March 11, 2002

by Richard Kogan

PDF of this report

Related reports

If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.

The "stimulus" bill that Congress has just approved carries lower costs than the stimulus legislation proposed in the President's budget. In addition, the economy has recently improved. Some Members of Congress believe that, with a smaller stimulus bill and an improved economy, it is both possible and desirable to balance the budget for the coming fiscal year, 2003. They infer this from Table 1 below, which shows that when the costs of the "stimulus" bill are removed from the President's budget, the result will be a deficit of only $3 billion, a relatively small sum.

But the conclusion that a balanced budget is within easy reach is far off the mark, for at least three reasons:

Table 1
OMB's Portrayal of the President's Budget for FY 2003

in billions of dollars
OMB's "baseline" surplus (unified budget, which includes Social Security) 51
Reduction in surplus because of cost of budget proposals:
increased expenditures for defense and homeland security -31
increased expenditures from pending farm bill -7
extension of expiring tax credits; new health and "charitable giving" tax proposals -6
cost of drug benefits and other Medicare/Medicaid provisions -3
cost of all other Presidential budget proposals -6
debt service (higher interest payments) resulting from above proposals -1
cost of President's "stimulus" proposal, including debt service -77
Total cost of Presidential proposals -131
Resulting unified budget deficit in President's budget (includes Social Security) -80
remove costs of President's "stimulus" proposal, including debt service +77
Resulting unified budget deficit w/o "stimulus" (includes Social Security) -3

Given the new stimulus legislation and other policies proposed in the President's budget — primarily major increases for defense, further tax cuts, the farm bill, and increased expenditures for homeland security — the 2003 budget will be in deficit by $87 billion, using CBO's estimates. (See Table 2; the CBO estimates used here are those that CBO issued on March 6.) Even using OMB's more optimistic estimating assumptions, the 2003 budget will be in deficit by $46 billion.

Table 2
CBO's Reestimate of the President's Budget

dollars in billions
OMB's estimates CBO's estimates
Starting "baseline" surplus/deficit in the total budget* +41 +6
Cost of the President's budget proposals, except "stimulus" -44 -49
Resulting unified budget deficit, without "stimulus" -3 -43
Additional cost of the stimulus legislation** -43 -43
Resulting unified budget deficit, including "stimulus" legislation -46 -87
May not add due to rounding.
* OMB documents show two baseline surpluses for 2003: the $41 billion figure used above (because it is comparable to CBO's) and the $51 billion figure used in Table 1. See Appendix for further explanation.
** The Joint Tax Committee and CBO estimate that the stimulus bill costs $43 billion in 2003. Of that amount, $2 billion represents the extension of expiring or expired tax provisions, as proposed by the Administration but not classified by OMB as "stimulus." The non-duplicated "stimulus" costs of the House-passed bill is therefore $41 billion. When higher interest payments on the debt are accounted for, the non-duplicated cost is $43 billion, as shown.

Furthermore, as noted, there are other costs not included in the President's budget that are virtually certain to occur and to add to the deficit for 2003.


Can Congress Balance the Budget Using CBO Estimates?

Enacting tax increases or budget cuts to reduce the projected deficit for 2003 from CBO's estimate of approximately $87 billion — given the stimulus bill — to zero would be a monumental task.


How About Using OMB Estimating Assumptions or Assumptions That Are Even More Optimistic?

As noted, OMB estimates are noticeably more optimistic than CBO's. Even they, however, show a deficit of $46 billion when adjusted for the House-passed stimulus bill. It is not possible to use even these optimistic OMB estimates as a basis for balancing the budget without relying on substantial gimmicks. (For a discussion of gimmicks, see the text box.)

For instance, if Congress adopts OMB's rosier assumptions and attempts to balance the 2003 budget by reducing the President's request for appropriated programs other than defense and homeland security, those "discretionary" programs would need to be cut 25 percent below the level the President has requested. As noted, the requested level itself already is 1 percent below the actual 2002 funding level, according to CBO.

Stated differently, a reduction of $46 billion in the budget year is larger than the first-year savings accomplished in the two largest deficit reductions Congress had enacted in the last few decades, the 1990 and 1993 budget packages. And as noted previously, the figures in the President's budget do not reflect the request for supplemental 2002 appropriations he is about to make or the increases in Medicare provider payments that Congress will inevitable adopt.

Consequently, to alter the policies in the President's 2003 budget to produce balance in 2003 seems impossible without massive gimmicks and or reliance on estimating assumptions even rosier than OMB's.

Admittedly, predicting how the economy will perform and how the budget will respond to the economy is not science, and is a very uncertain art. It is quite possible that 18 months from now we will discover that both CBO and OMB were too optimistic or too pessimistic. CBO's recent annual report shows that over the last two decades, the average size of its absolute estimating error for the coming fiscal year was 1.1 percent of GDP, or $120 billion in 2003 dollars. Thus, it should not be a shock if the 2003 budget deficit of $87 billion (CBO's estimate of the President's budget adjusted for the House-passed stimulus bill) turns into a surplus — or grows to a $200 billion deficit.

Using "Gimmicks" to Balance the Budget

Given the stimulus legislation, Congress could assert it had balanced the budget by a) making no further changes to the President's budget but adopting a set of estimating assumptions even rosier than OMB's; or b) adopting OMB's optimistic estimating assumptions and engaging in massive expenditure or revenue timing shifts. Experience demonstrates that Congress is not averse to timing shifts. For example, last June, Congress delayed the payment of some corporate taxes from the last two weeks of September into the first week of October; the result made the surplus for fiscal year 2001 look $23 billion smaller than it otherwise would have been and the deficit for fiscal 2002 look $23 billion smaller than it otherwise would have. Congress has enacted timing shifts of smaller magnitudes many times in the past, almost invariably for the sake of appearances. Timing shifts totaling $46 billion, however — the amount that would be needed to balance the budget in 2003 under OMB assumptions — would be unprecedented.

In any event, adopting rosy scenarios to hide prospective deficits or adopting massive timing shifts is widely, and appropriately, viewed as gimmickry. It would do nothing in reality to balance the budget in either the short term or the long term.

Congress has not always adopted CBO's estimates. Especially during 1986-1990, when the "Gramm-Rudman-Hollings" Deficit Control Act required Congressional budget plans to hit specified targets set in statute, Congressional budget plans would appear to hit those targets by using the more optimistic OMB estimating assumptions. During that era, the divergence between promised deficits and actual deficits grew much wider, and the deficit projections in Congressional budget resolutions become increasingly less credible and more unreliable. After repeated, embarrassing failures, the Gramm-Rudman-Hollings statute was replaced by a system of appropriations limits and a rule that all new tax and entitlement legislation had to be "paid for," or fully offset. History has demonstrated that actual budget restraint is far more effective than rosy scenarios in achieving fiscal prudence. Perhaps for this reason, Congress became increasingly committed to using CBO estimates after GRH collapsed. Indeed, at times the House leadership has argued that CBO's estimates are the only honest ones. In 1995, for instance, Speaker Gingrich said "we do not ask you to agree to anything but two principles, that the budget shall be balanced [by 2002] and that the scoring will be honest numbers based on the Congressional Budget Office."(1)

The risk in using optimistic budget estimates is not primarily that Congress will be embarrassed when promised balanced budgets turn into actual deficits. Rather, the risk is that official optimism will pave the way for irresponsible fiscal policy — for tax cuts and program expansions that could prove unaffordable. It is important to realize that between January 2001 and today, the projected surplus over the ten-year period 2002-2011 shrank by more than $1.4 trillion due solely to last year's overly optimistic estimating assumptions.



Given the stimulus bill, it will be politically impossible to attempt to balance the budget for 2003 using either CBO's estimates or the more optimistic OMB estimates, unless Congress engages in a massive and unprecedented use of timing shifts and other gimmicks — or of highly optimistic estimating assumptions — to pretend it is balancing the 2003 budget.

Estimating Difference Between CBO and OMB

Table 2 shows that CBO's estimates and OMB's estimates of the same budget policies differ by $35 to $40 billion, depending on the policies. Specifically, CBO and OMB have produced comparable baselines in which the results for 2003 differ by $35 billion — CBO shows a baseline surplus of $6 billion while OMB shows a baseline surplus of $41 billion. Virtually the entire difference derives from an OMB assumption that federal revenues under current law will be substantially higher than CBO believes.

In addition, CBO and OMB have different estimates for the costs of the President's policies proposals; CBO generally estimates those proposals will cost more or save less than OMB believes. The baseline difference of $35 billion widens to approximately $40 billion when the President's policy proposals are also accounted for.

At first blush, it appears as though CBO and OMB are $45 billion apart in their baseline estimates, not $35 billion: CBO estimated a $6 billion baseline surplus for 2003 in its March 6 testimony to the Senate Budget Committee whereas OMB estimated a $51 billion baseline surplus on February 4. (See An Analysis of the President's Budgetary Proposals for 2003, CBO testimony of March 6, 2002, and summary tables S-3 and S-9 in the President's Budget, February 4, 2002.) However, those two baselines are not strictly comparable; OMB's baseline figure omits the continuation of $20 billion in emergency funding enacted in December 2001; such continued funding is estimated to produce $10 billion of outlays in 2003. Table 15-1 in OMB's Analytical Perspectives (one of the budget documents) shows that an OMB baseline that extends such funding — as the CBO baseline does — would feature a surplus of $41 billion in 2003, a result that produces an OMB estimate $35 billion more optimistic than CBO's equivalent figure.

End Note

1. (Congressional Record, p. H12502, November 15, 1995).