’s FY2009 Budget . . . . . . . . . . . . . . 5
Discretionary Defense Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Discretionary Non-defense Spending . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Total Discretionary Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Discretionary Spending in the Long Term . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
List of Figures
Figure 1. Components of Federal Spending As Percentage of GDP,
FY1962-FY2008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Figure 2. Discretionary Spending By Type As Percentage of GDP,
FY1962-FY2008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
List of Tables
Table 1. Categories of Federal Spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

---

Trends in Discretionary Spending
What Does Discretionary Spending Include?1
Discretionary spending refers to outlays authorized through appropriations acts.
This funds many of the activities commonly associated with federal government
functions such as running executive branch agencies, congressional offices and
agencies, and international operations of the government. Essentially all spending
on federal wages and salaries is discretionary.2 On the other hand, essentially all
federal retirement and disability costs are classified as mandatory spending.
Discretionary spending is often contrasted with mandatory, or direct, spending,
which the Budget Enforcement Act (BEA) of 1990 (P.L. 101-508) defines as (1)
budget authority provided by law other than appropriation acts; (2) entitlement
authority; and (3) the food stamp program. Thus, while spending for some
entitlement programs, such as Medicaid and some veterans’ programs, is controlled
through appropriations acts, such outlays fall within the category of mandatory
spending in Congressional Budget Office (CBO) summaries. Table 1 illustrates
these budget categories.3
Table 1. Categories of Federal Spending
Budget Authority
Budget Authority Provided by
Provided by Law Other
Appropriation Acts
than Appropriation Acts
— Appropriated Entitlements
— Medicare
(e.g., veterans’ compensation,
Entitlement
— Social Security
Medicaid, TANF)
— Food Stamps (with caveats)
— Mandatory non-
— Discretionary Spendinga
Not an
entitlements
(defense, non-defense domestic
Entitlement
(e.g., Forest Service
discretionary, and international)
payments to states)
Note:
a. Discretionary spending programs. See discussion in text.
1 This report was originally written by Philip Winters, who has retired from CRS.
2 Salaries for Members of Congress and the President are classified as mandatory spending.
3 The Office of Management and Budget (OMB) uses a slightly different classification of
discretionary and mandatory spending.

---

CRS-2
Trends in Discretionary Spending
The composition of the federal budget has changed dramatically since the early
1960s. Total discretionary spending’s share in federal spending has fallen by almost
a fifth since FY1962, when discretionary spending accounted for just under half
(47.2%) of total outlays. While discretionary spending was the largest component
of federal spending until the mid 1970s, mandatory spending now accounts for about
3% more of GDP than discretionary spending. Figure 1 shows components of
federal spending as a percentage of GDP since FY1962. Measuring budget
components as a share of GDP compares their size with the economy as a whole.
Figure 1. Components of Federal Spending As Percentage
of GDP, FY1962-FY2008
Source: CBO data, CRS calculations. Data for FY2008 are estimated.
In the past decade, discretionary spending has grown more quickly than
mandatory spending. Social Security spending, measured as a share of GDP, has
been relatively stable in the last decade, while Medicare and Medicaid have
continued to grow faster than overall federal spending. Discretionary spending
increased 5.1% a year in real terms on average from FY1999 to FY2008, raising the

---

CRS-3
share of discretionary spending as a proportion of federal outlays from 33.6% in
FY1999 to 38.2% in FY2007.4
Discretionary Defense Spending. Since the late 1960s, discretionary
spending’s share in federal spending fell as entitlement spending grew rapidly and
defense spending grew more slowly than federal spending in other areas. Some
contend that defense expenditures as a proportion of GDP should be expected to fall
over the long term because the cost of defending the nation depends on factors that
are independent of economic growth.
Figure 2 shows discretionary spending (separated into categories of defense,
domestic non-defense, and international) as a share of GDP over the period FY1962-
FY2008. Defense spending, 73% of discretionary spending in FY1962, accounted
for 52.5% of discretionary spending in FY2007. Discretionary defense spending fell
from 9.5% GDP in FY1968 to 4.0% in FY2007. Defense spending increased,
however, during the early 1980s when Cold War tensions intensified.
Figure 2. Discretionary Spending By Type As Percentage
of GDP, FY1962-FY2008
Source: CBO historical data, Analysis of President’s Budget Proposals for FY2009, March 2008.
Note: FY2008 data are estimates and include requested and enacted supplemental funding.
4 CBO figures for discretionary and mandatory spending and their components are used in
all calculations. The Bureau of Economic Analysis GDP deflator is used to adjust for
inflation.

---

CRS-4
Defense spending increased sharply in the mid-1960s as the United States’
involvement in Vietnam deepened. After large-scale withdrawals of American troops
from Vietnam began in 1969, defense spending as a share of GDP fell for the next
decade. The Soviet invasion of Afghanistan prompted the Carter Administration, and
then the Reagan Administration, to boost military expenditures. After the Berlin
Wall was opened in November 1989 and communist governments in central and
eastern Europe collapsed, defense spending as a share of GDP dropped to historically
low levels, providing what some called a “peace dividend.”
Defense spending again rose after the attacks of September 11, 2001, and the
start of wars in Afghanistan and in Iraq. Discretionary defense spending, which had
fallen to 3.0% of GDP in FY1999, FY2000, and FY2001, rose sharply to 4.0% of
GDP in FY2005, FY2006, and FY2007. If the Administration’s defense proposals
and supplemental funding requests are approved, the defense share of discretionary
spending in FY2008 would reach 4.2% of GDP. The trajectory of future defense
discretionary spending depends in large part on the scale of future operations in Iraq
and Afghanistan.
Recent increases in discretionary spending are largely due to growing military
costs. Increased defense spending accounts for 75.6% of the increase in discretionary
spending in the past decade. Discretionary defense spending increased 8% a year in
real terms on average between FY1999 and FY2008.5 Defense spending increases,
however, were concentrated in the past six years, while the increases in non-defense
discretionary spending have been more gradual.
Discretionary Non-defense Spending. Trends for domestic non-defense
discretionary spending are less dramatic. Non-defense spending supports the largest
number of federal agencies and programs, including foreign affairs, science and
technology research, natural resources, energy, education, and numerous others.
None of the individual programs within the non-defense discretionary category have
approached 1% of GDP since FY1962. Most of these programs spent less than 0.5%
of GDP and many spent less than 0.2% of GDP during that period. Discretionary
spending for international programs since FY1968 has run 0.5% or less of total
outlays.
Domestic non-defense discretionary spending, 2.5% of GDP in 1962, rose to a
peak of 4.8% in 1980. Domestic non-defense discretionary spending’s share of GDP
fell during the Reagan Administration, reaching 3.1% of GDP in FY1987.6 Since
then it has fluctuated between 3.0% and 3.6% of GDP. FY2008 domestic non-
defense discretionary spending is estimated to account for 3.25% of GDP.
Although discretionary spending has increased rapidly in the past half decade,
spending in some government departments and agencies has grown very slowly or
has been cut, while spending in other areas has expanded rapidly. After the attacks
5 FY2008 data are estimated. Defense discretionary spending includes supplemental
requests.
6 For a summary of Reagan-era spending changes, see David Stockman, The Triumph of
Politics (New York: Harper&Row, 1986), pp.401-411.

---

CRS-5
of September 11, 2001, funding for defense and homeland security increased sharply
and has continued at levels much higher than in the 1990s. Funding for emergency
and disaster management increased sharply in the wake of hurricane Katrina,
boosting discretionary spending still further. In recent years, disaster funding has
dropped sharply, allowing non-defense spending as a share of GDP to fall as well.
Discretionary Spending in the President’s FY2009 Budget
The President’s FY2009 budget called for limiting growth in defense and non-
defense discretionary spending during the five years FY2009-FY2013. The proposals
would reduce total discretionary spending in dollars, in real terms, as a share of GDP,
and as a percentage of total federal spending over the five years.
Discretionary Defense Spending. Proposed discretionary defense
spending would rise from $549.2 billion in FY2007 to $670.2 billion in FY2009.
Although most of the defense budget comprises discretionary spending, it also
includes certain mandatory spending, such as contributions to military retirement
programs. Congress approved $86.8 billion in supplemental funding for FY2008.
Another $102.5 billion requested by the Administration has not been approved.7 In
addition, the Administration has requested $70 billion in emergency supplemental
funding for FY2009. The Secretary of Defense has stated that an additional
supplemental request for FY2009 will be made later in 2008.8
After FY2009, proposed defense outlays fall, although this is largely due to the
omission of war costs after FY2009 in the President’s budget proposals. Continuing
current levels of war funding over the coming five years would slow the fall in
defense spending as a percentage of GDP. Discretionary defense spending is
projected to fall to 3.1% of GDP in FY2013, according to the Administration.9
Discretionary Non-defense Spending. The Administration has made a
priority of holding down growth in “non-security” discretionary spending.10 Non-
defense spending is slated by the Administration to fall to $567.5 billion by FY2013.
7 U.S. Dept. of Defense, Office of the Assistant Secretary for Defense (Public Affairs), Press
Release 90-08, “FY2009 Department of Defense Budget Released,” available at
[http://www.defenselink.mil/comptroller/defbudget/fy2009/2009_Budget_Rollout_Relea
se.pdf].
8 Testimony of Secretary of Defense Robert M. Gates, in U.S. Congress, House Armed
Services Committee, 110th Cong., 2nd sess., Feb. 6, 2008, available at
[http://armedservices.house.gov/pdfs/FC020608/Gates_Testimony020608.pdf]
9 General Mike Mullen, Chairman of the Joint Chiefs of Staff, noted in an interview that the
defense budget was now “just under 4% of GDP,” and stated that “... I would see that in the
future as an absolute floor.” New York Times, Oct. 22, 2007. Transcript available at
[http://www.nytimes.com/2007/10/22/washington/22mullen-text.html]
10 The Administration defines “security” discretionary funding as spending on the
Department of Defense, Government-wide Homeland Security activities, and International
Affairs.

---

CRS-6
Total Discretionary Spending. CBO baseline projections show total
discretionary spending falling as a share of GDP between 2008 and 2017.11 Both
defense and non-defense discretionary spending are smaller as shares of GDP and
total outlays at the end of the projection period (2017) than they were at the
beginning. CBO projects that defense and non-defense discretionary spending would
both reach historically small shares of both GDP and total outlays over the next
decade. The falling share of discretionary spending in GDP reflects assumptions that
budget legislation has required for the calculation of CBO baseline projections.
Many independent analysts assume that discretionary spending’s GDP share will
remain roughly constant in the future.
Discretionary Spending in the Long Term
If discretionary spending were held constant in real terms, as the CBO baseline
presumes, then discretionary spending per capita would decrease as population grows
and discretionary spending as a share of GDP would shrink as the economy grows.
On average the U.S. population grew 1% a year and per capita GDP grew 2.25% per
year from 1962 to 2005.12 If those trends were to persist, then holding discretionary
spending constant in real terms implies per capita discretionary spending would
shrink by 1% a year and discretionary spending as a share of the economy would
shrink by 3.25% per year.
Choices of voters and political institutions that translate voters’ preferences into
policy outcomes ultimately determine spending decisions in a representative
democracy. Economists typically think that voters’ views about spending are
influenced by incomes, which represent the resources available to fund both private
consumption and publicly provided goods and services; the relative prices of inputs
which the government buys and the efficiency with which the government can
transform those inputs into outputs; and the willingness of voters to pay for publicly
provided goods and services through higher taxes. Discretionary spending, by
definition, flows from funds appropriated by Congress. The annual budget process,
and the appropriations decisions in particular, provide Congress with the ability to
change, continue, or reverse trends in discretionary spending.
Economists have well-developed tools for predicting future short- and medium-
term changes in national income, but less-well-developed tools for predicting other
determinants of voters’ choices. Therefore, many economists, as a rough
approximation, assume that future discretionary spending will increase at the same
rate as growth in national income. Put another way, this approach assumes that
discretionary spending per capita will grow as fast as the growth of per capita
income.
Over the longer run, growth in entitlement spending will present voters and their
representatives in Congress with severe fiscal challenges. Recent research on long-
term fiscal challenges has focused on continued increases in the per beneficiary cost
11 CBO. Analysis of the President’s Budget for FY2009, March 2008.
12 These figures based on CRS calculations derived from CBO and Census Bureau data.

---

CRS-7
of health care, as well as the more predictable demographic changes that will occur
as the baby boom generation retires. Projections from a variety of sources predict
that spending on Social Security and Medicare will increase sharply as a share of
GDP in coming decades.13 CBO projects that Medicare will expand from 5.6% of
GDP in 2030 to 14.8% in 2082 and that Social Security will grow from 6.1% of GDP
in 2030 to 6.4% of GDP by 2082. Federal Medicaid outlays, 1.4% of GDP in
FY2007, are projected to reach 17.6% of GDP in 2050.14 By contrast, FY2007
federal spending was 20% of GDP and revenues were 18.8% of GDP. Thus, Social
Security, Medicare, and Medicaid outlays as a proportion of GDP are projected to
equal or exceed the share of present federal revenues as a proportion of GDP
sometime after the middle of the 21st century. Maintaining current levels of
discretionary spending would then require either substantial tax increases or major
changes in those entitlement programs.

13 See CBO, “The Long-Term Budget Outlook,” Dec. 2007; point 10 of the United States of
America — 2006 Article IV Consultation, Concluding Statement of the IMF Mission, May
31, 2006; and CRS Report RL33623, Long-Term Measures of Fiscal Imbalance, by D.
Andrew Austin.
14 CBO, “The Long-Term Budget Outlook: An Update,” Dec. 2007.

---