The Perfect Tax

How the dismal science applies to your life.
Oct. 11 1996 3:30 AM

The Perfect Tax

If there must be taxes, at least they should be inefficient.

The expansion of government is limited by the consent of the governed. Once upon a time, that consent was harder to come by. In 1776, American colonists took up arms against a government far less oppressive than the one that now spends 40 percent of our incomes. We're more docile now, but the threat of revolution--or at least mass dissatisfaction--remains an important force for good. We can harness that force by designing laws and institutions that require each new government burden to be widely shared. Politicians are less likely to risk annoying 60 percent of the electorate than 10 percent.

Illustration by Robert Neubecker
Advertisement

That wisdom underlies the "takings clause" in the Fifth Amendment to the Constitution, which says, in effect, that the government can't take your front lawn and turn it into a park without paying you for it. The takings clause forces taxpayers to share in the cost of any taking, and so ensures that frivolous takings will meet broad opposition. Inspired by the same logic, I propose a constitutional amendment capping everyone's tax bill at (say) five times the average. Thus, if the average American pays $10,000 in taxes some year, no American could be required to pay more than $50,000 that year. That would force many taxpayers to share the cost of any new government spending, and so ensure broad opposition to the growth of government.

Illustration by Robert Neubecker

It's traditional to evaluate tax proposals according to the twin standards of "equity" and "efficiency." Before I subject my own proposal to those standards, let's talk a bit about the standards themselves.

A tax system is "inefficient" when it discourages beneficial economic activities. By discouraging working, saving, and investment, the United States tax code--like any tax system--meets that standard with ease. The only completely efficient tax would be one that does not depend on anything the taxpayer can control--such as a "head tax" of $5,000 per year. If your behavior can't affect your tax bill, your tax bill won't affect your behavior.

That kind of head tax is interesting not as a realistic policy proposal but as a benchmark for comparison: Economists like to say that a head tax has the advantage of being perfectly efficient (it does not discourage any economic activity), but the disadvantage of being perfectly inequitable (the rich and the poor pay equal amounts). But those economists are wrong on both counts.

Let's talk first about equity. It is an act of violence against the English language to describe as "inequitable" a tax that charges everyone an equal amount. In the rhetoric of tax policy, the word "inequitable" almost never means "inequitable"; rather, it means something like "less redistributionist than the speaker would prefer."

Illustration by Robert Neubecker

But I don't want to dismiss a substantive concern just because people frequently choose the wrong word to describe it. Instead, let me make a substantive response: Even if you believe that the tax system should be used to iron out income differentials, it's still perfectly easy to devise a head tax that is consistent with your redistributionist philosophy. The key is to make taxes dependent on variables that are good predictors of income but entirely outside the taxpayer's control. For example, whites (on average) earn more than blacks do. A direct tax on income might discourage work. But taxing whiteness would not discourage anything, while still redistributing income (on average) from the relatively rich to the relatively poor. A tax of "$10,000 per year if you're white and $5,000 per year if you're black" is a perfectly efficient (and highly progressive) head tax.

On similar grounds, we could tax people for being male or tall or beautiful; all these traits are positively correlated with income. In the case of beauty, though, we'd have to be careful to tax only natural beauty. Otherwise, we'd discourage expenditure on shampoo, cosmetics, and dentistry. (In fact, if we enjoy having beautiful neighbors, we might want to subsidize beauty rather than tax it; you can't always pursue two goals--in this case income redistribution and an attractive population--simultaneously.) Regardless of what you mean by "equity," therefore, you can always adjust the head-tax system to conform to your philosophy. Thus "inequity" (or, more accurately, "insufficient redistributive power") is not a good argument against head taxes.

But just as the alleged inequity of head taxes is no vice, so may their vaunted efficiency be no virtue. The problem with an efficient tax system is that it provides no built-in brake on the government's avarice. A government that takes $10,000 from everybody this year could decide to take $20,000 or $30,000 next year. That is much less likely under an inefficient tax system. The income tax, for example, is gloriously inefficient. The higher the tax rate, the less people work. If the government raised the income-tax rate to 100 percent, we'd all stop working, and tax revenue would be zero. To convince us to earn an income worth taxing, the government is forced to let us keep a substantial fraction of what we earn.

Illustration by Robert Neubecker

The challenge is to make taxes more efficient without making them too easy to raise. This brings me back to my proposed constitutional amendment: capping individual taxes and tying the cap to the average tax bill. Adding a cap to the current system would improve efficiency (there would be no disincentive to earn income beyond a certain level), while maintaining the natural safeguards against confiscatory government that are built into the income tax. In fact, those safeguards would be strengthened by the fact that any tax increase would have to be quite broad-based. As a bonus, a cap would bring the income tax closer to being equitable, in the true sense of the word. The offsetting disadvantages? None, as far as I can see.

Steven E. Landsburg, author of The Armchair Economist: Economics and Everyday Life, is a professor of economics at the University of Rochester.