Dr. Garrison is professor of economics at Auburn University. He wishes to thank David Laband, Jim Long, and Leland Yeager for helpful comments.
In modern American politics, advocating a flat tax is the surest way of labeling yourself as a supply-sider, a Jack Kemp/Steve Forbes Republican. Michael Evans made the case for the flat tax in his Truth About Supply-Side Economics (1983); Robert Hall and Alvin Rabushka have made it twice in their book-length treatment of The Flat Tax (1985 and 1995).
Libertarians, many of whom get their economics from the Austrian school and eschew the Republican label, also tend to favor a single rate. In explaining Why Government Doesn’t Work (1995), Harry Browne offers a flat tax as part of the fix, but he devotes barely more than a page to this issue. The space he allocated to the flat tax as compared to the space allocated to it by the supply-siders, as well as his attention to the size of the tax take rather than the shape of the tax schedule, suggests a significant difference in priority and perspective.
“How Much” and “Just How”
The primary concern of the libertarians is with how much the government might tolerably extract from income earners and only secondarily with just how it is best (i.e., least painfully) extracted. Browne, for instance, suggests a 10 percent rate, which might raise as much as $500 billion. Barely one-third of the current total tax take, this amount is to finance the correspondingly pared-down expenditures of the federal government. Supply-siders, by contrast, deal with the just-how question as if it can be answered independently of the how-much question. The most common proposal, for instance, is for a revenue-neutral reform: We should scrap our current progressive tax, which we know to be hopelessly complex and inefficient, and adopt a simple and efficient flat tax that would yield the same—or nearly the same—revenue. Some supply-siders (e.g., Evans) would hold out for even more revenue.
Opponents of the flat tax can easily point to perceived social inequities, exaggerated claims, and outright fallacies that conventional supply-side arguments entail. A more defensible case for the flat tax is one that keeps the questions of How much? and Just how? in proper perspective: A flat rate may do little to make a big tax simpler or more efficient, but it may be a near-perfect device for keeping a small tax small. The key issues are (1) the actual incentives created by eliminating deductions in the pursuit of simplicity and (2) the political alliances created by incorporating a large personal exemption for the sake of voter appeal. A healthy consideration of these and related issues suggests that reducing the total tax take should have priority over imposing a single tax rate.
TANSTAA . . .
We owe to Robert Heinlein the memorable if nearly unpronounceable TANSTAAFL (there ain’t no such thing as a free lunch), which expresses one of the most fundamental principles in all of economics. Each major field of study within economics would do well to find its own Heinleinian acronym so as to keep policy prescription anchored to the basics. Let me propose a suitable one for the field of public finance: TANSTAABST. There ain’t no such thing as a big simple tax. Head taxes, the only truly simple taxes, are never big; income taxes, the primary source of revenue for the welfare state, are never simple. The claim, made repeatedly by supply-siders, that with a flat tax our tax form would be the size of a postcard can easily be exposed as bad science fiction.
The gains in simplicity are supposedly achieved by the elimination of deductions. Instead of multiplying our income (minus a myriad of deductions) by the effective tax rate, we multiply our income (minus a single personal exemption) by the flat rate. The multiplicand that the tax reformers have in mind, of course, is the income routinely reported on W-2 forms (or on 1099s and the like). For taxpayers in the post-reform period who continued to earn W-2 income, filing would indeed be simple. We should realize, however, that the W-2 form remains a tolerable means of reporting precisely because it is only the starting point for calculating taxable income. To eliminate deductions, which give the taxpayers scope for bargaining with the tax collector, is to eliminate the acceptability to the taxpayer of receiving income on a W-2 basis.
Even under the current system, there are strong incentives for avoiding the W-2. In many areas of the business world the conventional employer-employee relationship is being replaced by the firm’s contracting with individuals for services rendered. The elimination of deductions that would accompany the institution of a flat tax would undoubtedly accelerate this trend toward self-employment—which has been driven from the start largely by tax considerations. Under a contractual arrangement, the payment by the firm to the individual is not W-2 income but gross receipts. Income is to be calculated by the individual, with advice from his or her tax accountant, as receipts minus expenses. Even a wholesale elimination of deductions, then, would not achieve a dramatic simplification; it would simply shift the battleground on which taxpayers and the tax collector confront one another. Tax-avoidance strategies would aim at minimizing receipts minus expenses rather than minimizing income minus deductions. TANSTAABST. And the very open-endedness of what might reasonably be counted, in each line of business, as an expense would quite likely make the tax system more complex rather than less.
The extent to which the taxpayers would resist clipping their checks onto a postcard-size tax form is measured by the tax rate itself. Current levels of government spending would require a high rate. Special features of the supply-siders’ flat tax would reduce the tax base and make the rate higher still. One of these features, not strictly implied by—and actually at odds with—the concept of the flat tax, is the source of the widely perceived inequity: Interest income is to be treated as if it were not income. This special reward to savers and hence to high-income earners (since they are the ones who can most easily save) derives from the belief that it is actually consumption and not income per se that should serve as the basis for taxation. According to this view, people whose current demands for consumer goods are being satisfied should pay the taxes. The preferential status accorded savings—and hence investment and economic growth—is what justifies naming supply-side policies for their one-sidedness.
The pro-saving feature of the flat tax means that the portion of income not saved will have to be taxed at a higher rate than would otherwise be necessary. At the same time, it constitutes one—possibly significant—way for taxpayers to avoid the tax. Instead of awarding raises to its employees, a firm may well offer them the opportunity to buy low-risk, high-yield bonds, whose coupon payments (interest-in-lieu-of-wages) are not taxable. The tax collector would, no doubt, attempt to police this and other such tax avoidance schemes, but the process through which the market tries to arrange them and the government tries to curb them is unlikely to contribute to either simplicity or efficiency.
With a tax base that includes salaries and pensions plus business income, the tax rate that achieves revenue neutrality would have to be about 19 percent, according to Hall and Rabushka. During his bid for the Republican nomination for president, Steve Forbes proposed a 17 percent rate, calculated to give most taxpayers a small tax cut—and to give them all a higher budget deficit. But any rate in this range (i.e., 17-19 percent) is certainly high enough to sustain a reconstituted tax-avoidance industry.
The Flatly Progressive Tax
All of the Republican proposals involve a second departure from a strictly flat tax applied to all income, namely, a relatively high threshold level below which no taxes are collected. A substantial personal exemption (Forbes would have allowed for about $36,000 for a family of four; Hall and Rabushka suggest $25,000) has the effect of blurring the distinction between a flat tax and a progressive tax. As a matter of terminology, flat means not progressive. How, then, could Hall and Rabushka argue that one advantage to their flat tax is its progressivity? The so-called single rate is actually two rates: 0 percent for income up to $36,000, using Forbes’s proposal for illustration, and 17 percent for all income above $36,000. Calculating the average tax rate for incomes up to ten times the personal exemption, we get the progressive pattern that rises from 0 percent at the threshold level to 8.5 percent at twice that level to 15.3 percent at ten times that level and that thereafter approaches 17 percent asymptotically.
Supply-siders do not consider this progressivity objectionable at all. What they do find objectionable is an unnecessarily high marginal rate, such as our current top rate of about 40 percent (on taxable income over $250,000). Why, then, do they allow for substantial inframarginal incomes to go untaxed? This, too, causes the top marginal rate of 17 percent to be higher than it needs to be. That is, if a positive rate of, say, 8 percent were applied to some portion of income below $36,000, then a rate of, possibly, 15 percent could be applied to all income above that level. And the lower the top marginal rate, the stronger the standard supply-side arguments about increasing employment, exploiting the Laffer curve, and reducing the federal budget deficit through economic growth.
It seems clear that the generous personal exemption is included in the flat-tax proposals largely if not wholly for its voter appeal. The prospects of earning lots of tax-free income and enjoying a progressivity in the average rate on incomes well over the threshold level is attractive to the so-called middle class—which is to say, to the median voter. But, whatever the benefits of a flat tax, political attractiveness achieved in this way is a double-edged sword. Once such a tax system is in place, that same political attractiveness would attach itself to government spending in the minds of low- and medium-income voters. An overly generous personal exemption creates an alliance between these voters and elected officials in their efforts to gain economically and politically at the expense of the higher-income taxpayers. Government spending could have (gross) benefits for us all or could benefit mostly the poor while being paid for by the rich. This pattern of benefits and costs and resulting conflict among the differently situated taxpayers is precisely what any worthwhile tax reform would have to preclude.
Actually Achieving Simplicity
The advertised simplicity of a flat tax cannot be achieved by the elimination of deductions. As already suggested, determining what constitutes income would be, if not more complex, just as complex as determining what counts as a deduction. Further, the flatness of a flat tax does not translate into simplicity in any relevant sense. Progressivity in the sense of multiple brackets with stepwise increases in the marginal rates eliminates big jumps in the tax schedule while adding little or no computational complexity. Taxpayers who look up their tax liability on a suitably constructed tax table, like the cashier who looks up the sales tax on a similar table taped to the cash register, may not even notice whether the table was constructed on the basis of one rate, two rates, or ten rates. And while reasonable people could disagree about the relative merits of having a single rate or having ten, the merits of having just two, as entailed by a flat rate with a generous personal exemption, are dubious. People may prefer living in a one-story house rather than having to cope with stairs. But it doesn’t follow that a two-story house can be simplified by removing the staircase. Similarly, replacing the several small steps in the current progressive tax schedule with one giant step at $36,000 is not an obvious improvement.
If tax simplicity is achieved, it will be achieved not by the tax rate’s flatness but by its lowness. TANSTAABST. There ain’t no such thing as a big simple tax. But a small tax can be simple—and for a simple reason: If taxpayers find it easier and less costly to pay the tax than to redesign their economic lives so as to avoid paying it, the incentives for creating and exploiting complexities are effectively blunted. The resulting simplicity, of course, is not a goal unto itself but rather a healthy indicator that we have achieved the prerequisite goal of low taxes.
As Hall and Rabushka have emphasized, the tax rate can be its lowest if the tax is applied broadly, although they would apply it broadly to all consumption rather than (even more) broadly to all income. Salary income encourages working; interest income encourages saving. With a broad base that encompasses both, the disincentive effects of taxing are minimized. Our current tax system has a strong anti-saving bias; Japan’s tax system has a strong pro-saving bias. Neither bias has a justification in economic theory, and both biases cut into the tax base. There is a strong and obvious case for avoiding a bias in either direction while at the same time broadening the base. A low rate applied to a broad base lets income earners make their decisions about working and saving on the basis of the actual—non-tax-related—tradeoffs that these decisions entail.
A lower rate still is facilitated by the elimination—or minimization—of the personal exemption. (Browne allows for none.) The paring down of government expenditures provides a double-barreled justification for eliminating this exemption. First, the rate would be low, so as not to significantly burden even the low-income taxpayer. Second, since the services actually provided by government would be only those considered as essential to our well-being as other necessities that low-income taxpayers buy in the private sector, no taxpayer would be unduly burdened. Subjecting the low- and medium-income taxpayers to a tax burden proportional to the burden of the high-income taxpayers is in full compliance with both the letter and the spirit of a flat tax. Stringent voting rules need to be in place to guard against uncalled-for increases in the flat rate, and even more stringent rules—possibly at the constitutional level—are needed to assure that the flat rate remains flat.
Importantly, a universal application avoids the perverse political alliance with respect to government spending mentioned earlier. In fact, a constitutionally guaranteed flat rate creates a healthy alliance among income earners at all levels and against elected officials who, under other tax arrangements, could more easily gain political advantage through targeted government spending to be financed by selectively adjusting the tax rates. Taxpayer solidarity as a check against increased taxing and spending should be seen as the sine qua non of the case for a flat tax.
The Flat Tax in Perspective
Currently we have a big, complex, and inefficient, progressive tax. It is folly to think that the complex and inefficient derive significantly from the progressive. The complex and inefficient derive from the big. Given the efficiency and adaptability of the market, there is probably no knee-of-the-curve below which the tax take can be declared small. But 10 percent can be declared smaller than 17 percent, and, at any rate, opportunities for further reform still exist.
TANSTAABST. Revenue-neutral tax reform is no solution. The smaller the tax, the greater the prospects for simplicity and efficiency. And a flat rate may be the best means of keeping a small tax from becoming a big one.
4. Not long after his campaign book was distributed, Browne began advocating a complete abolition of the income tax and arguing that essential government services could be funded by (1) existing tariffs and excise taxes and (2) the proceeds from the sale of government assets, such as land holdings in the western states.
5. Individuals would pay taxes only on wages, salaries, and pensions; business firms would be allowed to fully expense investment, which is the present-value equivalent of allowing them to exempt the competitive yield on those investments. Taken together, these provisions, which have the effect of excluding interest income (or, equivalently, saved in come) from the tax base, convert the income tax to a consumption tax.
6. Critics of interventionist policies may be tempted to embrace supply-side theory as the antithesis of Keynesian theory, which focuses almost exclusively on demand. However, a critical assessment of both theories suggests a more balanced view: On analytical issues (How do markets work?), we should be both-siders: supply and demand. On policy issues (What kind of bias should be built into our tax system?), we should be neither-siders.
7. Robert E. Hall and Alvin Rabushka, Simplify, Simplify, in Edwin Mansfield, ed., Leading Economic Controversies of 1996 (New York: W. W. Norton and Company, 1996). Reprinted from the New York Times, February 8, 1995. See also, Hall and Rabushka, The Flat Tax, p. 55 and passim.