Order Code RL30854
CRS Report for Congress
Received through the CRS Web
Uncertainty in Budget Projections
February 14, 2001
Philip D. Winters
Analyst in Government Finance
Government and Finance Division
Congressional Research Service ˜ The Library of Congress

---

Uncertainty in Budget Projections
Summary
January 2001 budget reports from the Office of Management and Budget (OMB)
and the Congressional Budget Office (CBO) project new 10-year (FY2002-2011)
cumulative total baseline surpluses ranging from $4.996 trillion (OMB) to $5.610
trillion (CBO). The projections assume no changes in current policy, follow the
constraints on baseline estimates imposed by law, include assumptions about
underlying economic (and some budget) conditions over the next 10 years, and
include assumptions about how the budget will interact with those underlying
economic conditions.
These projections, as acknowledged by both agencies, are not meant to predict
future budget outcomes. They provide a set of numbers against which proposed and
adopted policy changes can be measured. CBO in its January 2001 budget report is
very clear about the uncertainties surrounding its “midrange” baseline budget
projections for the fiscal years 2002 through 2011. In its chapter on budget
uncertainty, it states,
... the outlook for the budget can best be described not as the single row of
numbers presented in CBO tables but as a fan of probabilities around those
numbers. The fan widens as the projection extends.1
Budget estimates and projections cannot be more accurate than the information
that is needed to create them. This underlying information – future economic
conditions and how the budget and the economy will interact in the future – cannot
be known with certainty since it comes from assumptions about the future. The
outcome of the budget estimating process is the best guess of the likeliest outcome
for the government’s receipts, outlays, and surplus (or deficit) with the expectation,
nonetheless, that even the best guess is likely to differ from the eventual actual
amounts.
One can get a sense of how uncertain budget projections are by examining some
of those made previously. In January 1992, CBO projected a deficit of $322 billion
for fiscal year (FY) 2000. The budget in FY2000 actually produced a surplus of $236
billion. This substantial change resulted almost exclusively from differences between
the underlying assumptions – future economic conditions and the interaction of the
economy and the budget – used in the original projections and those that actually
occurred. CBO’s budget reports indicate that very little of the budgetary change over
those years resulted from legislation.
1CBO, The Budget and Economic Outlook: Fiscal Years2002-2011, Jan. 31, 2001, p. 93.

---

Contents
The Inevitability of Uncertain Budget Projections . . . . . . . . . . . . . . . . . . . . 2
Fiscal Year 2000 Projections and Estimates . . . . . . . . . . . . . . . . . . . . . . . . 3
Current Estimates and Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Reasonable Variations in Factors Affecting the Budget . . . . . . . . . . . . 7
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
List of Figures
Figure 1. CBO Projections for FY2000 Beginning in January 1992 . . . . . . . . . . 3
Figure 2. Administration Projections for FY2000 Beginning in February 1995 . 4
Figure 3. Surplus Projections, FY2000-2011 . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 4. CBO’s Alternative Scenarios, FY2000-2011 . . . . . . . . . . . . . . . . . . . 8
Figure 5. CBO Projections of the Percentage Change in Real GDP for Calendar Year
2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Figure 6. Total Receipts, Total Outlays, and Discretionary Spending as Percentages
of GDP, 1965-2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

---

Uncertainty in Budget Projections
Part of the budget debate over the last two years has centered on what should
be done with projected surpluses, particularly the on-budget surpluses, over the next
10 years.2 The official budget projections show potentially huge surpluses during the
coming decade. The Clinton Administration’s final budget document, released in mid-
January 2001 by the Office of Management and Budget (OMB), projected that
baseline total surpluses would rise from $237 billion in 2000 to $810 billion in 2011
or from 2.4% of gross domestic product (GDP) in 2001 to 4.6% of GDP in 2011.3,
4 The Congressional Budget Office (CBO) January 2001 budget report projects a
2011 baseline total surplus of $889 billion.5 The cumulative 10-year total surplus
(2002 through 2011) as projected by OMB and following the constraints built into the
baseline projections, reaches $4,996 billion, split between $2,445 billion in cumulative
on-budget surplus and $2,551 billion in cumulative off-budget surplus.6 CBO’s 10-
year cumulative surplus reaches $5,610 billion, split between $3,122 billion in on-
budget surpluses and $2,488 billion in off-budget surpluses.
Proposals to use the on-budget surpluses range from large tax cuts to various
spending increases, to no policy changes that would retain the whole surplus to
reduce federal debt.7 Any use of the surpluses for tax cuts or spending increases takes
funds away from debt reduction. Proposals that have appeared in the last two years
would use some portion of the projected on-budget surplus for spending increases or
tax cuts, but would leave all of the off-budget and a significant part of the on-budget
surplus for debt reduction. Opponents of the proposals to use large amounts of the
on-budget surpluses point to the likelihood that, among other things, the proposed
policy changes will have results different from those expected because the proposals
and the expected results are based on 10-year budget and economic projections that
are themselves inherently uncertain. They point out that CBO stated in its January
2001 budget report that:
2On-budget surpluses exclude the surpluses of Social Security and the Postal Service.
3The baseline projections assume no changes from current policy. They include adjustments
for inflation in discretionary spending and eligible population growth. They assume that most
expiring budget provisions (both tax and mandatory spending), will expire.
4Unless otherwise indicated, all years are fiscal years.
5CBO, Economic and Budget Outlook:FY2001-2011, Jan. 31, 2001.
6OMB, FY2002 Economic Outlook, Highlights from FY1994 to FY2001, FY2002 Baseline
Projections, Jan. 16, 2001.
7Surpluses will reduce federal debt held by the public. They do not affect federal debt held
by (federal) government accounts.

---

CRS-2
Looking forward five or 10 years allows the Congress to consider the longer-term
budgetary implications of policy changes. But it also increases the likelihood that
budgetary decisions will be made on the basis of projections that later turn out to
have been far wrong.8
There is concern that a large change in policy now, especially one that uses much
of the on-budget surplus, based on inherently wobbly 10-year budget projections,
might lead to serious budgetary or fiscal problems sometime during or after that
period. This part of the argument about whether or not to institute major policy
changes has little to do with the value of the proposed changes to society. Instead,
the focus is on an uncertain 10-year budget outlook on which to base the decision.
If one accepts the (relative) accuracy of the budget projections, the debate over policy
changes moves to whether they are good or bad for the Nation, along with the
debates over the details of the proposal(s). If one worries about the (relative)
accuracy of the budget projections, then the debate may remain focused on the
possible appropriateness or inappropriateness of major long-term policy change.9
The Inevitability of Uncertain Budget Projections
The future is unknown. Attempts to predict the future are inherently uncertain.
The budget estimates and forecasts produced by CBO and OMB, especially those
looking five to 10 years into the future are unlikely to be any more than
approximations of what that future might look like. Assuming that these projections
are accurate or that they approximate the most likely future outcome, they may lead
to substantial errors in policy decisions. CBO includes a chapter, The Uncertainty of
Budget Projections, in its January 2001 budget report.
The many difficulties of producing the estimates and projections are generally
acknowledged by both CBO and OMB in their respective budget reports. In 2001,
CBO included two alternative budget projections for its mid-range baseline estimate
in its budget report. The “alternative scenarios” were based on substantial differences
in the economic and technical assumptions underlying the projections.10 These
alternatives produce large differences in budgetary outcomes. OMB documents in the
President’s budget show the effects on the budget of changes in selected economic
variables such as real economic growth, the rate of inflation, or interest rates over five
years.
The alternative budget projections from CBO hint at the difficulties these
organizations face in producing budget estimates and projections. The complexities
of the budget itself need to be meshed with the budget’s interactions with and
dependence on the assumptions about the economy and the technical components of
8CBO. The Budget and Economic Outlook: Fiscal Years 2002-2011. January 2000. p. 94.
9The budget and the economy will come under severe pressure during the second decade of
this century as the baby-boom population begins to retire. Some have argued that keeping the
surpluses intact for debt reduction and for their positive effect on national savings may
mitigate, but not eliminate, the future pressure on the budget.
10See the discussion in chapter 5, The Uncertainty of Budget Projections, in the Budget and
Economic Outlook: Fiscal Years 2002-20110, Jan. 2001.

---

CRS-3
budgeting to produce budget estimates and projections. Both CBO and OMB have
developed budget models that attempt to reflect the budget and its interactions with
the economy and the population. They use past patterns of interactions combined
with economic and other expectations to produce the short-term estimates and longer-
term projections of the government’s budget.
Fiscal Year 2000 Projections and Estimates
Fiscal year 2000 ended on September 30, 2000. The actual budget results for
that year put the surplus at $326 billion, total receipts at $2,025 billion, and outlays
at $1,789 billion. These budget totals for FY2000 differed markedly from the
amounts that CBO
Figure 1. CBO Projections for FY2000
first projected for
Beginning in January 1992
FY2000 in January
(in billions of dollars)
1992. These initial
projections foresaw
$2,500
a deficit of $321
billion, total receipts
of $1,758 billion,
$2,000
and total outlays of
$2,079 billion.
Figure 1 presents
$1,500
the sequence of
CBO projections for
2000 since 1992.
$1,000
Outlays
Each period on the
Receipts
graph represents
Deficit/Surplus
CBO’s baseline
$500
estimate for 2000
from its annual
winter (usually
$0
J a n u a r y ) a n d
Source: CBO Budget
s u m m e r ( t h e s e
reports, various years.
reports have ranged
-$500
f r o m M a y t o
Jan 92Aug 92 Jan 93Sept 93 Jan 94Aug 94 Jan 95Aug 95Dec 95May 96 Jan 97Sept 97 Jan 98Aug 98 Jan 99July 99 Jan 00July 00Sept 00
September) budget
Date of Projection
reports for the
period 1992 through 2000. The rules CBO needed to follow at that time (and still
needs to follow) in constructing its baseline estimates and projections can affect the
estimate’s accuracy. These rules included assuming no policy change from existing
government policy, that the existing discretionary spending caps (the ones in place at
the time of the estimate or projections) would be achieved, and that various expiring
policy provisions would be allowed to expire whether or not that was actually likely
to happen.
The President’s budget documents provide a five-year history of FY2000
projections. The President’s FY1996 budget (February 6, 1995) projected a FY2000
deficit of $194 billion, receipts of $1,711 billion and outlays of $1,905 billion. Figure
2 parallels Figure 1 data from CBO, presenting the Administration’s projections of the
FY2000 budget totals from its various budget documents during the period from

---

CRS-4
February 1995 through June 2000. Unlike CBO’s estimates and projections, the
OMB budget estimates incorporated proposed or expected policy changes into its
forecasts. Other differences between the two agencies’ projections include estimating
techniques and assumptions, such as their respective economic outlooks, all of which
contribute to the differences between their projections over the years.
The underlying reason for the changes in budget estimates over time is that the
information needed to make the estimates is available only in the unknown future.
The projections and estimates rely on assumptions about future economic conditions
and policy decisions and the various interactions of the budget with these factors. As
the estimates are revised over time, CBO, in its budget reports, provides estimates in
three categories to explain the differences between the previous and current estimates.
T h e s e t h r e e
Figure 2. Administration Projections for FY2000
categories that CBO
uses are: the effects
Beginning in February 1995
on the budget of
(in billions of dollars)
r e c e n t p o l i c y
$2,500
changes; the effects
of change on the
b u d g e t f r o m $2,000
changes in forecast
e c o n o m i c
conditions; and the
$1,500
effects on the
budget of technical
revisions. The fiscal
$1,000
policy changes
Outlays
component reflects
Receipts
the effect on the
$500
Deficit/Surplus
budget of legislation
adopted since the
previous estimate.
$0
These can be tax or
Source: Administration
spending cuts or
Budgets, various years.
increases or any
-$500
other legislative
Feb 95 July 95 Feb 96 July 96 Feb 97 July 97 Feb 98
change that modifies
May 98 Feb 99 July 99 Feb 00 July 00 Sept 00
Date of Projection
s p e n d i n g o r
receipts. The effect
of economic changes, such as faster or slower real growth, or higher or lower
employment or interest rates, or change in productivity growth, makes up the second
category. The third category, technical revisions, is a catchall for the factors that do
not fit neatly into the first two groupings. CBO states that the technical revisions
are defined as any changes that are not ascribed to new legislation or to changes
in the macroeconomic forecast. Technical changes could be economic in nature but
are not tied to CBO’s economic forecast.... A variety of other factors could also
produce technical changes, such as revised assumptions about the number of
people who will qualify to receive various benefits, different estimates of the level

---

CRS-5
of benefits they will use, and adjustments to the rate at which discretionary
programs will spend their budget authority.11
Over the nine-year period 1992 through 2000, the economy consistently
performed better than expected. CBO assumed in 1992 that real economic growth
would stay at around 2% a year for the 1998-2002 period, that unemployment would
remain about 5.5%, and that short-term interest rates would stick at 5.6%. For the
years 1998 through 2000, the actual variables showed real economic growth rates
averaged almost 4.3%, the unemployment rate averaged 4.1%, and short-term interest
rates averaged 5.1%. The substantially better than originally forecast economic
conditions contributed significantly to the improved budget outcome for FY2000.12
Faster than projected economic growth combined with lower than expected
unemployment rates increased federal revenues (above the expected levels) reduced
federal spending (below the expected levels) in numerous federal unemployment-
related programs during these years.
The faster economic growth also contributed indirectly to the shrinking deficits
and growing surpluses by altering the technical revisions, especially for revenues.
Incomes grew larger, pushing large numbers of taxpayers into higher tax brackets.
The larger incomes combined with the movement into higher tax brackets rapidly
increased federal revenues above expectations, especially during the last several years.
CBO’s tables showing the causes of the changes in the budget estimates attribute
slightly more of the change to changes in economic conditions and somewhat less to
technical revisions. Only a small amount of the difference is shown coming from
legislative changes at least during the last few years.13
Current Estimates and Projections
The most recent 10-year budget projections from CBO and OMB, from the
January 2001 reports (no specific date has been set for the presentation of the Bush
Administration’s new budget proposals), continue to indicate extraordinarily positive
expectations for the budget throughout the forecast period. CBO shows the total
surplus rising from $236 billion in FY2000 to $889 billion in FY2011 (in its midrange
baseline projections “for the economy and the budget, based on past and current
11CBO, The Budget and Economic Outlook: Fiscal Years 2001-2010, Jan. 2000, p. 18.
12The continued projections of deficits during the late 1990s may have contributed to their
elimination by encouraging constraint in appropriations and other legislation that might have
increased spending or reduced receipts.
13The expected effects of earlier legislative changes, such as the Omnibus Budget
Reconciliation Act of 1993 (OBRA-93) would have been fully incorporated into the CBO
projections from January 1992.

---

CRS-6
trends and the
Figure 3. Surplus Projections, FY2000-2011
assumption that
(in billions of dollars)
current policies are
$1,000
not changed”).14
OMB projects the
CBO Baseline Jan. 2001
$900
b a s e l i n e t o t a l
OMB Baseline Jan. 2001
surplus rising to
CBO Baseline On-Budget Jan. 2001
$800
$810 billion in
OMB Baseline On-Budget Jan. 2001
FY2011. The on-
$700
budget surplus
becomes large in
$600
both projections,
rising to $479
$500
billion in the Clinton
Administration’s
$400
f i n a l b u d g e t
document and to
$300
$558 billion in the
CBO report in
$200
FY2011. Figure 3
provides the path of
$100
Source: Clinton
Adm. and CBO,
the total and on-
Jan. 2001.
budget baseline
$0
2000 2001 2002 2003 2004 2005
2006 2007 2008 2009 2010 2011
surplus projections
Fiscal Year
for the period 2000
through 2011.
CBO fully discusses the uncertainty of the 10-year budget projections in its
budget reports. In its summer 2000 budget report, CBO stated:
[The] long time horizon...increases the likelihood that substantial discrepancies
will emerge between actual results and projections. Since each year’s estimates
build on those of the previous year, longer projection periods imply a greater
chance that errors will compound and will produce more uncertainty in the
estimates used to make policy decisions.15
In its January 2001 budget report, CBO stated:
... considerable uncertainty surrounds...[budget and economic] projections for two
reasons. First, future legislation is likely to alter the paths of federal spending and
revenues. Second, the U.S. economy and the federal budget are highly complex
and are affected by many economic and technical factors that are difficult to
predict. As a result, actual budgetary outcomes will almost certainly differ from
the Congressional Budget Office’s baseline projections.16
14CBO. Budget and Economic Outlook: FY2002-2011, Jan. 31, 2001, p. xviii.
15CBO, The Budget and Economic Outlook: An Update, July 2000, p. 22.
16CBO, The Budget and Economic Outlook: Fiscal Years 2001-2010, Jan. 2000, p. 97.

---

CRS-7
OMB is generally less explicit than CBO in its reports about the uncertainty of
its budget projections, but the President’s budget contains information on the effects
of different economic conditions on the budget. In the Economic Assumptions
chapter in the Analytical Perspectives volume of the FY2001 budget (February 2000),
OMB states:
Both receipts and outlays are affected by changes in economic conditions.
This sensitivity seriously complicates budget planning, because errors in economic
assumptions lead to errors in the budget projections.17
The final Clinton Administration budget document included a chapter (Current
Services Estimates and the Pending Policy Agenda) that discusses some of the
anomalies of the budget baseline rules and how they create some of the inaccuracies
in the estimates. For instance, since errors should be expected in the economic
assumptions used in the budget projections, errors in the budget projections should
also be expected. This does not imply that budget projections are of no use, only that
they need to be used with caution with regard to their predictive ability.
Other observers noted problems that they found with the 10-year budget
forecasts. Alan Auerbach and William Gale, in a report on expected budget surpluses
and tax cut proposals in the publication Tax Notes, state that:
The short-term (10-year) on-budget surplus is based on what we regard as
unrealistically low assumptions regarding the level and path of real discretionary
spending. The surplus is also based on the assumption that recent increases in tax
revenues relative to GDP will largely prove permanent.18
A report from the Center on Budget and Policy Priorities contends that CBO’s
baseline projections overstate the size of the surplus expected over the next 10
years.19 The report argues, among other things, that discretionary spending has been
and continues to grow at rates greater than inflation and that the budget creation rules
under which CBO and OMB produce their baseline estimates do not reflect this. The
report further argues that the rules governing scheduled future changes in revenues
and mandatory spending are not realistic. The rules require that revenue provisions
that are scheduled to expire be allowed to, even if the provision has been reauthorized
repeatedly in the past. A smaller surplus slows the fall in federal debt held by the
public, which in turn slows the fall in the government’s interest payments. The higher
interest payments further erode the expected surpluses.
Reasonable Variations in Factors Affecting the Budget. There are an
almost endless number of reasonable variations in the multitude of factors underlying
the budget projections. These alternative assumptions are not necessarily more or
17OMB, The Budget of the United States Government for Fiscal Year 2001, Analytical
Perspectives, Feb. 2000, p. 14.
18Alan J. Auerbach and William G. Gale, “Does the Budget Surplus Justify Big Tax Cuts?:
Updates and Extensions,” Tax Notes, Oct. 18, 1999: 369-376
19Robert Greenstein, Can the New Surplus Projections Accommodate a Large Tax Cut,
Center on Budget and Policy Priorities, Jan. 4, 2001.

---

CRS-8
less reasonable or
Figure 4. CBO’s Alternative
accurate than the
Scenarios, FY2000-2011
assumptions used
(in billions of dollars)
by CBO and OMB
$1,600
in producing their
budget projections.
O f t h e m a n y $1,400
CBO - Baseline
p o s s i b i l i t i e s ,
CBO - Optimistic
different economic
CBO - Pessimistic
$1,200
assumptions that
underlie the budget
$1,000
e s t i m a t e s a n d
projections are
likely to produce
$800
t h e m o s t
substantial changes
$600
in the budget
p r o j e c t i o n s .
$400
Variations in the
technical aspects of
$200
budget projections
Source: CBO,
( s u c h a s t h e
Jan. 2001.
r e l a t i o n s h i p
$0
between national
2000 2001
2002 2003
2004
2005 2006
2007
2008 2009
2010
2011
Fiscal Year
i n c o m e a n d
personal income
tax collections) can also have substantial effects on the budget projections.
CBO’s January 2001 budget report (as in the one from January 2000) included
what CBO labeled as “alternative scenarios about future trends” in the budget. These
consisted of taking its midrange baseline and adjusting it for two “apparently
reasonable” alternative economic paths. As CBO put it, these are,
referred to as the optimistic and pessimistic trend scenarios...[and] are intended to
reflect assumptions that – although systematically different from the ones in the
baseline projection – still seem reasonable to CBO analysts.20
The surpluses over the 10 years of the optimistic and pessimistic alternatives and the
midrange baseline are shown in figure 4. The wide spread in the outcome is indicative
of the range of possibilities that might occur under reasonable differences in
underlying economic and technical assumptions. CBO’s budget report includes a
diagram containing an even wider range of possible outcomes for the surplus and
deficit through 2006. The extremes, with small probabilities of occurring, produce
larger surpluses than the optimistic scenario and smaller surpluses (and even deficits)
compared to the pessimistic scenario.
20CBO, The Budget and Economic Outlook: Fiscal Years 2002-2011, Jan. 2001, p. 98.

---

CRS-9
The differences in many of these underlying assumptions do not necessarily seem
very large. In the example from CBO that was used to produce figure 4, the
optimistic scenario
Figure 5. CBO Projections of the Percentage
assumes that trend
Change in Real GDP for Calendar Year 2000
productivity grows
(in percent)
at 3.2% a year
rather than 2.7% in
6.0%
the baseline and
the (historical)
1.5% a year in the
5.0%
p e s s i m i s t i c
scenario. The
p e s s i m i s t i c
4.0%
scenario assumes
that Medicare and
Medicaid spending
3.0%
grows 1% faster
t h a n i n t h e
baseline, while the
2.0%
optimistic scenario
has these programs
g r o w i n g 1 %
1.0%
slower than the
baseline. The
Source: CBO,
Various reports.
optimistic scenario
0.0%
a s s u m e s a
Jan
May
Jan
Sept
Jan
Aug
Jan
July
Jan
July
Jan
continuation of the
95
96
97
97
98
98
99
99
00
00
01
Date of Projection
i n c r e a s e i n
p e r s o n a l t a x
liabilities while the pessimistic scenario expects them to fade away over time. Even
this wide range of outcome possibilities is not necessarily wide enough. Among other
substantial variations that can occur in the underlying factors, CBO notes that the
current uncertainty over predicting future productivity growth and the large effect
differences in this variable can have on future budget outcomes. Figure 5 provides
a visual history of the variability in CBO’s estimates of the percentage change in real
GDP for (calendar year) 2000. The period covers six years beginning in January
1995.
There are other possibilities in addition to variations in the underlying economic
and technical factors. Congress and the President may agree on major changes to tax
or spending policies. This past year saw the adoption of appropriation legislation for
FY2001 that exceeded the amount needed to keep pace with inflation. A continuation
of this pattern, which some observers believe is likely, would increase discretionary
spending above the amounts in the current projections. In addition, as pointed out in
the Center on Budget and Policy Priorities report, increasing discretionary spending
at the rate of inflation would still let it “... decline in purchasing power on a per-

---

CRS-10
person basis in
Figure 6. Total Receipts, Total Outlays, and
coming years (since
Discretionary Spending as Percentages of GDP,
the U.S. population
1965-2011
will continue to
(in percent)
g r o w ) . . . . ” 2 1
24%
Whether the public
or Congress would
22%
be satisfied with
20%
such a continued
fall in per capita
18%
d i s c r e t i o n a r y
16%
outlays will partly
determine the future
14%
level of spending.
12%
O t h e r
10%
l e g i s l a t i v e
possibilities that are
8%
under discussion
6%
include adding some
form of prescription
4%
Total Outlays
drug benefit to the
Total Receipts
2%
Medicare program.
Discretionary Outlays
Source: CBO,
Jan. 2001.
If adopted, the
0%
increase in spending
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
would, without
Fiscal Year
e q u a l l y s i z e d
offsets, reduce the size of the projected surpluses. The same would be true of any
legislation that increases spending above or reduces revenues below the levels
foreseen in the existing projections. (Legislation could also be adopted that reduces
spending or increases revenues from the levels in the projections, thereby probably
increasing the surpluses and speeding the reduction in federal debt held by the public.)
In relative terms, several components of the federal budget are projected to reach
levels of GDP that have not been seen since the early 1950s. Total federal spending
is projected to fall to 15.6% of GDP from 18.2% in 2000. This would put total
federal outlays at their lowest level since 1951. At the same time, revenues would
remain at or near 20.0% of GDP throughout the decade, a level reached only during
World War II and approached in 1969 (19.7%) and in 1981 (19.6%). Figure 6 shows
total receipts and outlays as a percentage of GDP between 1965 and fiscal year 2011
(using the CBO January 2001 baseline projections). The figure also shows
discretionary spending as a percentage of GDP.
As has already been mentioned, the future direction of discretionary spending is
also in doubt. The CBO projections show discretionary spending continuing a decline
as a share of GDP. This follows a pattern that has occurred sporadically since the end
21Robert Greenstein, Can the New Surplus Projections Accommodate a Large Tax Cut,
Center on Budget and Policy Priorities, Jan. 4, 2001, p. 7.

---

CRS-11
of the Viet Nam War and accelerated in the late 1980s. The pattern is shown in
Figure 6. In some analysts’ opinion (see above), the likelihood of keeping
discretionary spending on this path, given pressures to increase defense and other
components of discretionary spending, may prove difficult.
Whether these historically unusual levels of spending and receipts will actually
occur (or can be maintained) may depend on how much the nature of the economy
and technical aspects of the budget have actually changed in the last decade. If the
last decade’s budget environment is only a deviation from the norm and that deviation
ends, the budget will shortly begin to reflect that norm and may revert to more
historic percentages of GDP. If the environment in which the budget exists has
undergone a fundamental shift in nature, these new spending and revenue levels as
percentages of GDP may be sustainable, if the public wishes to sustain them. The
uncertainty about whether a fundamental shift has occurred also contributes to the
uncertainty of the budget projections.
Conclusion
Federal budget projections and estimates have become a significant part of the
congressional budget debates and broader congressional debates over future policies.
The supporting assumptions that CBO and OMB and others must use to construct the
projected future paths for spending and revenue and their respective components are
themselves highly uncertain and changeable. The fundamental issue with all the
estimates and projections is that almost nothing about the future can be known with
any real certainty.
Certain aspects of what will happen in the future can be forecast with reasonable
accuracy, such as knowing, approximately, the number of people retiring in the next
20 years. This can provide a reasonably solid base from which to estimate the future
level of retirement benefits. But it cannot be completely accurate and it covers only
part of the budget. Too much remains unknown. Will a change in economic
conditions change retirement decisions? How will inflation affect the calculations?
How accurate is the forecast for the need for and cost of medical care for those
retiring? Longer life expectancy or medical breakthroughs can substantially alter
future spending levels. These are the problems that underlie both the budget
projections and the assumptions needed to make those projections. There is no way
of assuring that the underlying assumptions used in the projections, the assumptions
about real economic growth, productivity growth, inflation, interest rates, bad
weather for farmers, natural disasters, military emergencies, scientific breakthroughs,
or how any other thing that might affect the budget, will match what actually occurs.
Errors in any or all of these assumptions can throw off the budget projections by
large or small amounts, especially in the budget projections that run 10 years (or
more) into the future. In January 1995, CBO assumed that real economic growth
would be 2.3% in calendar year 2000. Real economic growth that year was 5.2%.
This mis-projection of an underlying assumption contributed to the differences
between the originally forecast budget numbers and those that actually occurred.
CBO projected a deficit of $297 billion for FY2000 in January 1995; the government

---

CRS-12
ended FY2000 with a surplus of $236 billion.22 It is reasonable to be concerned that
the current projections for FY2010 or FY2011 may not be any more accurate than
those of the past.
22As shown earlier in the report, CBO’s first budget projection for FY2000 was made in
January 1992, which showed a deficit of $322 billion for that year.

---