Skip Navigation

Mr. Chairman and Members of the task force, I appreciate the opportunity to be here this morning to discuss budget projections and budget baselines. In recent years, the Congressional Budget Office's (CBO's) baseline budget projections have been the most widely used yardstick for measuring budgetary proposals. When preparing a cost estimate of a bill, CBO and the Joint Committee on Taxation use baseline budget projections for calculating the cost or savings that would result from legislative proposals to change laws affecting entitlements and revenues. The Budget Committees generally use CBO's baseline budget projections as a starting point for formulating their recommendations for the annual Congressional budget resolution.
 

THE BASELINE CONCEPT

Twice a year--generally in January and August--CBO prepares baseline projections of federal revenues, outlays, and the surplus or deficit. Those projections are designed to show what would happen if current budgetary policies were continued as is--that is, they serve as a benchmark for assessing possible changes in policy. They are not forecasts of actual budget outcomes, since the Congress will undoubtedly enact legislation that will change revenues and outlays. Similarly, they are not intended to represent the appropriate or desirable levels of federal taxes and spending.

For tax revenues and entitlement programs, such as Social Security and unemployment insurance, CBO's baseline budget projections follow the rules set forth in section 257 of the Balanced Budget and Emergency Deficit Control Act of 1985. Current policies for those programs are defined by the laws that determine liability for taxes or eligibility for benefits. The baseline generally assumes a continuation of those laws as now on the statute books. It also assumes continuation of certain programs of limited duration that are routinely reauthorized.

For defense and nondefense discretionary programs, CBO's baseline budget projections assume that the Congress adheres to the statutory caps on budget authority and outlays through 2002. Once the caps expire, however, no overarching dollar total established in legislation will control discretionary appropriations. The concept of current policy for discretionary spending is therefore ambiguous after 2002. The benchmark established in the Deficit Control Act is the maintenance of real funding--that is, resource levels adjusted for inflation. The CBO baseline assumes that discretionary spending grows at the rate of inflation once the caps expire in 2002. An alternative is to fix the benchmark at a constant nominal (or dollar) level, and CBO publishes budget projections following that approach as well (see Table 1).
 

The budget includes two other categories of spending: offsetting receipts, which encompass Medicare insurance premiums and similar fees and collections, and net interest, which basically reflects the government's interest payments on the national debt. CBO's baseline for offsetting receipts represents the agency's best estimate of the amount that the government will collect under current laws and policies. Net interest, rather than being directly controlled by policymakers, is driven by market interest rates and the outstanding stock of federal debt held by the public. As a result, CBO estimates such spending based on its projections of those two determinants.

Baseline budget projections depend critically on the economic and other estimating assumptions on which they are based. CBO typically illustrates the sensitivity of the budget to the economy in two different ways. One way is through rules of thumb that involve changes in individual economic variables taken in isolation. Another approach is to prepare complete alternative economic forecasts, such as one marked by a recession. In addition, budget projections depend on a host of factors that are not included in the economic forecast. For example, budget estimators must take account of the expected effect on future spending of such factors as birth rates, mortality, disability, and changes in the circumstances of individuals and families. They must also make assumptions about how individual taxpayers, beneficiaries, state and local governments, and providers of health care and other services respond to federal legislation.
 

A BRIEF HISTORY OF THE BASELINE

The genesis of baseline budget projections can be found in the Congressional Budget Act of 1974. That act required the Office of Management and Budget (OMB) to prepare projections of federal spending for the upcoming fiscal year based on a continuation of the existing level of governmental services. It also required the newly established Congressional Budget Office to prepare five-year projections of budget authority, outlays, revenues, and the surplus or deficit. OMB published its initial current-services budget projections in November 1974, and CBO's five-year projections first appeared in January 1976. Today's baseline budget projections are very much like those prepared more than two decades ago, although they now span 10 years instead of five.

The Budget Act was silent on whether to adjust estimates of discretionary appropriations for anticipated changes in inflation. Until 1980, OMB's projections excluded inflation adjustments for discretionary programs. CBO's projections, however, assumed that appropriations would keep pace with inflation, although CBO has also published projections without these so-called discretionary inflation adjustments.

CBO's budget projections took on added importance in 1980 and 1981, when they served as the baseline for computing spending reductions to be achieved in the budget reconciliation process. The reconciliation instructions contained in the fiscal year 1982 budget resolution (the so-called Gramm-Latta budget) required House and Senate committees to reduce outlays by a total of $36 billion below baseline levels, but each committee could determine how those savings were to be achieved. The CBO baseline has been used in every year since 1981 for developing budget resolutions and measuring compliance with reconciliation instructions.

The Deficit Control Act of 1985 provided the first legal definition of baseline. For the most part, the act defined the baseline in conformity with previous usage. If appropriations had not been enacted for the upcoming fiscal year, the baseline was to assume the previous year's level without any adjustment for inflation. In 1987, however, the Congress amended the definition of the baseline so that discretionary appropriations would be adjusted to keep pace with inflation. Other technical changes to the definition of the baseline were enacted in 1990, 1993, and 1997.
 

CURRENT ISSUES

Baseline budget projections increasingly became the subject of political debate and controversy during the late 1980s and early 1990s. Some critics contended that baseline projections create a bias in favor of spending by assuming that federal spending keeps pace with inflation, increases in caseloads, and other factors driving the growth of entitlement programs. Those critics also observed that entitlement spending continued to rise, even as eligibility rules were tightened and payment rates reduced. Changes that merely slowed the growth of federal spending programs, they argued, should not be described as cuts in spending.

Such criticisms seem to be directed more against the concept of entitlements, however, than against the idea of the baseline. After all, the baseline provides useful information about entitlement programs--a projection of what spending would be in the absence of Congressional action. It would be difficult to formulate legislation to control entitlement spending without an estimate of its likely growth under current law.

To aid in understanding the dynamics of the budget, CBO supplements its baseline projections with an analysis of the major causes of growth in entitlements and other mandatory spending (see Table 2). Mounting caseloads account for about one-fifth of the projected growth in entitlement programs over the next 10 years, and automatic increases in benefits account for more than one-third. More than one-quarter of the growth stems from increases in Medicare and Medicaid that cannot be attributed to rising caseloads or automatic adjustments in reimbursement rates.
 

From the beginning, the major issue surrounding the baseline has been whether to include inflation adjustments for discretionary programs. The existence of statutory caps on discretionary spending since 1991 has made this issue less acute, although the problem persists for years after the caps expire. Because this question has at least two possible answers, CBO prepares baseline projections using both approaches. Members of Congress and other users of budget data can select the alternative they prefer.

For the same reason, CBO includes additional information with the cost estimates that it prepares for bills. First, as required by rule XI of the House, CBO's cost estimates include, when practicable, a comparison of estimated funding levels for the relevant program (or programs) with the level of spending under current law. For mandatory programs, such as Medicare, that information allows the reader to identify changes in the rate of growth of a program as well as changes from the baseline (see Table 3). Second, in cases in which a bill does not specify the dollar amount of an authorization, CBO projects the level of spending both with and without adjustments for inflation (see Table 4). CBO's baseline projections of discretionary spending are not used in cost estimates of legislation. New authorizations of discretionary spending are compared with existing appropriations and authorizations (or zero if the program is not authorized in a particular year).
 

Other criticisms of the baseline are more political than technical in nature. One complaint is that baselines promote playing budgetary games. Through the judicious choice of baselines, the same policy may be described either as an increase or as a reduction. Defining the baseline in law, however, has substantially reduced the scope for such gamesmanship.

Some experts on the federal budget process argue that baselines weaken the President's role in budgeting by providing a benchmark for Congressional action that is independent of the President's recommendations. Indeed, the baseline has supplanted the President's budget as the point of reference for measuring the effects of policy changes on tax revenues and entitlement spending. But the Appropriations Committees still generally use the President's request as the starting point for marking up the annual appropriation bills.
 

CONCLUSION

Despite such criticisms, the baseline remains the most commonly used benchmark for assessing the effects of proposed changes in budgetary policies. The baseline has existed in something close to its current form for over 20 years and has been defined in law for over 12 years. As a result, the baseline--and its limitations--are generally well understood.

CBO has responded to concerns about the limitations of the baseline by providing more and more information for our Congressional customers. We prepare projections with and without adjustments for discretionary inflation, analyze the sources of growth in entitlements and other mandatory spending programs, and show how the baseline is sensitive to changes in economic and other estimating assumptions. We also fully disclose the methods and assumptions underlying both our baseline budget projections and our cost estimates for pending legislation. The Congressional Budget Office is continually looking for ways to improve our budget projections and to explain them as fully and clearly as possible. We stand ready to assist the Budget Committee however we can as you look for ways to improve the federal budget process.