CRS Updates Report on Changes Required to Balance Budget

A balanced federal budget is a bipartisan goal of many Members of Congress. In addition, moving the budget closer to balance is a long-term necessity because the national debt cannot grow as a percentage of GDP indefinitely, as it would under current policy. The budget deficit in FY2007 is projected to be between $270 billion and $354 billion. Mathematically, the budget could be balanced by reducing total spending by 10%-13%, or mandatory spending by 19%-25%, or discretionary spending by 28%-37%, or non-military discretionary spending by 52%-68%, or by raising income tax rates by 23%-31%, or some combination of these options. The budget is unlikely to return to balance "on its own," as some have suggested, because higher productivity rates are already incorporated in the projections; research suggests that the revenue estimates of tax cuts are unlikely to be significantly overstated; and the decline in the deficit found in the CBO baseline or President's budget rests on assumptions that differ substantially from what is typically thought of as current policy. This report assumes a familiarity with basic budgetary terms and concepts and will be updated as events warrant.

Many Members of Congress of both parties support the goal of a balanced budget. In addition, moving the budget closer to balance is a long-term necessity because the projected deficits would cause the national debt to grow indefinitely as a percentage of GDP.¹ If this occurred, it would eventually result in financial insolvency. This report lays out generic scenarios for balancing the budget in the next fiscal year. Although these are not policy options that are likely to be enacted, they are meant to offer simple examples to gauge the scope of tradeoffs that would be required if policymakers eventually decide to bring the budget back to balance. If changes are postponed or made over a longer time period, they would need to be larger because of higher debt service. (CRS does not take policy positions on the appropriate time frame for balancing the budget.)

Under the Congressional Budget Office (CBO) baseline of current policy, the FY2007 budget deficit is projected to be $270 billion. In the Administration's budget proposal, the deficit would be $354 billion. However, these estimates are unlikely to match actual outcomes for a number of reasons. CBO is required to assume that discretionary spending would grow at the rate of inflation and all expiring tax provisions (including the recent tax cuts) would not be renewed, and to include only supplemental spending for military operations in Iraq that has already been enacted. The baseline is not meant to offer a "best guess" of future policy. The Administration's budget proposal depends on congressional enactment and Congress may have other priorities. The Administration's budget assumed a lower rate of non-supplemental discretionary spending growth than CBO. Even congressional resolutions often turn out to be different from actual results. For example, the actual budget deficit in FY2003 was $27 billion higher than called for in the conference budget resolution. Keeping these qualifications in mind, the actual deficit, absent policy changes and projection errors, is likely to be closer to the Administration's budget than CBO's baseline since the Administration includes the supplemental spending necessary to maintain military operations in Iraq and Afghanistan and extends some of the expiring tax provisions in 2007.

The deficit can be eliminated through higher tax revenue, lower spending, or some combination of the two. This report quantifies the scope of changes required to balance the budget using the CBO baseline and the Administration's budget under the following options:

• reduce total spending

• reduce mandatory spending

• reduce discretionary spending

• reduce non-military discretionary spending

• raise individual income tax rates

• reduce tax expenditures

• raise individual income tax rates and reduce spending equally

Rather than single out any specific spending area to bear a disproportionate burden of the reductions, one option would be to spread deficit reduction evenly across all policy areas by $258 billion under the CBO baseline or $339 billion under the Administration's budget.² (All policy options are smaller than the deficit because of the resulting reduction in debt service caused by the balanced budget.) Under this option, total non-interest spending would decrease by 10% from the CBO baseline and 13% from the Administration's budget.

About three-sevenths of total spending (excluding net interest) is discretionary spending (i.e., spending specifically appropriated by Congress) and four-sevenths is mandatory spending. Medicare, Social Security, and other retirement programs account for about two-thirds of mandatory spending, and income support programs account for another one-sixth. To balance the budget solely through reductions in mandatory spending, it would need to be reduced by 19% from the CBO baseline and 25% from the Administration's budget. If the deficit were eliminated solely through reductions in discretionary spending, it would need to be reduced by 28% from the CBO baseline and 37% from the Administration's budget.

Discretionary spending is split about evenly between military and non-military spending. Given current military operations abroad and political support for military spending, some policymakers would prefer to limit spending reductions to non-military discretionary spending. Non-military discretionary spending is spread across many policy areas; education and transportation are the largest. If the deficit were eliminated solely through reductions in non-military discretionary spending, it would need to be reduced by 52% from the CBO baseline and 68% from the Administration's budget.

Tax Increases

Tax increases could take many different forms and be pursued through many different parts of the tax system. One approach might be an across-the-board increase in marginal individual income tax rates. To balance the budget, average effective individual income tax rates would need to be increased by 23% (i.e., from 10.4% to 12.8%) under the CBO baseline or 31% under the Administration's budget. An approximately equivalent increase in all marginal income tax rates would be needed to raise average tax rates to that extent, assuming no behavioral responses. Because of interactions with the alternative minimum tax, non-refundable tax credits, and so on, marginal tax rates would probably need to be raised by a greater extent to actually achieve a 23%-31% increase in average effective tax rates.³

To raise revenue, economists often favor reforms that broaden the tax base rather than raise marginal tax rates.⁴ Under a theoretically "ideal" income tax system, the tax treatment of all net income would be the same, regardless of how it is earned or spent. In our current tax structure, tax expenditures (deductions, exemptions, and credits) give special preferences to certain types of economic behavior. While it is beyond the scope of this report to evaluate the effect on efficiency of any particular tax expenditure, it is useful to consider how much revenue could be generated by broadening the base as an alternative to raising marginal tax rates. Eliminating only the 5 largest tax expenditures, which include exclusions for pension contributions and health insurance premiums and deductions for home mortgage interest, would raise more than enough revenue under the Administration's budget to balance the budget without any change in marginal rates.⁵

The large sums involved in the previous examples suggest that some might find it desirable to balance the budget through a combination of tax increases and spending cuts. One option would be to split the revenue difference evenly between overall spending cuts and individual income tax increases. This solution would require a 5% decrease in total spending and a 12% increase in average effective tax rates under the CBO baseline and an 7% decrease in total spending and a 15% increase in average effective tax rates under the Administration's budget.

     Table 1. Summary of Selected Policy Options to Balance

Some commentators have argued that the drastic policy changes illustrated above will not be necessary because the budget deficit will shrink on its own. They make a number of arguments to support this claim.

First, they point to the improvement in the deficit that occurs over time in the CBO baseline and the President's budget proposal. In the CBO baseline, the deficit falls from $337 billion in 2006 to a surplus of $67 billion in 2016. In the President's budget proposal, which covers only five years, the deficit falls to $205 billion in 2011. It is accurate to characterize these projections as requiring significant changes from what is typically considered current policy, however. The CBO baseline assumes discretionary spending will decline 2 percentage points to 5.9% of GDP in 2016; discretionary spending has never been this low since World War II. The CBO baseline also assumes that many taxpayers will fall under the AMT and all expiring tax provisions will not be renewed, including the 2001 and 2003 tax cuts. The President's FY2007 budget proposal assumes that there will be little spending on operations in Iraq after 2007, AMT relief will be allowed to expire in 2007, and discretionary spending will fall to its lowest post-World War II level by 2011. If CBO's or the Administration's assumptions are relaxed, the deficit increases over the next 10 years.⁶

Second, it is argued that faster economic growth will lead to higher revenues than predicted, similar to the experience of the late 1990s. It is true that actual growth was higher than projected in the late 1990s, and this, in turn, caused revenues to be higher than projected. But at this point, much of the then-unexpected increase in economic growth has been built into CBO's projections. For example, CBO estimates that potential labor productivity grew by an average of 2.5% from 1996 to 2005 compared to 1.5% from 1974 to 1995. In the baseline, it assumes that potential labor productivity will grow by an average of 2.5%. In other words, the acceleration of productivity growth in the last decade is already incorporated in CBO's projection of the next 10 years.

The unexpectedly rapid increase in revenue from capital gains realizations was another important element of the "revenue surprise" of the late 1990s. Capital gains revenues rose from $54 billion in 1996 to $119 billion in 2000, but then fell, following the stock market crash, to a estimated $45 billion in 2003. Since the high realizations of the late 1990s were caused by the extremely rapid increase in equity prices, this source of the revenue surprise may also be unlikely to be repeated in the future. Adjusted for inflation, CBO does not project capital gains revenues to reach its previous peak at any point in the 10-year forecast.

It should also be noted that the improvement in the budget balance in the 1990s was not just good fortune, but also the result of underlying budgetary decisions. To achieve budget balance, taxes were raised in 1990 (P.L.101-508) and 1993 (P.L.103-66) and spending was reduced from 22.3% of GDP in 1991 to 18.4% of GDP in 2000. The largest reduction in spending was defense spending in response to the end of the Cold War. Defense spending fell from 6.2% of GDP in 1986 to 4.6% of GDP in 1991 to 3.0% of GDP in 2000. Spending on non-defense discretionary, Social Security, and interest payments on the debt (because the debt was declining) also fell as a percentage of GDP between 1991 and 2000. Since 2000, spending has risen as a percentage of GDP to an estimated 20.1% in 2005.

Third, it is argued that the deficit projections are based on faulty assumptions that overestimate the cost of tax cuts because they do not include "feedback effects." It is claimed that, in reality, the tax cuts will cost much less than originally projected because tax cuts spur higher growth, thereby "paying for themselves," at least in part. (The revenue-raising options laid out in this report also assume there will be no feedback effects.) Based on existing theory and empirical evidence, CBO and the Joint Committee on Taxation (JCT) have provided alternative estimates of how much the 2003 tax cuts will cost after allowing for feedback effects on GDP. Assuming that the Federal Reserve does not let inflation rise, JCT found that the tax cuts could cost 5.8%-16.1% less in the first five years, and 2.6%-11.8% less in the next five years. CBO found that the cost of the President's 2004 budget proposals could be between 29% lower and 10% higher over the first five years, and between 17% lower and 15% higher over the next five years.⁷ This indicates that, under the best case scenario, the feedback effects of the tax cuts would not generate enough revenues to move the budget significantly closer to balance and, under the worst case scenario, could increase the budget deficit more than under "static" revenue estimates. It should also be noted that, to date, there is no evidence that the tax cuts have resulted in less revenue loss than originally projected. After adjusting for economic conditions and temporary factors, revenues have fallen by more than the original "budget scores" for the tax cuts had indicated.⁸

Fourth, it is argued that budget projections are highly uncertain, and may prove to be too pessimistic. This is true: the degree of uncertainty surrounding budget projections dwarfs the projected deficits in the out years of the projections. For example, CBO estimates there is a 25% chance that the budget will be balanced in 2010 -- and a 25% chance that the deficit will be at least $274 billion larger than projected.⁹ This means that the permanency of the deficits now projected is far from being a "sure thing." But while the projections may prove to be too pessimistic, it is equally likely that the projections are too optimistic. Although projections made today will certainly prove to be incorrect, the probability that the budget deficit will turn out to be higher than predicted is equal to the probability that it will turn out to be lower than predicted.

The Burden of the Status Quo

Many would consider the policy options laid out in this report to be too burdensome to be feasible. But in mainstream economics, a budget deficit imposes a burden that is just as real as higher taxes or spending cuts. Deficits can be financed only by borrowing real resources. When these resources are borrowed out of American saving, the budget deficit pushes up interest rates and "crowds out" private investment spending that is necessary to increase future standards of living. In effect, this outcome shifts the burden of the deficit to future generations by causing future living standards to be lower than they otherwise would be. When the resources are borrowed from foreigners, the trade deficit widens and foreigners, rather than Americans, enjoy the returns from that borrowing. Balancing the budget shifts a burden, but it does not create one.¹⁰

² Researchers have singled out specific policy options for reducing the deficit that they think would be desirable. See, for example, Chris Edwards, "Downsizing the Federal Government," Cato Institute Policy Analysis No. 515, June 2004 and Alice Rivlin and Isabel Sawhill, eds., Restoring Fiscal Sanity: How to Balance the Budget (Washington, DC: Brookings Institution Press, 2004); Congressional Budget Office, Budget Options, Mar. 2003.

³ Some policymakers have argued that the fiscal position should be improved by repealing the tax cuts of 2001, 2003, and 2004. According to revenue estimates by the Joint Tax Committee, this policy would reduce the Administration's budget deficit by 61% in 2007.

⁴ The Tax Reform Act of 1986 (P.L.99-514) is a prominent example of an act that broadened the tax base and increased tax revenues (at least initially) -- despite reducing marginal tax rates.

⁵ Based on OMB estimates of the revenue lost to tax expenditures in 2007 relative to an ideal tax system. This can be considered only a rough estimate because if different tax expenditures were simultaneously reduced, there would be interactions between them that could be higher or lower than the cost of reducing them separately. For examples, see Leonard Berman, "Is the Tax Expenditure Concept Still Relevant?" National Tax Journal, Sept. 2003, p. 615.

⁶ See CRS Report RL3141, Baseline Budget Projections: A Discussion of Issues, by Marc Labonte.

⁷ Congressional Budget Office, An Analysis of the President's Budgetary Proposals for FY2004, Mar. 2004; Congressional Budget Office, How CBO Analyzed the Macroeconomic Effects of the President's Budget, July 2003; Joint Committee on Taxation, "Macroeconomic Analysis of H.R.2," Congressional Record, Doc 2003-11771, May 8, 2003. For a discussion, see CRS Report RL31949, Issues in Dynamic Revenue Estimating, by Jane Gravelle.

⁸ See CRS Report RS21786, The Federal Budget Deficit: A Discussion of Recent Trends, by Gregg Esenwein, Marc Labonte, and Philip Winters.

⁹ See Congressional Budget Office, The Uncertainty of Budget Projections, Apr. 2005, p. 17.

¹⁰ See CRS Report RL30520, The National Debt: Who Bears Its Burden? by Marc Labonte and Gail Makinen. A trade deficit is the necessary result of borrowing abroad because borrowing can only occur if Americans spend more abroad than foreigners spend on American goods.