The Congressional Budget Office (CBO) projects that if current policies remained unchanged, federal budget deficits--which reemerged in 2002 after four consecutive years of surpluses--would peak in 2003, decline steadily thereafter, and again yield to small but growing surpluses beginning in 2007. That improving outlook, however, is bound to the assumption that no policy will change, and as such should be viewed cautiously. For example, the major provisions of the tax cut enacted in 2001 are due to expire at the end of 2010. If policymakers extended those provisions, or made them permanent, projected surpluses would decrease significantly after 2010. Also, there is likely to be strong pressure in the 108th Congress for new initiatives to increase spending and reduce taxes--and a war in Iraq would necessitate additional outlays. Those changes could boost deficits considerably in the near term and delay or even prevent a return to surpluses over the next 10 years. Beyond that horizon loom budgetary pressures linked to the aging of the baby-boom generation, which could lead to unsustainable levels of deficits and debt over the longer term.

CBO's projections under current tax and spending policies show total budget deficits of $199 billion in 2003 and $145 billion in 2004--or, as a percentage of gross domestic product (GDP), 1.9 percent and 1.3 percent, respectively (see Table 1-1 and Table 1-2).⁽¹⁾ Those projections have been adjusted to incorporate the assumption that budget authority for discretionary appropriations for 2003 will total about $751 billion (see Box 1-1). That amount is about $12 billion more than the amount available for the year under the temporary continuing resolution that was in effect when CBO prepared this report.

Under CBO's adjusted baseline, deficits would continue to shrink after 2004, and a small budget surplus of $26 billion would emerge in 2007. Over the 2004-2008 period, by CBO's estimates, the cumulative deficit would total $143 billion, or 0.2 percent of GDP. Over the following five years, surpluses would steadily mount and, for the full 10-year projection period from 2004 to 2013, accumulate to $1.3 trillion. However, over 90 percent of that amount would be recorded in the years 2011 to 2013--that is, after the 2001 tax cuts are scheduled to expire and when the projections are the most uncertain.

Unlike total surpluses, on-budget surpluses--which exclude the off-budget transactions of Social Security and the Postal Service--would not reappear until 2012 in CBO's adjusted baseline. Although projections of off-budget transactions (which are dominated by Social Security) show net surpluses every year through 2013, the rest of the budget is projected to post deficits of $361 billion in 2003, $319 billion in 2004, and slowly declining amounts through 2011.

CBO developed its latest projections following a period of significant economic and fiscal change. As recently as January 2001, CBO was projecting record levels of surpluses for the 2002-2011 period--totaling $5.6 trillion--under its baseline assumptions. That estimate reflected years of robust economic growth and surging federal revenues--but later proved to be the high-water mark. The recession in 2001 (and a declining stock market) together with the terrorist attacks of September 11--and lawmakers' responses to those events--caused a sharp drop in federal revenues and a spike in spending in 2002, which led to similar changes in CBO's estimates for later years. Major new policies, including the tax cuts enacted in the Economic Growth and Tax Relief Reconciliation Act of 2001 (EGTRRA), a sizable boost in regular appropriations, and other initiatives, contributed to those trends. Now, just two years later, CBO estimates that the projected cumulative surplus for the 2002-2011 period has been all but eliminated.

Despite that dramatic reversal, the budget outlook over the next decade (2004 to 2013) under the assumptions of CBO's adjusted baseline remains relatively bright, by historical standards. Before 1998, the government had recorded deficits in every year since 1969. Moreover, the shortfalls for 2002 and 2003--1.5 percent and 1.9 percent of GDP, respectively--are relatively small when compared with the chronic deficits of the 1980s and early 1990s, which ranged from 3 percent to 6 percent (see Figure 1-1). Also, the amount of federal debt held by the public, which for the most part reflects government borrowing to finance past deficits, is projected to decline relative to GDP throughout the 2004-2013 period. (See the discussion of federal debt later in this chapter.) Nevertheless, the return of deficits after a decade of improving federal finances illustrates how quickly the nation's budgetary fortunes can change. It also shows how closely the budget is linked to the uncertain fiscal and economic circumstances that lawmakers now confront.
 

Figure 1-1.

Total Deficits and Surpluses as a Share of GDP, 1967-2013

(Percentage of GDP)

Source: Congressional Budget Office.

a. These projections incorporate the assumption that discretionary budget authority totals $751 billion for 2003 and grows with inflation thereafter.

 

Uncertainty and the Projection Horizon

Budget projections are always subject to considerable uncertainty. CBO's adjusted baseline shows future spending and revenues under current laws and policies--even though those laws and policies will almost certainly change. Thus, the actual budget totals for the projection period are virtually guaranteed to differ from the estimates in this report, and perhaps substantially. This year, however, the uncertainty that normally accompanies CBO's baseline projections is heightened.

Certain current policies as they are now reflected in the baseline may prove to be unrealistic. The major tax-cutting provisions of EGTRRA are scheduled to expire at the end of December 2010, and if they do, tax rates will rise to their pre-2001 levels. But many people contend that it is unrealistic to assume that lawmakers would permit that to happen. Allowing those provisions to expire, as current law provides, would significantly boost revenues for 2011 through 2013. And that upswing is the main reason that the baseline shows large surpluses for that period. If those and other expiring tax cuts were made permanent, the total 10-year surplus in CBO's adjusted baseline would be essentially eliminated. (Box 1-2 discusses the effects on federal revenues of extending expiring tax provisions.)
 

Box 1-2. The Expiration of Revenue Provisions The budget outlook for the next 10 years is strongly affected by the scheduled expiration of various revenue provisions.1 The Economic Growth and Tax Relief Reconciliation Act of 2001 (EGTRRA) plays a big role: three items are scheduled to end on or before December 31, 2006, and the rest of the law's provisions--which represent the bulk of its budgetary cost--expire on December 31, 2010. Another major impact would come from the economic stimulus law that policymakers enacted in March 2002 (the Job Creation and Worker Assistance Act of 2002). That law established new tax cuts for businesses; in most cases, those cuts end during the next two years. And many other provisions of the tax code that were enacted before EGTRRA are scheduled to expire over the next decade. By law, the Congressional Budget Office's (CBO's) baseline budget projections incorporate the assumption that almost all expiring tax provisions end as scheduled. (The only exception is for expiring excise taxes dedicated to trust funds.) An alternative measure of the long-term budgetary effects of current policy could incorporate a different assumption: that all of those expirations do not occur as scheduled and instead the provisions are immediately and permanently extended. Under that assumption, as the Joint Committee on Taxation (JCT) and CBO estimate, federal revenues would be $1.2 trillion lower during the 2004-2013 period than the amount projected in CBO's adjusted baseline (see the table below). About two-thirds of that estimated decline ($785 billion) would come from extending EGTRRA. And about 85 percent of that EGTRRA-related drop would occur from 2011 to 2013, immediately after most of the law's provisions are scheduled to expire. Some effects, however, would be felt earlier. For example, extending the changes that the law made to estate and gift taxes could reduce revenues as early as 2003--because if taxpayers knew that those changes would become permanent in 2011, some people might postpone until then making some taxable gifts that they would otherwise have made earlier in the decade.                                   Effects on Revenues of Extending Expiring Tax Provisions (In billions of dollars)   2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Total, 2004- 2008 Total, 2004- 2013 Economic Growth and Tax Relief Reconciliation Act of 2001     Provisions expiring in 2010 * -1 -1 -1 -1 -2 -2 -2 -131 -230 -240 -5 -610   Provisions expiring before 2010a n.a. n.a. -3 -12 -17 -22 -26 -29 -25 -18 -21 -55 -175     Subtotal * -1 -4 -13 -19 -24 -28 -32 -156 -249 -260 -60 -785 Job Creation and Worker Assistance Act of 2002b n.a. * -28 -42 -40 -35 -30 -26 -22 -20 -19 -145 -262 Other Expiring Tax Provisionsc * 1 -3 -8 -12 -15 -17 -20 -23 -27 -30 -36 -152 Estimated Interaction Effects from Enacting All Provisions Simultaneously 0 0 1 1 1 1 1 1 -4 -12 -12 4 -23       Total Effect on Revenues * * -34 -61 -69 -73 -74 -76 -206 -308 -321 -237 -1,222 Memorandum:   Total Effect on Revenues Excluding the Job Creation and Worker Assistance Act * * -6 -19 -30 -37 -44 -50 -184 -288 -302 -93 -960 Sources: Congressional Budget Office; Joint Committee on Taxation. Notes: The estimates incorporate the assumptions that the expiring provisions are extended immediately rather than when they are about to expire and that they are extended at the rates or levels existing at the time of expiration. The estimates do not include effects on debt-service costs. When this report went to press, JCT's estimates were unavailable for several expiring tax provisions--most significantly, for EGTRRA's major individual income tax provisions that expire in 2010 and for the provisions of the alternative minimum tax (AMT) that expire in earlier years. CBO estimated the effects of extending those provisions and of the interaction from extending all expiring tax provisions simultaneously. As a result, cost estimates by JCT for legislative proposals to extend the EGTRRA and AMT provisions might not match the figures shown here. n.a. = not applicable; * = between -$500 million and $500 million. a. Includes the increased exemption amount for the alternative minimum tax (expires in 2004), the deduction for qualified education expenses (expires in 2005), and the credit for individual retirement accounts and 401(k)-type plans (expires in 2006). b. New provisions in the Job Creation and Worker Assistance Act that are scheduled to expire include special depreciation-expensing allowances for certain property and tax benefits for the area of New York City that was damaged in the September 11 terrorist attacks. The provisions that allowed a special five-year carryback of net operating losses have already expired and are not included in these estimates. The estimates also do not include provisions in the law that had existed and been extended in previous years. The effects of extending those provisions again are included in the line for other expiring tax provisions. c. Includes numerous items, such as the tax credit for research and experimentation. Under a more limited alternative measure, all expiring tax provisions would be extended except the ones created by the economic stimulus law, which were not intended to be permanent. (Those provisions include allowing businesses to take an additional first-year deduction for depreciation of certain property and targeting tax benefits to the area of New York City that was damaged in the September 11 terrorist attacks.) If all but those expiring provisions were extended, federal revenues would be $960 billion lower during the 2004-2013 period, JCT and CBO project. 1.  The provisions' expiration can also be expected to affect the economy, but only some of those effects are reflected in the estimates presented here--for example, the estimates do not reflect macroeconomic changes. (For a discussion of those effects, see Box 2-1.)

Other factors might also create strong budgetary pressures this year and in later years, leading to changes in current spending or revenue policies that could increase deficits or diminish surpluses. For example, the nation continues to fight the war on terrorism, which may lead to additional spending. The possibility of war with Iraq clouds the budgetary picture as well, with its uncertain costs and possible economic effects (see Box 1-3). Lawmakers are also under pressure to enact new tax and spending legislation to stimulate the sluggish economy. And there is interest in enacting other costly initiatives, such as a prescription drug benefit for Medicare beneficiaries and changes in the alternative minimum tax.
 

Box 1-3. An Estimate of the Costs of a Potential Conflict with Iraq Recently, the Congressional Budget Office (CBO) was asked to gauge the costs of activities related to possible military operations in Iraq.1 Estimates of the total cost of a military conflict with Iraq and such a conflict's aftermath are highly uncertain. They depend on many factors that are unknown at this time, including the size of the force that is deployed, the strategy to be used, the duration of the conflict, the number of casualties, the equipment lost, and the need for reconstruction of Iraq's infrastructure. Of the many force levels that might be used to prosecute such a war, CBO examined two representative examples. Both alternatives were based to some extent on the forces that the Department of Defense (DoD) had previously indicated it would require for a major theater war. The first of CBO's examples emphasized U.S. ground forces. This so-called Heavy Ground option would include about five Army divisions and five Air Force tactical fighter wings. The second option relied more on air power. Termed the Heavy Air option, it would comprise two and one-third Army divisions and 10 Air Force tactical fighter wings. Using those forces, CBO employed various methods to develop its estimates, including the use of data on the cost of prior and current military operations--most notably, those in the Balkans, Afghanistan, and Desert Shield/Desert Storm. Using those two examples, CBO estimated that the incremental costs of deploying a force to the Persian Gulf (that is, the costs that would be incurred above those budgeted for routine operations) would be between $9 billion and $13 billion. Prosecuting a war would cost between $6 billion and $9 billion a month--although how long such a war might last could not be estimated. After hostilities ended, the costs to return U.S. forces to their home bases would range between $5 billion and $7 billion, CBO estimated. Further, the incremental cost of an occupation following combat operations would vary from about $1 billion to $4 billion a month. CBO had no basis for estimating any costs for reconstruction or for foreign aid that the United States might choose to extend after a conflict had ended. Many alternative force structures--other than the two options that CBO used in its estimates--could be fielded. And whatever forces were used, multiple unknown factors would characterize any scenario of how a conflict with Iraq might actually unfold. On the one hand, if the Iraqi leadership or selected elements of its military forces quickly capitulated, ground combat could be of short duration, as in Desert Storm. On the other hand, if the leadership and military chose to fight, Iraq's use of chemical or biological weapons (CBW) against regional military or transportation facilities could extend the war, as could the need to engage in protracted urban fighting. Given those uncertainties, CBO's estimates of the monthly costs of operations exclude expenditures for decontaminating areas or equipment affected by CBW attacks. A war in Iraq could lead to substantial costs in later years that were not included in CBO's estimates, either because their magnitude could not be assessed even roughly or because they depended on highly uncertain decisions about future policy. For example, the United States might leave troops or equipment in Iraq, which could require the construction of new military bases. Sustaining the occupation over time could require either increases in overall active-duty and reserve force levels or major changes in current policies on basing and deployment. The United States might provide Iraq with funds for humanitarian assistance and reconstruction. And substantial aid might be provided in the future to allies and other friendly nations in the region. 1.  See CBO's letter to Senator Kent Conrad and Congressman John M. Spratt, Jr., on September 30, 2002, Estimated Costs of a Potential Conflict with Iraq.

Another source of considerable uncertainty in the budget outlook is the accuracy of the economic and technical assumptions that underlie CBO's adjusted baseline. The economy is recovering slowly from the 2001 recession. CBO's baseline budget projections hinge in part on estimates of the timing and strength of that recovery (see Chapter 5 for more details). And technical factors that influence revenue collections--such as the behavior of the stock market and changes in taxable income--could also determine whether federal revenues bounce back as projected (see Chapter 3).

Uncertainty compounds as the projection horizon lengthens. Even small annual differences in the many key factors that influence CBO's budget projections--factors such as inflation, increases in productivity, economic growth, the distribution of income, and rates of growth for Medicare and Medicaid spending--can add up to substantial differences in budgetary outcomes 10 years from now. For details of how changes in several key assumptions would affect the budget outlook, see Appendix C.

Given such uncertainty, five-year projections may be more useful than 10-year numbers. As noted earlier, CBO's current 10-year projections of revenues are significantly influenced by the scheduled expiration of EGTRRA at the end of 2010. Also, the budget horizon has now shifted forward one year, which eliminates the year in which the deficit is estimated to peak (2003) and adds a year in which the baseline surplus is projected to be large and perhaps artificially high (2013). To provide a more complete budgetary picture, many of the tables in this report show both five-year (2004 to 2008) and 10-year (2004 to 2013) totals for the adjusted baseline.

Nonetheless, the longer term (beyond the 10-year horizon) is a critical consideration for lawmakers as the baby-boom generation ages. The worsening of the budget outlook since January 2001--along with its heightened uncertainty--exacerbates the budgetary challenges that lurk beyond the 10-year projection period. Toward the end of that span, the baby-boom generation will begin qualifying in large numbers for Social Security and Medicare benefits, putting increased pressure on those programs. And by 2030, the number of workers paying Social Security and Medicare taxes is expected to rise by only about 15 percent while the number of beneficiaries of those programs is projected to balloon by about 80 percent. Growth in the number of beneficiaries, combined with increases in life expectancy, will boost spending for long-term care, about half of which is financed by Medicaid and Medicare.⁽²⁾ Together, demographic changes and the growth of medical costs are projected to push total federal spending for Medicare, Medicaid, and Social Security as a share of GDP sharply higher over the next few decades.
 

The Return of the Deficit in 2002

The $158 billion budget deficit in 2002 marked the end of a period of surpluses--four consecutive years--the likes of which had not been seen since the late 1920s. The total shortfall for 2002 was a net reversal of $285 billion from the $127 billion surplus recorded for 2001. The on-budget deficit was $317 billion, and the off-budget surplus was $160 billion.

Revenues fell for the second consecutive year in 2002, following annual increases from 1994 through 2000 that averaged more than 8 percent. The decline in 2002 revenues of nearly 7 percent ($138 billion) was the largest percentage drop since the mid-1940s; it stemmed primarily from the weak economy, fewer realizations of capital gains, and, to a much smaller extent, the tax cuts enacted in the past two years. Declines in the two major sources of revenues were even greater, on a percentage basis, than the overall drop. Revenues from individual income taxes in 2002 were 14 percent lower than in the previous year. (Although the tax cuts enacted in 2001 and 2002 held down the growth of revenues from that source, those revenues would have fallen by approximately 10 percent over the year, by CBO's estimates, even without the cuts.) In recent years, revenues from corporate sources have followed a similar path. After growing at an average annual rate of almost 7 percent from 1994 through 2000, they fell off sharply after corporate profits began declining in 2000.

While revenues dwindled in 2002, outlays grew by $147 billion, topping $2 trillion for the first time. Large increases in appropriations for both defense and nondefense programs, a steep rise in payments for unemployment benefits, and substantial growth of Medicaid outlays led to the largest percentage jump in noninterest spending since 1981--about 11 percent. Defense outlays (including a shift in payment dates) grew by 14 percent in 2002; more than half of that growth was due to initiatives that were in place before the September 11 terrorist attacks, CBO estimates. The rise in nondefense discretionary spending was spread among numerous programs--three areas with the largest increases were health, education, and transportation. The slowdown in the economy caused the unemployment rate to peak at 6.0 percent in late 2002, which resulted in a record amount of spending for unemployment compensation--$51 billion (including $8 billion in extended benefits.) Medicaid spending also grew rapidly, increasing by more than 13 percent over the previous year's level.
 

The Concept Behind CBO's Baseline

The projections that make up CBO's baseline are not intended to be predictions of future budgetary outcomes but rather CBO's best judgment about how the economy and other factors will affect federal revenues and spending under current laws and policies. CBO constructs its baseline according to rules set forth in law, mainly in the Balanced Budget and Emergency Deficit Control Act of 1985 and the Congressional Budget and Impoundment Control Act of 1974. In general, those laws instruct CBO and the Office of Management and Budget to project federal spending and revenues under current policies. Lawmakers can use the baseline as a neutral benchmark to measure the effects of proposed changes in tax and spending policies.

For revenues and mandatory spending, the Deficit Control Act requires that the baseline be projected under the assumption that current laws continue without change. In most cases, the laws that govern revenues and mandatory spending are permanent. The baseline projections reflect anticipated changes in the economy, demographics, and other relevant factors that affect the implementation of those laws.⁽³⁾

The baseline rules are different for discretionary spending, which is governed by annual appropriation acts. The Deficit Control Act states that after the current year, projections of discretionary budget authority should be adjusted to reflect inflation--using specified indexes--as well as other factors (such as the cost of annualizing adjustments to federal pay). That approach to developing baseline projections can be problematic when lawmakers do not complete action on all of the appropriation acts, as is the case this year. Programs that have not yet received full-year funding are operating, as discussed earlier, under a continuing resolution that expires on January 31, 2003. However, the President and the Republican leadership in the Congress have apparently agreed on a total funding level of about $751 billion for all of the regular appropriations for 2003. CBO therefore has adjusted its baseline to incorporate that assumption--pending enactment of the remaining discretionary appropriation bills--and extrapolated that funding level over the next 10 years (adjusting it for projected rates of inflation and other specified factors).

By convention, CBO has prepared another benchmark for discretionary spending. Lawmakers sometimes use a freeze in appropriations--a set amount of budget authority without an adjustment for inflation--to gauge the impact of proposed levels of discretionary spending. The budget outlook under an effective freeze of $751 billion per year is shown in Box 1-4.
 

Box 1-4. The Budgetary Effects of Freezing Total Discretionary Appropriations at $751 Billion Some lawmakers view a freeze in discretionary spending as the most logical starting point from which to measure the effects of appropriations--rather than a baseline for such spending based on the assumption that spending would grow with inflation. If total discretionary appropriations were effectively frozen at $751 billion and current policies remained unchanged, by CBO's estimates the budget would return to surplus in 2006. Under that scenario, the total budget surplus would equal 4.5 percent of gross domestic product (GDP) by 2013 (see the table below). At that point, discretionary outlays would be 4.4 percent of GDP, down from the share of 7.4 percent that CBO's adjusted baseline anticipates for 2003. Under the adjusted baseline, discretionary spending would be 5.7 percent of GDP in 2013.                                 The Budget Outlook Assuming That Discretionary Appropriations Are Frozen at $751 Billion (In billions of dollars)   Actual 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Total, 2004- 2008 Total, 2004- 2013 On-Budget -317 -361 -307 -238 -177 -127 -75 -23 30 186 384 470 -925 121 Off-Budget 160 162 175 195 212 232 251 269 287 304 318 332 1,064 2,574   Total Surplus or Deficit (-) -158 -199 -133 -43 35 104 176 245 316 490 702 802 139 2,695 Memorandum: Total Surplus or Deficit (-) as a Percentage of GDP -1.5 -1.9 -1.2 -0.4 0.3 0.8 1.3 1.7 2.0 3.0 4.1 4.5 0.2a 1.9a Source: Congressional Budget Office. a. As a percentage of cumulative GDP over the period.

 

Changes in CBO's Projections Since August 2002

CBO's projection of the cumulative surplus for the 2003-2012 period has fallen by $385 billion since last summer (see Table 1-3). By convention, CBO attributes changes in its projections to three factors: recently enacted legislation; modifications to its outlook for the economy; and changes in other conditions that affect the budget (a category labeled technical).⁽⁴⁾

Revisions that are technical in nature account for essentially the entire decline in the projected surplus relative to CBO's previous estimates; changes that fall into the other two categories are much smaller and almost completely offset each other. Legislative actions (including the apparent agreement to set the level of total appropriations at $751 billion for 2003) have lowered the projected cumulative surplus by $64 billion for the 2003-2012 period. However, changes stemming from revisions in CBO's economic forecast add $67 billion to the 10-year surplus estimates.

Legislative Changes
Relatively little legislation affecting the budget has been enacted since CBO last published its baseline.⁽⁵⁾ Legislative actions have increased CBO's projections of revenues and outlays over the 2003-2012 period by $5 billion and $68 billion, respectively. Included in the projection of outlays is the adjustment to CBO's baseline to account for the level of discretionary spending for 2003--$751 billion--that appears to have been agreed to by the President and the Congress's Republican leadership. As a result, discretionary budget authority for nondefense programs totals $369 billion in CBO's adjusted baseline--or $17 billion below the level that CBO had projected in August by inflating 2002 appropriations. Using the adjusted level as the basis for projections through 2013 results in a cumulative drop in nondefense outlays of $112 billion.

Two of the 13 regular appropriation acts--defense and military construction--have already been enacted, and they provide funding for 2003 that is about $13 billion above August's baseline levels. However, some defense programs are funded in other appropriation acts. Under CBO's adjusted baseline, those programs are funded at the levels in the current continuing resolution, which are marginally lower than the levels projected in the August baseline. Over the next decade, additional appropriations for defense discretionary programs are projected to boost outlays by $137 billion. Combining that addition with the lower level of spending for nondefense programs brings total discretionary outlays in CBO's adjusted baseline for the 2003-2012 period to a cumulative $25 billion above the amounts projected in August.

Other legislative changes have raised CBO's projection of mandatory outlays (excluding debt-service costs) by about $24 billion through 2012. About one-third of that amount will be spent in 2003; it stems from the five-month extension of certain unemployment benefits enacted in Public Law 108-1.⁽⁶⁾ The Terrorism Risk Insurance Act of 2002 (P.L. 107-297), which would provide financial assistance to insurers for certain losses from future terrorist acts, will also increase projected mandatory outlays over the next 10 years by $6 billion. (CBO based that projection on assumptions about various outcomes of terrorist attacks--ranging from no damages to very large effects.) During the 10-year period, approximately half of that cost would be offset by revenues collected from assessments on the insurance industry.

Another change in projected outlays that CBO has attributed to legislation is the additional interest payments on the government's debt. Because legislative actions since August have decreased projections of the cumulative surplus over the 2003-2012 period, debt-service costs in the adjusted baseline would be $20 billion higher over that decade, CBO estimates.

Economic Changes
Economic revisions to the baseline have added a relatively small amount to the projection of the cumulative surplus. In light of recent developments, CBO has lowered its forecast for short- and long-term interest rates, inflation, wages and salaries, and corporate profits. (For a detailed discussion of CBO's new economic forecast, see Chapter 2.) Those revisions in turn reduce projections of both revenues and outlays, leading to an increase of $67 billion in the projected cumulative surplus over the 2003-2012 period.

Revenues. A dimmer outlook for nominal income has reduced CBO's projections of revenues by $146 billion over the 10-year period. Over half of that drop stems from the assumption, beginning in 2011, of a slightly slower rate of growth of aggregate income than CBO had previously used. Over the 2003-2012 period, lower projections of personal income reduce revenues from both individual income and social insurance taxes by $168 billion. But partially offsetting that decline is an upward reestimate of corporate profits in the near term. That change increases projected revenues from corporate income taxes by $30 billion over the decade.

Outlays. Revisions to CBO's economic forecast reduce its projection of spending over the 2003-2012 period by $213 billion--which more than offsets the change in revenues that was attributed to economic factors. The impact of lower interest rates on net interest payments explains a large part of the decline in projected spending. An additional factor is lower projections of certain measures of inflation, which reduce estimated outlays for Social Security and Medicare.

Compared with its August outlook, CBO has lowered its forecast for interest rates on three-month Treasury bills by nearly 110 basis points for 2003 and 165 basis points for 2004. (A basis point is one-hundredth of a percentage point.) Similarly, CBO has lowered its forecast for rates on 10-year Treasury notes by almost 100 basis points for 2003 and about 70 basis points for 2004. Those lower estimated rates decrease projections of net interest costs by $90 billion over the 2003-2012 period; nearly 70 percent of those savings would accrue through 2005.

Although mandatory spending flows from the provisions of permanent laws, the growth or contraction of many mandatory programs is keyed to the economy. Thus, lower estimated wage growth and cost-of-living adjustments in large part have led CBO to reduce its 10-year projections of spending for Social Security (by $49 billion) and Medicare (by $18 billion). For unemployment compensation, revisions to CBO's economic forecast did not result in a substantial change in projected spending over the decade. In the near term, however, CBO now projects $2 billion less in unemployment compensation for 2003, $3 billion more in such spending for 2004, and $2 billion more for 2005.

Because changes in CBO's economic forecast increase projected surpluses, debt-service costs are projected to decline by $31 billion over the 10-year period, with most of the change occurring over the latter half of the projection horizon.

Technical Changes
Reestimates that cannot be ascribed either to legislative actions or to changes in CBO's economic assumptions have reduced the projected cumulative surplus for the 2003-2012 period by $388 billion. Almost a quarter of that change is the additional debt-service costs that result from all technical revisions.

Revenues. Since August, CBO has cut its projection of revenues for 2003 through 2012 by $67 billion. The largest revision--$140 billion over the 10-year period--flows from the smaller amount of revenues projected for individual and social insurance tax collections. Offsetting $65 billion of the decline, however, are higher projections of revenues from corporate sources.

The reestimate of revenues is based on several factors. First, the weak performance of the stock market in 2002 led CBO to reduce its projection of revenues from capital gains realizations in the near term. (CBO has not changed its assumptions about the long-term relationship of capital gains realizations to GDP.) Second, current revenue collections are running below the amounts that might be expected given the level of economic activity, capital gains, retirement distributions, and other factors that influence the effective tax rate. CBO's projections incorporate the assumption that the shortfall will continue in the near term but diminish in later years. Third, CBO has reduced its projections of revenues from social insurance taxes largely because of new information about the composition of recent receipts.

Higher projections of income taxes paid by corporations partially offset the downward reestimate for revenues. Last summer, CBO recognized that corporate tax receipts were lower than anticipated, given economic conditions, and projected that shortfalls would continue. CBO now believes that some of the weakness will be temporary. Evidence suggests that a portion of the drop-off in corporate revenues occurred because corporations had been receiving larger-than-expected "carryback" refunds, mainly as a result of temporary provisions enacted last year in the Job Creation and Worker Assistance Act of 2002 (P.L. 107-147).⁽⁷⁾ That high level of refunds will persist in 2003, CBO expects. However, as provisions in that act expire, refunds are likely to return to more typical levels.

Outlays. Technical reestimates increased projections of spending for both discretionary and mandatory programs by a total of $321 billion over the 2003-2012 period. Of that amount, discretionary outlays account for $34 billion, mostly for nondefense programs. Revisions in the projections for the Section 8 housing program, which derive from higher-than-anticipated costs for rent subsidies, are the largest contributor to the rise in nondefense discretionary spending. For defense discretionary outlays, increases are mainly related to the accrual charge that pays for the health care of future military retirees, their dependents, and surviving spouses. Because the estimated payments for that accrual charge add to other costs for military personnel, CBO has adjusted its projection of the inflators applied to personnel spending to more accurately reflect the charge's future cost.

On the mandatory spending side, technical reestimates have increased projections of outlays for many programs. For example, expectations of faster growth in numbers of participants have contributed to higher projected outlays for both Social Security and veterans' compensation over the 10-year period. CBO also increased its projections of Medicare outlays over the decade by $68 billion, mostly because higher-than-anticipated spending was recorded in 2002 for hospice care, outpatient services furnished by facilities or nonphysician professionals, and ancillary services (such as prosthetics, orthotics, and durable medical equipment; laboratory tests; ambulance services; and outpatient prescription drugs).

Since the summer, CBO has also increased its projection of spending for the Commodity Credit Corporation (CCC), raising it by $15 billion over the 10-year period. (The CCC makes loans and payments to farmers to support farm income and prices.) In the near term, the projection is lower than it was last August because drought has spurred recent increases in crop prices; over the longer term, however, CBO expects that those prices will fall and push CCC outlays higher. In addition, CBO has modified its baseline estimating procedures to account for variations in future commodity prices, which should provide more-accurate projections of agricultural spending over the next decade.

CBO's projections for unemployment insurance and spectrum-related transactions have also risen. Outlays for unemployment insurance are projected to be $17 billion higher during the 2003-2012 period because of an upward adjustment in the estimated average benefit. Contributing to that change were revised estimates of the impact of legislation previously enacted in California, which nearly doubles the state's maximum benefit by 2005. (Unemployment insurance is a joint federal/state program, and federal outlays are tied to the eligibility requirements and benefit levels set by each state.) CBO has lowered its projection of the amounts that are likely to be paid for licenses to use the electromagnetic spectrum; that change results in net federal outlays that are an estimated $12 billion higher over the period. Roughly half of the rise stems from a recent ruling by the Federal Communications Commission that allowed companies to withdraw their offers to pay for certain disputed licenses. Most of the remaining amount derives from recent trends in the price and quantity of spectrum that is likely to be auctioned in the future.

Adjustments that CBO has made to its projections of net interest reflect new data on the stock of outstanding federal debt and revised assumptions about the future composition of debt held by the public. (CBO now assumes that more longer-term debt will be issued than it had estimated in August.) Those changes boost projected net interest outlays over the 10-year period by $31 billion. In addition, debt-service costs attributable to technical changes boost net interest outlays by another $95 billion from 2003 through 2012.
 

The Outlook for Federal Debt

Federal debt consists of two main components: debt held by the public and debt held by government accounts. Debt held by the public--the most meaningful measure of debt in terms of its relationship to the economy--is issued by the federal government to raise cash. Debt held by government accounts is purely an intragovernmental IOU and involves no cash transactions. It is used as an accounting device to track cash flows relating to specific federal programs (for example, Social Security).

Debt held by the public and debt held by government accounts follow different paths in CBO's projections. The holdings of government accounts have risen steadily for several decades and are expected to continue doing so through the projection period. Debt held by the public, in contrast, fluctuates according to changes in the government's borrowing needs. As a percentage of GDP, publicly held debt had reached 50 percent as recently as 1993. Since 1994, it had been falling, but it rose to about 34 percent of GDP in 2002 (see Table 1-4). If current policies remained the same--that is, discretionary appropriations of $751 billion for 2003 grew with inflation and the tax cuts enacted in EGTRRA expired as scheduled--debt held by the public would fall below 15 percent of GDP by 2013. Indeed, publicly held debt is projected to decline even before EGTRRA is due to expire--dropping to approximately 24 percent of GDP in 2010--because under CBO's projections, the amount of debt would remain roughly stable while the economy grew steadily.

Debt Held by the Public
When revenues are insufficient to cover spending, the Department of the Treasury raises money by selling securities in the capital markets to investors. Debt held by the public represents the accumulation of those sales. For example, between 1969 and 1997, the Treasury sold debt to finance deficits, and debt held by the public climbed each year, peaking at $3.8 trillion in 1997. That trend reversed in 1998 with the onset of budget surpluses. By the end of 2001, debt held by the public had dropped to $3.3 trillion.

Under current tax and spending policies, debt held by the public, as projected by CBO, would grow over the next few years as deficits necessitated additional borrowing. The level of publicly held debt would reach a high of over $4 trillion in 2006, by CBO's estimate, before beginning to decline again. However, after 2003, debt held by the public as a percentage of GDP would begin to fall again because projected deficits in the near term are relatively small.

The Composition of Debt Held by the Public. Over 85 percent of publicly held debt consists of marketable securities, such as Treasury bills, notes, and bonds, and inflation-indexed notes and bonds. The remainder of that debt comprises nonmarketable securities (such as savings bonds and state and local government securities), which are nonnegotiable, nontransferable debt instruments that are issued to specific investors.

The Treasury sells marketable securities in regularly scheduled auctions, although the size of those auctions varies according to fluctuations in the government's cash flow. (It also sells cash management bills periodically to cover shortfalls in cash balances.) For some time, the Treasury has been shifting its borrowing toward shorter-term bills and notes. For example, in 2001, it introduced a four-week bill and eliminated the 30-year bond; as a result, the Treasury securities that are now sold to the public range in maturity from four weeks to 10 years. Those changes may alter the composition of outstanding public debt in the future. However, the trend toward shorter average maturity may be slowed if the Treasury curtails its program to buy back bonds before they reach maturity.

Why Changes in Debt Held by the Public Do Not Equal the Size of Surpluses and Deficits. In most years, the amount that the Treasury borrows or redeems approximates the total surplus or deficit. However, a number of factors broadly labeled "other means of financing" also affect the government's need to borrow money from the public. Over the 2004-2013 period, CBO projects that public debt will increase by more than the amount of deficits and decrease by less than the amount of surpluses as other means of financing increase the Treasury's borrowing needs.

In most years, the largest component of those other means of financing is the capitalization of financing accounts used for federal credit programs. Direct student loans, rural housing programs, loans by the Small Business Administration, and other credit programs require the government to disburse money in anticipation of repayment at a later date. Those initial outlays are not counted in the budget, which reflects only the estimated subsidy costs of such programs. For the 10 years of CBO's current baseline, the amount of the loans being disbursed will typically exceed the repayments and interest. Thus, the government's annual borrowing needs will be $9 billion to $16 billion greater than the annual budget surplus or deficit would indicate.

In 2002, other means of financing led to a net rise of $63 billion in the government's borrowing--an abnormally large amount. About one-quarter of that total reflected capitalization of financing accounts for credit programs. The remaining $47 billion reflected higher-than-average increases in a host of financing activities, including cash held by the Treasury, cash balances held in commercial banks as compensation for financial services, and premiums paid in the Treasury's bond buyback program.

In CBO's projection of other means of financing for 2003, borrowing rises by $27 billion, or about $10 billion to $15 billion more than in the other years of the projection period. Two factors account for most of that net difference. Purchases of private securities and Treasury debt by the National Railroad Retirement Investment Trust are expected to total about $18 billion; such purchases are counted as a means of financing in the budget. That amount will be partially offset by a decline in the Treasury's cash balance. (CBO assumed that the Treasury would decrease its cash balance by nearly $11 billion over the course of the year to reach its desired year-end target of about $50 billion.) The rest of the difference between the amount estimated for 2003 and the amounts projected for future years is largely attributable to lower projections of the cash flows into financing accounts for credit programs.

Debt Held by Government Accounts
In addition to the securities it sells to the public, the Treasury has issued almost $2.7 trillion in securities to various federal government accounts. All of the major trust funds and many other government funds invest in special, nonmarketable Treasury securities known as the government account series. In practical terms, those securities represent credits to the various government accounts and are redeemed when funds are needed to pay benefits and other expenses. In the meantime, the government pays interest to itself on that debt (that is, it credits interest earnings to the funds holding those securities).

Debt issued to government accounts is handled within the Treasury and does not flow through the credit markets. Because those transactions and the interest accrued on them are intragovernmental, they have no direct effect on the economy and no net effect on the budget. The largest balances of such debt are in the Social Security trust funds (more than $1.3 trillion at the end of 2002) and the retirement funds for federal civilian employees ($574 billion). If current policies remained unchanged, the balance of the Social Security trust funds would rise to $4.1 trillion by 2013, CBO estimates, and the balance of all government accounts would climb to $6.7 trillion.

Gross Federal Debt and Debt Subject to Limit
Gross federal debt and its companion measure, debt subject to limit, comprise debt issued to government accounts as well as debt held by the public. The future path of gross federal debt will be determined by the interaction of those two components. In CBO's projections, gross debt increases every year through 2012 as the growth of debt held by government accounts outpaces the future redemption of debt held by the public. In 2013, the last year of the projection period, slightly more debt could be redeemed (by using the projected surplus) than would be issued to government accounts. However, in developing that estimate, CBO assumed that all provisions of EGTRRA would expire at the end of 2010.

The Treasury's authority to issue debt is restricted by a statutory limit set by the Congress. (The debt subject to limit is nearly identical to gross federal debt, except that it excludes securities issued by agencies other than the Treasury, such as the Tennessee Valley Authority.) The current debt ceiling, which was enacted in June 2002, is $6.4 trillion (see Figure 1-2). By CBO's estimates, debt would exceed that limit sometime this year--possibly as early as the end of February--if current laws remained in place.
 

Trust Funds and the Budget

The federal government has more than 200 trust funds, although fewer than a dozen account for the bulk of trust fund dollars. Among the largest are the two Social Security trust funds (the Old-Age and Survivors Insurance Trust Fund and the Disability Insurance Trust Fund) and those dedicated to Civil Service Retirement, Hospital Insurance (Part A of Medicare), and Military Retirement (see Table 1-5). Trust funds have no particular economic significance; they do not hold separate cash balances and function primarily as accounting mechanisms to track receipts and spending for programs that have specific taxes or other revenues earmarked for their use.

When a trust fund receives payroll taxes or other income that is not currently needed to pay benefits, the excess is loaned to the Treasury. As a result, the government borrows less from the public, collects less in taxes, or spends more on other programs or activities than it would in the absence of those excess funds. The process is reversed when revenues for a trust fund program fall short of its expenses. In that case, the government raises the necessary cash by borrowing more, collecting more in taxes, or spending less on other programs or activities than it otherwise would.

Including the cash receipts and expenditures of trust funds in the budget totals with other federal programs is necessary to assess how federal activities affect the economy and capital markets. CBO, the Office of Management and Budget, and other fiscal analysts therefore focus on the total surplus or deficit.

In CBO's current baseline, trust funds as a whole are projected to run a surplus of $193 billion in 2003. That balance is somewhat misleading, however, because trust funds receive much of their income in the form of transfers from other parts of the budget. Such intragovernmental transfers reallocate costs from one part of the budget to another; they do not change the total surplus or the government's borrowing needs. Consequently, they have no effect on the economy or on the government's future ability to sustain spending at the levels indicated by current policies. For 2003, those intragovernmental transfers are estimated to total $352 billion. The largest of them involve interest credited to trust funds on their government securities ($156 billion in CBO's projections); transfers of federal funds to Medicare for Hospital Insurance, or Part A ($9 billion), and Supplementary Medical Insurance, or Part B ($83 billion); and contributions by government agencies to retirement funds for their current and former employees ($41 billion). When intragovernmental transfers are excluded and only income from sources outside the government is counted, the trust funds as a whole are projected to run deficits every year in the projection period; those shortfalls grow from $158 billion in 2003 to $273 billion in 2013.

Although the budgetary impact of the baby-boom generation's aging will not be completely realized during the 2003-2013 period, CBO's current projections provide initial indications of the coming budgetary pressures. Charting the differences between projected receipts and outlays for the Social Security and Medicare Hospital Insurance trust funds (excluding intragovernmental interest payments) illustrates that point (see Figure 1-3). Under current policies, receipts would exceed expenditures throughout the period, but after reaching nearly $130 billion between 2008 and 2011, the excess of revenues over outlays would fall to about $110 billion in 2013. At that point, outlays would be increasing by almost 7 percent per year, but annual growth of noninterest receipts would be only slightly higher than 5 percent. Thus, in CBO's projections, the capacity of the Social Security and Medicare Hospital Insurance trust funds to offset some of the net deficit in the rest of the budget--as they currently do--will begin to dwindle during the coming decade. Shortly thereafter, those programs are projected to begin adding to deficits or reducing surpluses.
 

Figure 1-3.

Surpluses or Deficits (Excluding Interest) of the Social Security and Medicare Hospital Insurance Trust Funds

(In billions of dollars)

Source: Congressional Budget Office.

Note: Hospital Insurance surpluses are calculated with adjustments for shifts in the timing of payments to Medicare+Choice plans in 2005, 2006, 2011, and 2012.

 

The Expiration of Budget Enforcement Procedures

The rules that formed the basic framework for budgetary decisionmaking for more than a decade--the annual limits on discretionary appropriations and the pay-as-you-go requirement for new mandatory spending or revenue laws--expired on September 30, 2002. That framework was established by the Budget Enforcement Act of 1990 (and later extensions) to enforce a series of multiyear budget agreements aimed at reducing and eliminating budget deficits. In general, the procedures were meant to ensure that the net budgetary effects of new laws would not increase projected deficits (or lower projected surpluses).

Although the effectiveness of the Budget Enforcement Act was mixed, lawmakers are facing the issue of whether that framework should be revived or something similar to it instituted. CBO's adjusted baseline shows the return of deficits as short-lived. However, the uncertainty of those estimates and the near and long-term budgetary pressures that confront lawmakers may necessitate some type of statutory framework of constraints. (For details on the expiration of budget enforcement procedures, see Appendix A.)