All Commentary
Sunday, December 1, 1996

The Flat Tax

How Would a Flat Tax Work?

To the Editor:

In his article “The Flat Tax: Simplicity Desimplified” (The Freeman, October 1996), Roger Garrison implies that those who favor the flat tax do not care about the size of the tax burden. Since the vast majority of flat-tax supporters are big advocates of lower taxes, and since all the major flat-tax proposals include a significant tax reduction, this claim is somewhat confusing. Moreover, evidence from the states shows that single-rate tax systems make it harder for states to raise taxes. As such, adoption of a flat tax presumably would impose limits on the growth of taxes on the federal level (primarily because politicians would have a harder time using divide-and-conquer tactics).

Garrison argues that the tax system cannot be simplified. Given that the flat tax eliminates all the most difficult and confusing aspects of the current system, this assertion is quite puzzling. No longer would individuals or businesses have to worry about capital gains, depreciation, estates and gifts, alternative minimum tax, foreign tax provisions, inventory accounting, phase-outs, itemized deductions, and so forth. It is certainly true that there is no free lunch, but there certainly are ways to reduce the cost of the lunch, and tax reform provides those efficiencies.

Garrison also claims that huge problems would be created as taxpayers reclassify W-2 income as business income in order to take advantage of business deductions. The incentive to play that game, however, depends on the tax rate. Since the tax rate will come down under a flat tax, there actually will be less income shifting.

Perhaps the most glaring error is Garrison’s claim that income from savings is not taxed under a flat tax. Even liberal economists admit that the core principle of the flat tax is to tax income only once. This means either taxing once when the income is first earned, but then leaving the returns alone (the Hall-Rabushka approach), or not taxing income that is saved, but taxing the interest and principal when spent (the IRA approach). Liberals admit that doing neither is double taxation, but justify it on pure income-redistribution grounds. It is difficult to imagine why anyone who believes in markets would adopt that position.

Garrison envisions a tax scam where employers give employees money to buy bonds, the interest to which would be nontaxable. He forgets, however, that the money provided to the employees under the flat tax would either be considered income to the worker (and thus taxable) or a fringe benefit (and thus taxable at the firm level). Either way the compensation is taxed (but not double taxed, since the interest properly is left alone).

Garrison believes that one rate has little to do with simplicity. In reality, one rate is critical if we want to tax all income at the source. One rate, for instance, allows us to tax AT&T one time on their income at the single rate of 17 percent. This is vastly preferable to tracking down all 2.2 million shareholders and imposing separate tax rates depending on their total income. The same thing with interest income. Not only will the single rate eliminate the one billion 1099 forms in the economy, it will also eliminate income shifting designed to have income declared to the low-tax person and deductions attached to the high-tax person.

All believers in limited government agree that the tax burden should be reduced to the maximum extent possible. This goal is not in conflict, however, with the idea of making sure whatever level of taxes is collected is taken in the least destructive, least intrusive manner possible.


—Daniel J. Mitchell
McKenna Senior Fellow
The Heritage Foundation Washington, D.C.

Roger Garrison replies:

Dan Mitchell’s challenging remarks, particularly his reference to a “glaring error” concerning the tax status of saved income, provide an opportunity for dealing with a common point of confusion. I take Robert Hall and Alvin Rabushka’s Flat Tax (2nd ed., Hoover Institution Press, 1995) to be ground zero for the modern resurgence of interest in simplifying our tax system. Hall and Rabushka leave little doubt about the status of saving in their proposed system:

Here is the logic of our system, stripped to the basics: We want to tax consumption . . . . We can measure consumption as income minus investment. A really simple tax would just have each firm pay tax on the total amount of income generated by the firm less that firm’s investment in plant and equipment. (p. 55)

Saving, then, which stands in contrast to consumption and underlies investment, is not taxed. Hall and Rabushka’s only significant departure from this “really simple tax” is one that exempts some consumption: the part of the firm’s (gross) income that is paid out in wages is untaxed until it is in the hands of wage earners, each of whom is allowed a generous personal exemption. This provision causes a substantial amount of consumption to go untaxed and gives a progressive character to average tax rates, but it does not bring saving into the tax base.

Given the ex post macroeconomic identity between saving and investment, Hall and Rabushka could hardly fail to recognize the nature of their proposal: “Our proposal is based squarely on the principle of consumption taxation. Saving is untaxed . . . .” (p. 54). Yet the authors themselves are at least partly responsible for the current confusion. At critical points, they misleadingly write “income” instead of “consumption,” and sometimes they write as if there were no difference between these two magnitudes. Although there are many such instances, I will cite just two: “The business tax . . . is carefully designed to tax every bit of income outside wages but to tax it only once” (p. 61). In fact, it is actually designed, as Hall and Rabushka had already stipulated, to tax only income net of investment, effectively converting the income tax to a tax on consumption. And virtually guaranteeing confusion, the authors explicitly affirm their “goal of taxing all income once at a common, low rate and achieving a broad consumption tax” (p. 63). This compound goal is simply at war with itself. Consumption and income are not the same thing; they differ precisely by the amount of income saved.

The confusion that has its roots in the original Hall and Rabushka proposal has caused Mr. Mitchell and undoubtedly others to see my exposition as involving a “glaring error.” Some supply-siders leverage the confusion by insisting that “consumed income” is, in fact, what “income” actually means. Others offer the all- too-facile claim (not supported by Hall and Rabushka’s basic logic) that saved income (or, alternatively, the yield on saved income) has already been taxed. These and other confusing claims stem from their using the rhetoric about taxing income once and only once in defense of a consumption tax. It is consumption, not income, that (beyond the generous personal exemptions) is taxed once and only once.

Several other points of disagreement raised by Mr. Mitchell are resolved once the tax status of investment (and hence saving) is established. For instance, there undoubtedly would be efforts in the private sector to disguise part of the (taxable) net income as (nontaxable) investment as well as efforts by the government to counter such attempts at tax avoidance.

Mr. Mitchell points to the vulnerability of our current system to the “divide-and-conquer tactics” of politicians trying to raise tax rates but fails to acknowledge that Hall and Rabushka’s generous personal exemption, which converts flatness into progressivity, would seriously weaken taxpayer solidarity and expose their proposed system to those same divide-and-conquer tactics.

Remaining differences between Mr. Mitchell’s views and my own are matters of perspective and judgment. Yes, some—maybe most—supply-siders would prefer tax reductions, but their willingness—even eagerness—to propose revenue-neutral or revenue-enhancing reform suggests their priorities lie elsewhere. And yes, given the complexities of the current system, there is plenty of room for reform in the direction of tax simplification. TANSTAAFL does not deny that some lunches are cheaper than others; TANSTAABST (There ain’t no such thing as a big simple tax) should be interpreted analogously. My arguments do suggest that a tax system involving (1) postcard-size tax forms and (2) the transferring of hundreds of billions of dollars from the private sector to the public sector is (not-so-unhappily) outside the realm of possibility.


Professor of Economics

Auburn University

We will print the most interesting and provocative letters we receive regarding Freeman articles and the issues they raise. Brevity is encouraged; longer letters may be edited because of space limitations. Send your letters to: The Freeman, FEE, 30 S. Broadway, Irvington-on-Hudson, New York; 10533; fax (914) 591-8910; E-mail: [email protected].

  • The Foundation for Economic Education, founded 1946, works for a free and prosperous world.