FRBSF Economic Letter
2002-08; March 22, 2002
The Changing Budget Picture
In January 2001, the Congressional Budget Office (CBO) reported that
the federal government was projected to run a $313 billion surplus in
2002. Now, just a little over a year later, the CBO's latest reports indicate
a major deterioration in the budget projections. In its February 2002
report, the CBO projected a deficit of $21 billion in 2002; in a revision
released in March 2002, to account for the release of President Bush's
federal budget for 2002-2003, the CBO projected a small surplus in 2002
under its baseline scenario and a $90 billion deficit under the President's
proposals. Finally, when the economic stimulus bill passed Congress last
week, the CBO again revised its baseline projection for 2002 to show a
$46 billion deficit.
Although large changes in the budget outlook from year to year are not
uncommon, the magnitude of the change for 2002 is unusually large. This
Economic Letter examines the change in the budget outlook and discusses
the causes of the revisions in the projections. Because the complete CBO
baseline analysis of the Economic Stimulus Bill is not yet out, this Letter
focuses on the projections released in early March.
The budget revisions
Figure 1 shows the large downward revisions in both the current and future
surpluses that occurred between early 2001 and 2002. In January 2001,
the CBO projected that the surplus would be $313 billion for 2002 and
would rise significantly over the decade, reaching an annual level of
$889 billion in 2011 and totaling $5.6 trillion for the entire decade.
In the CBO's February 2002 projections, the surplus for 2002 was gone.
In its place was a projected deficit of $21 billion. The CBO still projected
a surplus over the entire decade, but the total fell from $5.6 trillion
to just $1.6 trillion. In early March, the CBO revised its projections
again and showed a small surplus of $5 billion for 2002.
What accounts for this large swing in the budget
To answer this question, it is important to recognize that the CBO's
budget projections are different from forecasts of future budget surpluses
and deficits. (See Walsh
1999 for a full discussion of the differences.) One of the most important
differences is that the CBO projections assume that current laws and policies
will remain unchanged. For example, though Congress was debating a tax
cut in the first half of 2001, the CBO did not change its budget projections
until the legislation actually passed. Therefore, the January 2001 projections
assumed no tax cut, while the February 2002 projections incorporate the
effects of the new tax bill. Similarly, the CBO's baseline projection
released in March 2002 does not incorporate the proposals contained in
the President's 2002-2003 budget. The CBO does provide estimates of the
impact of the President's budget, however, and they show the budget swinging
from a $5 billion surplus to a $90 billion deficit in 2002. Whenever major
legislative changes in either taxes or spending occur, there will be large
revisions in the CBO's projections.
Legislative changes did play an important role in altering the budget
outlook between January 2001 and early 2002, reducing the projected surplus
by $91 billion for 2002. The effect of these legislated changes grows
over time; by 2011 they are projected to reduce the surplus by $319 billion.
Figure 2 focuses on these legislative changes, distinguishing among tax,
spending, and debt service factors. The tax cut passed by Congress in
2001 is the primary legislative reason for the downward revisions in the
surplus. Increased discretionary spending also contributes to the smaller
surpluses; for the period 2002-2011, the CBO has raised its projection
for discretionary spending by $550 billion. Of this total, $301 billion
is for defense spending and $249 billion is for nondefense outlays.
Because these legislative changes reduce the surplus available to repay
the government's outstanding debt, the debt will fall more slowly than
had been projected earlier. As a result, federal interest payments are
now likely to be significantly higher than were assumed in the CBO's January
2001 projections. This debt service component grows to become a major
source of the changing outlook for the surplus, accounting for $124 billion
of the $319 billion reduction in the surplus projected for 2011.
While the CBO projections assume current laws and policies remain unchanged
over the horizon of the projection, the CBO does incorporate their forecasts
of economic conditions. And the near-term economic outlook has changed
significantly since January 2001. The long economic boom the U.S. enjoyed
through most of the 1990s officially ended in March 2001. An economic
recession has major implications for budget projections. As overall income
declines, the government's tax revenues fall, contributing to a fall in
the budget surplus. During February 2002, the outlook for economic activity
improved. As a result, between February and March 2002, the CBO revised
its baseline projection for 2002 from a deficit of $21 billion to a surplus
of $5 billion.
A final source of changes in the budget projections arises from what
the CBO classifies as technical changes. This category is a catch-all
for any changes in the budget outlook that are related neither to legislative
changes nor to the CBO's economic forecast. In the February projections,
most of the technical changes are due to downward revisions in revenue
estimates. One major factor contributing to these lower revenue forecasts
is the weak performance of the stock market during 2001 and early 2002.
This weakness has caused the CBO to reduce its estimate of the government's
revenue from such sources as capital gains realizations.
Figure 3 puts together these three sources of revisionslegislative,
economic, and technicalto show how each has contributed to the reduction
in projected surpluses.
The relative importance of each component varies over the projection
horizon. For 2002, for example, changing economic conditions and related
factors, such as the decline in capital gains realizations, account for
more than 70% of the deterioration in the budget projection. Of the $308
billion reduction in the projected surplus for 2002, $125 billion is the
result of changes in the CBO's economic forecasts. Technical changes account
for a further $92 billion reduction, while legislative changes are estimated
to reduce the 2002 surplus by $91 billion. As the decade progresses, legislative
changes become the major reason for the downward revisions in the surplus
projections. In 2010, for example, $357 billion of the $488 billion change
in the projection is due to legislative changes, and, as Figure 2 showed,
most of these legislative changes reflect the impact of the tax cut.
Changes in the CBO's budget projections help explain why the outlook
for future federal surpluses has changed dramatically over the past year.
The projections allow one to separate the impacts of the 2001 recession
from the 2001 tax cut. What the projections do not provide, though, is
any assessment of whether the changes are desirable or not. For that,
one needs some basic principles for assessing the government's financial
For instance, most economists believe governments should aim to balance
their budgets, not year to year but over longer time horizons. When the
economy is in a recession and tax revenues decline, it is appropriate
to run a deficit (or a smaller surplus). When the economy booms and tax
revenues rise, the government should run a larger surplus. These short-run
fluctuations that occur as the economy experiences business cycles can
cloud the longer-run budget picture. The decline in the projected surpluses
caused by the economic slowdown in 2001, and reflected in the changes
in the CBO's economic forecasts, does not signal any fundamental change
in the government's financial health.
To abstract from budget fluctuations associated with the business cycle,
economists often focus on what is variously called the full-employment,
high-employment, or structural budget. This measure is designed to show
what the budget surplus would be if the economy maintained full employment.
Currently, the full-employment budget is still in surplus; if economic
activity had not declined during 2001, the federal government would be
running a budget surplus.
That still raises the issue of whether, at full employment, the government's
surplus is too large or too small. To assess this issue, a second budget
principle is helpful. The government should take into account all the
implications of current policies, not just their implications for this
year or the next few years. For example, no assessment of the federal
government's budget outlook can ignore the looming retirement of the baby
boom generation and the effects this will have on the Social Security
System. Because of the need to build reserves to fund the projected Social
Security payments to the baby boomers, the Social Security Trust Fund
currently takes in more revenues than it pays out in current benefits.
Figure 4 shows that when these Social Security surpluses are separated
out of the overall budget, the remaining budget, referred to as the on-budget
surplus, was actually in deficit in 2001, despite an overall surplus of
$127 billion, and the on-budget balance is projected to remain in deficit
until 2010. Thus, the Social Security System masks deficits in the rest
of the budget that are projected to continue for almost the entire decade.
Finally, any assessment of the budget outlook must bear in mind the uncertainty
inherent in projecting future revenues and spending. The events of September
11, 2001, and their impact on defense and security-related federal expenditures
are telling reminders of the difficulties of accurately looking into the
future. Economic conditions will change as well and, as they do, so will
the budget outlook.
Projections like those shown here, with a single line extending into
the future, do not provide a sense of the range of possible outcomes.
To overcome this difficulty, the CBO has begun to produce "fan charts"
like those used by the Bank of England for its inflation forecasts. A
fan chart shows a range of projections emanating from the most recent
actual data, and the width of the fan is based on an estimate of the uncertainty
in the projection. For example, it can show the range of outcomes that,
under current laws and policies, will include the actual budget outcome
with 90% probability. For 2002, this range extends from a surplus of $108
billion to a deficit of $149 billion. As one looks further into the future,
the range gets even wider. For 2007, the 90% range goes from a surplus
of $901 billion to a deficit of $568 billion. These wide bounds serve
as useful reminders that the CBO budget projections are like snapshots,
based on current policies and economic forecasts. But these projections
are likely to be subject to major revisions in the future, just as they
have been in the past, as Congress legislates changes in taxes and spending
and as economic conditions evolve.
Carl E. Walsh
Professor of Economics, UC Santa Cruz,
and Visiting Scholar, FRBSF
Congressional Budget Office. 2002. The Budget and
Economic Outlook: Fiscal Years 2003-2012 (January).
Congressional Budget Office Testimony: An Analysis
of the President's Budgetary Proposals for 2003. 2002 (March 6).
Walsh, Carl E. 1999. "Projecting
Budget Surpluses." FRBSF Economic Letter No. 99-27 (September