Digesting the Congressional Budget Office’s Latest Outlook

August 27, 2002

IntroductionThe Congressional Budget Office (CBO) today released its latest projections of federal budget surpluses/deficits for the fiscal years 2002 through 2012. As widely expected, CBO is now projecting a current year budget deficit of $157 billion. This is a dramatic change from CBO’s past projections of the FY2002 budget outlook, which have ranged over the past twenty months from a surplus of $405 billion to a small deficit of $21 billion.

CBO’s yet-again revised deficit projection should not come as a surprise. The fact of the matter is that, while drastic, such swings in fiscal projections are par for the course. Fiscal forecasting is fraught with difficulties. Margins of error of 50 percent or greater are typical.

Consider:

  • Since George Bush’s inauguration on January 20, 2001, CBO has issued six official updates to their forecast of the FY2002 budget. These updates have taken CBO’s projection of the FY2002 budgetary outlook from an estimated surplus of $405 billion to an estimated deficit of $157 billion. The combined tax relief of EGTRRA and the Job Creation and Worker Assistance Act of 2002 (the economic stimulus bill) account for only 13.5 percent of this $562 billion downward re-estimate.
Why were the Projections of FY 2002’s Surplus Wrong?
Billions of Dollars Percentage Change in Estimate
CBO surplus Estimate for 2002 as of January 20, 2001 + $ 405
(1) Technical Changes — $ 270 48.0%
(2) Economic Changes — $100 17.8%
(3) Legislative Changes

Tax Relief

— $ 76 13.5%

Discretionary Spending

— $ 79 14.1%

Mandatory Spending

— $ 37 6.6%
CBO Current Estimate of FY 2002 deficit — $ 157
Note: Totals may not add due to rounding

  • In moving their FY2002 forecast from a $405 billion surplus to a $157 billion deficit, CBO analysts made total adjustments of $759 billion. Some of these adjustments—made in the categories of tax law changes, discretionary spending, mandatory spending, technical corrections, and economic assumptions—had a positive impact on the budget surplus forecast and others had a negative impact. The $562 billion overall decline in CBO’s projected surplus/deficit between January 1, 2001 and today represents the net decline from adding all these adjustments together.
  • In March 1997, CBO estimated that the FY2002 budget would be $188 billion in deficit. This estimate, made five and a half years ago was closer to what the deficit actually will be for FY2002 than CBO’s March 2002 estimate of a $5 billion budget surplus, made just five and one half months ago.

Most of the adjustments made by CBO are due to changes in the economy or technical changes made by CBO statisticians, not legislative actions. For example, of the $759 billion in total adjustments made by CBO between January 1, 2001 and March 2002 to the FY2002 budget deficit projection, only $192 billion, or 25.3 percent, were due to legislative changes. The remaining 74.7 percent were due to either changes in the economy (27.5 percent of the total adjustments) or technical changes (47.2 percent of the total adjustments).

Far from gospel, the fiscal forecasts released by CBO and other government agencies are at best guesses. That is why lawmakers should not place undue weight on these assessments, and, instead, should base fiscal policy decisions on sound principles of economics.

Legislative ChangesChanges in the law obviously affect fiscal forecasts. A change in tax policy, either a tax increase or a tax decrease, will directly affect tax collections. Changes in annual appropriations from baseline estimates will also have an effect. Finally, even small changes in mandatory federal programs can have a large impact on the overall fiscal outlook of the federal government. All of these changes will affect the level of outstanding debt and therefore annual interest payments and debt service charges.

Of the $759 billion in adjustments CBO has made since January 1, 2001 in reaction to new laws, only $76 billion, or 10.0 percent of total adjustments, were due to changes in the tax code. The primary components of these adjustments were $31 billion in tax relief provided by the Economic Growth and Tax Relief Reconciliation Act of 2001 (EGTRRA) and the $43 billion in tax relief included in the Job Creation and Worker Assistance Act of 2002 signed into law on March 9, 2002.

An additional $79 billion, or 10.4 percent of total adjustments, were made in response to increased discretionary spending. Each year, CBO adjusts its forecast of future discretionary spending based on recently enacted spending bills. Because discretionary spending has far outpaced expectations over the past three years, fueled only in part by increased spending in reaction to the terrorist attacks of September 11th, CBO has reacted by increasing its projections of future year spending.

Major adjustments to discretionary spending over the period included an increase of $29 billion in response to FY2001 appropriations bills passed in 2000, an increase of $10 billion in response to the Supplemental Appropriations Act passed on July 24, 2001, and an increase of $42 billion in response to spending legislation passed in response to FY2002 appropriations bills passed in 2001 and emergency supplemental funds passed in response to the attacks of 9/11.

The remaining $37 billion, 4.9 percent of total adjustments over the period, were adjustments in response to legislative changes in mandatory spending programs including servicing the national debt. The driving factors in this category of changes have been modifications made in 2001 to Medicare benefits ($8 billion), the refundability of the child tax credit to certain individuals who have no income tax liability enacted as part of EGTRRA ($6 billion), and temporary unemployment assistance passed as part of the economic stimulus bill of 2002 ($8 billion).

Economic ChangesChanges in economic conditions include changes in gross domestic product (GDP), inflation, unemployment, and interest rates. These economic factors affect all areas of spending and collection estimates but are most pronounced on tax collections and mandatory spending programs. For example, a slight change in the expected rate of inflation affects medical costs and therefore Medicare outlays. Changes in inflation expectations also affect cost of living adjustments (COLAs) and therefore federal government employee pay, Social Security payments, and outlays through several other entitlement programs.

Since President Bush was inaugurated on January 20, 2001, CBO has made a total of $209 billion in changes to its projection of the FY2002 surplus/deficit due to changing economic conditions. The vast majority of these adjustments, $179 billion or 23.6 percent of total adjustments over the period, have been made in expected revenues. The largest single economic change came in CBO’s January 2002 outlook, in which deteriorating economic conditions accounted for a downward revision in the projected FY2002 surplus/deficit of $100 billion.

Technical ChangesCBO makes technical corrections several times per year to its estimates and projections of federal deficits and surpluses. Technical corrections include many one-time adjustments such as re-estimates of revenue raised through electromagnetic spectrum auctions conducted by the Federal Communications Commission. However, technical corrections, as defined by CBO, also include such important adjustments as re-estimates of tax collections not accounted for by changes in economic conditions and changes in the number of entitlement beneficiaries not due to legislative actions. CBO also includes changes in interest payments and debt service costs associated with technical corrections in this category.

Since the beginning of 2001, CBO has made a total of $357 billion in technical adjustments to their FY2002 surplus/deficit forecast, 47.2 percent of total adjustment over the period. The single largest technical correction ($104 billion) made by CBO over the period came this August when CBO adjusted its individual income, corporate income, and capital gains tax bases in response to the terrorist attacks and the slumping stock market. Other significant technical adjustments over the period have included revised estimates of collections from electromagnetic spectrum sales, technical adjustments to CBO’s method for calculating how tax revenues will change in response to economic changes, technical changes to CBO’s underlying assumption of average Social Security benefit payments.

Another large technical adjustment of note, totally $95.9 billion, came in late January 2001 when CBO stopped including statutory spending limits in their baseline estimates. As CBO noted in their January 2001 Outlook, “Throughout most of the 1990s, CBO’s baseline for discretionary spending assumed adherence to the statutory limits that were originally enacted in 1990 (and extended in 1993 and 1997). However … it is clear from appropriations enacted in recent years that they are no longer a useful measure of current policy or a viable guideline for projecting discretionary spending in the future.” In other words, federal lawmakers were spending so far beyond the caps by January 2001, that CBO felt no need to include the statutory limits in their estimates of future spending levels.

An Important NoteMany commentators and lawmakers place recent legislative changes in the context of CBO’s long range budget forecast. For example, a Center on Budget and Policy Priorities study issued just after CBO’s January 2002 Outlook was released claimed that “The tax cut enacted last June is the largest single component of that $4.0 trillion reduction in projected total surpluses [over the ten-year period 2002-2011].” Kent Conrad, Chairman of the Senate Budget Committee, made the same point more recently, if only more politically, “Despite the notable impact of the recession and war on terrorism, there is simply no escaping the fact that the largest single factor causing the drop in surplus over the ten years is the tax cut – accounting for nearly 40 percent of the loss…”

Such claims are misleading and inaccurate for two primary reasons:

  1. The fact that CBO forecasts a surplus for a future point in time does not mean that a surplus actually exists. Although a basic fact, it is easy to forget that CBO’s forecasts (indeed fiscal forecasts made by any agency or organization) are just that, forecasts, not factual statements about known future occurrences. There are literally thousands of factors that can change between the time CBO estimates a future fiscal position and the actual occurrence of that position. The fact of the matter is that a deficit occurs only once lawmakers spend more than the federal government collects in taxes. Therefore, it is misleading to characterize the effect of any legislation as eroding future surpluses or deficits.
  2. The manner in which CBO creates its forecasts guarantees that legislative changes will always account for a greater percentage of total adjustments to estimates of future fiscal years. CBO, in preparing its estimates of future year deficits and surpluses, incorporates the full impact of changes in legislation but relies upon what is known as mean reversion in making assumptions about future economic activity. Simply stated, CBO incorporates changes in legislation but not economic booms or recessions in its future year estimates. This mix of assumptions is well-established in the macroeconomic forecasting industry, but can be misleading if not properly understood.

Technically, mean reversion relies upon a baseline level of economic growth. For example, in CBO’s newly released Outlook, the agency assumes that nominal GDP will increase by an average of 5.0 percent per year between FY2002 and FY2012. Because CBO’s estimates of FY2002 and FY2003 nominal GDP growth are 2.9 percent and 4.4 percent respectively, the out year estimates must be higher than the average. In fact, for fiscal years 2004 through 2012 CBO’s baseline anticipates an average annual growth rate of 5.3 percent. No year between 2004 and 2012 varies more than a tenth of a point from this mean. The result of incorporating changes in legislation but not economic cycles in future year estimates is that legislative changes are an overstated source of error in those future years. CBO essentially assumes that it will not make any economic or technical mistakes leading up to future years, an assumption that is always proved wrong.


Topics


About the Author


Related Research