A Framework for Deficit Reduction: Principles and Cautions
The nation is on an unsustainable fiscal course, and policymakers need to make major changes in policy. As a number of bipartisan panels have recommended over the past year, policymakers should aim to stabilize the debt as a share of the economy (the Gross Domestic Product) so the debt does not rise relentlessly as a share of the economy. That would put the nation on what economists define as a sustainable budget path.
To achieve this goal, policymakers should aim to balance the primary budget — the budget other than interest payments on the debt. As these panels have explained, stabilizing the debt — and avoiding the specter of a debt explosion in future decades — is the key, not balancing the total budget (i.e., the budget including interest payments). As a rough rule of thumb, if the budget excluding interest payments is in balance, then the debt will not grow faster than the economy. That means running total budget deficits of no more than about 3 percent of GDP.  In short, balancing the total budget isn't necessary to put us on a sustainable course and reassure financial markets. Stabilizing the debt is.
Policymakers should meet this goal in a reasonable period of time, but it isn't necessary to meet it in the next few years. Indeed, it would be unwise to put austerity measures into effect now while the economy is still growing too slowly to bring unemployment down to more normal levels in the next few years. Putting substantial deficit-reduction measures into effect now would cause the loss of hundreds of thousands of jobs over the next year or two by slowing the already inadequate economic growth. Ideally, policymakers would enact legislation this year that begins to take effect once the economy is stronger — probably in fiscal year 2013, not before — and puts us on track to stabilize the debt as a share of GDP by the end of this decade. Doing so would involve very tough choices, both substantively and politically, but meeting that goal would be a huge accomplishment and greatly allay the fears of financial markets. Reducing the deficit more precipitously, however, is neither necessary nor sound as fiscal or economic policy.
In pursuing this goal, policymakers should follow a series of principles that would make deficit-reduction efforts equitable and more likely to be effective and sustainable over time. At the same time, they should avoid a series of steps that not only would make deficit cutting harder to achieve and sustain but also would hurt the economy down the road. As explained in detail in later sections of this paper, policymakers should:
- Craft a deficit-reduction plan that is balanced and inclusive, affecting all parts of the budget and with the savings split about 50-50 over time between program reductions and revenue increases. (Deficit reduction may need to rely more heavily on revenue increases in the early years and more heavily on program savings — especially in health care programs — over the longer run.) A substantial share of the new revenues should come from scaling back "tax expenditures," the more than $1 trillion a year in tax breaks that the tax code provides each year for particular taxpayers or groups of taxpayers.
- Enact annual caps on funding for discretionary programs but only in the context of an overall deficit-reduction plan that includes increases in revenues and savings in mandatory programs. The caps should be reasonable and attainable, with separate caps for security and non-security discretionary programs and with discretionary savings split about 50-50 between those two categories.
- Recognize that, while the single largest spending contribution to deficit reduction over the long run must come from slowing the growth of health care costs system-wide (in both the public and private sectors), policymakers cannot reasonably get sizable savings from federal health care programs over the next five to ten years. The new health reform law includes nearly all of the steps that we know how to take now to slow health care costs without reducing health care quality or access to care or pushing more people into the ranks of the uninsured; going further now by just slashing Medicare and Medicaid will not work well. Similarly, though steps to restore Social Security's long-term solvency can contribute to deficit reduction in future decades, policymakers should not expect to reap significant savings from Social Security over the next decade; there is broad bipartisan agreement that changes in benefits should not significantly affect anyone who is now at least 55 years old and that any changes in benefits and revenues should be phased in gradually.
- Meet the goal of reducing deficits to 3 percent of GDP over the coming decade through a combination of letting President Bush's tax cuts expire on schedule at the end of 2012 —however politically unachievable that seems at the moment — and securing reasonable savings in discretionary programs, curbing unproductive tax breaks, and reforming entitlement programs. Over time, a growing share of the public may conclude that alternatives that do not include letting the tax cuts expire would produce outcomes that are considerably more undesirable than returning to the tax rates of the Clinton era, when the economy performed well.
- Avoid misguided proposals such as those that would place a statutory cap on total annual federal spending or write a balanced budget requirement into the U.S. Constitution — either of which would diminish the government's ability to respond effectively to recessions (and, in fact, would make recessions worse) while largely or entirely shielding taxes from deficit-reduction efforts.
- Avoid making the problems of poverty and inequality, both of which are higher in the United States than in most other Western industrialized nations, still worse. Policymakers should adopt and adhere to the principle espoused in the Bowles-Simpson deficit-reduction plan to protect the disadvantaged. The major deficit-reduction packages of 1990, 1993, and 1997 all generally protected programs for low-income Americans; those packages, in fact, reduced poverty and inequality even as they reduced deficits.
A Basic Principle for Deficit Reduction
It is crucial to recognize that a great deal of spending occurs through the tax code, in the form of tax expenditures. The Budget Act of 1974 defines tax expenditures as revenue losses attributable to any provisions in federal tax law that provide special benefits to particular taxpayers or groups of taxpayers. Deductions, exemptions, exclusions, and preferential tax rates on certain forms of income such as capital gains and dividends are the principal forms of tax expenditures, which Alan Greenspan once referred to as "tax entitlements."
In 2010, the tax code included over $1 trillion a year in tax expenditures. This far exceeded the cost of Medicare and Medicaid combined ($719 billion), or Social Security ($701 billion), or non-security discretionary programs, which stood at $589 billion or a little over half the cost of tax expenditures. (See Figure 1.)
Martin Feldstein, the Harvard economist who served as Chairman of President Ronald Reagan's Council of Economic Advisers, wrote last summer that tax expenditures are the single largest source of wasteful and low-priority spending in the federal budget and should be the first place that policymakers go to restrain spending.
Tax Expenditures Can Be Like Other Spending: The Case of Child Care
A parent with low or moderate income may be able to obtain a subsidy to help her defray child care costs, with the subsidy being provided through a government spending program. A parent higher on the income scale also can receive a government subsidy that reduces her child care costs, but this parent's subsidy is delivered through the tax code, via a tax credit.
The two types of subsidies differ, however, in their availability to eligible families. The low- or moderate-income parent may fail to get any subsidy to help with her child care costs, because the spending programs that provide these subsidies are not open ended; they can serve only as many people as their capped funding allows. By contrast, the child care subsidies for higher-income households are guaranteed, because the child care tax subsidy operates as an open-ended entitlement (and there is no limit on how large a family's income can be to claim this tax credit).
It does not make much sense to make the tax-code subsidies sacrosanct and the program subsidies a target for deficit reduction merely because one type of subsidy is delivered through a "spending" program and the other is delivered through the tax code.
Policymakers should aim for deficit-reduction packages that, over time, are split about 50-50 between outlay reductions (i.e., reductions in programs) and revenue increases, with much of the new revenues coming from scaling back tax expenditures. This is the approach taken in the plan produced in November by the Bipartisan Policy Center commission co-chaired by former Republican senator Pete Domenici and former Clinton White House budget director Alice Rivlin.
Caps on Discretionary Spending
To ensure that all parts of the budget are on the table and that adequate deficit reduction is achieved, policymakers also need to avoid certain steps. For example, while multi-year caps on discretionary spending will need to be part of a deficit-reduction package, policymakers should not enact such caps on their own and separate from a larger deficit-reduction package that also includes measures to raise revenues and secure savings in mandatory programs. Enacting multi-year caps by themselves would undercut broader deficit reduction by making it harder to achieve. If policymakers who oppose raising any revenues for deficit reduction can get multi-year discretionary caps without those caps being part of a broader package that raises revenues as well, they will have much less incentive to agree to a larger, more inclusive package.
For example, if the discretionary caps included in the 1990 bipartisan deficit-reduction package had been enacted on their own in 1989, the 1990 package likely never would have been passed. President George H.W. Bush and Republican congressional leaders surely wouldn't have agreed to the tax increases in the package if doing so hadn't gotten them the discretionary caps in return, and Democratic congressional leaders wouldn't have agreed to cuts in Medicare and other entitlements without the tax increases. For the same reason, enacting multi-year caps now, on their own, would likely prove counterproductive to deficit reduction. Nor would it be likely to assure financial markets, which rightly understand that cuts in discretionary programs alone can't yield the amount of deficit reduction that will be required.
It also is important that multi-year discretionary caps instituted as part of a larger deficit-reduction package be set at reasonable, attainable levels. Because such caps come with no specifics regarding the actual cuts to be made (the specific cuts would come when subsequent annual appropriation bills are enacted), it can be tempting for policymakers to set very austere caps that require overly deep cuts in discretionary programs in the years that follow. History shows that such an approach is self-defeating. In 1990 and 1993, Congress enacted multi-year discretionary caps as part of larger deficit-reduction packages and set the caps at levels that produced substantial savings but were reasonable and achievable. Those caps were adhered to, and the savings were realized. In 1997, however, policymakers gave in to the temptation to write caps into that year's deficit-reduction legislation that would require deep discretionary cuts in order to show very large discretionary savings on paper. Cuts of that magnitude proved unsustainable, and subsequent Congresses — on a bipartisan basis — chose not to adhere to the caps. 
Another critical issue in fashioning multi-year caps on discretionary programs is to make sure that savings in both non-security and security programs are included — and to aim for a 50-50 split here as the Bowles-Simpson plan would do. In so doing, it is essential to have separate caps on security and non-security discretionary programs for all years that the caps are in place. If there is only a single cap on total discretionary funding, the risk will be high that most of the assumed defense savings will not materialize and non-security discretionary programs will have to bear overly severe cuts in order to make up the difference.
The Timing of Health Care Savings
In the long run, the single largest contribution to deficit reduction will need to come from slowing the rate of growth of health care costs throughout the U.S. health care system, in the public and private sectors alike. A slower rate of health care cost growth will produce substantial budgetary savings in areas ranging from Medicare and Medicaid to the tax exclusion for employer-based health coverage.
But we need to recognize that major savings are not likely to be achievable here in the next five or ten years. The recently enacted health reform law includes nearly all of the steps we know how to take now to reduce expenditures in these areas; that is how the Affordable Care Act is able to produce modest deficit reduction even as it extends coverage to 32 million uninsured Americans. Going further now and slashing Medicare and Medicaid — for example, by attempting to mandate that costs grow more slowly in Medicare and Medicaid than they are growing in the U.S. health care system overall — will not work. For over 30 years, Medicare, Medicaid, and private-sector health care costs all have grown at just about the same rate per beneficiary, which shouldn't be surprising since they all use the same doctors, hospitals, and medical procedures. Trying to hold Medicare and Medicaid to significantly lower rates of cost growth than private-sector health care will lead to either or both of two highly undesirable outcomes: 1) our health care system will become more of a two-tier system, in which Medicaid and Medicare beneficiaries (except for those Medicare beneficiaries who can afford to buy ample supplemental coverage) are denied some needed treatments and medical advances that other Americans get and in which health care thus increasingly is rationed on the basis of income; and 2) extensive cost shifting will occur from Medicare and Medicaid to private payers, with the result that costs for employer-based coverage (and other private coverage) go up substantially to cross-subsidize doctors and hospitals who are being underpaid by Medicare and Medicaid.
The Affordable Care Act recognizes this problem and contains an extensive array of demonstration projects, pilots, and research to test and identify cost-saving reforms in health care delivery and payment systems that could produce substantial savings throughout the health care system. These reforms, however, will take some time to identify, test, and institute on a broad scale. There is a potential for large and growing savings here in future decades, and these efforts need to be nurtured and adequately funded so they can produce the desired results. But there is little prospect of securing large savings here between now and 2020.
Measures to restore long-term Social Security solvency also can make a contribution to deficit reduction in future decades, but here, too, significant savings cannot be secured in the decade ahead. There is bipartisan agreement both that changes in Social Security benefits generally should not affect people now 55 and over and that changes in both Social Security benefits and taxes generally should be phased in gradually over a considerable period of time.
How to Address the Timing Problem
So how then can sufficient savings be achieved in the coming decade to stabilize the debt as a share of the economy, and thereby buy us time for the reforms — especially in health care delivery and payment systems — that are the most important component of longer-term deficit reduction? There is an obvious answer to this question, which stands out when one examines the analysis of the nation's fiscal problems that the Congressional Budget Office issued in January.
The CBO report shows that if we continue on the current policy path (including extension of all of the current tax cuts, relief from the Alternative Minimum Tax, and relief from the scheduled deep cuts in Medicare physician fees), deficits will run close to 6 percent of GDP even after the economy recovers, reaching 6.1 percent of GDP in 2021 — and the debt will climb by 2021 close to 100 percent of GDP.
Yet the data and projections in the CBO report also indicate that if policymakers simply let all of the tax cuts enacted in 2001 and 2003 (not just the tax cuts for people with incomes over $250,000) expire on schedule at the end of 2012, or if they paid for any of those tax cuts that they wish to extend with offsetting revenue increases or spending reductions — deficits would be cut nearly in half. (See Figure 2.) The deficit would stand at 3.6 percent of GDP by 2021. This is a course that both Reagan economic adviser Martin Feldstein and former OMB director Peter Orszag have called for, in light of the large fiscal challenges the nation faces and the realities regarding the timing of Medicare, Medicaid, and Social Security savings.
Federal taxes are now at historically low levels (see Figure 3). Nonetheless, letting all the Bush tax cuts expire enjoys scant political support at the moment. Yet this approach is likely to be the only way to stabilize the debt as a share of the economy over the coming decade without draconian cuts that would cause serious damage and that the public likely would not stand for. As that reality becomes clearer, the politics in this area could change.
It is worth noting that revenue increases have been important ingredients of almost all of the major deficit-reduction packages enacted over the past 30 years, including those enacted in 1982, 1984, 1987, 1990, and 1993. Presidents and lawmakers of both parties have recognized — or at least conceded — that a mix of program cuts and revenue increases is both desirable on policy grounds and essential on political grounds to achieve major deficit-reduction success.
Unsound Measures That Should be Avoided
Some proposals that are now floating around would be exceedingly ill-advised. Chief among them are proposals to place a statutory cap on total federal spending and proposals to write a balanced budget requirement into the U.S. Constitution. Many economists warn that such proposals would risk doing serious damage to the economy.
(To be sure, proposals for a cap on total federal spending or a balanced budget amendment often contain mechanisms allowing the cap or the balanced budget requirement to be suspended upon the vote of a supermajority of both the House and Senate. But such supermajorities could prove impossible to obtain until long after the economy had begun to weaken; it often takes close to a year after a recession has started before data become available showing that we are, in fact, in a recession. Even then, determined minorities could resist voting to lift the spending cap or balanced budget requirement, or might demand destructive measures — potentially including new, permanent tax cuts that increase deficits and ultimately necessitate even deeper budget cuts — as the price for their votes.)
Measures capping total federal spending at 20 percent or 21 percent of GDP also are designed to serve another function: such proposals would largely or entirely shield revenues — including tax expenditures — from making any significant contribution to deficit reduction by focusing solely on the expenditure side of the budget. That is why such proposals have been a long-standing goal of right-wing advocacy organizations and anti-tax activists. They are antithetical to the goal of producing a balanced, equitable deficit-reduction package.
That this is the case is shown by a report issued a year ago by an expert committee on the deficit convened by the National Academy of Sciences that outlined four possible paths to stabilize the debt. As panel co-chair and former Congressional Budget Office director Rudolph Penner explained, the panel designed paths at two "extremes" — one that achieved all its deficit reduction by cutting programs and another that got nearly all its deficit reduction by raising taxes — and two intermediate paths (which Penner and most other NAS panel members saw as more realistic) that blended program and tax changes. The extreme low-spending path — the path that got all of its deficit reduction by cutting programs while including tax changes that would, on net, reduce revenues over the long run — included very deep cuts in Social Security, Medicare, and Medicaid, and reductions of about 20 percent in all other spending, including defense, veterans' programs, education, and the like. Under this extreme path, federal spending would be 21 percent of GDP.
Indeed, federal spending under President Ronald Reagan averaged 22 percent of GDP at a time when no baby boomers were retired and health care costs were more than one-third lower as a share of the economy than they are today. As commentator and former OMB official Matt Miller has written, "As a matter of math, if you run the government at a smaller level than did Ronald Reagan while accommodating this massive increase in the number of seniors on our health and pension programs, you have to decimate the rest of the budget."
In short, measures like imposing a cap on total federal spending or trying to write fiscal policy into the Constitution would do more harm than good. Such approaches also suffer from being utterly devoid of any specific policy changes to actually achieve deficit reduction. There is no substitute for making the specific changes in discretionary and mandatory programs and the tax code that will move us to a sustainable fiscal course.
A Key Final Principle: Deficit Reduction Should Be Designed So It Doesn't Increase Poverty Or Inequality
The United States has higher degrees of poverty and inequality than most other Western industrialized nations. Deficit reduction ought not to make these problems worse. (The United States also has more modest retirement benefits and larger burdens from out-of-pocket health costs than most other Western nations.) Erskine Bowles and Alan Simpson made the need to protect the disadvantaged and to avoid increasing poverty and inequality (see Figure 4) one of the fundamental principles of their commission's work (although some of their specific policy recommendations do not comply with this principle.)
History shows this principle can be honored if there is a will to do so. The three major deficit-reduction packages of the last two decades — the 1990, 1993, and 1997 packages — all adhered to this principle. In fact, all three of these packages reduced poverty and inequality even as they shrank deficits, as a result of their inclusion of substantial increases in the Earned Income Tax Credit (in the 1990 and 1993 packages) and food stamps (in the 1993 package), and their creation of the Children's Health Insurance Program (in the 1997 package). This principle also was reflected in the Gramm-Rudman-Hollings law, the Budget Enforcement Act of 1990, and last year's Pay-As-You-Go law — all of which exempted means-tested entitlement programs from automatic across-the-board cuts when deficit targets were missed or pay-as-you-go standards were violated. (Sadly, the recent Corker-McCaskill bill to cap total federal spending stands this principle on its head: it would cut entitlement programs — including the basic low-income assistance programs and Social Security — disproportionately if the spending cap were exceeded, one of a number of unfortunate aspects of this misguided legislation.)
 The size of deficits that will stabilize the debt-to-GDP ratio depends on the starting level of debt, real economic growth, and real interest rates. Under projected circumstances, total deficits of about 3 percent of GDP starting in the middle of this decade would keep the debt from increasing relative to the size of the economy. Since interest payments are expected to total about 3 percent of GDP after the economy is back to normal, the primary budget would be roughly in balance if the total deficit is equal to 3 percent of GDP.
 For a further discussion of how tax expenditures can be reformed in ways that increase economic efficiency and improve tax progressivity while contributing to deficit reduction, see the testimony of Robert Greenstein, President, Center on Budget and Policy Priorities, before the Senate Budget Committee, March 9, 2011.
 The achievement of a balanced budget starting in 1998 made evading the caps almost inevitable, but many observers of the 1997 agreement believed that the caps were set so low that Congress and the President were unlikely to comply with them, whether or not the budget actually reached balance.
 Congressional Budget Office, The Economic and Budget Outlook: Fiscal Years 2011 to 2021, January 2011.
 Matt Miller, "A spending goal too small for aging America," Washington Post, January 28, 2010.
 James Horney, Paul Van de Water, and Robert Greenstein, "Bowles-Simpson Plan Commendably Puts Everything on the Table But Has Major Deficiencies Because It Lacks an Appropriate Balance Between Program Cuts and Revenue Increases," Center on Budget and Policy Priorities, November 16, 2010; "Statement of Robert Greenstein and James Horney on the Final Report from the Co-Chairs of the Deficit Commission," Center on Budget and Policy Priorities, December 1, 2010.