-- Former President Clinton, Sept. 5
-- President Obama, Sept. 6
-- Republican presidential candidate Mitt Romney, Sept. 9
The now-famous August study from the Urban-Brookings Tax Policy Center (TPC) presents the data, and it shows that achieving all these goals simultaneously is, under reasonable assumptions, mathematically impossible. (Citations are at the end of this article.) That allowed the Democrats to score some points. We won't dwell on that here. Instead we are going to look beyond the election and explore what could be done in 2013 to help President Romney draft a decent tax plan that would be roughly consistent with his campaign promises.
The TPC Study
The figure below illustrates the main points of the TPC study. A 20 percent rate cut on households with income over $200,000 lowers their taxes by $251 billion. Despite repeated inquiries from the press, Romney has not specified which tax expenditures he would cut. To complete its analysis, the TPC assumed that all tax breaks except those that favor investment income (shown in the table below) would be eliminated.
Under these assumptions, the study found that households with income over $200,000 would have offsetting tax increases of only $165 billion. That's $86 billion less than the $251 billion of relief from rate cuts. So the Romney approach results in a tax cut for high-income households. And because his overall reform is revenue neutral, a 20 percent rate cut for incomes over $200,000 means that taxes must rise on households with income below $200,000 to make up the difference.
Option 1: Rates Adjustment
One way to salvage the Romney plan would be to eliminate all tax expenditures (except for investment income), and then instead of cutting statutory rates uniformly, adjust them as necessary to maintain distributional neutrality. As already noted, Romney cannot reduce rates for the wealthy by 20 percent given that he put an off-limits sign on tax breaks for saving. Back-of-the-envelope calculations suggest a reduction of something like 13 percent might be possible.
Because households with income below $200,000 have a lot of tax breaks, eliminating all of them would permit a much larger revenue-neutral rate cut -- one of probably more than 30 percent. In other words, if Romney were to cut all non-investment tax expenditures, he could cut rates substantially, but he would have to make the rate structure steeper.
Most economists would prefer that outcome: a much broader base and much lower rates. The huge fly in the ointment is that there is no way in the real world that tax expenditures like those for home mortgages, charitable contributions, and employer-provided health insurance will be eliminated.
But it is all a matter of degree. Instead of being eliminated, these tax benefits could be scaled back. And what we have in mind here are very mechanical cuts. For example, whatever credits, exclusions, and deductions (that are tax expenditures) that taxpayers currently receive could be cut by one-quarter. As long as that scaling back is not on the basis of income (more on that below), it is possible to design a plan that is economically beneficial and distributionally neutral. But to make that possible, Romney would have to scrap his idea of cutting rates uniformly.
Option 2: Deductions Become Credits
Most tax expenditures come in the form of deductions and exclusions from income and therefore provide more benefit for higher-income taxpayers in the upper brackets. That tilt in favor of upper-income households can be corrected by converting deductions to credits. (For example, President Bush's 2005 tax reform panel proposed converting the mortgage interest deduction to a 15 percent credit.) An alternative that produces similar results would be a percentage cut in deductions and exclusions for the upper-income households so that the value of their deductions is no greater than it is for those households facing lower statutory rates. (President Obama has proposed in all four of his budgets limiting the tax benefit of itemized deductions to 28 percent.)
In a Romney tax reform plan, converting all tax deductions and exclusions to, let's say, 15 percent tax credits would make the system more progressive. And from an economic growth perspective, there are many positive aspects to that approach. Not only would it reduce distortions that result from tax expenditures favoring one activity over the other, but it would do so in a manner that minimizes marginal tax rates on income. In other words, it would have minimal detrimental effects on the incentive to work.
In effect, under a system in which all deductions are converted to credits, taxpayers just below the next highest bracket are pushed up into that bracket and face a higher marginal tax rate. That hurts growth. But these marginal rate increases occur only in proportion to current deductions and exclusions that reduce marginal rates for some and not for others. Converting to credits eliminates that disparate treatment. All taxpayers would be on the same rate schedule -- and therefore face the same marginal rates -- as their economic income rises. A tax system in which all individuals face the same rate schedule would be more efficient than a tax system that raises the same revenue but under which individuals face different schedules. The conversion of deductions to credits does not create a new distortion. It eliminates one that already exists.
This approach has its limitations. High-income tax benefits are only scaled back, not eliminated. That limits the amount of rate reduction for high-income taxpayers. If deductions were converted to 15 percent credits, high-income households almost certainly would not be able to get rate reductions of more than 10 percent in a plan like Romney's.
Option 3: Income Phaseouts
Another way to salvage the Romney approach and maintain distributional neutrality would be to phase out deductions, exclusions, and credits as income rises. Phaseouts are used in more than a dozen individual provisions in the current code. And from 1991 through 2000, up to 20 percent of itemized deductions were disallowed above specific income thresholds. (That provision, commonly referred to as the Pease amendment, will be reinstated in 2013 if the Bush tax cuts are allowed to expire.) Pease-type phaseouts are a clearly inferior approach to using rates to achieve distributional goals.
To see why, let's begin by looking at a stylized best-case scenario in which a phaseout is economically equivalent to a rate increase. For argument's sake, suppose all taxpayers are eligible for a $5,000 tax credit, the tax rate on income over $150,000 is 30 percent, and nobody's income exceeds $250,000. Now also suppose that we want to raise taxes on income over $150,000. We could raise the top rate to 35 percent. Or we could phase out the tax credit by $5 for every $100 over $150,000. The result is exactly the same in both cases. For all income over $150,000, the effective marginal rate is 35 percent and the tax increase is 0.05 times the amount that income exceeds $150,000.
Although in this case the two approaches are economically equivalent, the phaseout is still objectionable for two reasons. First, it is a sneaky way of raising taxes. The public can far more easily comprehend a tax increase that comes through a rate cut than through a phaseout. To minimize whatever political damage that changes to the highly visible rate schedule may cause, politicians will find refuge by replacing them with phaseouts of deductions, credits, and exemptions. Congress's 1990 adoption of the Pease amendment, which phased out up to 20 percent of itemized deductions for upper-income taxpayers in lieu of a politically unacceptable rate increase, is a classic example. If only for the sake of transparency, phaseouts should be replaced by rate adjustments to the statutory rate schedule that have the same distributional effects.
Tax Benefits for Investment Not Used as
Revenue Raisers in the TPC Study
- Preferential rate on capital gains and dividends
- Exclusion of interest on tax-exempt municipal bonds
- Deduction for health savings account contributions from taxed income
- Keogh, SEP, and simple contribution deduction
- Penalty on early withdrawal of savings
- Self-employment IRA deduction
- Investment interest
- Savers credit
- Capital gains exclusion on home sales
- Step-up basis of capital gains at death
A second reason to favor rate adjustments over phaseouts is the considerable added complexity and uncertainty. (See sidebar.) A taxpayer with uncertain income will not know if he qualifies for tax benefits, diluting any incentive effects the tax breaks may be expected to deliver and making planning difficult.
SIDEBAR: Phaseouts Have No Place in Tax Reform
- There is widespread agreement among experts that income phaseouts should be repealed. Below are two quotes from leading authorities. The first is an excerpt from the 2008 national taxpayer advocate report to Congress (p. 342).
More than half of all individual income tax returns filed each year are affected by the phase-out of certain tax benefits as a taxpayer's income increases. Like tax sunsets, phase-outs are largely used to reduce the cost of tax provisions for budget-scoring purposes. However, phase-outs are burdensome for taxpayers, reduce the effectiveness of tax incentives, and make it more difficult for taxpayers to estimate their tax liabilities and pay the correct amount of withholding or estimated taxes, possibly reducing tax compliance. Phase-outs also create marginal "rate bubbles" -- income ranges within which an additional dollar of income earned by a relatively low-income taxpayer is taxed at a higher rate than an additional dollar of income earned by a relatively high-income taxpayer. Because Congress could achieve a similar distribution of the tax burden based on income level by adjusting marginal rates, phase-outs introduce unnecessary complexity to the Code. The National Taxpayer Advocate recommends repealing phase-outs or at least taking another look at phase-outs to ensure that they are really necessary to accomplish their intended objective.
The second is from a 2001 study by the Joint Committee on Taxation, "Study of the Overall State of the Federal Tax System and Recommendations for Simplification Pursuant to 8022(3)(B) of the Internal Revenue Code of 1986," April 2001, Vol. 1, p. 12:
These phase-outs require taxpayers to make complicated calculations and make it difficult for taxpayers to plan whether they will be able to utilize the tax benefits subject to the phase-outs. Eliminating the phase-outs would eliminate complicated calculations and make planning easier. These phase-outs primarily address progressivity, which can be more simply addressed through the rate structure.
TPC Analysis of Romney-Like Tax Reform Plan
(billions of dollars; for fiscal 2015)
But that's not all. Now let's get a little more realistic and look at cases in which rate changes and phaseouts are not economically equivalent. First, let's relax our prior assumption and let incomes exceed $250,000, as they do in the real world. That does not alter the phaseout, but it does expand the range of income to which a rate increase applies. This expansion in the number of taxpayers allows there to be a lower marginal tax rate for a given revenue target compared with what would occur with the phaseout. In other words, the marginal rate changes are spread out more evenly and at lower levels with a rate change than with a phaseout. In order to improve efficiency, economists are always looking to spread the pain as broadly as possible. A rate change allows for a broader base and lower marginal tax rate.
The inferiority of income phaseouts to statutory rate changes is also evident if you consider the real world, where otherwise equal taxpayers enjoy significantly different amounts of tax credits. If phaseouts are in force, taxpayers with large credits will bear higher marginal tax burdens than taxpayers with none. Again, a revenue-neutral rate increase would be superior because the required marginal rate changes would be spread equally across all taxpayers and therefore more efficiently.
Alternative 4: Raise Taxes on Investment
Another modification to the Romney approach that would allow him to achieve his distributional goals would be to cut tax expenditures on investment income for high-income taxpayers. That would contradict not only his campaign promises, but one of the most fundamental tenets of Republican tax orthodoxy: Always seek to cut taxes on capital. Most prominent among these possible changes would be permitting an increase in the current 15 percent rate on capital gains and dividends. (These rates are scheduled to increase in 2013 to 20 percent for capital gains received by most taxpayers and to 39.6 percent for dividends received by taxpayers in the top tax bracket. Also, as a result of the Affordable Care Act, there will be a 3.8 percent surtax on investment income for households with income over $250,000.)
However, history tells us that raising rates on capital and dividends and reducing other tax breaks on investment income can happen. Ronald Reagan -- the undisputed champion of tax cuts for capital during his first term in office -- was the moving force behind the Tax Reform Act of 1986, which largely eliminated the preferential tax treatment of capital gains.
Unfortunately for tax reformers, the Romney approach is not a fast track to lower rates and a broader base. That is because of the simple fact that the tax benefits that Romney might have to cut disproportionately benefit the middle class. If elected, Romney could address these issues by making the rate structure more progressive, by converting deductions to credits, or by phasing out tax breaks as income rises. The first two approaches are alternatives Romney should consider. But income phaseouts are bad economics and have no place in a tax reform plan that is supposed to improve the code.
A fourth alternative would be to reduce tax benefits for investment income, including dividends and capital gains. That could be part of a grand compromise with Democrats on tax reform if Romney chose to govern from the middle rather than from the far right. If Romney remains as fervently committed to distributional neutrality as he is in the quote at the beginning of this article, only tax reform that includes tax increases on investment income makes it possible to reduce the top rate by 20 percent.
- Daniel Baneman, Joseph Rosenberg, Eric Toder, and Roberton Williams, "Curbing Tax Expenditures," Urban-Brookings Tax Policy Center, Jan. 30, 2012.
Samuel Brown, William G. Gale, and Adam Looney, "On the Distributional Effects of Base-Broadening Income Tax Reform," Urban-Brookings Tax Policy Center, Aug. 1, 2012, and "Implications of Governor Romney's Tax Proposals: FAQs and Responses," Aug. 16, 2012.