INTRODUCTION: ECONOMIC ANALYSIS OF TAX EXPENDITURES

Transcription

1 National Tax Journal, June 2011, 64 (2, Part 2), Introduction INTRODUCTION: ECONOMIC ANALYSIS OF TAX EXPENDITURES James M. Poterba Many economists and policy analysts argue that broadening the tax base is one of the most effi cient ways to raise income tax revenues. Current tax deductions, credits, and exclusions, which are collectively known as tax expenditures, reduce tax revenues and create complicated incentives that affect the ways taxpayers earn and spend their incomes. Because these tax provisions distort behavior relative to a neutral tax code, it is possible that eliminating some or all of them could simultaneously raise revenue and reduce tax-induced distortions of economic activity. Sweeping changes in tax expenditures have been recommended in several recent high profi le reform proposals. The papers in this special issue of the National Tax Journal examine the revenue effects of, the distribution of benefi ts from, and the effi ciency costs of current tax expenditures. These papers will be a valuable input for the analysis of potential reforms, and will help to identify unresolved issues that deserve further research attention. Keywords: tax expenditures, federal income tax JEL Codes: H22, H24 Long-term deficit reduction will be one of the major federal legislative challenges of the next decade. It is likely to be accomplished both by reducing outlays and raising revenues. Many economists and policy analysts argue that broadening the tax base is one of the most efficient ways to raise income tax revenues. Current tax deductions, credits, and exclusions, which are collectively known as tax expenditures, reduce tax revenues and create complicated incentives that affect the ways taxpayers earn and spend their incomes. Because these tax provisions distort behavior relative to a neutral tax code, it is possible that eliminating some or all of them could simultaneously raise revenue and reduce tax-induced distortions of economic activity. James M. Poterba: Department of Economics, MIT and National Bureau of Economic Research, Cambridge MA, USA (poterba@mit.edu)

2 452 National Tax Journal Two Presidential commissions have recently called for sharp reductions in tax expenditures. The President s Advisory Panel on Federal Tax Reform (2005) recommended sweeping changes, including limiting the deductibility of mortgage interest and including a share of employer-provided health insurance in the tax base. The panel also proposed eliminating the federal tax deduction for taxes paid to state and local governments. The National Commission on Fiscal Responsibility and Reform (2010) recommended eliminating all itemized deductions. It recommended adopting a 12 percent tax credit for mortgage interest payments, including in taxable income employer-provided health insurance premiums exceeding the 75th percentile of the premium distribution, consolidating and reducing the deduction for contributions to retirement saving accounts, and eliminating the tax exclusion for interest on newly-issued state and local government bonds. Both groups focused on finding ways to lower marginal tax rates without reducing income tax revenues. Base broadening is an essential component of any reform that seeks to achieve this goal. The Office of Management and Budget (2011) presents estimates of tax expenditures in the Analytical Perspectives report that accompanies the President s budget. The Joint Committee on Taxation (2010) also presents tax expenditure estimates. Table 1 lists the largest individual income tax expenditures based on the Office of Management and Budget (2011) analysis. A tax expenditure estimate measures the amount of federal income tax revenue that is lost as a result of a particular tax provision, given current taxpayer behavior. It is not necessarily an estimate of the revenue that might be collected if the tax provision were modified, since such a change might affect taxpayer behavior. Because of the way tax expenditures are calculated, they cannot be aggregated. Changing one tax provision could affect taxpayers marginal tax rates and itemization status, and consequently the magnitude of other tax expenditure estimates. The limitations of tax expenditure estimates notwithstanding, Table 1 illustrates that the current collection of tax expenditures has a very substantial effect in narrowing the tax base, and in reducing the revenue collected by the federal income tax. The U.S. Government Accountability Office (2005) notes the highly concentrated nature of tax expenditures, as three-quarters of the revenue loss is accounted for by only fourteen tax expenditures. While calls to repeal or restrict tax expenditures are often featured in proposals for tax reform, and were important in the last major U.S. reform in 1986, they invariably encounter opposition from those who receive tax relief from the provisions that might be eliminated as part of a base-broadening initiative. Many tax expenditure provisions provide benefits to a large fraction of taxpaying households, and they are supported by politically savvy interest groups. Steuerle (2008) discusses the rise of tax expenditures in the years following the Tax Reform Act of 1986, and points out that when social policy is carried out by allowing a tax deduction for a particular activity, the tax reduction shows up as a smaller tax-to-gdp ratio, whereas if the same activity had been supported with a government spending program, it would have appeared as an increase in the share of government spending in GDP. This created a strong political bias toward tax expenditures and away from traditional expenditure programs in the period of fiscal restraint that spanned the decade following 1986.

3 Special Issue on Economic Analysis of Tax Expenditures 453 Table 1 Individual Income Tax Expenditures, ($Billion) Exclusion of employer-provided health insurance 1,071.2 Mortgage interest deduction for owner-occupied homes Exclusion of pension contributions and earnings, employer plans and 401(k) plans Deduction for state and local income and property taxes Step-up in basis at death for assets with accrued capital gains Exclusion of imputed rental income on owner-occupied homes Deduction for charitable contributions Earned income tax credit (Revenue plus expenditure effect; revenue effect alone is 45.1) Reduced tax rate on capital gains Exclusion of capital gains on home sales Exclusion of contributions to and earnings in individual retirement accounts and keogh plans Exclusion of interest on state and local government public-purpose bonds Source: Office of Management and Budget (2011), Table Some entries have been constructed by aggregating tax expenditure estimates in several categories, such as state and local income and property taxes. In light of the potential importance of tax expenditures in future tax reform discussions, the National Bureau of Economic Research undertook a research project, graciously supported by the Smith Richardson Foundation, to enhance our understanding of the economics of tax expenditures. The research addressed three issues. First, how much revenue is lost because of each tax expenditure? Revenue estimates must recognize potential behavioral responses that might be associated with changes in income tax provisions. Second, what are the efficiency costs of particular tax expenditures? Estimates of the efficiency costs associated with current tax expenditures can help guide policy-makers as they focus on particular tax expenditure provisions. Finally, who benefits from current tax expenditure policies, and who would be adversely affected by an expansion of the income tax base that eliminated or curtailed current tax expenditures? The full incidence analysis of tax expenditure reforms is complicated, since the burdens may not fall on those who currently pay fewer taxes as a result of the tax expenditure. The 10 studies that are included in this research compendium address all of these issues in various contexts. The first two papers address broad issues of tax expenditure

4 454 National Tax Journal analysis, while the remaining eight focus on specific tax expenditure policies. In Reconsidering Tax Expenditure Estimation, Rosanne Altshuler and Robert Dietz describe the strengths and limitations of standard tax expenditure measures. They draw attention to conceptual issues, such as the ambiguity of defining the normal tax structure against which tax expenditures are calculated, and describe how a number of current measurement conventions affect reported tax expenditures. In Income versus Consumption Tax Baselines for Tax Expenditures, Robert Carroll, David Joulfaian, and James Mackie explore how standard tax expenditure measures would be affected if a consumption tax, rather than an income tax, were the baseline tax system. They explain that some of the largest current tax expenditures, such as those for contributions to, and income accruing in, pension plans, would not be tax expenditures with a consumption tax baseline, although they are tax expenditures relative to an income tax baseline. Because a consumption tax does not tax capital income, while a comprehensive income tax taxes such income at the same rate as wage and salary income, the definitions of deviations from the normal tax base with regard to capital income are quite different in the two cases. The remaining studies focus on specific tax expenditures that are not affected by this choice of baseline tax system. The studies of specific tax provisions have been organized roughly in order of the estimated size of the associated tax expenditure. Jonathan Gruber s paper, The Tax Exclusion for Employer-Provided Health Insurance, describes arguments that have been advanced for, and against, the current exclusion of employer-provided health insurance premiums from the income tax base. The paper then presents the results from a detailed micro-simulation model that analyzes how the current tax exclusion affects household insurance demand. The paper also uses this model to evaluate how a number of potential reform options, such as limiting the amount of the exclusion or eliminating it, would affect health insurance coverage and the distribution of income tax burdens. The paper illustrates the complexity of the analysis of tax expenditures. In the health insurance area, many different behavioral parameters interact in determining the revenue and market outcome effects of tax changes. Todd Sinai and I examine the tax treatment of owner-occupied housing. A number of tax expenditures, including the tax deduction for home mortgage interest payments, the tax deduction for property tax payments, and the tax exemption of a substantial share of capital gains on owner-occupied homes, affect taxpayers decisions with regard to housing consumption. In Revenue Costs and Incentive Effects of the Mortgage Interest Deduction for Owner-Occupied Housing, we focus exclusively on the effects of restricting or eliminating the current deduction for home mortgage interest. The paper describes the distributional effects of such reforms on taxpayers in different age and income categories, and points out that large differences in housing debt-to-equity ratios generate important disparities in the effects of such reforms. The paper also describes how changing the mortgage interest deduction would affect the user cost of owneroccupied housing, and the reduction in housing demand that would be associated with such a tax change.

5 Special Issue on Economic Analysis of Tax Expenditures 455 The next two papers analyze federal income tax provisions that affect state and local governments. Gilbert Metcalf s paper, Assessing the Federal Deduction for State and Local Tax Payments, studies how the deductions for state and local income and property taxes affect total spending by state and local governments. Metcalf uses the NBER TAXSIM program to compute state-by-state estimates of the marginal federal tax subsidy to state and local taxes, and he then exploits this cross-state variation to estimate how the after-tax cost of tax-financed state and local government services affects the size of sub-federal governments. This paper also examines the distributional effects of changes in the structure of marginal income tax rates, for example the set of changes that would follow from repeal of the 2001 and 2003 Bush administration tax cuts, and how the Alternative Minimum Tax affects the net subsidy to state and local governments. The next study, by Arturo Ramírez Verdugo and me, is Portfolio Substitution and the Revenue Cost of Exempting State and Local Government Interest Payments from Federal Income Tax. The current exclusion of interest on state and local government bonds from the federal income tax base allows state and local governments to borrow at a lower interest rate than they would face if their bonds generated taxable interest. The ownership of tax-exempt bonds is concentrated among high-income taxpayers, who pay an implicit tax in the form of a reduced yield by holding these bonds. The paper explores the potential revenue gain associated with changing the tax treatment of interest on these bonds, and demonstrates that assumptions about the way current tax-exempt bond holders would respond to the tax change have a large influence on the resulting revenue estimates. It is unlikely that the current holders of tax-exempt debt would continue to hold state and local government bonds if the interest on these bonds became taxable; they would be more likely to shift their portfolios to other relatively lightly taxed assets. The next two papers focus on tax policy toward charitable giving. Jon Bakija and Bradley Heim s paper, How Does Charitable Giving Respond to Incentives and Income? presents new estimates of a key behavioral elasticity for evaluating the current tax deduction for charitable gifts. A substantial prior literature suggests that charitable giving is sensitive to the after-tax cost of making gifts, but estimates of this behavioral response vary across studies. This analysis exploits differences across states in the net federal and state tax subsidy to charitable gifts, as well as variation in tax subsidies associated with predictable changes in federal tax policy, to develop new estimates of the price elasticity of charitable giving. The results suggest that there are incomerelated differences in this elasticity, with higher-income households displaying more sensitivity to tax rates than their lower-income counterparts. The findings underscore the importance of recognizing changes in taxpayer behavior when assessing the revenue effects of changes in tax expenditures. While most policy discussions of tax subsidies toward charitable gifts focus on cash gifts, Deena Ackerman and Gerald Auten present interesting new results on in-kind gifts in Tax Expenditures for Noncash Charitable Contributions. After describing the significance of non-cash gifts and presenting summary information on the taxpayers

6 456 National Tax Journal who make them, the authors investigate the in-kind gift provisions of the American Jobs Act of This legislation limited deductions for vehicles donated to charity beginning with tax year 2005 to the amount that the charity received when it sold the vehicle, and it required the charity to provide this information to the taxpayer. The paper demonstrates a striking change in the relationship between the value of vehicles claimed on tax returns, and the market-based transaction prices for such vehicles, between 2004 and The results provide interesting evidence on compliance issues associated with tax expenditures, and demonstrate that the revenue cost of specific provisions can depend critically on their administrative implementation. In Redistribution and Tax Expenditures: The Earned Income Tax Credit, Nada Eissa and Hilary Hoynes examine one of the tax provisions that has had the largest impact in transforming the distribution of federal income tax burdens in the last three decades. The Earned Income Tax Credit (EITC) was introduced in the late 1970s and expanded substantially in 1986 and It is currently larger than most other traditional antipoverty programs. The EITC generates a complicated non-linear budget set for qualifying households, subsidizing marginal hours of work for some and levying a substantial tax penalty on work for others. This paper examines the distributional impact of the current EITC structure as well as six potential reforms, and analyzes how changes to the structure of the credit could affect labor supply incentives, hours worked, and the welfare of potential beneficiaries. By bringing together empirical evidence on how the EITC affects labor supply as well as detailed modeling of the effects of potential reforms on the structure of tax rates and tax liabilities, the analysis offers valuable insights on not just distributional but also efficiency consequences of potential reforms. The final paper, by Susan Guthrie and James Hines, examines a corporate income tax expenditure. U.S. Defense Contracts During the Tax Expenditure Battles of the 1980s investigates the impact of tax rules that allowed defense contractors to account for their profits, for tax purposes, using the completed-contract method of accounting. This convention defers the date at which earnings are booked for tax purposes, and consequently reduces the present discounted value of the tax burden on a given contract. This study finds that when the Tax Reform Act of 1986 and other tax bills limited the applicability of the completed contract method, defense contractors shortened the length of their contracts. The findings raise interesting issues about the net budgetary effect of repealing this tax expenditure, since contractors may have passed some of their higher taxes back to the government agencies that were procuring their products. The papers in this special issue of the National Tax Journal illustrate the role that the economic analysis of tax policies can play in contributing to the policy design process. The wide array of exclusions, deductions and credits that are part of the federal tax code create a complex web of behavioral incentives and distributional effects. If and when federal tax reform focuses on base broadening, the economic analysis of the revenue effects of, the distribution of benefits from, and the efficiency costs of current tax expenditures will be an essential input for reform analysis. These papers will contribute to that debate, and will help to identify unresolved issues that deserve further research attention.

7 Special Issue on Economic Analysis of Tax Expenditures 457 ACKNOWLEDGEMENTS I am extremely grateful to Therese J. McGuire, the former editor of the National Tax Journal, who joined me in editing the 10 papers that appear in this special issue. Her input has substantially improved the quality of this project on many dimensions. Therese and I are both grateful to Maja Butovich, who has assisted us from the start to the finish of the review process. REFERENCES Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years U.S. Government Printing Office: Washington, DC. National Commission on Fiscal Responsibility and Reform, The Moment of Truth. The White House, Washington, DC. Office of Management and Budget, Analytical Perspectives on the Budget of the United States Government: Fiscal Year U.S. Government Printing Office, Washington, DC. President s Advisory Panel on Federal Tax Reform, Simple, Fair, and Pro-Growth: Proposals to Fix America s Tax System. President s Advisory Panel on Federal Tax Reform, Washington, DC. Steuerle, C. Eugene, Contemporary U.S. Tax Policy, Second Edition. Urban Institute Press, Washington, DC. U.S. Government Accountability Office, Tax Expenditures Represent a Substantial Federal Commitment and Need to be Re-Examined. U.S. Government Accountability Office, Washington, DC.

8