Taxes on Income, Red and Green

Prof. Landsburg recently posted a draft of his contribution to a FestschriftThe reader is urged to read the whole thing because, quite apart from the subject of this post, it has much news even for some who fancy themselves well-informed on the subject. For example, the author, while a reader and admirer of both McCloskey and the Greats of the Chicago School, had not known that he (as she then was) had been such a prominent and electrifying figure in it. for Prof. McCloskey:

Journalists—even bright and thoughtful journalists like Michael Kinsley—frequently justify the taxation of capital income with an appeal to the principle that everything ought to be taxed equally.

…

The notion of a general principle that everything should be taxed equally is pure blather, of exactly the sort that McCloskey warned us against at every turn. And if there were such a principle, it would still be incumbent on us to figure out exactly what the principle entails before jumping to any conclusions.

This remark is both startling and perhaps overgenerous. For most writers on taxation do not even appeal to this principle, but just make the assumption that a good income tax system should tax all income at the same rate.Apart, of course, from the equally unspoken because so obviously correct assumption that larger amounts of income should be taxed at a higher rate than smaller ones. While the author may at times been guilty of erroneously making the former unstated assumption, at least he has never committed the latter sin. But it is not at all obviously correct and Landsburg, as he has done before, does a service by pointing to this unstated assumption.

Yet the author would argueAs he has done before. that this assumption, while when taken as a universal premise is a fallacy, is still a very good heuristic.

Consider an income tax code which for some reason (good or bad) decided to tax different forms of income, let’s call them red and green income, at rates lower for green than for red income. This would likely, but not certainly, be a bad decision for reasons both administrative and economic.

Administratively, the income tax system relies, as it must, on the tax-payer’s self-reports combined with occasional audits. Self-reporting means that the tax code must be written in such a way that it cannot plausibly be misinterpreted by a clever adversary.The tax-payer is clearly the adversary of the tax codifier and collector. And some tax-payers, or their tax lawyers, are decidedly clever. That is an enormously difficult task even for so basic a concept as income alone. Inevitably the following ensues:

1. The tax codifier writes what he considers to be a simple and straight-forward definition.
2. Either:
   1. Some persons whose transactions are unambiguously covered by the definition prevail upon the codifier that this is undesirable and the codifier adds an exception to the code.
   2. Some tax-payers discover that they can plausibly escape that definition by merely restructuring the form of their transactions while keeping their substance, either entirely or mostly, the same. Eventually, the codifier extends the definition so as to unambiguously cover those restructured transactions.
3. Go back to step 2 with a new set of tax-payers suffering from the extension or exploiting the exemption.

The enormous volume and complexity of the tax code is explained, not by the codifier’s malice or inexplicable love for complexity, but by endless repetitions of this cycle at ever-higher levels of complexity.

This is the unfortunate result of even a tax code that just has to define income. A tax code which also had to distinguish red income from green income would face the same problem at an even more severe level. The raging controversy over whether fund manager’s carried interest is or should be taxed as labor or capital income is a real-world example of this problem.

That is the administrative problem of red and green income tax rates. Perhaps, if tax codifiers’ and tax lawyers’ time was free,It is not. the administrative problem would be of no great concern. But that would still leave this economic problem:

Tax-payers who restructure their transactions so as to obtain a more favorable red/green classification will sometimes choose substantially less efficient forms. For the tax-payer, the economic loss of this choice can, on a post-tax basis, be more than offset by the more favorable classification. But from the standpoint of overall efficiency this is still a, possibly very large, dead-weight loss.

So Prof. Landsburg is right; one should not just assume that an income tax system with fewer different rates is necessarily better than one with more—it may very well not be. However, differential rates will always have administrative and economic cost; any such feature of the tax system should be examined very carefully to make sure that these costs are justified by greater benefits. Usually, they are not.