Download in PDF format
One measure of the federal government's impact on the economy is its annual budget, which expresses the totals of revenues, outlays, and surpluses or deficits. Congressional consideration of the budget is influenced by the procedures and elements of the budget process, some of which are quite controversial. For example, the idea of "tax expenditures" - tax provisions that are presented as equivalent to governmental outlays - has evolved as part of the budget process in recent decades. However, the notion of "tax expenditures" is controversial because tax payments are viewed from the viewpoint of the government as opposed to the viewpoint of taxpayers. The "tax expenditure" concept rests on the assumption that tax rates should be applied to an expansive definition of taxpayer income so as to maximize tax revenue at any given tax rate.
Thus, tax provisions that shield components of this broad definition of income are viewed as depriving the government of its rightful revenues; these lost revenues are regarded as properly belonging to the federal government. Tax provisions that shield taxpayer income, expansively defined, from exposure to prevailing income tax rates are regarded as analogous to government expenditures, hence the term, "tax expenditure."
This violates the deeply ingrained principle that income, at least initially, belongs to those who generate it and that only through the democratic process becomes subject to taxation. It also contradicts with common sense perceptions of many taxpayers. For example, if a taxpayer were asked if the amount of his or her IRA deduction or 401(k) deferral should be properly viewed as the taxpayer's property or as the property of the government, the practical problems of the tax expenditure budget would become even more evident.
The theory underlying the concept of tax expenditures represents only one side of a very lively debate over what constitutes an appropriate tax base. In truth, our tax system is a compromise between income and consumption tax bases. Since the democratic process has not resolved many of these issues, it is inappropriate to institutionalize one point of view and mandate its use in official measures of tax policy. As a guide to tax policy the tax expenditure budget is one-sided and thus introduces a bias into the decision-making process.
The abstract principle underlying the tax expenditure concept presumes that the multiple taxation of saving is the norm, ignoring the compromise between income and consumption bases present in our tax code. Tax code provisions that eliminate or mitigate the multiple taxation of saving are listed as tax expenditures, and hence, as revenue losers. A tax policy guided by the tax expenditure budget will be biased against increasing tax neutrality between saving and consumption.
The remainder of this paper gives an in-depth review of the problems associated with the use of an expansive view of income and with the tax expenditure concept in general. Section II gives a brief overview of the mechanics of the tax expenditure budget. Section III covers conceptual problems dealing with the definition of the tax baseline. Section IV deals with the question of whether many tax expenditures represent subsidies or tax relief. Section V concludes with a summary of the arguments given in the paper.
The tax expenditure budget enumerates the tax incentives or tax subsidies that are a part of our income tax system. Stanley S. Surrey helped institute the "tax expenditure budget" in 1967 while he was Assistant Secretary for Tax Policy in the Treasury Department. In 1972, the Joint Committee on Taxation (JCT) began preparing an annual tax expenditure budget for the Committee on Ways and Means. In 1974, the Congressional Budget and Impoundment Control Act (PL 93-344) required that a list of tax expenditures be included in the annual budget in order to control spending and make tax provisions more transparent. Table 1 lists the top 25 expenditures (in terms of revenue loss) for fiscal year 2000.
Tax expenditures are defined in the Congressional Budget Act of 1974 as "revenue losses attributable to provisions of the Federal tax laws which allow a special exclusion, exemption, or deduction from gross income or which provide a special credit, a preferential rate of tax, or a deferral of tax liability."1 Surrey himself stated that the tax expenditure budget is "essentially an enumeration of the present tax incentives or tax subsidies, contained in our present income tax system."2 He has also defined tax expenditures as spending programs embedded in the Internal Revenue Code.3
The Budget Act of 1974 does not specify guidelines for determining which provisions constitute tax expenditures.4 Consequently, the Office of Management and Budget (OMB) states that judgment is needed in determining which provisions are receiving preferential treatment and which are not.5 The JCT has stated that the legislative history of the Budget Act indicates that tax expenditures are to be defined with reference to normal income tax law, although this definition has not been codified.6 The Treasury Department, when preparing the tax expenditure estimates for the annual budget, uses both the normal tax law employed by the JCT and another baseline called reference tax law. Both baselines are modeled after the Comprehensive Income Tax (CIT).7
The classic book Blueprints for Tax Reform best sets forth the CIT as well as an alternative tax system.8 Prepared by U.S. Treasury Tax Policy Staff, Blueprints for Tax Reform presents the results of a year-long study on tax reform. It presents two model tax systems, one of which is the Comprehensive Income Tax. By using two model systems, the Treasury Tax Policy staff acknowledged the fundamental split between economists that favor an expansive view of income and economists who are concerned with the neutrality between consumption and saving.
The CIT uses the accretion concept of income as its guide. This method defines income as the sum of consumption and change in net worth over a given period, usually a year. Income that is consumed is taxed and income that is saved is taxed (and the earnings on that saving is taxed as well), with subtractions from income of expenditures that are neither consumption nor additions to net worth (i.e., business expenses). The remaining income forms the tax base, and generally speaking, exclusions or deductions from this definition of taxable income constitute a tax expenditure. The normal income tax baseline and the reference tax baseline both use the accretion income base as a guide.
A drawback of an accretion income base is that it leads to the multiple taxation of saving. Saving occurs after the payment of income taxes, and the earnings on those savings are then taxed again, resulting in the same income stream being taxed twice. By taxing the returns to saving, the price of saving increases relative to consumption. As the relative cost of saving rises, people reduce the amount they save.
Economic research indicates that this bias is a partial explanation of the low saving levels in the United States. The bias towards consumption can be viewed as the intellectual source for the designation of tax provisions that treat income saved and income consumed in a neutral manner as "tax expenditures." The exclusion of contributions to employer-sponsored pension plans, different treatment of capital gains, and exclusion of contributions and earnings on IRA's can all be viewed as tax-neutral or tax-mitigating devices that help to alleviate the multiple taxation of saving.
In addition to the use of an accretion income base in calculating the tax expenditure budget, the pay-as-you-go (PAYGO) rules in the federal budget process work against these tax-neutral or tax-mitigating provisions. The PAYGO rules require that new direct spending and revenue legislation be deficit neutral. Therefore, any new entitlement spending requires a spending decrease elsewhere or revenue increase to offset the new spending. The elimination of "loopholes" is one policy option by which policymakers find money for new spending. The Administration's fiscal year 2000 budget calls for the elimination or curtailment of at least 72 "unwarranted benefits" or "loopholes."9 This becomes problematic when provisions trimmed or eliminated are tax-neutral provisions, resulting in additional tax biases against personal saving.
Examination of the provision for the net exclusion of employer pension-plan contributions and earnings provides evidence to support this concern. Exclusion of employer pension-plan contributions has been the fastest growing tax expenditure in the budget, rising from $5.6 billion in fiscal year 1975 to $84.3 billion in fiscal year 2000. For over two decades, employer-provided pension plans have been one of the largest tax expenditures in terms of revenue losses. Over that time, Congress has enacted tax law changes that have reduced revenue losses attributable to employer-provided pension plans.10
However, a tax expenditure budget prepared with a definition of taxable income based on tax neutrality between saving and consumption would look drastically different from the list of tax expenditures shown in Table 1. Nine of the twenty-five largest tax expenditures (in terms of revenue loss) are provisions aimed at mitigating or eliminating the multiple taxation of income. These nine items are enumerated in Table 2. The inclusion of these items overestimates the revenue losses attributable to tax subsidies because they are not subsidies.
By using a tax baseline patterned after the CIT, the current tax expenditure budget places unwarranted scrutiny on provisions of the tax code that treat saving and consumption equitably. The current tax expenditure budget has the effect of encouraging the elimination of these provisions - provisions that address our low savings levels in the United States - because they are a source of revenue for new spending programs. A tax expenditure budget based on tax neutrality would lessen the tendency to target saving provisions in an effort to comply with PAYGO rules.
A proper tax expenditure concept must have a clearly defined set of laws to determine whether a specific provision is or is not a tax expenditure.11 The lack of an operationally precise definition of taxable income requires tax officials to exercise subjective judgment, a reality that potentially limits the tax expenditure approach of objectivity and leaves it open to attacks on the grounds of arbitrariness.12 The classification of a specific provision as a tax expenditure under normal and reference tax law requires such judgment on the part of tax officials.
The OMB statement that judgment is needed to establish a tax baseline provides further evidence that the current system is not a "clearly defined set of laws.13 Noted tax scholar and Yale Law Professor Boris I. Bittker has objected to the tax expenditure concept on the grounds that it is entirely subjective. He argues that it is nothing more than each individual analyst's list of departures from what he or she thinks is the tax baseline.14 Critics often point to the Treasury Department's use of two different baselines as evidence of this charge.
Surrey responded in 1973 to these charges by stating that tax expenditures were not subjective, rather their designation entailed consensus. He stated that a tax expenditure existed if a provision departed from a tax structure that is "generally accepted" by professional tax analysts and economists.15 It is not clear, however, that using a baseline based on the CIT reflected what the consensus was then or is now. In a joint paper written two decades ago, two future Chairmen of the President's Council of Economic Advisers, Joseph Stiglitz and Michael Boskin, stated that there was "widespread sentiment among economists" that a consumption tax base was better than an income tax base at the time of the inception of the tax expenditure budget.16
Norman Ture, former Undersecretary of the Treasury, has persuasively argued that the concept of neutrality should be the essential criterion when choosing a tax base for the tax expenditure budget:
The neutrality criterion is essential if tax expenditures are to be identified as tax subsidies equivalent to those provided
by government outlays (and other government actions and policies). The distinguishing
attribute of a subsidy is that it reduces the cost or the price of the subsidized product below the level that
would prevail in a market unaffected by governmental policies or activities. A subsidy, therefore, alters the
relationship among costs and prices that would otherwise prevail. A neutral tax system, accordingly,
would include no provisions that provided subsidies: by the same token, it would contain no provisions that
raised the price or cost of any product, service, or activity relative to the prices or costs of others.17
Thus, if tax expenditures are to be defined as "revenue losses due to preferential provisions of the Federal tax laws" as the Budget Act of 1974 declares,18 then Ture's approach to tax neutrality as stated above leads to the conclusion that neutral tax provisions should not be classified as "preferences."
The tax expenditure budget includes several tax-mitigating provisions with those provisions that abate the multiple taxation of saving often are listed among the 25 largest revenue losers. The IRA exclusion, the Keogh plan exclusion, and the exclusion of employer pension-plan contributions can all be viewed as provisions that mitigate the multiple taxation of saving. The tax expenditure for the "graduated corporation income tax," which is estimated to cost the government over five billion dollars in fiscal year 2000, is another example of this error. A tax-neutral baseline would not identify the corporate income tax as a tax expenditure.
The lack of a clearly defined set of tax laws is illustrated by the difference in the treatment received by corporate income relative to the treatment received by individual income. The corporate income tax rate schedule is graduated, as is the individual income tax rate schedule. Under normal income tax law this qualifies the graduated corporate income tax as a tax expenditure, yet the graduated individual income tax is not. Logic would dictate that if the graduated corporate income tax qualifies as a tax expenditure, then the graduated individual income tax schedule should as well.19 Yet it does not, presumably for political reasons.
Therein lies the main issue concerning the lack of a clearly defined set of tax laws. The moment that judgment enters the analysis the potential for internal inconsistency and the appearance of arbitrariness increases. The more judgment that is exercised, the less useful the tax expenditure budget becomes for analytical purposes. A rigorous definition of tax expenditures would define a comprehensive tax base (i.e., all labor income, all capital income, all wealth transfers, all capital and labor income, etc.,) and the applicable tax rate. By that definition, there is no room for subjective concerns. Yet, the current tax expenditure budget includes several such examples that seem at odds with a baseline derived from a comprehensive income tax, most notably the exclusion of the personal exemption and standard deduction.
According to the JCT, personal exemptions and the standard deduction are not treated as exclusions from the income tax base for the following reason: "one may consider these amounts as approximating the level of income below which it would be difficult for an individual or family to obtain minimal amounts of food, clothing, and shelter."20 From this viewpoint they can be seen as being analogous to business expenses in that they approximate the "costs of doing business" for individuals. This may be a reasonable consideration, but the determination of the appropriate amounts is based on normative judgment, not an objective or empirical fact. Such considerations are the concern of those that write the tax code, not those that are providing an analytical guide to an aspect of taxation.
Furthermore, the calculation of tax expenditure estimates has measurement limitations that raise concern over the appropriateness of their use in tax policy. Calculations of tax expenditure estimates by the JCT and Treasury department do not incorporate any changes in taxpayer behavior.21 Consequently, a revenue loss estimate for a given tax expenditure is not identical to an estimate of the revenue gain if the tax provision was repealed. A 1994 General Accounting Office study came to the conclusion that, as a result of the exclusion of behavioral and economic effects from their calculation, tax expenditure estimates are at best useful as broad guidelines.22 Combined with the conceptual difficulties described above, these estimates are not a reliable guide to policymakers.
Regardless of whether an income or a consumption tax base is used, the tax expenditure concept suffers from another problem, namely that of closing so-called tax "loopholes" or tax preferences. The concept of a "loophole" is problematic as it has a pejorative connotation that unnecessarily complicates budgetary review and tax relief.23 Implicit in the notion of closing loopholes is the idea that virtually the total income of the country constitutes the tax base. A deduction or exemption can only be defined as a "loophole" with this presupposition.
The untenable nature of this premise is illustrated by the following reductio ad absurdum. Imagine if capital gains tax rates were raised to 100 percent while tax credits were given that kept individuals' effective capital gains rate constant. Though nothing of substance has changed, an enormous tax expenditure would now exist. From this viewpoint, the untenable nature of viewing tax payments from the position of the government becomes obvious.
This view fails to take into account the ways in which the tax base can affect tax rates. The raison d'ętre for many exemptions, deductions, and credits is to provide tax relief for eligible taxpayers. As one law professor remarked, "It could just be that tax preferences are surrogates for low tax rates."24
This paper has documented the problems associated with the tax expenditure budget. In light of the compromise between income and consumption tax bases inherent in our current tax code, there appears to be little evidence to support the institutionalization of an expansive view of income. If the current tax expenditure budget, using an expansive view of income to define tax expenditures, provided an accurate enumeration of tax subsidies, then a plausible argument could be made for its continued use. However, the evidence presented in this paper shows that the current tax expenditure budget does not accurately quantify tax subsidies, a necessary condition if tax expenditures are to be equated with governmental outlays.
The tax expenditure budget is an unfortunate component of the current federal budget process. It overstates the number of tax expenditures and the revenue losses attributed to those tax provisions. This misleading identification places a weighty obstacle in the path of those policymakers who desire tax neutrality between consumption and saving. The balance of evidence presented here strongly suggests that the statutory provision requiring the listing of the tax expenditure in the budget, the costs of which outweigh any potential benefits, be repealed.
Joshua Hall
Policy Analyst
Committee on the Budget, United States Senate, prepared by the Congressional Research
Service. Tax Expenditures: Compendium of Background Material on
Individual Provisions. Washington, DC: Government Printing Office,
1998.
Congressional Budget Office. "Comparing Income and Consumption Tax Bases."
Washington, DC: Government Printing Office, July 1997.
Joint Committee on Taxation. Estimates of Federal Tax Expenditures for Fiscal Years
1999-2003 (JCS-7-98), December 14, 1998.
Murray, Samuel H. "Current and Quotable." Tax Notes. (March 10, 1997): 1341-1342.
Office of Management and Budget. Analytical Perspectives of Budget of the United
States Government, Fiscal Year 1977. Washington, DC: Government
Printing Office, 1976.
-----. Analytical Perspectives of Budget of the United States Government, Fiscal
Year 2000. Washington, DC: Government Printing Office, 1999.
-----.Budget of the United States Government, Fiscal Year 2000.
Washington, DC: Government Printing Office, 1999.
Schuyler, Michael A. and Norman B. Ture. Federal Revenue Estimates: What's
Wrong; What Should Be Done. Washington, DC: Institute for Research
on the Economics of Taxation, 1996.
Shoup, Carl. "Surrey's Pathways to Tax Reform – A Review Article." Journal of
Finance 30, no. 5 (December 1975): 1329-1342.
Stiglitz, Joseph E. and Michael J. Boskin. "Some Lessons from the New Public Finance."
American Economic Review 61, no.1 (February 1977): 295-301.
Surrey, Stanley S. Pathways to Tax Reform: The Concept of Tax Expenditures.
Cambridge, MA: Harvard University Press, 1973.
Surrey, Stanley S. and Paul R. McDaniel. Tax Expenditures. Cambridge, MA:
Harvard University Press, 1985.
Thuronyi, Victor. "Tax Expenditures: A Reassessment." Duke Law Journal
1998, (December 1988): 1155-1206.
Ture, Norman B. Tax Expenditures: A Critical Appraisal. Washington, DC:
Institute for Research on the Economics of Taxation, 1990.
U.S. General Accounting Office. Tax Policy: Tax Expenditures Deserve More
Scrutiny. Washington, DC: U.S. General Accounting Office, 1994.
Wagner, Richard E. The Tax Expenditure Budget: An Exercise in Fiscal
Impressionism. Washington, DC: Tax Foundation, 1979.
2
Stanley S. Surrey, Pathways to Tax Reform: The Concept of Tax Expenditures (Cambridge, MA: Harvard University Press, 1973): 7.
3
Stanley S. Surrey and Paul R. McDaniel, Tax Expenditures (Cambridge, MA: Harvard University Press, 1985): 1.
4Norman B. Ture, Tax Expenditures: A Critical Appraisal (Washington, DC: Institute for Research on the Economics of Taxation, 1990), i.
5
Office of Management and Budget, Analytical Perspectives of Budget of the United States Government, Fiscal Year 2000 (February 1999): 119.
6Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 1999-2003 (JCS-7-98, December 14, 1998): 2. Ture, Tax Expenditures, 3.
7
Analytical Perspectives, 119.
8
David F. Bradford and the U.S. Treasury Tax Policy Staff, Blueprints for Basic Tax Reform (2nd edition, Arlington, VA: Tax Analysts, 1984). The Comprehensive Income Tax was itself developed from the "Haig-Simons" (or accretion) definition of income, perhaps the most commonly used definition of income. The Haig-Simons (H-S) definition of income, while useful for analytical purposes, has many shortcomings as a tax base. For example, the H-S definition is subject to measurement problems with many items that are required for the calculation of net income being guessed at or determined by arbitrary rules. For more on the shortcomings of an accretion income tax base, see chapter two of Blueprints.
9
Office of Management and Budget, Budget of the United States Government, Fiscal Year 2000 (February 1999): 371-373.
10
The Tax Reform Act of 1986 (TRA 86) restricted the availability of IRA's and imposed a cap on elective contributions to 401(k) plans. The Omnibus Budget Reconciliation Act of 1987 (OMBRA) reformed pension funding and imposed 150% of current liability cap on pension funding. The Omnibus Budget Reconciliation Act of 1989 (OMBRA 89) restricted ESOPs and imposed a mandatory penalty for violations of ERISA. TRA 86, OMBRA, OMBRA 89 are but three examples of the restrictions placed on private employer provided pension plans.
11
Ture, Tax Expenditures, i.
12
See supra note 8.
13See supra note 5.
14
Cited in Victor Thuronyi, "Tax Expenditures: A Reassessment," Duke Law Journal 1988, (December, 1988): 1181.
15
Surrey, Pathways, 186.
16
Joseph E. Stiglitz and Michael J. Boskin, "Some Lessons from the New Public Finance," American Economic Review 61, no.1 (February 1977): 297.
17Ture, Tax Expenditures, 5.
18Analytical Perspectives, 105.
19
The JCT makes the argument that concern about the ability of individuals to pay warrants the exclusion of graduated individual income taxes. This can be found in Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 1999-2003 (JCS-7-98), December 14, 1998. However, concern over ability to pay was the reason behind the graduated corporate income tax. See page 232 of Tax Expenditures: Compendium of Background Material on Individual Provisions for the rationale behind the graduated corporate income tax.
20
Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 1999-2003 (JCS-7-98), 3.
21
Behavioral and economic effects of any modification or elimination of tax expenditures are taken into account when making revenue estimates for proposals that are included in the Administration's annual budget.
22See U.S. General Accounting Office, Tax Policy: Tax Expenditures Deserve More Scrutiny (Washington, DC: U.S. General Accounting Office, 1994): 53.
23
In tax terminology, a loophole is defined as the following: a technicality making it possible to circumvent the law's intent without violating the letter of the law. Clearly, there exists a difference between a tax expenditure and a tax loophole, yet the two terms are often used interchangeably in common usage, unnecessarily complicating budgetary review.
24
Thuronyi, 1179.
Endnotes
1Congressional Budget and Impoundment Control Act of 1974 (PL 93-344), sec. 3(3).
Return
Home
