Skip Navigation

Chairman Kasich and members of the Committee, I am glad to be here this morning to discuss the current outlook for the economy and the budget. You have before you the Congressional Budget Office's (CBO's) new annual report, The Economic and Budget Outlook: Fiscal Years 1999-2008, which describes in detail the economic and budget projections released in a preliminary report last month. In my statement today, I will provide a brief overview of CBO's current outlook through 2008. I will also present a simple comparison of CBO's baseline budget projections and the current-policy projections that were included in the budget the President submitted on Monday. CBO is working on its own estimate of the effect of the policy proposals in that budget and expects to issue a preliminary report on that estimate in early March. Finally, I want to discuss updated projections of the federal budget and the debt in the decades after 2008. Those long-term projections, which were not completed in time to be included in the annual report, will be discussed in detail in our report on long-term budgetary pressures, scheduled to be released in the next few months.

CBO's new estimates point to an even brighter outlook for the budget over the next 10 years than we anticipated last September in our economic and budget update. Moreover, that improved near-term outlook has a positive effect on the long-term outlook. Like all projections, however, our new estimates are subject to considerable uncertainty. That uncertainty is a point I would particularly like to emphasize today.
 

ECONOMIC AND BUDGET PROJECTIONS

CBO's revisions to the budget outlook since last September are partly based on evidence that the economy continues to grow at stronger-than-expected rates. In addition, federal revenues have continued to grow faster than the size of the economy, suggesting that some of the factors that have boosted revenue growth over the past few years are likely to remain in play longer than previously anticipated.

CBO now estimates that under current policies, the federal budget deficit will be in the single digits for fiscal years 1998, 1999, and 2000, followed by a small surplus in 2001 and growing surpluses through 2008 (see Table 1). CBO's new projections point to a decline in total outlays as a share of gross domestic product (GDP)--from 20.1 percent in 1997 to 18.3 percent in 2008. That expected 2008 level would be substantially below the norm for the past 30 years. Over the same period, revenues are projected to decline only modestly as a share of GDP--from 19.8 percent in 1997 to 19.3 percent in 2008. But even with that decline, revenues as a percentage of GDP would be high in historical terms.
 

The economy continued to surprise observers with an impressive performance in 1997. Real GDP grew at the highest rate since 1988 (3.7 percent), unemployment fell to a 24-year low (5 percent), and inflation dropped to levels last seen in the 1960s (2.3 percent, as measured by the consumer price index). Such a combination cannot be sustained indefinitely. In fact, CBO forecasts slower, but still solid, economic growth and slightly higher inflation over the next two years, but we do not foresee a recession in the near future (see Table 2). The currency devaluations and turmoil in the financial markets in Asia will contribute to the slowing of the U.S. economy this year and help prevent it from overheating. A significant worsening of the Asian crisis, however, could slow economic growth in the United States more than CBO now expects.
 

UNCERTAINTY OF THE PROJECTIONS

With total revenues and total spending each approaching $1.7 trillion, small percentage deviations from the amounts that CBO projects can swing budgetary outcomes by tens of billions of dollars. Because the difference between revenues and spending is expected to be so little in 1998 through 2001, such small deviations could easily produce surpluses (or larger deficits) in 1998 through 2000, or a deficit (or a larger surplus) in 2001, without any change in budget policies or a dramatic change in the performance of the economy.

Recent experience has vividly demonstrated how hard it is to make accurate projections of federal revenues and spending, even for the current fiscal year. For example, in projections released last winter, when the fiscal year was already one-fourth over, both CBO and the Office of Management and Budget (OMB) overestimated the 1997 deficit by more than $100 billion. Most private forecasters were similarly off the mark. Large estimating errors are not uncommon for the federal government's complex budget, which is greatly affected by the economy and numerous other factors that are difficult to predict. The reason the 1997 deficit was so much lower than expected was partly that mandatory outlays were smaller than anticipated, but mainly that revenues turned out to be $72 billion higher than CBO had projected last January.

Surprisingly rapid growth in individual income tax receipts explains most of the unexpected strength of revenues in 1997. Those receipts rose by more than 12 percent, in part because personal income grew more rapidly than expected, but mainly because realizations of capital gains were unusually high and because a larger share of income was earned by people at the top of the income ladder, who are taxed at higher rates. Those last two factors caused individual income tax receipts to grow twice as fast as personal income.

The special factors responsible for 1997's revenue surge cannot grow at the current pace indefinitely, but they could persist for a while, contributing to uncertainty in CBO's latest projections. The level of capital gains realizations in 1997 will not be known until 1997 tax returns are filed. It is particularly uncertain because of large swings in the stock market and because the Taxpayer Relief Act of 1997 cut the tax rate on gains during the year. CBO is estimating that realizations rose by 45 percent last year, the same percentage increase as in 1996. If they actually grew by 65 percent, revenues in 1998 would be $10 billion higher than estimated; if they grew by only 25 percent, revenues would be $10 billion lower. Similarly, if bonuses and stock options continue to increase rapidly and push up income in high tax brackets more than estimated, revenues will be higher than CBO expects. Other changes in the pattern of income growth, however, could result in lower revenues than expected.

The differences between projected and actual outcomes may not be as great in the next few years as in 1997, but even much smaller deviations than those seen last year can still seem significant. Actual revenues in 1997 came in nearly 5 percent higher than anticipated last January; actual outlays were almost 2 percent lower than expected. It would be surprising if actual 1998 revenues or outlays were 5 percent higher or lower than CBO is currently projecting, but an examination of the historical record shows that a 2 percent error is not unlikely. If both revenues and outlays were 2 percent higher (or lower), the errors would be roughly offsetting and would have little effect on the budget's bottom line. However, if revenues were 2 percent higher and outlays were 2 percent lower (or vice versa), the bottom line would swing by more than $60 billion (see Figure 1). Because CBO is projecting very small deficits for the next three years, such a swing could lead to a moderate surplus in any of those years. Or a swing in the opposite direction could keep a balanced budget from being achieved, even if a legislative package was enacted that was supposed to balance the budget before 2001.
 

Figure 1. Deficit (-) or Surplus Under Alternative Assumptions (By fiscal year)

SOURCE: Congressional Budget Office.

Although fairly typical, a 2 percent change in revenues and outlays over the next three years does not represent the full range of possible outcomes. As already noted, the error in CBO's revenue projection for 1997 was 5 percent. Unexpected changes in economic performance could have an effect on the budget's bottom line that was as great or even greater. CBO has estimated that even a moderate recession, like the one experienced in the early 1990s, could cause the budget outlook to deteriorate by more than $100 billion for a year or so. Likewise, unexpectedly strong growth for a few more years could improve outcomes by $100 billion in a given year. Such cyclical disturbances would have little effect on the longer-term outlook, but if potential growth was just 0.5 percentage points higher or lower than CBO projects for the next 10 years, budget outcomes would be about $150 billion better or worse than projected in 2008.
 

COMPARISON WITH THE ADMINISTRATION'S PROJECTIONS UNDER CURRENT POLICY

CBO obviously has not had time to reestimate the budget that the President submitted to the Congress on Monday. However, a basic comparison shows little difference between CBO and the Administration in their projections of budget outcomes under current policies.

CBO currently expects a deficit of about $5 billion in 1998; the Administration estimates a deficit of almost $10 billion (see Table 3). After 1998, the Administration's current-policy estimates become slightly more optimistic than CBO's: the Administration projects surpluses in 1999 through 2003 if policies are unchanged and discretionary appropriations comply with the statutory caps; CBO expects small deficits in 1999 and 2000 and surpluses after that. In 2003, OMB projects a surplus of $89 billion, which is $35 billion higher than CBO projects. Although that $35 billion is not insignificant, it is equal to less than 2 percent of projected outlays in 2003. It is also much smaller than the differences that have existed in many previous years between CBO's projections and the projections of various Administrations.
 

The largest difference between the two current-policy projections is in the estimates of Medicare spending. The Administration expects that total outlays for Medicare over the next six years (including premiums paid to the government by Medicare beneficiaries) will be $50 billion lower than CBO projects, largely because the Administration believes that policies enacted in last year's Balanced Budget Act will produce larger savings than CBO estimates. Again, $50 billion is a large amount of money, but it is only 4 percent of CBO's projection of total Medicare spending over the 1998-2003 period and is not large in historical terms.

Projections of revenues under current policies, which often produce very large differences between CBO and the Administration, are remarkably similar; the total difference is only $3 billion through 2003. One reason the revenue projections are so similar is that the Administration's assumptions about the performance of the economy over the next six years are not very different from CBO's. The Administration projects slightly lower nominal GDP, inflation, and interest rates than CBO does, but for the years after 1998, it expects total taxable incomes--which directly affect the revenue projections--to be a little bit higher than CBO projects (see Table 4).
 

According to the Administration's own estimates, if the President's policies are enacted, both spending and revenues will increase by about $80 billion above the current-policy baseline in 1999 through 2003. As I indicated, CBO is currently working on its own estimate of the effect of the President's policies and plans to have that estimate available to the Congress in early March.
 

LONG-TERM BUDGET OUTLOOK

Although the current bright outlook for the budget over the next 10 years is expected to have a positive impact on the long-term picture, CBO nonetheless projects that deficits will reemerge and grow in the years after that if policies remain unchanged. The retirement of the large baby-boom generation will pinch the growth of revenues and boost outlays for Social Security, Medicare, and Medicaid. Moreover, because costs per enrollee in Medicaid and Medicare are expected to continue growing faster than inflation, projected spending for those programs will rise at an especially rapid rate.

In examining the long-term effects of demographic changes and growing health costs on the budget, CBO uses simple projections of components of the budget and takes into account how the budget affects national saving, growth, and interest rates. For years after 2008, CBO simply uses the growth projections for Social Security and Medicare outlays from the official reports of the trustees of those programs, adjusting the numbers for differences between CBO's economic assumptions and those of the trustees. CBO follows the trustees in assuming that the growth of health care costs per enrollee will eventually slow from current rates. CBO's assumptions about taxes and discretionary spending are even simpler: the effective tax rate is assumed to remain constant at its 2008 level, so revenues grow at the same rate as GDP in the long run; and discretionary spending is assumed to grow as fast as the economy, keeping it at a roughly constant share of GDP.

Although any long-term projection is inherently uncertain, CBO's base scenario indicates that the deficit could rise to about 5 percent of GDP in 2030 and to over 20 percent in 2050. In that year, the federal debt would reach about 200 percent of GDP--an unprecedented level for the United States. The long-term imbalance in the budget can be measured by the size of the tax increase or spending cut that would be needed to keep the ratio of debt to GDP at or below today's level through 2070. CBO estimates that a permanent tax increase or spending cut of 1.6 percent of GDP would be necessary to put the budget on that sustainable path.

Without a doubt, the improved budget outlook for the next decade has substantially improved the long-term outlook. Last March (before enactment of last year's Balanced Budget and Taxpayer Relief Acts, and before the recent good news on the economy, tax collections, and the slower growth of federal entitlement programs), CBO projected that the deficit would equal 2.2 percent of GDP in 2007 without changes in policy. Moreover, CBO estimated that federal debt would exceed 100 percent of GDP during the 2020s (see Figure 2). Today, we are projecting a surplus of 1 percent of GDP in 2008. And the estimated time when federal debt will become larger than gross domestic product has been pushed back two decades to the 2040s. One reason for the change is that the Balanced Budget Act of 1997 has lowered projected Medicare outlays in 2007 by slightly more than 10 percent. Because Medicare is one of the fast-growing programs, cutting its size significantly improves the long-run budget outlook.
 

Figure 2. Long-Term Projections of Federal Debt

SOURCE: Congressional Budget Office

a. The long-term projection that CBO made in March 1997.

b. A projection that assumes that the budget is balanced from 2001 to 2008.

c. CBO's current long-term projection.

The projected near-term surpluses also play a key role in brightening the picture. If, instead of running surpluses, the Congress kept the budget balanced over the next decade (by increasing spending or cutting taxes), the long-term budget outlook would be more pessimistic. In a "no surplus" scenario, federal debt would exceed GDP in the 2030s rather than in the 2040s, and the size of the long-term imbalance would increase from 1.6 percent of GDP to 2.3 percent.

Those scenarios represent CBO's current view of the long run, but the uncertainty about any long-term projection is considerable. CBO's projections may well be too optimistic for a variety of reasons. CBO uses the population projections developed by the Social Security Administration (SSA). However, the technical panel of the 1994-1995 Advisory Council on Social Security, as well as a number of private demographers, argues that the SSA understates the probable decline in mortality rates among the elderly, especially among people who are very old. As a result, CBO's projections of the elderly population in the next century could be too low. CBO also assumes a slowdown in the growth of health costs per enrollee between 2008 and 2020; if those costs did not slow, CBO's long-term projections would be considerably bleaker.

Other assumptions may make the long-term projections too pessimistic. CBO's current base scenario assumes that discretionary spending will grow as fast as the economy after 2008 (reflecting both real growth and inflation), rather than remain constant in real dollars. By contrast, if those outlays were held constant in real terms, the long-term budget picture would be much brighter. However, such a policy would reduce discretionary spending from 7 percent of GDP in 1997 to 3 percent (the lowest level since before World War II) in 2050, which could be difficult to maintain in the face of rising incomes and a growing population.
 

CONCLUSION

As a result of dramatic improvements in the past year, the current outlook for the budget is quite bright through 2008. Although CBO's baseline projections provide a useful benchmark for policymakers, neither they nor any other projections can be used to fine-tune fiscal policies or hit a precise budgetary target. Despite CBO's projection of small deficits for the next three years, the budget could end up in surplus in any of those years even if the Congress and the President did nothing to reduce spending or increase revenues. Similarly, deficits could persist even if legislation was enacted that achieved significant savings in those years.

The outlook for years after 2008 has also improved in the past year, but CBO still projects that the retirement of the baby-boom generation, together with expected further growth in per-enrollee costs for Medicare and Medicaid, will eventually lead to rapidly growing deficits if current policies are not changed. A major issue facing the Congress and the President is how best to begin preparing for the budgetary pressures that this demographic phenomenon will generate.