June 12, 2002

by Robert Greenstein and Richard Kogan


PDF of this report

Related Reports

If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.

The House of Representatives is expected to vote June 12 on an amendment to the U.S. Constitution to require a two-thirds vote by the House and Senate for any measure that would raise federal tax revenues (H.J. Res. 96). This amendment would have far-reaching effects.


I. The Constitutional Amendment and the Long-Term Fiscal Forecast

The federal deficit has returned with a vengeance. Current estimates show that the deficit could reach $125 billion to $150 billion this year. In addition, once the requested increases in defense and homeland security and the costs of such items as the extension of expiring tax provisions are taken into account, the forecast is not a pretty one for a number of years. CBO's estimate of the President's budget shows deficits outside of Social Security for each of the next ten years.

Furthermore, by the end of this decade, the baby boom generation will begin to retire in large numbers, increasing the costs of Social Security, Medicare, and Medicaid long-term care. CBO's estimates suggest that even when the budget as a whole (rather than the budget outside Social Security) is considered, any surpluses that reappear will be modest and short-lived. The prospect of paying off the debt before the baby boomers retire seems to have vanished.

Working from CBO estimates and projections made by the Social Security and Medicare actuaries, the General Accounting Office recently issued long-term budget projections. GAO projects that if last year's tax cut is extended and appropriated programs do not shrink further as a share of the economy, deficits will reappear in the unified (or total) budget by 2011 and grow to dangerous and unsustainable levels once the retirement of the baby boom generation is complete. The GAO projects that by 2029, the deficit will exceed six percent of GDP — apparently the largest peacetime deficit in U.S. history — and grow exponentially in years after that.(1)

Deficits of the magnitude that GAO projects would cause the national debt to rise rapidly and be unhealthy for the U.S. economy. The interest payments on the rapidly rising debt would threaten to overwhelm the budget. To avoid such developments, major deficit reduction measures are likely to be needed sooner or later.

Urban Institute president and former CBO director Robert Reischauer, when testifying some years ago about long-term budget problems, observed that the public would be unlikely to accept the steps that would be required either to extract all of the needed deficit reduction in the decades ahead just from government programs or to extract all of the needed deficit reduction from revenues. In the long run, Reischauer predicted, policymakers will need to reach agreement on some mix of program cuts and revenue increases to prevent deficits of a magnitude that would significantly damage the economy. This prediction is supported by history; all but one of the deficit reduction packages enacted over the last two decades have combined program reductions and revenue increases.

The proposed constitutional amendment is designed to ensure, however, that virtually no future deficit reduction comes from raising revenues and virtually all of it comes from cutting programs. That the amendment is likely to bar all or nearly all revenue increases can be seen by examining House votes for the five principal deficit reduction measures enacted between 1982 and 1993, each of which included increased revenues as part of the package.

Most States Do Not Have Supermajority Requirements

Only seven states require the approval of at least two-thirds of their legislatures for any tax increase. Eight other states either require such approval for some taxes but not others, require a three-fifths rather than a two-thirds vote, or both. The other 35 states generally require simple majority approval for revenue increases of all sorts.

A 1993 General Accounting Office study of recent state budget trends found that a majority of states surveyed had used both spending cuts and revenue increases to balance their budgets. Revenue increases accounted for about one-third of the deficit reduction these states instituted to balance their budgets during the period studied.

As a result, the proposed constitutional amendment would likely lead to one or more of several possible outcomes:

Such stark outcomes would be much less likely if a balance of budget cuts and revenue-raising measures could be considered in coming decades. But such a balance is what the constitutional amendment appears designed to prevent.

(Some supporters of the amendment claim it would lead to another outcome — stronger economic growth. These supporters sometimes cite data purporting to show that states which require supermajority votes to raise taxes have experienced faster growth than other states. An appendix at the end of this paper shows these claims are not valid. Careful examination of the data does not support the claim that supermajority tax requirements boost state economic performance.)


II. The Amendment Would Make it Difficult to Close Tax Loopholes

The requirement for a two-thirds supermajority would apply not only to measures to raise tax rates but also to measures to cut unproductive "tax expenditures" that grant subsidies to powerful special interests. A 1995 Congressional Budget Office study found that more than half of the corporate subsidies the federal government provides are delivered through the tax code. Curbing "corporate welfare" provided through the tax code could contribute to addressing future deficits, but doing so would require a two-thirds vote under the proposed amendment. This would likely rule out closing unproductive corporate tax breaks to help shrink deficits when they re-emerge.

The constitutional amendment would effectively place a substantial share of the federal budget off limits for deficit reduction in future decades. Provisions of the tax code that the Joint Committee on Taxation classifies as "tax expenditures" — spending programs that operate through the tax code by selectively reducing the tax liability of particular individuals or businesses — now appear to cost in the neighborhood of $830 billion a year.(2) This is substantially more than the federal government spends on Social Security or defense.

In testimony before the Bipartisan Commission on Entitlement and Tax Reform in 1994, Federal Reserve Board chairman Alan Greenspan referred to these tax code provisions as "tax entitlements" because they entitle those who qualify for them to government subsidies provided in the form of a tax reduction. Greenspan testified that the tax entitlements should be looked at, along with the spending entitlements, in developing measures to address the nation's long-term deficit problems.

If anything, however, the proposed constitutional amendment would lead to more tax expenditures over time, since such measures would take only a majority vote to enact but a two-thirds vote to remove. California's experience with a constitutional supermajority requirement for raising taxes bears out this concern. The California Citizens Budget Commission, a distinguished bipartisan panel of business and community leaders formed in 1993 to study the state's budget problems, reported that the state's supermajority requirement "makes it relatively easy to enact tax breaks but difficult to repeal them." The Commission recommended the state Constitution be changed to allow the legislature to narrow or eliminate tax breaks by a simple majority vote, so that it would be as easy to eliminate tax breaks as to create them.(3)

Moreover, if Congress passed a series of tax changes that were thought to be deficit-neutral, but tax lawyers and accountants found ways to convert some of the measures into tax shelters at a greater-than-anticipated cost to the Treasury, it would take a two-thirds vote to scale back the shelters so the original measure did not produce an unintended revenue loss. Even measures to prevent corporations from gaining tax advantages by moving plants, jobs, or their corporate addresses overseas would require a two-thirds vote.

Legal Scholar Warns of Perverse Effects, Including Increased Difficulty in Stanching Unintended Revenue Losses

In testimony before the Subcommittee on the Constitution of the House Judiciary Committee on March 18, 1997, Samuel C. Thompson, then Dean of the University of Miami Law School, warned of potential perverse consequences from a constitutional amendment requiring a supermajority vote to raise revenues. Thompson stated:

"[The amendment] will have the effect of making it much harder for Congress to close tax loopholes, because any such legislation could be blocked with a mere 34% vote in either house of Congress....

"This amendment would also penalize the American public for mistakes made in the tax legislative process. For example, assume that after an adoption of this Constitutional amendment, Congress adopts a flat tax. Assume that before enactment it is estimated that the flat tax will reduce revenues by $100 billion. It turns out, however, that the revenue estimates are wrong and the actual revenue loss is $200 billion, which will lead to a significant increase in the budget deficit. The Treasury immediately proposes legislation to increase the revenues by $100 billion, thereby restoring fiscal responsibility. The legislation is, however, opposed by powerful special interest groups who will prevail if they can convince just 34% of the members of either the House or the Senate to vote against the amendment....

"The core problem with this proposed constitutional amendment is that it would give special interest groups the upper hand in the tax legislative process. Once a group of taxpayers receives either a planned or unplanned tax benefit with a simple majority vote of both houses of Congress, the group will then be able to preserve that tax benefit with just a 34 percent vote in one house of Congress. Thus this amendment would create an unlevel playing field in the tax legislative process. Indeed, I believe that if enacted this amendment would become known as the 'Tax Loophole Preservation Amendment to the Constitution.'"

Thompson has served in the Treasury Department's Office of Tax Legislative Counsel and has taught tax law at UCLA, the University of Chicago, the University of Miami (serving as Dean of the Law School), Northwestern University, the University of Virginia, and Yale.


III. The Amendment Would Tilt Policy Toward the Wealthy and the Powerful at the Expense of Average Families and the Poor

Most government benefits that low- and middle-income Americans receive come through government programs, such as Social Security, Medicare, Medicaid, unemployment insurance, student loan and grant programs, the school lunch program, and the food stamp program. By contrast, most government subsidies that wealthy individuals and large corporations receive come through tax subsidies. A constitutional amendment that makes it difficult to scale back tax subsidies in future decades when major deficit reduction is needed would tilt the playing field in favor of the wealthy and powerful over Americans of average or lesser means.

The amendment also would be likely to affect the middle class and the poor adversely for another reason. If the federal government is unable to raise revenue when needs for public expenditures rise, one likely result will be to shift more of the burden of raising revenue and meeting public needs to state and local governments. Most state tax codes are regressive — that is, the taxes they impose consume a larger percentage of the income of low- and middle-income households than of more affluent households. Compared with the federal government, state and local governments extract a larger proportion of the revenues they raise from the middle class and the poor and a smaller proportion from the well-to-do. If revenue-raising burdens are shifted from federal to state and local levels, the share of the overall tax burden borne by the middle class and the poor is likely to rise.

The amendment's bias towards protecting the wealthiest individuals is particularly ill-advised, given that disparities in income and wealth already are at exceptionally wide levels. Data from CBO show that in the 1980s and the 1990s, the top one percent of the population experienced extraordinary after-tax income gains that vastly exceeded the gains in after-tax income the rest of the population secured. From 1979 to 1997 (the latest year for which these data are available), the average after-tax income of the richest one percent of households climbed a remarkable 157 percent, or $414,000, after adjusting for inflation. By comparison, among the middle fifth of households, average after-tax income rose a far more modest 10 percent, or $3,400.

The CBO data also show that the share of the after-tax national income that the top one percent of households received in the late 1990s was at the highest level on record. (These CBO data go back to 1979.) When these CBO data are combined with data from a study conducted by researchers at the National Bureau for Economic Research, it appears that the top one percent of the population is now receiving a larger share of the nation's after-tax income than at any time since at least 1941.


IV. The Amendment Could Make it More Difficult to Restore Social Security and Medicare Solvency Without Overly Deep Benefit Cuts

The constitutional amendment could lead to larger reductions in Social Security and Medicare benefits for low- and middle-income elderly and disabled beneficiaries than otherwise would be needed to put these programs in long-term actuarial balance. This would be the case for several reasons.

Expanding coverage in this manner would improve the actuarial balance in the Social Security trust fund while benefitting many workers who spend only part of their careers in the government sector. But expanding coverage would entail a revenue increase, since the newly covered employees and their state and local employers would become subject to the Social Security payroll tax. Accordingly, this provision would require a two-thirds vote and be much more difficult to pass. Failure to pass proposals such as this would likely mean that larger Social Security benefit cuts eventually would have to be instituted instead.

The constitutional amendment also would effectively rule out small adjustments in Medicare and Social Security payroll taxes as part of the effort to bring these programs into long-term actuarial balance. Modest increases in the ceiling on wages subject to the Social Security payroll tax or modest increases of a fraction of a percentage point in the Medicare payroll tax would require a two-thirds vote, making them virtually impossible to achieve.

This would be an especially serious problem for Medicare. It is likely to prove exceedingly difficult to restore long-term solvency to the program without some modest increase in payroll tax contributions, along with other significant program changes, unless the health insurance that Medicare provides is scaled back substantially. A politically diverse panel of Medicare experts that the National Academy of Social Insurance convened in 1999 concluded unanimously that the provision of new revenues for Medicare must be part of the debate on long-term Medicare solutions.(4)


V. The Amendment Could Increase Pressures for Pork-barrel Projects

The proposed supermajority requirement could result in the expansion of special-interest projects. The two-thirds requirement would empower individual legislators to demand federal funding for pet projects in return for agreeing to provide the necessary votes so legislation containing revenue-raising measures could obtain two-thirds support and pass. To be sure, such vote-swapping can occur whether revenue-raising measures require a simple majority or a supermajority. But as the California Citizens Budget Commission noted, the degree of vote-swapping tends to intensify as the level of difficulty needed to obtain the necessary votes to pass a budget increases. The level of difficulty is much greater when a two-thirds supermajority is required.

The Commission found evidence that California's requirement for a two-thirds majority to pass budgets has led to enactment of substantial "pork-barrel" legislation that individual legislators have promoted.(5) Supermajority requirements present legislators with tempting opportunities to threaten to block revenue-raising measures that a majority favors unless the majority accepts various pet projects or programs these legislators are pushing.


VI. The Amendment Would Weaken Our System of Democracy

As the foregoing discussion indicates, the amendment would weaken the principle of majority rule that has been at the heart of our system of representative democracy for more than 200 years. As Rep. James Moran observed during previous House debate on this measure, it would partially restore the system we had in the 1780s under the Articles of Confederation, a system that functioned poorly and was soon abandoned.

The Articles of Confederation required the assent of nine of the 13 states to raise revenue. At the Constitutional Convention in 1787, the Founding Fathers recognized this as an insurmountable defect. They fashioned instead a national government that can impose and enforce laws and collect revenue through simple majority rule.

The proposed constitutional amendment would end the ability of a majority of the American people, acting through their duly elected representatives, to decide whether they want to raise more revenues so the federal government can address needs the majority finds legitimate. The amendment would deny the majority this right both now and in future generations. As Rep. Nancy Johnson pointed out on the House floor during debate on a similar constitutional amendment in April 1996, future generations should be "free to establish that balance between taxing and spending that they believe is in their interest."

Aside from giving an unfair advantage to one side of a public policy issue, the amendment also could make gridlock more likely. Faced with a substantial deficit, a solid majority might favor deficit reduction through a balanced package of budget cuts and revenue increases. Under the amendment, a determined minority might block that package without being able to obtain majority support for its preferred alternative. It was precisely this sort of gridlock that bedeviled the Continental Congress under the Articles of Confederation.

At its core, the amendment is rooted in deep distrust of the ability of the majority of the American people to make decisions that the authors of the amendment believe to be ideologically correct. The amendment seeks permanently to deny the majority that right. Powerful, well-connected minorities would gain power at the expense of the majority.

Data Do Not Show Better Economic Performance in States with
Supermajority Requirements Grover Norquist's Claim

A recent article in the Washington Times by Grover Norquist claims that requiring a supermajority vote to raise taxes contributes to stronger economic growth.(6) Norquist cites data that applies to nine states that have had tax limitations mechanisms in place since 1993, saying per capita Gross State Product was 10 percent higher in supermajority states, and employment growth was 20 percent higher.

Norquist, however, is citing data in a highly selective manner to bolster his case. Looking carefully at economic data for the nine states that have had supermajority requirements since 1993 that cover most or all tax increases — Arkansas, Arizona, California, Delaware, Louisiana, Mississippi, Oklahoma, South Dakota, and Washington — one finds significant evidence of relatively weak economic growth in most supermajority states. The evidence does not support the conclusion that supermajority requirements increase economic performance.

Norquist's claim that employment growth was 20 percent higher in supermajority states is similarly flawed. Between 1993 and 2000, employment grew at a 2.5 percent annual rate in the supermajority states as compared to a 2.1 percent rate in all other states. But, again, Arizona's 4.6 percent rate of growth pulls up the average. Employment growth was below the national average in four of the nine supermajority states.

This is not to say that supermajority requirements hinder economic growth, but rather to point out that the data do not support the claim that supermajority requirements boost economic growth. Different measures of economic growth and difference analyses of the results suggest that supermajority states are just as likely to have lower-than-average economic growth as they are to have higher growth. This should serve as a strong caution that no valid conclusions about the effects of supermajority requirements can be drawn from the type of simplistic analysis Norquist has conducted.


The Heritage Foundation Study

In the past, the Heritage Foundation also has contended that states in which a supermajority vote of the legislature is required to raise taxes have experienced faster economic growth than other states. A March 1996 Heritage report looked at the seven states that had had supermajority requirements in place for a number of years — Arkansas, California, Delaware, Florida, Louisiana, Mississippi, and South Dakota — and found that five of the seven states experienced faster economic growth than the average state. The Heritage report suggested a causal link between supermajority limits, lower taxes, and faster economic growth.(7)

The Heritage study, however, was seriously flawed. For example, it compared state economic data for 1980 — a year in which the economy was just beginning to turn down after reaching the peak of an economic expansion — with data for 1992, a year at the beginning of a recovery from a deep recession. Economists and analysts frown upon comparisons such as this that use years representing different points in the business cycle. If one examines years that represent similar points in the business cycle and looks at various measures of economic growth, conclusions very different from those Heritage presented may be drawn. By some measures, the supermajority states in the Heritage study had less economic growth.

Five of the seven states with supermajority requirements experienced lower-than-average economic growth, as measured by changes in per capita personal incomes, between 1979 and 1989. (Both of these years were business-cycle peaks.)

End Notes:

1. From data presented in "Medicare: Financial Outlook Poses Challenges for Sustaining Program and Adding Drug Coverage" (GAO-02-643T) Testimony of David M. Walker, Comptroller General of the United States, before the House Ways and Means Committee, April 17, 2002

2. The Joint Committee on Taxation publishes each year a compendium of all tax expenditures and their costs. The Joint Committee does not total these expenditures, but if one adds the cost for each of these tax expenditures, the total is approximately $830 billion for fiscal year 2003. Since various tax expenditures interact, the total cost will vary somewhat from the $830 billion estimate. The Joint Committee indicates that due to interactions, the total cost could be either higher or lower.

3. Sources: A 21st Century Budget Process for California: Recommendations of the California Citizens Budget Commission (Los Angeles, Center for Governmental Studies, 1998) and Reforming California's Budget Process: Preliminary Report (1995)

4. Members of the National Academy of Social Insurance Study Panel on Medicare Financing, The Financing Needs of a Restructured Medicare Program, September, 1999.

5. Reforming California's Budget Process: Preliminary Report and Recommendations of the California Citizens Budget Commission, (Los Angeles: Center for Governmental Studies, 1995), pp. 43-47.

6. June 11, 2002, p. A17.

7. Daniel J. Mitchell, "Why a Supermajority Would Protect Taxpayers," The Heritage Foundation, March 29, 1996.