How Big Are Tax Preferences?

By :: April 9th, 2012

The tax code is chock full of credits, deductions, deferrals, exclusions, exemptions, and preferential rates. Taken together, such tax preferences will total almost $1.3 trillion this year.

That’s a lot of money. But it doesn’t necessarily mean that $1.3 trillion is there for the picking in any upcoming deficit reduction or tax reform.  In fact, even if Congress miraculously repealed all of these tax preferences, it would likely generate much less than $1.3 trillion in new resources. 

Where did I come up with that number? For a short piece in Tax Notes, I simply added together all the specific tax expenditures identified by the Department of Treasury; these were reported in the Analytical Perspectives volume of the president’s recent budget.

Treasury doesn’t report this total for a good, technical reason: some provisions interact with one another to make their combined effect either larger or smaller than the sum of their individual effects. As a result, simple addition won’t give an exact answer. That’s an important issue. In the absence of a fully integrated figure, however, I think it’s useful to ballpark the overall magnitude using basic addition.

In your travels, you may find other estimates that do the same thing but come up with a figure of “only” $1.1 trillion. Why is mine higher? Because it includes some important information that Treasury reveals only in footnotes. Treasury’s main table estimates how tax expenditures reduce individual and corporate income tax receipts; those effects total $1.1 trillion. But they also have other effects. Refundable credits like the earned income tax credit increase outlays, for example, and some preferences, like those for employer-provided health insurance and alcohol fuels, lower payroll and excise taxes. I include those impacts in my $1.3 trillion figure.

Budget hawks and tax reformers have done a great job of highlighting tax expenditures in recent years. I fear, however, that we have lifted expectations too high. Just because the tax code includes $1.3 trillion in tax preferences doesn’t mean it will be easy to reduce the budget deficit or pay for lower tax rates by rolling them back. Politics is one reason. It’s easy to be against tax preferences when they are described as loopholes and special interest provisions. It’s another thing entirely when people realize that these include the mortgage interest deduction, the charitable deduction, and 401(k)s.

Basic fiscal math is another challenge. Tax expenditure estimates do not translate directly into potential revenues. Indeed, there are several reasons to believe that the potential revenue gains from rolling back tax preferences are less than the headline estimates. One reason is that the estimates are static—they measure the taxes people save today but do not account for the various ways that people might react if a preference were reduced or eliminated; those reactions may reduce potential revenues. Second, most reforms would phase out such preferences rather than eliminate them immediately. That too reduces potential revenues, at least over the next decade or so.

Finally, the value of tax preferences depends on other aspects of the tax code, most notably tax rates. If a tax reform would lower marginal tax rates, the value of deductions, exclusions, and exemptions would fall as well. Suppose you are in the 35 percent tax bracket. Today, each dollar you give to charity results in 35 cents of tax savings—a 35-cent tax expenditure. If the top rate were reduced to 28 percent, as some propose, your savings from charitable donations would be only 28 cents. The 20 percent reduction in tax rates would thus slice the value of your tax expenditure by 20 percent. That means that the revenue gain from eliminating the deduction—or any other similar tax expenditures—would also shrink by 20 percent, thus making it harder for tax expenditure reform to fill in the revenue gap left by reducing tax rates.

My message is thus a mixed one. Tax expenditures are very large—$1.3 trillion this year alone if you add up all the individual provisions – and deserve close scrutiny. But we need to temper our aspirations of just how much revenue we can generate by rolling them back. It isn’t as though there’s an easy $1.3 trillion sitting around. In coming months, the Tax Policy Center will explore how to translate tax expenditure figures into more reasonable estimates of the potential revenues that tax reformers and budget hawks can bargain over.

P.S. For an interesting analysis of how individual tax preferences interact with each other, see this piece by TPC’s Dan Baneman and Eric Toder.


  1. Michael Bindner  ::  3:59 pm on April 9th, 2012:

    These proposals are best estimated in terms of actual tax reform proposals, which should include trade-offs among preferences, rates and revenues desired. For example, one proposal is to consolidate the child tax credit, earned income credit and child deduction into a single larger child tax credit – with the original child tax credit held at its current policy rather than its permanent law rate. It could be further increased by ending the mortgage interest deduction and property tax deductions and using the additional revenue to justify a more generous credit.

    Other preferences are for special rates which should simply be ended, although the price for doing so is a lower marginal rate and the expectation that some additional revenue would be raised in the process – such as eliminating the special capital gains and dividend rates and lowering the top corporate and individual rates to 28%. If distributions from inheritances were also taxed at normal income, but inherited assets were not, it would also have an impact and would likely raise revenue.

    Other proposals would be to raise the floor on tax payment entirely and put in consumption taxes to replace the lost income, including shifting payroll taxes to a VAT-like net business receipts tax and shifting the aforementioned child tax credit to that tax as well – essentially forcing employers to pay lower wages to the childless and higher income workers in order to offset the higher tax rate that pays these taxes. Indeed, the NBRT might replace the corporate tax and be set higher, but with a lower effective tax rate for most firms once deductions for healthcare and children are taken into account.

  2. AMTbuff  ::  8:01 pm on April 9th, 2012:

    When considering tax deferrals, conventional tax expenditure methodology is faulty, looking only at the short term. Ending any tax deferral creates an immediate revenue gain followed much later by revenue losses, when the income would have been taxable under the old rules. Advocates of higher spending never point this out, instead implying that the extra revenue will continue forever rather than reversing to a loss.

    For example, if you tax 401k’s and such today, you can’t collect tax on the same income when it’s withdrawn. Most of that foregone revenue is decades in the future, so that offsetting cost vanishes from this accounting.

    It’s the same story with stimulus spending, where advocates appear not to know that adding debt creates anti-stimulus in the long run, reducing growth rates in the out years as the initial stimulus is paid back. In stimulus and in eliminating tax deferrals there is no free lunch. There is only eating your dessert before your vegetables, which is one thing at which our government excels.

  3. barbour  ::  9:36 pm on April 26th, 2012:

    In sixteen twenty-one, a big celebration of barbour jacket took place at Plymouth Colony in what is now the state of Massachusetts. European settlers known as the barbour jackets were celebrating their autumn harvest after a winter of struggle.
    Other colonists held earlier ceremonies of thanks. But the Pilgrims’ three-day feast is often called the nation’s first barbour outlet Thanksgiving. President Abraham Lincoln declared a national holiday in eighteen sixty-three during the Civil War.
    Today families and friends gather on the fourth Thursday in November. And, thanks to the barbour quilted jacket United States Census Bureau, here are some facts about where their Thanksgiving meal comes from.
    The main dish is traditionally a turkey. About two-thirds of the nation’s turkeys are produced in Minnesota and five other quilted barbour jackets tates. Producers are expected to raise two hundred forty-eight million birds this year, two percent more than last year.
    Popular side dishes include cranberries and sweet potatoes. Last year North Carolina barbour sale grew more sweet potatoes than any other state. Wisconsin is expected to be the top cranberry producer this year.
    People often finish the meal with pumpkin pie. Last year Illinois grew the most pumpkins. California, New York, Ohio and barbour uk were also major pumpkin producers.
    Thanksgiving is a big event, but buying a barbour mens traditional holiday meal can be a struggle for the poor. So charity groups often hold food drives to collect food forbarbour ladies low-income families.