Over the past two and a half years the official U.S. federal budget outlook has deteriorated in spectacular fashion. The federal budget shifted from a surplus of $127 billion in fiscal year 2001 to a projected deficit of about $400 billion in fiscal 2003, which ended on September 30. This decline is due mostly to short-term factors like the economic slowdown and the war on terrorism. More ominously, the Congressional Budget Office's projected baseline budget for 2002-11 changed from a surplus of $5.6 trillion in January 2001, when President George W. Bush took office, to a deficit of $2.3 trillion as of August. The worsening budget projections for later in the decade are due mostly to the 2001 and 2003 tax cuts and to spending increases.
The official figures, moreover, paint far too optimistic a picture of the nation's fiscal health. The official numbers count current surpluses in the Social Security and Medicare trust funds in the budget, even though both face substantial long-term deficits. And the projections of the future make unrealistic assumptions regarding current policy. Making more realistic assumptions about current policy and taking the Social Security and Medicare trust funds "off-budget" reveal large and persistent imbalances between spending and revenues totaling about $8 trillion over the next decade.
Worse, these deficits are projected for the decade that should be the easy time for federal finances. The first baby boomers will become eligible for Social Security in 2008 and for Medicare in 2011. The growing number of new beneficiaries, combined with lengthening life spans, technological changes that boost health care spending, and slow growth of the labor force, will place mounting pressure on federal finances in the decades to come.
Solutions to these difficulties are available, but they will be mostly unpleasant. Policymakers have not completely ignored the problems, but they have not actively sought answers either, and recent policies have made matters worse. Every year of delay makes the problems more difficult and more expensive to resolve.
What Is the Real Budget Outlook?
CBO's baseline budget is intended to be a benchmark against which legislative changes can be measured, not a prediction of likely outcomes. In making its 10-year projections, the CBO makes three assumptions about current policy that we believe make these projections unrealistic. First, although it assumes that Congress will extend some expiring spending programs, CBO assumes that almost all temporary tax provisions will "sunset" (expire) as scheduled in the law. The 2001, 2002, and 2003 tax cuts, for example, are scheduled to sunset at various points before 2011.
Substantial sunsets in the tax code are a recent and dramatic departure from historical practice by federal policymakers intent on manipulating budget rules and hiding the true likely costs of new tax cuts. These tax sunsets, which make it possible to increase the size of annual tax cuts while staying within the budget rules, put fiscal policy on an increasingly unsustainable course, because once a tax cut is in place, Congress will be sorely tempted to extend it past its official sunset. The alternative—not extending it—will be denounced by opponents as a tax increase, a step that policymakers traditionally find distasteful, especially in election years. With President Bush and many congressional leaders already pushing to make the tax cuts permanent, it would be more realistic for the CBO baseline to assume that the tax cuts will be extended.
The second unrealistic assumption involves spending on programs that require annual appropriations decisions by Congress. The baseline assumes that such so-called discretionary spending grows each year only because of inflation. Although judgments may reasonably differ about future spending choices, we believe that current services will be difficult to maintain unless spending keeps pace with population growth too. George W. Bush endorsed the same criterion as a presidential candidate.
The third unrealistic assumption involves the alternative minimum tax (AMT), which runs parallel to the regular income tax system. Although originally designed to raise taxes on wealthy households that aggressively use tax shelters, the AMT will increasingly apply to middle-income households over the next decade, raising their tax bills and plaguing them with mind-boggling, pointless tax complexities. Unlike the ordinary income tax, the AMT is not adjusted for inflation and thus covers ever more taxpayers as prices (and incomes) rise—ensuring that policymakers will come under increasing pressure to cut back the AMT. For that reason, our projection, unlike the CBO baseline, assumes the AMT is adjusted so that the share of taxpayers who face the tax in the future is 3 percent—about the same as today.
Officially, the CBO baseline projects a 10-year deficit of $1.4 trillion for 2004 through 2013, with surpluses rising over time, as figure 1 shows. Adjusting the CBO baseline for our three assumptions regarding current policy tells quite a different story. Extending all expiring tax provisions would cost the federal budget $2.4 trillion. Adjusting the AMT would add another $400 billion. Adjusting discretionary spending implies another $500 billion in outlays.
These changes leave a federal budget with a deficit of $4.6 trillion over the next decade. Numerous uncertainties about what the future holds make it impossible to take these budget figures as exact predictions, but several basic trends are clear. First, the CBO baseline suggests that the budgetary future features rising surpluses within the 10-year window, whereas our adjusted unified budget baseline implies continual deficits through 2013, as shown in figure 1. Second, the differences grow over time. By 2013 the difference between the official projected unified budget and our alternative unified deficit is more than $700 billion each year. Third, our adjustments, which do not include the costs of a Medicare prescription drug benefit or other new initiatives, may themselves understate the severity of the problem.
Finally, all these figures include cash-flow surpluses of $3.2 trillion over the next decade accruing in trust funds for Social Security, Medicare, and government pensions. But—as will come as a surprise to no one—these trust funds, now in surplus, face tremendous long-term shortfalls. In various pieces of legislation between 1983 and 1990, Congress took Social Security off-budget to help clarify the state of the rest of the federal budget. We follow that approach and extend it to Medicare and government pensions. Combining this adjustment with the ones made above results in a 2004–13 projected shortfall in the nonpension portion of the budget of roughly $7.8 trillion. Notably, this shortfall exists in every year through 2013, the end of the budget window.
That the official budget window ends in 2013 itself makes the projections misleadingly optimistic, because the budget outlook deteriorates rapidly thereafter. Indeed, few observers dispute that the long-term forecast involves increasing deficits as a share of the economy.
Some have lately claimed that a previously unrecognized "pot of gold" in future revenue from tax-deferred retirement accounts will be large enough to eliminate most or all of the long-term budget shortfalls. But the underlying calculations of the long-term budget shortfalls already include almost all the projected revenue from withdrawals from tax-deferred accounts. As a result, incorporating the new projections has trivial effects on the long-run budget outlook.
Are the Deficits a Problem?
None of this would matter if deficits didn't matter, but they do. Under conventional views of how the economy operates, an increase in budget deficits (or a reduction in surpluses) will reduce future national income. This chain of events begins because the deficit reduces national saving. Theoretically, the decline in public saving represented by a budget deficit could be offset by an increase in private saving. Numerous tests, however, conclude that households do not raise saving enough to offset fully a drop in public saving induced by tax cuts.
Lower national saving, in turn, reduces national investment. Whether the fall in investment comes in the form of lower domestic investment or lower net foreign investment by Americans, it reduces the capital stock owned by Americans and therefore diminishes the flow of future capital income received by Americans. That is, either the domestic capital stock falls or the nation is forced to mortgage some of its future capital income by borrowing from abroad. In the latter case, the foreign capital inflow may help keep domestic production at the level it would have been without the deficit, but having to repay such borrowing means that Americans will have a smaller claim on the income. Either way, Americans' future national income is lower than it would otherwise have been. In short, deficits reduce future national income.
This is not a controversial theory. It is espoused, for example, by Federal Reserve Chairman Alan Greenspan, by the chair of President Bush's Council of Economic Advisers, Gregory Mankiw, and by economic leaders in the earlier Clinton, Bush, and Reagan administrations.
A related concern is that the projected future budget deficits will raise interest rates. Along with many other economists, including those noted above, we believe that deficits do raise interest rates. Indeed, President Bush's Council of Economic Advisers recently reported that a sustained deficit of 1 percent of gross domestic product would raise rates by almost one-third of a percentage point. One implication is that the Bush tax cuts will eventually raise rates by about two-thirds of a percentage point—increasing payments on a $150,000 mortgage by about $750 a year. Another implication is that despite the reduction in marginal tax rates, the 2001 tax act raised interest rates enough to reduce investment and hurt long-term growth.
To see the real cost of long-term deficits in everyday terms, think of a family that borrows lots of money to take a vacation. It may run up its credit card balances so much that lenders charge higher interest rates on additional borrowing. But even without a higher rate, the family has a debt to repay. The borrowing effectively places a mortgage on the family's future income and thereby reduces its future living standards. The same is true for the country as a whole.
Just how much do deficits affect growth? Conventional estimates, based on models developed by CEA Chair Mankiw, indicate that the decline in the fiscal outlook since January 2001 will reduce gross domestic product in 2012 at least 1 percent and reduce national income per household in 2012 about $2,300. The costs will grow even larger thereafter. To put it differently, controlling deficits is a pro-growth policy.
Large and persistent budget deficits may also create broader problems. As economist Edwin Truman has noted, a substantial fiscal deterioration over the longer term may cause "a loss of confidence in the orientation of U.S. economic policies and?undermine the strength of the U.S. economy and confidence in U.S. economic and financial policies." Such a loss in confidence could then put upward pressure on domestic interest rates as investors demand a higher "risk premium" on U.S. assets. Long-term deterioration in a country's fiscal position can create difficult and lasting economic problems.
Not Having a Policy Is a Policy
Policymakers face three sets of related budget challenges: near-term deficits over the next few years, medium-term deficits over the next 3–10 years, and long-term deficits thereafter.
If realistic budget projections showed rising surpluses as far as the eye could see, the near-term deficits would be no problem at all. In the presence of projected shortfalls, however, current deficits are a mixed bag. The deficits eat away at the capital stock, and the resulting costs will have to be borne over time. Nevertheless, in a slack economy, deficit spending can give the economy a needed boost, and unusual events like the Iraq War should be at least partially funded by means of deficits.
The medium- and long-term deficits are much more troubling. Even though the economy is projected to have reached full employment within the next few years, the adjusted unified budget shows a deficit in each of the next 10 years. The disjuncture between revenues and spending indicates a fundamental imbalance persisting up to and beyond the time when baby boomers begin to retire, putting unprecedented pressure on the federal budget that will only increase over the years.
The source of the imbalance for this year is mainly a decline in revenues. Federal tax revenues in fiscal 2003 are at a generational trough of 16.5 percent of gross domestic product, the lowest share since 1959. Corporate and personal income taxes are at their lowest share of the economy since before World War II. In contrast, spending is about 20 percent of GDP in 2003, slightly less than its average since 1975.
If government leaders cannot reestablish fiscal responsibility in the remainder of this decade, it will prove far more difficult to do so after the baby boomers start retiring. The administration says that its approach to dealing with the deficit is to reduce spending and to cut taxes to make the economy grow and thereby bring in more revenue. The president asserted in his 2003 State of the Union address, "We will not deny, we will not ignore, we will not pass along our problems to other Congresses, to other presidents, and to other generations."
That rhetoric is empty at best. Leaving the problem to future generations and making the problem worse are exactly what the administration is doing. Not only is the administration raising spending, not cutting it, its tax cuts won't raise revenue—in fact, the tax cuts won't even generate much, if any, increased long-term growth.
Economic growth induced by something other than tax cuts makes the economy bigger and generates more revenue. But economic growth induced by tax cuts does not necessarily raise more revenue—the economy is bigger, yes, but a smaller tax rate applies to it. Thus, for example, if a 10 percent cut in taxes makes the economy grow 1 percent, revenues do not rise. Instead, they fall about 9 percent. Indeed, in 2001 the administration claimed that the purpose of tax cuts was to reduce the surplus. In 2003, in a remarkable turnaround, administration officials claimed that the same tax cuts, accelerated and made permanent, were needed to raise the surplus (reduce the deficit). Even the administration's own estimates show that its new rhetoric is flawed and that tax cuts do not raise revenue—consistent with estimates from Congress's Joint Committee on Taxation, the Congressional Budget Office, and every other major macroeconomic model of which we are aware.
Cutting taxes is not a way to reduce the long-term deficit—even if the tax cuts do generate more economic growth, they do not generate enough to offset their direct revenue-reducing effects. Furthermore, estimates from the CBO show that the administration's budget plans will likely have a zero or negative effect on growth. And estimates from the Joint Tax Committee show that the 2003 tax cut will likely cut jobs and growth in the long term because the adverse effect from larger budget deficits outweighs the positive effects of the tax cuts. The reductions in tax rates may increase saving, work, and investment, but the higher deficit induced by the tax cut reduces national saving and may drive up interest rates. The net effect on long-term growth, according to most studies, is negative.
Over the past several decades, national policymakers of both parties had been making a sustained push toward fiscal responsibility. Ronald Reagan agreed to income tax hikes in 1982 and 1984 as it became clear that his 1981 tax cuts, combined with increased defense spending and a slowing economy, were wreaking havoc on federal finances. In 1990, faced with continuing projected deficits, Democrats and Republicans put together a budget agreement that raised taxes, cut spending, and imposed new restrictions on spending increases and tax cuts. In 1993, with economic and fiscal problems persisting, Democrats in Congress raised taxes, cut spending, and extended the 1990 budget restrictions. That laudable recent history makes the administration's determination to ignore the underlying problems both cynical and disappointing. Rather than raise taxes and cut spending, the Bush administration has proposed spending hikes and continued tax cuts aimed at high-income families. What is needed is a return to a semblance of fiscal sanity.
An Alternative Approach
Congress and the administration can and should reduce "waste, fraud, and abuse" in spending programs and improve efforts to collect taxes that are currently unpaid. But even if these efforts are spectacularly successful, they will not resolve the underlying imbalance between projected revenues and expenditures.
Cutting spending and raising taxes are not happy alternatives, but in the end some combination of the two will be required. The fiscal problems ahead are unlike any other the country has faced in origin and nature. We will likely have to find a new way of dealing with them. In particular, the notion that federal spending can be held to its post-1980 average of about 21 percent of GDP seems impossible to maintain without either severely cutting major entitlement programs or virtually eliminating the rest of government. By 2030, spending on Social Security, Medicare, and Medicaid alone is expected to reach 14 percent of GDP—up from about 8 percent today. Even by 2013, under the adjusted budget projections noted above, balancing the budget would require a 41 percent cut in spending on Social Security and Medicare, a 47 percent cut in discretionary spending, or a 17 percent cut in all noninterest spending. Such severe cutbacks seem unlikely for political reasons and would be unattractive for economic and social reasons, unless one believes that virtually all government spending is wasteful.
The most imminent threat to fiscal sanity is the administration's proposal to make the 2001, 2002, and 2003 tax cuts permanent. The administration prefers to label Social Security and Medicare as "the real fiscal danger." But making all the temporary tax provisions in the tax code permanent would cost $430 billion—or 2.4 percent of GDP—in 2013 alone. Over the next 75 years, making the recent tax cuts permanent would reduce long-term revenue by more than three times the actuarial deficit in Social Security and by more than the combined deficit in Social Security and Medicare's Hospital Insurance program over the same period. By these measures, the administration's tax cuts deserve at least equal billing on the list of policies accounting for "the real fiscal danger."
In addition, legislation that boosts Medicare prescription drug benefits without simultaneously reforming the health care system will not only dramatically raise spending but also mean missing a rare chance to reach legislative agreement on reforming Medicare.
The single most useful policy change to prevent the creation of new tax sunsets or new entitlements or the removal of existing sunsets would be for Congress to reinstate permanently the pay-as-you-go budget rules from the 1990s. Those rules required that mandatory spending increases or tax cuts be financed by other changes in taxes or spending. Permanently reinstating those rules would require us to pay for any prescription drug benefit expansion or removal of the tax sunsets.
The bottom line is that avoiding an unsustainable explosion in government debt without destroying the role of the federal government in American society will require considerable increases in tax revenues as a share of the economy. Although some policymakers would like nothing better than to gut the role of the government, the more responsible debate should focus on how and when to raise revenue—and how to balance the required revenue increases against responsible reductions in spending. The sooner that debate begins in earnest, the more likely we will be to avoid the wrenching changes that will otherwise be necessary.