Although the federal government has run budget deficits every year for nearly three decades, recent government estimates indicate that revenues will exceed spending by 2001. Indeed, both President Bill Clinton and congressional leaders have stated their intention to balance the budget as early as 1999. Even if lawmakers do nothing more than control future spending to the limited degree called for in last year's budget deal-admittedly, a bold assumption-the projected budget surplus will climb to more than $100 billion by 2006. This emerging surplus has created a three-way battle in Washington, D.C. Some policymakers would like to spend the surplus money on new government programs. Others recommend using it to reduce the national debt. A third group, meanwhile, prefers returning the money to U.S. taxpayers.

If policymakers wish to increase economic growth and improve living standards, they should dedicate the surplus revenues to the option that will generate the highest returns. In simpler terms, they should ask themselves how the surplus money could get the "most bang for the buck." Economic research continues to show that tax cuts, particularly reducing marginal tax rates on work, savings, and investment, would generate large returns. Debt reduction also would have positive returns, but the benefit would be modest because the government's inflation-adjusted borrowing costs are relatively low. Increasing spending, by contrast, would be the least desirable way to dispose of surplus revenues. Indeed, most government programs have negative returns, meaning that the economy's overall performance falls as government spending increases.

The analysis by the country's lawmakers should not take place in a vacuum, however. In the real world of politics, interest groups actively lobby to increase spending on their favored programs. And because politicians can curry favor with constituents and supporters by boosting spending, it will be almost impossible to achieve the debt reduction option. As such, the only pro-growth way to deal with excess tax revenues is to cut taxes. More specifically, policymakers could bring about significant increases in the country's wealth if they used the surplus to facilitate either tax reform or Social Security reform.

Why the Size of Government Is the Real Issue
Two decades of budget battles in Washington, D.C., have created a bipartisan myth that balancing the budget is the most important goal of fiscal policy. This is a deeply flawed assumption. Budget deficits are neither good nor bad. They simply measure the extent to which government is financed through borrowing instead of taxes. To be sure, there are some good reasons to avoid large and persistent deficits, such as a moral concern about imposing costs on future generations and a political concern about whether taxpayers recognize the true cost of government. Nonetheless, a myopic obsession with balancing the budget distracts policymakers from more important issues-such as the size of government and the burden of the tax system.

In reality, the battle over the surplus is a battle over the size of government. Advocates of bigger government want the tax burden to remain high so that excess tax revenues can be used to create new programs and expand existing ones. The Clinton Administration, for example, has proposed expanding the Medicare program and creating new federal childcare programs. In fact, the Administration's budget proposals include $100 billion in additional taxes over five years in order to finance addiional increases in the size of government. Some policymakers, by contrast, want the tax burden to remain high in order to keep government on its current growth path while balancing the budget faster and/or paying off some of the national debt.

Why Cutting Taxes Is the Better Option. Advocates of tax reduction believe that surplus tax revenues should be returned to those who earned the money in the first place-the taxpayers. This view also holds that a tax cut should be equal, at the very least, to any projected surplus, thus precluding politicians from spending the money. In this scenario, tax cuts would keep government from growing any larger. Many tax cut proponents today hope to go beyond such a point, however, and urge tax reductions that exceed the estimated surplus. The benefit of this approach is that lawmakers would be under pressure to reduce federal spending in order to limit or preclude increases in government borrowing.

In addition to constraining the growth of government, tax cuts are desirable because lower tax rates increase incentives to work, save, and invest. The degree to which the economy benefits, however, will depend on how the tax cut is structured. In order to maximize the increase in family income and improvement in standards of living, tax reductions should be designed to move the tax code toward a single rate consumption-based tax, such as the flat tax. Alternatively, tax cuts could help facilitate the transition to a private Social Security system that would boost retirement income, increase national savings, and reduce the unfunded liabilities of the current system. Furthermore, cutting taxes has a more beneficial impact on the economy, dollar for dollar, than reducing the debt. Although scholars have failed to find any significant relationship between government borrowing and growth, the academic literature is rife with studies that illustrate the ways in which high tax rates reduce incentives to engage in productive economic behavior.¹

Finally, the tax burden should be reduced because Americans are overtaxed. Despite the tiny tax cut approved last year, federal taxes are expected to consume 19.9 percent of economic output in 1998, a peacetime record.² To put this issue in perspective, taxes totaled 19.7 percent of gross domestic product (GDP) in 1981, the year Ronald Reagan took office in part because of a nationwide tax revolt.

Not only are taxes at record highs today, the trend is in the wrong direction. Since Bill Clinton took office in 1993, the tax burden as a percent of GDP has climbed by 2.1 percentage points. This may not sound like a large amount, but 2.1 percent of an $8.461 trillion economy is $177.7 billion. Just reducing taxes to their level when Clinton took office would mean that the average family of four would receive more than $2,500 in annual tax relief.

HOW TO CUT TAXES The burden of government in the United States is smaller than in many other countries. U.S. labor markets are much more flexible, and the country has comparatively small amounts of economic regulation. Inflation is at very low levels, and Americans enjoy the benefits of expanded international trade. These factors help to explain why the United States is prosperous, with reasonable growth and low unemployment. Nevertheless, several reforms could accelerate the economy's performance. Two of these reforms involve taxes, and in both cases the reforms almost certainly would require a reduction in the country's tax burden.

First, the income tax system suffers from serious moral and economic shortcomings. With respect to the moral question, the current income tax code fails a simple test of justice and fairness because it does not treat everyone equally. In terms of the economic shortcomings, the present system undermines the economy's performance by levying punitive tax rates, imposing double taxation on savings and investment, and burdening taxpayers with more than $150 billion of compliance costs. The best solution to the tax code's myriad problems is the flat tax. In order to minimize political opposition to a flat tax, however, the tax rate under a flat tax should be set at a sufficiently low level so that a substantial majority of taxpayers would receive a tax cut. The upcoming budget surplus makes such a tax cut more feasible.

Second, the Social Security system is a financial disaster. Not only is it actuarially bankrupt, with trillions of dollars of unfunded liabilities, it also is a bad deal for workers, offering them meager retirement benefits in exchange for the huge amount of taxes they pay into the system.³ The only way to solve both problems is to reduce payroll taxes substantially and then require workers to place that money in private retirement accounts. Workers who chose this option would have no problem foregoing their promised Social Security benefits, because their private retirement investments would generate a nest egg that would give them much more income in retirement than would be possible in the government's system. Surplus tax revenues could help facilitate the transition to a private system while ensuring the benefits payments to current retirees would continue

Regardless of whether fundamental reform of either Social Security or the internal revenue code is politically practical in the near future, incremental changes could move the United States closer to a tax code that treats all citizens equally or to a Social Security system that provides workers with more retirement security.

In the case of Social Security, lawmakers could begin the process of reform by allowing workers to divert a portion of the existing payroll tax into private pension accounts. This option would reduce the government's long-term debt problem because workers who choose this option would agree to forego a portion of the future benefits they currently are promised. Workers would be better off under this option because private pension accounts earn better returns, thus accruing more income for retirement. Future taxpayers would be better off as well, because even partial privatization would reduce the huge unfunded liability of the system.

Using surplus tax revenues to begin moving to a flat tax is somewhat more complicated, but only because there are many problems with the current tax code. Its high tax rates, pervasive double taxation of savings and investment, and mind-numbing complexity cry out for attention. Policymakers need to be sure, however, that their incremental changes to the tax code are consistent with their efforts to move to a fair and simple flat tax. In other words, any revisions should move the country closer to a system that taxes all income, but only taxes it one time and at one low rate.

Among the reforms that would satisfy these objectives are:

• The repeal of the marriage penalty. The current tax system penalizes marriage. A married two-earner couple will pay more in taxes than an otherwise identical couple of two income earners who choose to live together. For the 21 million couples affected, this marriage penalty averages about $1,400 annually. Because there is no marriage penalty under the flat tax, repeal of the marriage penalty would be an important step toward fundamental reform.

• The repeal of the death tax. A core principle of tax reform is that the Internal Revenue Service (IRS) should get only one bite of the apple. Once taxpayers pay tax on their earned income, the government should not be allowed to impose an additional layer of tax on parents who choose to save that after-tax money to leave a nest egg for their children. Repealing the death tax would be a major step toward establishing a tax code that treats all income and all taxpayers the same.

• An end to double taxation of savings. The current tax code does not wait until a taxpayer's death before imposing double taxation on his or her income. With some exceptions, such as individual retirement accounts (IRAs), the current system imposes a second layer of tax on income that is saved by taxing the interest earned. Because there is no second layer of tax on income that is consumed, this creates a bias against saving. The ideal way to end that bias would be to extend IRA treatment to all savings.⁴ To the extent that comprehensive IRA expansion is not feasible, lawmakers could move in this direction by eliminating the double taxation on certain kinds of savings, such as money set aside for purposes of higher education.

THE WEAK CASE FOR DEBT REDUCTION

Many lawmakers believe that surplus tax revenues should be used to pay down the national debt. They believe that this approach is morally proper because it would reduce the burden on future generations. At the very least, reductions in the national debt would lower the amount of interest future taxpayers would have to pay to service the debt. In addition, those who favor debt reduction hope that budget surpluses would increase national savings and bring down interest rates, thereby generating positive impact on the economy.

All these arguments have merit. Reducing the national debt would generate benefits for taxpayers and the economy. The real question, however, is whether the benefits are greater than or less than the benefits that would accrue from using surplus revenues to cut taxes. Properly answering this question requires a comparison of the cost of maintaining the current tax code and Social Security system versus the cost of maintaining the national debt at present levels. Both the tax code and the Social Security system impose enormous costs on the economy, and there is substantial evidence that fundamental reform would mean significant increases in family well-being. Could debt reduction yield similar benefits?

An analysis shows that debt reduction would produce only minor benefits. The cost of carrying debt is modest, which means the benefit of reducing debt would also be very modest. According to projections from the Congressional Budget Office,⁵ taxpayers currently pay 6.5 percent interest on the government debt, and that rate is expected to fall to about 6.0 percent over the next five years. With prices rising at about 2.5 percent to 3 percent each year, this would mean that the real (or inflation-adjusted) cost of paying interest on the national debt in the future is only between 3 percent and 4 percent annually. Historical numbers also verify that carrying the cost of the government debt is relatively small.⁶ Inflation-adjusted interest on the national debt over the past 30 years has averaged only 1.57 percent.

With a national debt of nearly $3.8 trillion, expenses add up even when interest rates are low. Interest payments are expected to cost about $244 billion this year. Paying down the national debt, however, will not change that annual interest cost significantly. A $100 billion reduction in the national debt next year, for example, would reduce interest payments by less than $6.6 billion. Although $6.6 billion in lower interest payments is desirable (as is the return of $100 billion to private capital markets), consider the alternative: That same $100 billion of surplus tax revenues would be more than enough money to abolish the capital gains taxes, the death tax, and the alternative minimum tax completely, and have enough left over to cut income tax rates across the board.

Just as it would be in the case of debt reduction, tax cuts would return $100 billion to the productive sector of the economy, but the elimination of these three taxes (capital gains taxes, death taxes, and alternative minimum taxes) would have a huge impact on economic growth. Different scholars have produced varying estimates,⁷ but it is safe to say that increased incentives to save and invest would yield benefits that dwarf the $6.6 billion interest savings from debt reduction. The reduction in compliance costs alone from eliminating the three taxes almost certainly would exceed the benefits generated by debt reduction.⁸

Social Security reform provides another example. With $100 billion of surplus, lawmakers could reduce the Social Security payroll tax by about 3 percentage points, and require taxpayers to place that money in private retirement accounts. Although this percentage may sound modest, it initially would represent a huge step toward privatizing the system. Australia's private retirement system is based on a 9 percent savings requirement, for example, while the Chilean system mandates 10 percent savings. It is difficult, however, to measure precisely the economic benefits that would accrue from new savings and lower tax rates on labor. Nonetheless, using $100 billion of the surplus revenues as the first step toward privatization clearly would produce economic benefits that are several times in excess of the $6.6 billion in interest savings resulting from paying off some of the debt.⁹

Interest savings from reducing the national debt are modest, but what about broader macroeconomic benefits? Advocates of debt reduction argue, for example, that budget surpluses will lower interest rates, and that this will spur additional investment. These assertions probably are true, but the evidence indicates that the positive effects are very small. There is not a significant relationship between interest rates and budget deficits and/or debt.¹⁰ Simply stated, in world capital markets comprising trillions of dollars, even large shifts in U.S. government borrowing would be relatively minor, and any potential small impact on interest rates can be swamped by other factors, such as monetary policy and changes in the demand for credit. It is also not true that interest rates are a primary determinant of investment choices.¹¹ Investors primarily seek to maximize after-tax return when they put their money at risk. Although interest rates can be part of that calculation, tax policy and expected profits play much bigger roles in the investment decision.

A simple analogy may help explain the reason that debt reduction is not necessarily the wisest financial choice. Consider the case of a family that reaps an unexpected financial windfall. Assuming its members do not want to spend the money right away, they have two choices for the ways in which to use it: They can pay off a portion of their mortgage or they can save for retirement by investing the money in a mutual fund. Because Americans over the years have been told that debt is bad, the family might be tempted to pay off a portion of its mortgage. Choosing this option certainly would have a beneficial impact. One again should compare the alternatives, however. Assume that the mortgage carried a 7.5 percent interest rate. Once inflation is factored into the equation, the real cost of the mortgage is 5 percent or lower. This calculation is verified by historical data from the February 1997 Economic Report of the President, which reveals that real home mortgage interest rates over the past 30 years have averaged just 4 percent. And because mortgage interest is tax-deductible for almost 30 percent of the population, the actual cost could be even lower. By contrast, consider the impact on the family's fortunes if it took the money and invested it in stocks. Long-term returns average 10 percent, with inflation-adjusted returns averaging 7 percent. In other words, the family would be sacrificing long-term wealth by choosing to reduce the size of its mortgage.¹²

Some may respond that economic growth should not be the sole criterion when deciding the ways in which to dispose of surplus revenues. This is a legitimate point. Perhaps it is worth sacrificing potential improvements in standards of living in exchange for the moral satisfaction of reducing the amount of debt on future generations. This approach assumes, however, that politicians would be able to resist pressure or the temptation to use the surplus revenues for new spending. President Clinton already has proposed some $40 billion to $50 billion of new and expanded government programs. Many Republicans in Congress appear equally profligate in proposing to increase spending on everything from highways to science. In the final analysis, it is probable that the only way to limit the growth of government is to return the money to taxpayers.

WHY INCREASING FEDERAL SPENDING IS THE WORST OPTION

Government spending this year is expected to consume $1.67 trillion. Nonetheless, there are those who argue that more spending is the best way to use the surplus tax revenues. More specifially, they argue that certain types of government spending may have a positive rate of return. New highways, for example, will facilitate commerce. More education outlays will boost human capital, creating a more productive work force. Government-sponsored scientific research will expand knowledge, spinning off commercial benefits. Virtually every interest group promoting new spending makes claims that their programs have a positive effect on the economy.

Although it is theoretically possible that certain types of government spending can promote growth, real world evidence shows that the federal government does a very poor job of correctly identifying those theoretical possibilities.¹³ Transportation infrastructure is desirable, for example, but there is every reason to believe that better returns would come from choices made at the state and local levels. Similarly, education has a positive effect, but the federal government has been notoriously poor at judging how best to allocate those resources. Instead, states and localities should be encouraged to de-monopolize their school systems. Likewise, scientific research produces some economic benefits, but most observers agree that Silicon Valley is much more likely to make successful breakthroughs than the federal government.

In other words, although particular types of government spending may generate some benefits, the question is how much of a return taxpayers are getting for their money. If $1.00 of new highway spending produces $0.50 of economic benefit, taxpayers and the economy still suffer. Even if a new highway produced $1.05 in economic benefits, the spending still would be counterproductive because private-sector investment generates average yields of about 10 percent.¹⁴

Advocates for various programs will continue to argue that their spending proposals would yield very large returns. Examining the veracity of each individual claim is beyond the scope of this paper. At the very least, however, government should not be allowed to grow faster than current projections. Advocates for new spending should be forced to compete against other constituencies when seeking funding for their programs, within the context of the excessively generous spending levels agreed to in last year's budget agreement.

CONCLUSION

The government is collecting near-record amounts of tax revenue. This windfall, combined with rather modest levels of fiscal restraint, could generate a budget surplus. The potential existence of a surplus, however, is not nearly as important as the questions of whether government already is too big or the U.S. tax code is too destructive. Thorough analysis of these questions strongly suggests that the tax burden should be reduced, preferably by reforming the tax code and/or privatizing the Social Security system.

ARE DEFICITS AND DEBT ALWAYS BAD?

Many Americans recognize that politicians have used government borrowing to finance useless and counterproductive programs. This has created a well-justified suspicion of debt and deficits. It is important to realize, however, that government borrowing is sometimes preferable to the other options. Consider the following examples:

• World War II: Defeating Nazi Germany and Imperial Japan may have been impossible if the government did not have access to private credit markets. Yes, the government did incur a huge amount of debt, pushing the debt-to-GDP ratio to more than 100 percent, but the benefits of maintaining freedom certainly were worth the cost. Moreover, because subsequent generations also enjoyed the benefits of freedom, using debt to spread the cost of victory over several generations was a reasonable approach.

• Winning the Cold War and Saving the Economy: As the 1980s began, the economy was wracked by stagflation, and the United States was in retreat around the world. President Ronald Reagan undertook several steps to improve the country's stability. Two of these steps, restoring the military and combating inflation, caused long-run increases in the deficit. And although the tax-rate reductions eventually did pay for themselves (total revenues increased by 99.44 percent over the decade), short-term revenue losses during the early part of the decade added to a growing national debt. Just as was the case in World War II, however, the benefits of Reagan's policies clearly exceeded the modest cost of additional debt. The defense buildup was an integral part of the collapse of communism, and the dramatic economic reforms of the 1980s largely are responsible for the economy's remarkable performance over the past 15 years.

Compare these examples to what happened during the Great Depression of the 1930s. Driven in part by concerns about debt, President Herbert Hoover raised tax rates from 25 percent to 63 percent in 1930. President Franklin D. Roosevelt went further, increasing the top tax rate to 77 percent in 1936.^* Notwithstanding-or perhaps because of-these huge rate increases, tax collections were stagnant, and the economy remained moribund throughout the decade.

The analysis also has direct applications to the private sector. Most individuals and businesses incur debt as part of financial strategies to improve standards of living and profitability. Most households go into debt to buy a house, and many use debt for other large purchases such as automobiles. Consumers often take these steps because they expect the benefits of home ownership and car purchases will exceed the costs of the debt. Likewise, businesses borrow when they feel they have investments that will generate enough income to cover their costs.

To be sure, the fact that debt and deficits have positive uses, particularly in the productive sector of the economy, does not mean that politicians should have an unlimited ability to borrow (for that matter, individuals and businesses sometimes get into debt trouble as well). In short, debt and deficits are justified only if the long-run benefits of a policy exceed the long-run costs. Because elected officials oftentimes respond to political pressures, however, it is reasonable to remain constantly vigilant to ensure that lawmakers do not abuse the privilege of taking on debt.

---

* The Tax Foundation, Facts & Figures on Government Finance (Baltimore, Md.: Johns Hopkins University Press, 1988, 1989).

Daniel J. Mitchell, Ph.D., is the McKenna Senior Fellow in Political Economy at The Heritage Foundation.

Endnotes

1. Roger H. Gordon and Dale Jorgenson, "The Investment Tax Credit and Countercyclical Policy," Harvard Institute of Economic Research Discussion Paper No. 373, Cambridge, Mass., June 1974; James M. Poterba and Lawrence Summers, "Dividend Taxes, Corporate Investment, and `Q'," National Bureau of Economic Research Working Paper No. 829, December 1981; Martin Feldstein, "Inflation, Tax Rules, and the Accumulation of Residential and Non-Residential Capital," Institute for International Economic Studies Seminar Paper No. 186, University of Stockholm, Sweden, November 1981; Dale Jorgenson, "Taxation and Technical Change" in Ralph Landau and N. Bruce Hannay, eds., Taxation, Technology, and the U.S. Economy (New York: Pergamon Press, 1981); Robert E. Hall and Dale Jorgenson, "Tax Policy and Investment Behavior," American Economic Review, Vol. 58, No. 3, pp. 391-414; Charles W. Bischoff, "The Effect of Alternative Lag Distributions," in Gary Fromm, ed., Tax Incentives and Capital Spending (Washington, D.C.: The Brookings Institution, 1971); Keith Marsden, "Links Between Taxes and Economic Growth: Some Empirical Evidence," World Bank Staff Working Paper No. 605, Washington, D.C., 1983.  

2. U.S. Government Printing Office, Historical Tables, Budget of the United States Government, FY1999, February 1998.

3. William W. Beach and Gareth G. Davis, "Social Security's Rate of Return," A Report of the Heritage Center for Data Analysis, No. CDA98-01, The Heritage Foundation, January 15, 1998.

4. This can be achieved through traditional IRAs, which allow a deduction when the income is first earned, but then impose the one layer of tax on withdrawals, or back-ended (or Roth) IRAs, which impose the one layer of tax when income is first earned, but then do not impose a second layer of tax on subsequent withdrawals.

5. Congressional Budget Office, The Economic and Budget Outlook, Fiscal Years 1999-2008, Washington, D.C., January 1998.

6. U.S. Government Printing Office, Economic Report of the President, February 1997.

7. See William W. Beach, "The Case for Repealing the Estate Tax," Heritage Foundation Backgrounder No. 1091, August 21, 1996, and "Balanced Budget Talking Points #8: How a Capital Gains Tax Cut Would Boost State Revenues," Heritage Foundation F.Y.I. No. 82, December 29, 1995.

8. Arthur P. Hall, "The Compliance Costs and Regulatory Burden Imposed by the Federal Tax Laws," Tax Foundation Special Brief, January 1995.

9. Martin Feldstein, "The Missing Piece in Policy Analysis: Social Security Reform," American Economic Review, Vol. 86, No. 2 (May 1996).

10. U.S. Department of the Treasury, "Government Deficit Spending and Its Effects on Prices of Financial Assets," Washington, D.C., May 1983.

11. Aldona Robins, Gary Robins, and Paul Craig Roberts, "The Relative Impact of Taxation and Interest Rates on the Cost of Capital," in Dale Jorgenson and Ralph Landau, eds., Technology and Economic Policy (Cambridge, Mass.: Bellinger Press, 1986).

12. By contrast, if the family had large credit card debts that were being carried at a high interest rate, say 18 percent, then the family's economic well-being would be best served by using any financial windfall to reduce that debt.

13. The evidence also is abundant that the overall size of government imposes costs on the economy. See, for instance, James R. Barth and Michael D. Bradley, "The Impact of Government Spending on Economic Activity," The National Chamber Foundation, Washington, D.C., 1988; Robert J. Barro, "A Cross-Country Study of Growth, Saving, and Government," National Bureau of Economic Research Working Paper No. 2855, February 1989; Robert J. Barro, "Economic Growth in a Cross-Section of Countries," Quarterly Journal of Economics 56 (1991), pp. 407-443; Daniel Landau, "Government Expenditures and Economic Growth: A Cross-Country Study," Southern Economic Journal 49 (1983), pp. 783-792; Kevin B. Grier and Gordon Tullock, "An Empirical Analysis of Cross-National Economic Growth, 1951-1980," Journal of Monetary Economics 24 (1989), pp. 259-276; R.C. Kormendi and P.G. Mequire, "Macroeconomic Determinants of Growth: Cross-Country Evidence," Journal of Monetary Economics 16 (1985), pp. 141-163; Michael Marlow, "Private Sector Shrinkage and the Growth of Industrialized Economies," Public Choice 49 (1986), pp. 143-154; John McCallum and Andre Blais, "Government Special Interest Groups and Economic Growth," Public Choice 54 (1987), pp. 3-18.

14. For specific evidence on the negative net impact of government spending, particularly on infrastructure programs, see Lawrence H. Thompson, "Anti-Recessionary Job Creation: Lessons From the Emergency Jobs Act of 1983," Testimony of, U.S. General Accounting Office, GAO/T-HRD-92-13, February 6, 1992; James Bovard, "JTPA: Another Federal Training Fraud," Policy Analysis, No. 131, The Cato Institute, Washington, D.C., March 27, 1990.