Last Friday, the American Enterprise Institute’s Alex Brill and Alan Viard published this column on “The Folly of Obama’s Tax Plan,” in which they point out that all of Senator Obama’s proposed tax cuts for the middle class have the unfortunate side effect of raising marginal tax rates–effective tax rates on the next dollar earned–over certain ranges of the “middle-income” spectrum. They were chastised by pgl over on Angry Bear for their “basic dishonesty.” (UPDATE: here’s the link to Jonah Gelbach’s post that inspired the thread on AB.)
But in defense of Alex and Alan, their column is not dishonest, and is probably not even intended to “distort” or even “confuse” (characterizations of the column used on Angry Bear and the Economists for Obama site that pgl quotes from). It’s just that Alex and Alan are trying to make a big deal out of a small point, and a small point that has hardly any policy relevance.
That point is that adding greater progressivity to a tax system (raising relative tax burdens on the rich or lowering relative tax burdens on the poor)–something Senator Obama is clearly trying to do–necessarily means raising marginal tax rates somewhere along the income distribution, in order to increase how steeply average tax rates rise with income. In the case of refundable tax credits, these introduce progressivity by creating negative average tax rates (i.e., net income received rather than net taxes paid) at low income levels, but as income rises and the credits are phased out, the effective marginal tax rate equals the rate of reduction of the credit. And if you boost the value of any existing tax credit but keep its qualifying ranges of income the same, well, you’re necessarily going to be phasing out more benefits faster, i.e., boosting the effective (positive) marginal tax rate. Economists sometimes obsess about these phaseout marginal tax rates, because we like to worry about the adverse incentive effects of these tax rates on labor supply and saving–the idea that people might not work as hard or as many hours if they know that in earning more income, the value of their refundable tax credits will go down. Even if the credits don’t go down by anywhere close to the amount of extra income–i.e., even without an effective marginal tax rate that’s less than 100%–economists worry that the “substitution effect” of this higher marginal tax rate (caused by the negative average tax rate) could be a decline in how much people work.
But this is a “problem” with any tax preference that’s targeted to a certain range of income (it doesn’t just apply to refundable tax credits), so there’s no way to avoid it without “de-targeting” the preference and spending a heck of a lot more money, on tax cuts for everybody. The adverse economic effects of a much more costly tax cut, going to more people than is really justified given the purpose of the tax cut, far outweigh any potential benefit from reducing marginal tax rates and the possible substitution effects associated with the phaseouts of tax benefits.
This marginal tax rate criticism is often levied against the Earned Income Tax Credit (EITC) when economists feel like getting negative (i.e., insist on staying “dismal”) about a tax credit that otherwise delights most economists. The EITC is a refundable tax credit that phases in at lower income levels, plateaus over a range of income, and then phases out over a higher range of income, as explained on this Tax Policy Center page. Economists like the phase-in range of the EITC, where the rising credit with income means the marginal tax rate is negative–creating an incentive, rather than a disincentive, to work. Economists dislike, however, or at least fret about, the phase-out range of the EITC, where the phasing out of the credit adds to the effective marginal tax rate.
But I say it’s much ado about nothing. Theoretically, for the phaseout of a tax preference to adversely affect work incentives, it must be true that the worker is able to choose his or her hours of work in a continuous manner–that is, that I could choose to work 30 hours per week instead of 35, for example. Some workers can do this, but many cannot. It must also be the case that the worker is aware of the positive marginal tax rate faced at each decision point–aware that if he or she works an hour more, or 5 hours more, that his or her tax credit will go down by $__. Most people are much more keenly aware of their average tax rates and their take-home pay, than their marginal tax rates and their net wage rates “at the margin.”
Empirically, studies on the EITC have shown that there’s just not much to worry about in terms of disincentive effects associated with the phasing out (i.e., the targeting) of tax credits. A summary of recent empirical research on the EITC by the National Bureau of Economic Research explains that (emphasis added):
In Behavioral Responses to Taxes: Lessons from the EITC and Labor Supply (NBER Working Paper No.11729), NBER researchers Nada Eissa and Hilary Hoynes review a large number of economic studies of the EITC and conclude that the main lesson from the accumulated evidence is that real responses to taxes are important. The second lesson is that, while the EITC stimulates people to join the work force, there is no evidence that it prompts them to work fewer hours. This difference, the authors write, “has several important implications for the design of tax-transfer programs and the welfare evaluation of taxation.“
In other words, the implications for policy design are that we don’t need to worry about these potentially high marginal tax rates when tax credits are phased out, because: (i) there’s no way to avoid them when targeting a tax credit, and (ii) there’s no evidence that the disincentive effects are significant (regardless of the theoretical potential).
So the point that Alex Brill and Alan Viard make in their column is not dishonest. It’s just not very interesting (outside of a course in the microeconomics of public finance) and certainly not very policy relevant.