Regarding America’s Public Debt: How Do We Keep It From Rising?

September 28, 2006 • Testimony

Mr. Chairman and members of the committee, thank you for inviting me to testify today on the topic of controlling growth in the federal public debt. Federal debt continues to rise as spending growth keeps running ahead of the increased tax revenues the government is enjoying as a result of the strong economy. I will discuss some of the relationships between federal debt, spending, and taxation in light of recent budget developments.

Background: The Cost of Federal Spending

To support its large budget, the federal government will extract $2.4 trillion in taxes and about $300 billion in borrowed funds from families, businesses, and investors in fiscal 2006. That extraction transfers resources from the more productive private sector to the generally less productive government sector of the economy. Many studies have shown that, all else equal, the larger the government’s share of the economy, the slower economic growth will be.1 That is true regardless of whether higher spending is financed by increased taxes or higher deficits, which can be considered deferred taxes on future generations.

It is clear that a larger federal budget results in slower growth when you consider that a big share of spending is aimed at “social” goals, not at spurring growth. Indeed, 50 percent of the federal budget goes to transfers, which are typically justified on “fairness” grounds, not economic grounds.2 For example, the largest federal program, Social Security, has a negative impact on growth the way it is currently structured. People may support the current Social Security system for non‐​economic reasons, but economists believe that its pay‐​as‐​you‐​go structure reduces national savings and economic growth.

An additional problem is that extracting the current and future taxes needed to support federal spending is a complex and economically damaging process. As a result, substantially more than one dollar of private activities are displaced for every added dollar of spending. Those added costs are called “deadweight losses,” which are inefficiencies created by distortions to working, investment, and entrepreneurship. Those distortions reduce the nation’s standard of living

The Congressional Budget Office found that “typical estimates of the economic [deadweight] cost of a dollar of tax revenue range from 20 cents to 60 cents over and above the revenue raised.“3 Studies by Harvard’s Martin Feldstein have found that deadweight losses are even larger. He noted that “the deadweight burden caused by incremental taxation … may exceed one dollar per dollar of revenue raised, making the cost of incremental governmental spending more than two dollars for each dollar of government spending.“4

What this means is that the large increases in federal spending of recent years will create a substantial toll on the economy because current or future taxes will be higher than otherwise to fund the expansion. There is no free lunch on the spending side of the federal budget, but we can minimize the damage of raising federal funds by continuing to reform the most distortionary aspects of the income tax system.

Deficits and Tax Cuts

Policymakers opposed to recent tax cuts have argued that tax cuts that are “financed by deficits” don’t do much good for the economy. It is true that recent tax cuts have not benefited the economy as much as they would have if they had been matched by spending cuts.6 To the extent that recent tax cuts have added to federal deficits, a burden is imposed on future taxpayers (assuming that federal spending is not affected).6

However, there is a crucial point to consider with regard to the debate over recent tax cuts and budget deficits‐not all tax cuts are created equal. “Supply‐​side” tax cuts that reduce distortions in the tax code will spur economic growth and will not create as large a revenue loss as static calculations suggest. Any added debt from such tax cuts can be compared against the larger gross domestic product that will be generated. Supply‐​side tax cuts that represent long‐​term reforms of the federal fiscal system should be implemented regardless of the current budget balance. By contrast, further “social policy” tax cuts that do not simplify the tax code or make it more efficient should be avoided, or at least not considered unless they are matched by equal spending cuts.

Numerous studies have found that supply‐​side tax cuts on capital income are particularly beneficial to the economy. A 2005 Joint Committee on Taxation study presented the results of a macroeconomic simulation of hypothetical personal and corporate income tax cuts.7 They found that a corporate tax rate cut (matched by spending cuts) boosted U.S. output twice as much in the long run as an individual rate cut of the same dollar magnitude. The JCT also found that there are much larger positive growth effects when tax cuts are offset by spending cuts to prevent the deficit from increasing.

Federal tax legislation since 2001 has been a mix of supply‐​side and social policy cuts. About 55 percent of recent tax cuts have been supply‐​side tax cuts, including the reductions in individual rates, the dividend and capital gains tax cuts, small business expensing, and the liberalization of savings accounts.8 The other 45 percent of recent tax cuts have been social policy tax cuts, including the new 10 percent income tax bracket, the expansion of the child tax credit, and various education tax benefits.

The economic impact of social policy tax cuts, if combined with higher deficits, is mixed at best because those cuts generally do not reduce the deadweight losses of the tax system. By contrast, supply‐​side tax cuts boost long‐​term economic growth.9 The dividend and capital gains tax cuts of 2003, for example, have helped to reduce long‐​recognized distortions caused by the double taxation of corporate equity. The markets have responded strongly to the dividend and capital gains cuts, indicating that the prior high rates were creating substantial distortions.

Spending Increases, Not Tax Cuts, Are the Problem

Have tax cuts or spending increases caused today’s large budget deficits? Federal outlays have increased from $1.9 trillion in fiscal 2001 to $2.7 trillion by fiscal 2006, an increase of $800 billion. By contrast, the tax cuts enacted in 2001 and 2003 have reduced federal revenues by roughly $200 billion this year.10 Thus, recent spending increases are four times more important in explaining the current budget deficit than are recent tax cuts.11

Another way to think about recent tax cuts is that they have helped reverse the large tax increases of 1990 and 1993. CBO data shows that those tax increases increased federal revenues by a combined 1.1 percent of GDP over the first five years after each was enacted. The 2001 and 2003 tax cuts reduced revenues by a similar magnitude of 1.2 percent of GDP over the first five years after each was enacted.12

Regardless of whether or not one supports recent tax cuts, it is clear that there are gigantic long‐​term fiscal problems on the spending side of the budget. The Government Accountability Office has projected a long‐​range business‐​as‐​usual scenario for the budget.13 The projections assume that entitlement programs are not reformed, and that other programs and taxes stay at the same size as today relative to GDP. Under that scenario, federal spending would grow from 20 percent of GDP today to a staggering 45 percent of GDP by 2040. Such a European‐​sized government would bring with it slow growth, lower wages, a lack of opportunities, and many other pathologies.

Unfortunately, the long‐​term fiscal situation could be even worse than that. The GAO’s “static” estimates ignore the economic death spiral that would occur if taxes were raised in an attempt to fund higher spending. Higher taxes would result in greater tax avoidance, slower growth, less reported income, and thus less than expected tax revenue, perhaps prompting policymakers to jack up tax rates even higher.

Consider Social Security and Medicare Part A, which are funded by the federal payroll tax. On a static basis, the cost of these two programs as a share of taxable wages is projected to rise from 14 percent in 2005 to 25 percent in 2040.14 But as tax rates rise, the tax base will shrink. To get the money it would need to pay for rising benefits, and taking into account this dynamic effect, the government would have to hike the payroll tax rate to about 30 percent by 2040.15 That would be a crushing blow to working Americans, who would have to pay this tax in addition to all the other federal and state taxes they pay.

Note that on top of these federal costs, state and local governments are also imposing large and unfunded obligations on future generations. State and local governments have rapidly rising levels of bond debt, and they have unfunded costs for their workers’ pension and health plans that could total more than $2 trillion.16

Reform Options

These figures suggest a bleak fiscal future awaiting young Americans and taxpayers without major reforms. There are many actions that should be taken right away to reduce deficits and unfunded obligations.

  • Social Security should be cut by indexing future initial benefits to the growth in prices rather than wages.
  • Medicare deductibles and premiums should be increased. Those changes could be phased‐​in over time, but it is important to get the needed cuts signed into law to reduce the exposure of taxpayers.
  • Medicaid should be block‐​granted and the federal contribution to the program restrained or cut. This was the successful strategy behind the 1996 welfare reform.
  • Federalism should be revived and federal aid to the states cut sharply. Aid to the states does not make any economic sense. It has been a bastion of “pork” spending, and it has created massive bureaucracies at all three levels of government. With the coming entitlement crunch, the federal government simply cannot afford to be Santa Claus to the states any longer.

Of course, such cuts are politically difficult for Congress to make. That is why new budgeting structures are needed to get a handle on rising spending and deficits. Considering that federal outlays have increased 45 percent in the last five years and the government has run deficits in 33 of the last 37 years, it is obvious that current budget rules are not working very well.

That is why budget reform proposals, such Senator Gregg’s “Stop Over Spending Act of 2006” (S. 3521) are important.17 The Act contains new rules to control deficits, restrain entitlement spending, cap discretionary spending, limit “emergency” spending, and create a commission to eliminate waste in federal programs.

Some people argue that such new budget restrictions are not needed because Congress has the power to restrain spending anytime it wants. But political scientists have long recognized that the self‐​interested actions of individual policymakers often lead to overall legislative outcomes that undermine the general welfare. Indeed, frequent statements by many policymakers make it clear that their top priority is to target spending to interests in their states, not to legislate in the national interest. If left to their own devices, many members become activists for narrow causes, while broader concerns such as the size of the federal debt are ignored.

New and improved federal budget rules are needed to channel the energies of members into reforms that are in the interests of average citizens and taxpayers. Without tight budget rules, Capitol Hill descends into an “every man for himself” spending stampede‐​a budget anarchy that creates unsustainable budget expansion and soaring deficits. That is why there have been numerous, and often bipartisan, efforts to create new budget procedures, such the 1974 Budget Act, the 1985 Gramm‐​Rudman‐​Hollings Act, and the 1990 Budget Enforcement Act.

Consider also that the 50 states generally have much tighter budget rules than does the federal government.18 Virtually all the states have statutory or constitutional requirements to balance their budgets. Governors in 42 states have line‐​item veto authority. Most state constitutions include limitations on government debt. More than half the states have some form of overall tax and expenditure limitation (TEL).19 Also, the states are fiscally constrained by the need to prevent their bond ratings from falling.

Capping Total Federal Spending

Senator Gregg’s proposals are a good starting point for discussing budget reforms, but Congress should also consider a more comprehensive budget control idea. That is to impose a statutory cap on the annual growth in total federal outlays, including discretionary and entitlement spending.20 Deficits are a byproduct of the overspending problem, and such a cap would target that core problem directly. The basic principle of a budget growth cap is that the government should live within constraints, as average families do, and not consume an increasing share of the nation’s output.

Prior budget control efforts have imposed caps on discretionary spending, but not entitlement spending. Yet the rapid growth in entitlement spending may cause a major budget crisis, and thus should be included under any cap. There has been interest in capping entitlements in the past. In 1992, the bipartisan Strengthening of America Commission, headed by Sens. Sam Nunn (D-GA) and Pete Domenici (R-NM), proposed capping all non‐​Social Security entitlement spending at the growth rate of inflation plus the number of beneficiaries in programs.21 The Entitlement Control Act of 1994 (H.R. 4593) introduced by Rep. Charles Stenholm (D-TX) would have capped the growth in all entitlement programs to inflation plus one percent plus the number of beneficiaries. Both of those proposals included procedures for sequestering entitlement spending with broad cuts if the caps were breached.

A simple way to structure a cap is to limit annual spending growth to the growth in an economic indicator such as personal income. Another possible cap is the sum of population growth plus inflation. In that case, if population grew at 1 percent and inflation was 3 percent, then federal spending could grow at most by 4 percent. That is the limit used in Colorado’s successful “TABOR” budget law. Whichever indicator is used should be smoothed by averaging it over about five years.

An interesting alternative would be to simply cap total federal spending growth at a fixed percentage, such as four percent. That would make it easy for Congress to plan ahead in budgeting, and would prevent efforts to change caps by fudging estimates of economic indicators. Another interesting advantage of a fixed percentage cap is that it would provide an incentive for Congress to support a low inflation policy by the Federal Reserve Board.

With a spending cap in place, Congress would pass annual budget resolutions making sure that discretionary and entitlement spending was projected to fit under the cap for upcoming years. Reconciliation instructions could be included to reduce entitlement spending to fit under the cap for the current budget year and to reduce out‐​year spending to fit under projected future caps.

The Office of Management and Budget would provide regular updates regarding whether spending is likely to breach the annual cap, and Congress could take corrective actions as needed. If a session ended and the OMB determined that outlays were still above the cap, the president would be required to cut, or sequester, spending across the board by the amount needed. The GRH and the BEA included sequester mechanisms that covered only portions of the defense, nondefense, and entitlement budgets, but a sequester on the overall budget would be a better approach.

A shortcoming of a statutory spending cap and other budget rules is that Congress would always have the option of rewriting the law if it didn’t want to comply. But a cap on overall spending would be a very simple and high‐​profile symbol of restraint for supporters in Congress and the public to rally around and defend. An overall cap on spending growth of, say, four percent is easy to understand, and watchdog groups would keep the public informed about any cheating by policymakers. Over time, public awareness and budgetary tradition would aid in the enforcement of a cap.


Federal policymakers need a change in mindset and tougher budget rules to ward off large tax hikes and rising debt as entitlement costs soar in future years. Policymakers need to scour the budget for programs and agencies to cut.22 A cap on total federal spending should be part of the solution to get the budget under control. Clearly, current budget rules have not worked very well, and we should experiment with new rules to try and get a grip on the overspending problem.

Thank you for holding these important hearings. I look forward to working with the committee on its agenda for federal budget reform.

Chris Edwards
Director of Tax Policy Studies
Cato Institute

1See James Gwartney and Robert Lawson, “Economic Freedom of the World: 2004 Annual Report,” Fraser Institute, 2004, and see James Gwartney and Robert Lawson, “Economic Freedom of the World: 2005 Annual Report,” Fraser Institute, 2005. For a summary of academic studies, see Daniel J. Mitchell, “The Impact of Government Spending on Economic Growth,” Heritage Foundation, March 15, 2005. To state this relationship more precisely, if the government increases its share of the economy beyond a certain modest level of about 15 percent, then growth begins to suffer.

2Transfers are 50 percent of total program outlays (outlays excluding interest). See Chris Edwards “How to Spend $2.8 Trillion,” Cato Institute Tax & Budget Bulletin no. 39, August 2006.

3Congressional Budget Office, “Budget Options,” February 2001, p. 381. For a general discussion, see Chris Edwards, “Economic Benefits of Personal Income Tax Rate Reductions,” U.S. Congress, Joint Economic Committee, April 2001. See also William Niskanen, “The Economic Burden of Taxation,” presented at a conference at the Federal Reserve Bank of Dallas, Texas, October 22–23, 2003.

4Martin Feldstein, “How Big Should Government Be?” National Tax Journal, Volume 50, no. 2, June 1997, pp. 197–213.

5Tax cuts matched by spending cuts produce much stronger growth effects in the long run. See the various simulations in Joint Committee on Taxation, “Macroeconomic Analysis of Various Proposals to Provide $500 Billion in Tax Relief,” JCX-4–05, March 1, 2005.

6If higher deficits create a “starve the beast” effect resulting in lower spending, then tax cuts now will not lead to equally large tax increases later.

7Joint Committee on Taxation, “Macroeconomic Analysis of Various Proposals to Provide $500 Billion in Tax Relief,” JCX-4–05, March 1, 2005.

8Based on the dollar values of extending the cuts between 2012 and 2016. See Office of Management and Budget, Midsession Review Fiscal Year 2007, July 11, 2006, Table S-6. The estate tax is not included. 9For example, see Joint Committee on Taxation, “Macroeconomic Analysis of Various Proposals to Provide $500 Billion in Tax Relief,” JCX-4–05, March 1, 2005.

10Based on CBO’s estimate of the revenue loss from EGTRRA and JGTRRA in fiscal 2012 as a share of GDP, then applied to GDP in fiscal 2006. I have not included the alternative minimum tax.

11Note that this estimate of federal revenue losses is on a static basis. The actual loss is likely to be smaller because of the positive economic effects of the cuts.

12Chris Edwards, “Social Policy, Supply‐​Side, and Fundamental Reform: Republican Tax Policy, 1994 to 2004,” Tax Notes, November 1, 2004, p. 691.

13Government Accountability Office, “21st Century Challenges: Reexamining the Base of the Federal Government,” GAO-05–325SP, February 2005, Figure 2, p. 8.

14The 2005 Annual Report of the Board of Trustees of the Federal Old‐​Age and Survivors Insurance and the Federal Disability Insurance Trust Funds (Washington: Government Printing Office, April 5, 2005), p. 166. These are the intermediate assumptions.

15Estimate based on Martin Feldstein, “Prefunding Medicare,” National Bureau of Economic Research, Working Paper no. 6917, January 1999, p. 4.

16Chris Edwards and Jagadeesh Gokhale, “Unfunded State and Local Health Costs: $1.4 Trillion,” Cato Institute Tax & Budget Bulletin, September 2006.

17U.S. Senate, Committee on the Budget, “The Stop Over Spending Act of 2006,” Senate Report 109–283, July 14, 2006.

18For background on state budget processes, see National Association of State Budget Officers, “Budget Processes in the States,” January 2002.

19Michael New, “Limiting Government through Direct Democracy,” Cato Institute Policy Analysis no. 420, December 13, 2001.

20For background, see Chris Edwards, “Capping Federal Spending,” Cato Institute Tax & Budget Bulletin no. 32, March 2006. Also see Brian Riedl, “Restrain Runaway Spending with a Federal Taxpayers’ Bill of Rights,” Heritage Foundation, August 27, 2004.

21The commission was sponsored by the Center for Strategic and International Studies.

22For detailed discussion of federal programs that should be cut, see Chris Edwards, Downsizing the Federal Government (Washington: Cato Institute, 2005), www​.down​siz​ing​gov​ern​ment​.com.

About the Author
Chris Edwards

Director of Tax Policy Studies and Editor, Down​siz​ing​Gov​ern​ment​.org