Skip Navigation

The Congressional Budget Office (CBO) has prepared this paper as background for a joint hearing on budget estimation procedures to be held by the House and Senate Committees on the Budget. The paper reviews current methods for estimating the budgetary effects of proposed changes in revenues and spending and examines the pros and cons of alternative approaches.

The paper was written by the staffs of CBO's Budget Analysis, Macroeconomic Analysis, and Tax Analysis Divisions, and the General Counsel. Sherry Snyder edited the manuscript, and Kathryn Quattrone prepared the final version of the paper.
 

Robert D. Reischauer
Director
January 1995
 
 

SUMMARY AND INTRODUCTION

CURRENT ESTIMATING PRACTICES

• Macroeconomic Assumptions
• Estimates of Spending Proposals
• Estimates of Tax Proposals

BROADENING THE SCOPE OF BUDGET ESTIMATING: PRACTICAL ISSUES

• Types of Macroeconomic Effects
• Degree of Consensus
• Magnitude of Macroeconomic Effects
• Resources and Timeliness
• Changing the Budget Process

BROADENING THE SCOPE OF BUDGET ESTIMATING: ALTERNATIVE APPROACHES

• Provide Additional Information
• Include Only Structural Effects
• Include Both Cyclical and Structural Effects

CONCLUSION

 

SUMMARY AND INTRODUCTION

The Congressional Budget Act of 1974 requires the Congressional Budget Office (CBO) and the Joint Committee on Taxation (JCT) to provide estimates of the budgetary effects of all legislative proposals reported by a Congressional committee. The act assigns JCT the responsibility for preparing the estimates of most revenue legislation, and CBO does the estimates of spending proposals.

Since the inception of the Congressional budget process in 1975, the House and Senate budget committees have used these estimates to assess whether a bill would breach the spending or revenue totals in the budget resolution or would be subject to a point of order on the floor of the Congress.¹ More recently, the budget committees have also used them to monitor Congressional compliance with the requirements of the Budget Enforcement Act of 1990. That act placed dollar limits on discretionary budget authority and outlays and established a pay-as-you-go (PAYGO) requirement; under PAYGO, changes in legislation affecting revenues and mandatory spending, in total, may not increase the deficit in any year. In 1993 and 1994, the Senate used CBO's and JCTs estimates to enforce procedural rules requiring that legislation not increase the deficit in the budget year, the first five years (starting with the budget year), and the second five years. As additional constraints have been added to the Congressional budget process, the estimates have become increasingly important.

The Balanced Budget Act of 1985, as amended by the Budget Enforcement Act, also assigns important estimating responsibilities to the Office of Management and Budget (OMB) in the executive branch. OMB, in turn, depends on the Department of the Treasury for its revenue estimates. At the end of each Congressional session, OMB tabulates the estimated effect of all legislation subject to the pay-as-you-go requirement. If OMB determines that legislation has added to the deficit, it must order an across-the-board reduction (or sequestration) in all nonexempt mandatory spending programs sufficient to eliminate the excess. Similarly, if OMB estimates that the discretionary spending limits have been breached, it must order a sequestration of spending authority for nonexempt discretionary programs.

CBO, JCT, OMB, and the Treasury employ the same basic estimating conventions. Those conventions are simple and practical but not without their limitations. They have served best when changes in policy have been small and when concern has been focused on the budget totals. Their limitations are most apparent when policymakers are considering substantial changes or when they are interested in identifying the full effects of individual proposals.

Current budget enforcement procedures reflect these basic estimating assumptions. As specified by the budget committees, CBO's estimates of spending proposals and JCT's revenue estimates have been consistent with the economic assumptions used in preparing the annual budget resolution. As required by the Balanced Budget Act, OMB's PAYGO estimates must use the economic assumptions underlying the President's budget submission. The economic assumptions associated with the budget resolution and the President's budget have been reasonably consistent with the respective tax and spending proposals, as well as the resulting fiscal policy. If the proposals are adopted, that approach produces estimates of total revenues and outlays that correctly reflect the new fiscal policies. But the procedure does not attempt to measure the full incremental cost or savings of an individual proposal that would affect the economic aggregates--for example, gross domestic product (GDP) or the price level.

The budget estimates are also based on numerous assumptions about the microeconomic effects of the proposed policies--that is, how those policies might change individual behavior in response to new economic incentives. These behavioral and other technical estimating assumptions cover a wide variety of effects and reflect recent research and the best available estimating practices. For example, the estimate of a proposal to subsidize health insurance for early retirees would include the cost associated with the increase in the number of Social Security beneficiaries that would occur. Similarly, the estimate of a proposal to increase the excise tax on tobacco products would take into account the resulting decrease in consumption of cigarettes.

For the vast majority of bills the Congress considers, these estimating conventions are accepted and noncontroversial. Disputes occasionally arise, however, in two sorts of situations. First, estimators sometimes differ in their assumptions about the size of the microeconomic responses. In 1990, for example, the Bush Administration and JCT had different estimates of the Administration's proposal to reduce the tax rate on capital gains. Although the Administration and JCT assumed similar types of behavioral responses, the former estimated an increase in receipts and the latter a reduction. The Administration believed that reducing the tax rate on capital gains would permanently increase realizations of gains by enough to offset the revenue loss from the lower rate. JCT judged that after an initial surge, the increase in realizations would offset much but not all of the rate reduction.²

Second, although enactment of some proposals might affect the overall economy, the estimates traditionally exclude macroeconomic responses. Some tax or spending bills might affect aggregate demand (that is, total spending in the economy) and, if not offset by changes in monetary policy, restrain or stimulate the economy in the short run. Other proposed legislation could alter the supplies of labor, capital, or technology that determine the potential growth of the economy in the long run. In both cases, use of the economic assumptions underlying the budget resolution or the President's budget could lead to an over- or underestimate of the budgetary effects of the proposal. Whether the full budgetary effects of a proposed policy change are attributed to the bill or included in the economic assumptions can make the difference between whether the bill can pass on a simple majority or whether it needs 60 votes in the Senate. It can also determine whether OMB is required to order a PAYGO sequestration.

In theory, estimators could incorporate macroeconomic effects into budget estimates, thereby providing more information to the Congress and a more comprehensive basis for pay-as-you-go scoring. But in practice, such a change would also raise several difficult issues.

• The results could depend greatly on assumptions about the behavior of the Federal Reserve Board.
  
  
• In situations in which little consensus exists on the magnitude of the macroeconomic effects, the estimates would be subject to considerable controversy and uncertainty.
  
  
• The macroeconomic effects occurring within the usual five-year estimating period might not accurately represent the proposal's long-term economic gains or losses.
  
  
• For many bills, preparing estimates would take substantially more time and require more resources.
  
  
• Counting macroeconomic effects for only some bills could reduce delays, but that approach would raise serious problems of consistency and fairness,
  
  
• To make the budget process consistent with the new estimating approach, the Congress would have to change the Congressional Budget Act and the Balanced Budget Act to reflect the interrelated effects of tax and spending proposals.

Including a proposal's macroeconomic effects in budget estimates is sometimes described as "dynamic," as opposed to "static," estimating. As has been indicated, however, the estimates of CBO, JCT, and OMB are not static in that they already incorporate a wide variety of behavioral changes in response to changes in economic incentives. Therefore, the labels "dynamic" and "static" are misleading and are not used in this paper.

1. A point of order is an objection raised on the House or Senate floor to an action that would violate the body's rules. Usually, a point of order may be waived by a majority vote. In the Senate, however, waiving a point of order for a violation of the Budget Act usually requires a three-fifths vote.

2. Congressional Budget Office, How Capital Gains Tax Rates Affect Revenues: The Historical Evidence (March 1988); Joint Committee on Taxation, Explanation of Methodology Used to Estimate Proposals Affecting the Taxation of Income from Capital Gains, JCS-12-90 (March 27,1990).

This document is available in its entirety in PDF.