|Tax Policy Outlook logo 130|
A number of recent proposals have called for broadening the individual income tax base while lowering statutory income tax rates. Such proposals would eliminate or curtail various preferential income tax provisions and use some or all of the resulting revenue to lower statutory tax rates. In this Outlook, we analyze the economic effects of this approach to tax reform.
Key Points in this Outlook
President Obama, the Bowles-Simpson Commission, and others have called for broadening the individual income tax base by eliminating or curtailing tax preferences and using the resulting revenue to lower statutory tax rates.
Lowering statutory tax rates while broadening the income tax base generally does not reduce work disincentives because it leaves the relevant effective tax rates unchanged. Such reforms may promote economic efficiency by providing neutral tax treatment that allows market forces to allocate resources across different economic sectors.
Some types of base broadening, however, may actually move the tax system away from neutrality, impeding economic efficiency.
Income tax base broadening does not remove, and may even reinforce, the income tax’s central distortion: its penalty on saving.
Concerned that the US individual income tax system is distorted by a myriad of preferential tax provisions, President Obama, key members of Congress, and bipartisan groups have called for curtailing these tax preferences and thereby broadening the income tax base. Curtailing such tax preferences can reduce the tax system's interference with taxpayers' choices, allowing resources to be allocated according to their economic value rather than tax considerations and promoting economic efficiency. In conjunction with reduced statutory tax rates, the right kind of base broadening can offer significant gains along these lines. Nevertheless, base broadening must be approached with caution because some base-broadening measures make the tax system less neutral, impeding economic efficiency. Also, income tax base broadening cannot eliminate, and may even reinforce, the income tax's central flaw--its penalty on saving.
To begin, we discuss the major tax preferences now in place and briefly describe some of the recent proposals. We then look at the potential economic effects of base broadening.
Numerous tax code provisions narrow the income tax base. These provisions, intended to benefit specific types of taxpayers or activities, are known as "tax expenditures" because they are comparable in some respects to direct government spending programs. Table 1 lists the top ten tax expenditures in 2010 and their associated revenue losses.
Two broad types of tax expenditures exist: tax credits and tax deductions/exclusions. A tax credit, which may be nonrefundable or refundable, is applied after an individual's tax is computed. A nonrefundable credit (for example, a $1,500 tax credit for the purchase of energy-efficient windows) is applied up to only the amount of a taxpayer's tax liability. If the value of a nonrefundable credit is more than the amount of tax, the taxpayer's tax liability is reduced to zero but does not become negative. In contrast, an individual eligible for a refundable credit receives a check from the US Department of the Treasury for any credit in excess of tax liability. The Making Work Pay tax credit (now expired), the earned income tax credit, and the child tax credit--all refundable credits--were three of the top ten tax expenditures in 2010.
Exclusions exempt certain types of income from taxation, while deductions reduce taxable income. Both types of tax expenditures provide no benefit to those who do not pay income tax. The value of these tax preferences depends on the taxpayer's marginal tax rate, generally reducing tax liability more for a high-income taxpayer than a low-income taxpayer. For example, a high-income taxpayer in the 35 percent bracket receives a $35 tax saving by deducting a $100 charitable contribution while a low-income taxpayer in the 15 percent bracket receives only a $15 tax saving for the same level of contribution.
The Joint Committee on Taxation lists nearly 200 individual income tax expenditures for 2010. Estimates for these tax expenditures total nearly $1 trillion, a large number relative
to the $2.2 trillion in individual income tax revenue collected that year. If the federal government eliminated or curtailed income tax expenditures--particularly the top few--statutory tax rates could be drastically reduced across the board without a net revenue loss.
The Tax Reform Act of 1986 adopted this approach, eliminating or restricting many tax preferences and lowering statutory tax rates in a roughly revenue-neutral manner. Yet, as we have already mentioned, many tax preferences remain in place today. Some were retained by the 1986 act, and others have been added to the tax code over the last 25 years.
Within the last year, various groups and policymakers have made proposals to again broaden the income tax base and lower statutory tax rates. Some of these proposals are intended to be revenue-neutral, while others are intended to raise revenue for deficit reduction. In November 2010, the Bipartisan Policy Center proposed a plan to address the long-term fiscal imbalance that would have eliminated many income tax preferences and reduced the top tax rate to 27 percent.1 In December 2010, an eleven-member majority of the eighteen-member bipartisan Fiscal Commission appointed by President Obama and congressional leadership voted in favor of a fiscal plan that included income tax base broadening. The plan presented an illustrative proposal, the Modified Zero Plan, which eliminated many income tax preferences and lowered the top statutory rate to 28 percent.2 The fiscal 2012 budget resolution passed by the House of Representatives in April 2011 called in general terms for income tax base broadening and statutory tax rate reductions. President Obama also expressed support for such reforms, in general terms, in his September 2011 budget plan.3
As table 1 illustrates, the major tax expenditures are popular provisions used by millions of middle-class tax-payers. Any effort to significantly lower income tax rates through base broadening cannot rely solely on curtailing tax loopholes used by small groups of taxpayers but must tackle provisions that benefit large numbers of ordinary taxpayers. Of course, efforts to reduce income tax rates by cutting spending face a similar political complication, as real spending restraint requires cuts to popular programs such as Social Security and Medicare.
Although the popularity of the major tax preferences may make base broadening politically difficult, that does not necessarily mean such reforms are undesirable. The question we examine here is whether base broadening offset by rate reduction is likely to offer net economic gains. Stated differently, is it economically better to raise revenue by restricting income tax preferences than by raising tax rates? As we will see, the answer depends on which preferences are restricted.
Simplification and Distribution
We begin the economic analysis by arguing that two considerations have only a limited impact on the desirability of base broadening: simplicity and distributional issues. Simplification is frequently equated with base broadening, but the two are often unrelated. And distributional objectives are best advanced by altering tax rates, not by keeping or removing particular tax preferences.
Although tax simplification is often treated the same as base broadening, the two are quite different. Broadening the base by eliminating the exclusion of a particular type of income may actually increase complexity by requiring additional measurement and reporting. For example, taxing employees on employer-provided fringe benefits would require valuing each worker's benefits, and taxing the imputed rental value of owner-occupied homes would be an administrative nightmare. Conversely, many of the biggest simplification gains can be achieved without broadening the tax base. For example, consolidating and simplifying the numerous intricate education and savings preferences provided under current law would help, even if the overall generosity of the preferences remained unchanged and the tax base became no broader.4
Because some tax preferences largely benefit lower-income taxpayers while others largely benefit higher-income taxpayers, base broadening can also change the distribution of the tax burden across income classes. Although distributional concerns are important, they should not be given much weight when considering base-broadening options. Sound tax policy generally calls for distributional concerns to guide the selection of the rate structure and for efficiency to guide the design of the tax base.5 Statutory tax rates, as well as per-person exemptions and standard deductions available to all taxpayers, can be adjusted to undo any undesired distributional effects that may arise from base broadening.
The Wrong Case for Base Broadening: Work Incentives
One of the major arguments some supporters of base broadening make--that such reforms reduce work disincentives--is generally mistaken. Work incentives depend on effective marginal tax rates rather than statutory tax rates, and revenue-neutral base broadening leaves the former roughly unchanged.
Tax Rates and Work Incentives
We can illustrate the distinction between statutory and effective tax rates with a simple example. Consider an income tax with a statutory tax rate of 40 percent that allows a deduction for apple purchases, which comprise half of consumption. As a result, the 40 percent tax applies only to purchases of oranges, the remaining half of consumption.
Now, suppose that a tax reform lowers the statutory rate to 20 percent while repealing the apples deduction. As discussed below, this reform does enhance efficiency by eliminating the distortion between apples and oranges. It does not, however, reduce work disincentives.
At first glance, this revenue-neutral reform appears to encourage work, because the worker now pays only 20 cents of tax on each additional dollar of income earned. Didn't the worker pay 40 cents of each additional dollar before the reform? No, actually not. Because the worker spent half of each dollar on apples and deducted that amount, only 20 cents of tax was paid on each additional dollar. Although the statutory rate has fallen from 40 to 20 percent, the effective tax rate is unchanged at 20 percent.
When people work, they give up leisure in favor of consumption. Work incentives therefore depend on how the tax system affects the trade-off between leisure and consumption, not on the statutory tax rate assigned to income. In this case, reform reduces the tax on working to buy oranges from 40 percent to 20 percent but also increases the tax on working to buy apples from zero to 20 percent, and this combination leaves work incentives roughly unchanged.
In this example, we assume neither apples nor oranges have any special relationship to work. Revenue-neutral base broadening does reduce work disincentives if it raises taxes on goods that are substitutes for work. A tax increase on vacation spending might have such an effect. On the other hand, revenue-neutral base broadening increases work disincentives if goods complementary to work are targeted for tax increases. For example, curtailment of the child and dependent care credit would likely increase work disincentives in view of the statistical evidence that cheaper child care increases parental labor supply.6
Because the Tax Reform Act of 1986 dramatically reduced statutory tax rates, including the top rate from 50 to 28 percent, many observers concluded that it would reduce work disincentives. As Alan Auerbach and Joel Slemrod noted, though, because the rate reductions were offset by base broadening, the reform left effective tax rates and work incentives roughly unchanged for the economy as a whole, although it significantly raised or lowered disincentives for particular taxpayers.7
Computing Effective Marginal Tax Rates
Calculating effective tax rates is not always simple. As we discussed above, such calculations must account for the fact that some of each additional dollar earned is spent on items that receive a tax deduction or credit. For example, if a taxpayer in the 35 percent bracket claims itemized deductions equal to 20 percent of income, then his or her effective marginal tax rate is only 28 percent. Conversely, effective tax rates increase if some of the additional dollar earned through work is spent on items subject to excise taxes. If there is a 20 percent gasoline tax and a worker spends five cents of each additional dollar on gasoline, then the gasoline tax increases the worker's effective marginal tax rate by one percentage point.
Income-based phaseouts of tax breaks also increase effective marginal tax rates. For example, for an unmarried Social Security recipient with a yearly income just above $25,000, earning an additional dollar of income may cause an additional 50 cents of Social Security benefits to become taxable. Her taxable income rises by $1.50 as a result of earning the extra dollar; if she is in the 15 percent bracket, her effective marginal tax rate is now actually 22.5 percent. In two earlier articles, we computed effective marginal tax rates for taxpayers in 2008 and found a wide range of rates.8
The Right Case for Base Broadening:
Leveling the Playing Field
In our example, replacing the 40 percent tax on oranges with a 20 percent tax on apples and oranges does not reduce work disincentives. Nevertheless, such a reform offers different significant economic benefits by promoting tax neutrality between apples and oranges.
The economy is generally most efficient when the free market determines the allocation of resources between goods, based on production costs and consumer preferences. When different goods are taxed at different rates, efficiency is impeded because the allocation of resources is based partly on tax considerations, rather than costs and preferences. For example, if apples, but not oranges, are tax-deductible, the economy produces too many apples and too few oranges. Switching to a neutral 20 percent tax on both goods corrects this misallocation, increasing the production of oranges and reducing the production of apples and yielding a set of goods consumers find more attractive. The economic gains from moving to a 20 percent uniform tax arise from a more efficient mix of goods, not from an increase in work effort.
Base broadening is likely to be most useful when it addresses the major distortions of the current tax system, such as the preferences for employer-provided health insurance and owner-occupied housing. The exclusion from income tax of employer-provided health insurance--the largest tax expenditure in 2010--leads workers to consume more health insurance and less cash wages. Similarly, the mortgage interest deduction--the second-largest tax expenditure--prompts taxpayers to own larger houses.
The Wrong Kinds of Base Broadening
Although base broadening often promotes economic efficiency, it does not always do so. As we noted above, a reform increases efficiency if it makes the tax system more neutral across alternative uses of economic resources. Curtailing deductions, credits, and exclusions can make the tax system more neutral by reducing the preferential treatment of tax-favored activities. It is possible, though, for base broadening to tilt, rather than level, the playing field across different uses of economic resources. In that case, base broadening impedes economic efficiency.
Undermining Tax Neutrality
Eliminating or curtailing a tax preference that offsets other preferences can undermine tax neutrality and economic efficiency if the other preferences are left untouched. For example, eliminating the deduction for out-of-pocket medical expenses may appear to promote neutrality by correcting an incentive for excess consumption of health care. But if the tax preference for employer-provided health insurance, which is more generous than the deduction of out-of-pocket costs, remains in place, the change may increase the tax bias toward employer-provided insurance.9
Even when completely eliminating a provision would promote neutrality, a policy that removes it in some cases while preserving it in other cases may impede neutrality because this inconsistent policy may itself create inefficiency. Although completely eliminating the exclusion for interest on municipal bonds might promote neutrality by curbing excessive state and local government borrowing, eliminating it for only private-purpose bonds (bonds issued for projects in which the state or local government works with private firms or nonprofits) may inefficiently induce state and local governments to shift away from those projects and back toward less efficient public projects.10
In short, an excessive focus on statutory rates can lead to the wrong kind of base broadening. Neither base broadening nor rate reduction is an end in itself. The objective should be to make the economy more efficient by removing tax provisions that favor some economic activities over others. Base broadening is helpful when it promotes this objective and harmful when it deviates from it.
Increasing the Tax Penalty on Saving
One way base broadening can impede tax neutrality is by amplifying the income tax's inherent distortion between current and future consumption. Income taxation distorts resource allocation by penalizing saving and investment, thereby favoring current consumption over future consumption. True neutrality requires consumption taxation, not income taxation. Broadening the income tax base may increase the saving penalty, undermining the pursuit of economic efficiency.
We can illustrate this penalty with a simple example, borrowed from an earlier Outlook.11 Consider two individuals, Patient and Impatient, each of whom earns $100 in wages today. Impatient wishes to consume only today and Patient wishes to consume only "tomorrow," which is many years in the future. Saving yields a 100 percent rate of return between today and tomorrow. With no taxes, Impatient consumes $100 today. Patient saves the $100, earns $100 interest, and consumes $200 in the future.
What happens with a 20 percent income tax? Impatient pays $20 tax on his wages today and consumes the remaining $80, which is 20 percent less than in the no-tax world. Patient also pays $20 tax and saves the remaining $80, earning $80 interest. However, $16 tax is also imposed on the $80 interest. That leaves Patient with $144, which is 28 percent less than in the no-tax world, compared to a mere 20 percent reduction for Impatient. The income tax imposes a higher percentage tax burden on Patient solely because she consumes later. The income tax's penalty on saving causes an inefficient distortion of consumer choice and lowers the accumulation of national wealth and the long-run rise of living standards.
The current income tax system includes certain provisions that mitigate the saving penalty. As shown in table 1, three of these provisions--the reduced tax rate on dividends and capital gains, the tax preferences for defined benefit plans, and the tax preferences for defined contribution plans--are among the top ten tax expenditures. These "preferential" provisions tend to promote efficiency by treating current and future consumption more neutrally. Base broadening that removes or curtails them may, therefore, damage economic efficiency.
Because the penalty on saving is an inherent feature of income taxation, it cannot be eliminated by income tax base broadening. The only way to eliminate the penalty is to replace the income tax with a consumption tax; in the above example, a consumption tax with a constant tax rate over time would have the same percentage impact on Patient and Impatient. In that respect, income tax base broadening is an imperfect substitute for consumption taxation, a reform that lies outside the scope of this Outlook.
Even if base broadening cannot remove the tax penalty on saving, it must not increase the penalty by removing provisions that ameliorate the penalty. To be sure, many of these provisions are poorly designed; the bewildering variety of tax-preferred savings plans creates needless complexity and is not the most effective way to ease the tax penalty on saving. Base broadening measures can be useful if they improve the design of these provisions without increasing the penalty on saving.
The Way Forward
Base broadening and statutory rate reductions are not ends in themselves. Like other tax policies, these reforms should be evaluated by whether they promote efficiency, simplicity, and fairness.12 Revenue-neutral base broadening can be useful if it is done properly. Priority should be given to reducing the largest distortions, such as the preferences for employer-provided health insurance and owner-occupied housing. Tax preferences should be preserved for work-related costs, and it is particularly important to avoid increases in the tax penalty on saving. In fact, policymakers should consider using part of the revenue raised by base broadening to reduce the tax penalty on saving.
1. Bipartisan Policy Center, Debt Reduction Task Force, Restoring America's Future: Reviving the Economy, Cutting Spending and Debt, and Creating a Simple, Pro-Growth Tax System (Washington, DC: Bipartisan Policy Center, November 2010), 31-40, www.bipartisanpolicy.org/sites/default/files/BPC%20FINAL%20REPORT%20FOR%20PRINTER%2002%2028%2011.pdf (accessed September 19, 2011).
2. National Commission on Fiscal Responsibility and Reform, The Moment of Truth (Washington, DC: Author, December 2010), 29-32, www.fiscalcommission.gov/sites/fiscalcommission.gov/files/documents/TheMomentofTruth12_1_2010.pdf (accessed September 19, 2011).
3. Office of Management and Budget, Living Within Our Means and Investing in the Future: The President's Plan for Economic Growth and Deficit Reduction (Washington, DC: September 2011), 45-52, www.whitehouse.gov/sites/default/files/omb/budget/fy2012/assets/jointcommitteereport.pdf (accessed September 21, 2011).
4. For a recent description of ways to simplify the income tax, with references to the work of the Joint Tax Committee and the National Taxpayer Advocate, see Alan D. Viard, "The Tax Code's Burdens on Families and Individuals," Testimony before the House Ways and Means Committee, April 13, 2011, www.aei.org /article/103457.
5. For a rigorous explanation of why distributional concerns should be pursued through the rate schedule rather than giving preferential treatment to particular uses of income, see Louis Kaplow, The Theory of Taxation and Public Economics (Princeton, NJ: Princeton University Press, 2008), 122-35.
6. See Alan D. Viard, "The Child Care Tax Credit: Not Just Another Middle-Income Tax Break," Tax Notes, (September 27, 2010): 1397-1403, www.aei.org/article/102577.
7. See Alan J. Auerbach and Joel Slemrod, "The Economic Effects of the Tax Reform Act of 1986," Journal of Economic Literature 35, no. 2 (June 1997): 589-632.
8. See Alex Brill and Alan D. Viard, "Effective Marginal Tax Rates, Part 1: Basic Principles," Tax Notes (September 8, 2008): 969-73, www.aei.org/article/28596; Alex Brill and Alan D. Viard, "Effective Marginal Tax Rates, Part 2: Reality," Tax Notes (October 20, 2008): 327-32, www.aei.org/article/28829. These computations account for income-based phaseouts and the effects of itemized deductions, but not for excise taxes.
9. The alternative minimum tax limits the deduction for out-of-pocket medical costs, which may not promote efficiency. See Alan D. Viard, "The Alternative Minimum Tax: A Better System?" AEI Tax Policy Outlook (November 2006), www.aei.org/outlook/25110.
10. The alternative minimum tax eliminates the exclusion for most private-purpose bonds. See Viard, "The Alternative Minimum Tax."
11. Robert Carroll, Alan D. Viard, and Scott Ganz, "The X Tax: The Progressive Consumption Tax America Needs?" AEI Tax Policy Outlook (December 2008), www.aei.org/outlook/29082.
12. Alan D. Viard, "Two Cheers for Corporate Tax Base Broadening," National Tax Journal 42, no. 3 (September 2009): 399-412, www.aei.org/article/101260, makes a similar point with respect to corporate tax base broadening