Wednesday, March 06, 2013

Skepticism about "fundamental tax reform"

Today we had a special lunchtime event at NYU Law School.  Victor Thuronyi of the International Monetary Fund, an old friend with whom I worked on the Tax Reform Act of 1986 (he was on the Treasury staff, while I was at the Joint Committee on Taxation), discussed his plan for enacting a "supplemental expenditure tax" (SET) as part of an overall tax reform plan.  The SET is basically a consumed income tax, aka an individual-level cash flow consumption tax, that he would use to pay for rate reductions in the existing income tax.  You can read about it here.

I myself find it easier to imagine a VAT being added to supplement the income tax than as SET (as in the Michael Graetz tax reform plan).  But I question the premise that both Thuronyi and Graetz advance, to the effect that the revenues raised by the new tax should be "spent" on reducing income tax revenues.  Considered in isolation, that might be a good idea, but what about other budgetary uses for the funds when we appear to have a long-term fiscal gap to worry about once the horrible employment situation of the last few years has passed into history.

But both plans are certainly welcome as expansions to the list of ideas that are people are talking about.

I was a commentator at Victor's session today, as was David Miller, and here is a fleshed-out portion of the part of my remarks in which I addressed, not so much Victor's plan as such, as the issue of whether we need 1986-style "fundamental tax reform."  This, of course, is a topic that I have also addressed here.

Shaviro remarks (3/6/13) on 1986-style "fundamental tax reform"

Everyone loves tax reform in theory, though not so much in practice. And of course tax policy types tend to love it, and why wouldn’t we? But I want to verge on heresy by throwing a bit of cold water on the idea.

Obviously I’d like to “reform” the tax system, in the sense of changing it to be better. And there is so much wrong with the current system that identifying things we could improve is a bit like shooting fish in a barrel.

But I have become a real skeptic about what I call 1986-style tax reform. Let me first define it, then say why I have become so skeptical about it.

1986-style reform typically involves lowering the rates and broadening base, under a constraint of maintaining revenue neutrality and distributional neutrality relative to prior law. By the way, in 1986 the top group for purposes of assessing distributional neutrality was people with income of $250,000 or more.  Today, even adjusting for 27 years of price-level changes, we might want to look distinctively at much higher-up groups (such as the top 1%, 0.1%, and 0.01%).

In 1986, I agree that this tax reform model was a good one.  The parties disagreed about the revenue and distributional parameters, but were capable of agreeing about some other stuff. So they did what they agreed about, while agreeing to a ceasefire in place for the rest.  Note, by the way, that the top rates they were lowering rates were 50% and 46% for corporations - considerably above where we are today.

Today, I don’t see the politics working as well, for two reasons.  First, the relationship between the parties is obviously very different than it was in the mid-1980s. But second, consider what Congressman Richard Gephardt notoriously said in 1986 (as quoted in the subsequent Birnbaum-Murray book, "Showdown at Gucci Gulch."  Biographical note: Gephardt was a cosponsor of the well-known Bradley-Gephardt tax reform plan, but then went completely MIA when the 1986 Act started marching towards enrollment,  Anyway, he said: “It’s not that good an issue.” And he was right as a political matter.  The 1986 Act only went through, despite widespread public hostility or indifference, due to what economist Henry Aaron called the "dead cat" problem.  None of the main leaders responsible for shepherding it through (President Reagan, James Baker when he became the Treasury Secretary, Dan Rostenkowski in the House, or Bob Packwood in the Senate) wanted the "dead cat" of having killed tax reform on his doorstep.  Even though the public disliked the tax reform bill, it was all too ready to assume that anyone who killed it was a tool of the interest groups and/or an ineffective leader.  So the political dynamic worked, but it was a very odd one and hard to replicate.

That's just about the feasibility of 1986-style reform, which flies in the face of the political science maxim that concentrated special interests tend to beat out diffuse general interests.  But again, perhaps so what.  Here are several reasons for being, at a minimum, only mildly rather than greatly upset (and, in some cases, affirmatively relieved) if 1986-style reform doesn't happen:

(1) Let’s not exaggerate the benefits.  At a first approximation, if you broaden the base but lower the rates, work incentives are the same as before.  The economic payoff is to distortion between rival consumption and/or investment choices.  That can be a nice payoff, but it doesn't, for example, project to greatly increased ecnomic growth.

(2) Complexity costs for average individual taxpayers, which these plans often seek to reduce (e.g., by repealing the alternative minimum tax) often are not as great a concern as they used to be.  Two words (or is it actually just one?): TurboTax.

(3) 1986-style reform often treats a high rate and a broad base as substitutes. You use revenue gain from the latter to buy out of the former.  Economically speaking, however, they are complements.  The broader the base and thus the harder the tax is to avoid, the lower the efficiency costs of having a higher rather than a lower rate.

(4) What does it even mean to say, 1986-style, that revenues will remain the same? Given the point that targeted tax breaks serving allocative purposes may (more or less rightly) be viewed as tax expenditures, any such claim is based on form, not substance.  It might be more accurate to say that both "taxes" and "spending" have declined (although these terms aren't very meaningful to begin with) if we, say, repeal the home mortgage interest deduction to pay for lower rates.

(5) Should everything have the same tax rate?  The theory of optimal commodity taxation explains why the answer to this question is no under realistic parameters that are relevant here. Rather, unless you've done more fundamental reform, items that are more elastic typically should be taxed at lower rates than those that are less elastic.  The revenue-maximizing rate for capital gains, at least under present law (with a realization system and tax-free basis step-up at death) is probably below 30%.  That for wage income may be a lot higher.  Or consider the argument that, so long as we have an income tax that's collected at the entity level for income earned through a corporation, the corporate rate should be lower than the individual rate.  (This reflects the fact that corporate income is a proxy, albeit an imperfect one, for globally mobile capital.)  Finally, consider the argument I will make in my forthcoming international tax book that the optimal U.S. rate for foreign source income of U.S. companies is lower than the domestic rate, even if not so low as zero (as it would be under a territorial system), and that everyone kind of knows this although plenty of people (including experts) confuse themselves by relying on the foreign tax credit to lower the effective tax rate on the foreign source income.

(6) Given the long-term fiscal gap that we face, it's questionable whether we should do a whole lot of heavy lifting in our tax system design that has zero net payoff in terms of returning to sustainability.  In addition to diverting scarce "reform capacity," a 1986-style might give away "low-hanging fruit" by "spending" the budgetary gain from base-broadening (or new taxes such as the VAT or SET) on wish lists other than restoring long-term sustainability.

(7) If you are concerned about high-end distribution, why would you want to cut the top individual rate?  Arguably we should be thinking instead about increasing it.

1 comment:

PJS Pereira said...

I have been thinking about rates vs. base broadening for a while. You are right, that increasing top rates will tend to reduce income inequality, but I think that the effect may be more subtle. I think that the key parameter is not the top rate, but the difference in average tax rate between the top ~10% of income earners and the bottom 90% of earners (as opposed to top 1% or 0.01%), i.e. the overall level if income transfer between the top and the bottom.

If this level increases, then one would expect the income of the bottom 90% to increase and the top 10% to decrease as the level of inequality changes. The ~10% threshold is related to the Pareto coefficient for the US income distribution.

In summary, I suspect one may reduce inequality with relatively flat tax rates, so long as one monitors the level of transfer between the top ~10% and bottom ~90%. Whether the top 0.01% are taxed at a much higher rate than the top 1% does not matter.