Taxation, A Preface

Economists specializing in public finance have long enumerated four objectives of tax policy: simplicity, efficiency, fairness, and revenue sufficiency. While these objectives are widely accepted, they often conflict, and different economists have different views of the appropriate balance among them.

Simplicity means that compliance by the taxpayer and enforcement by the revenue authorities be as easy as possible. Further, the ultimate tax liability should be certain. A tax whose amount is easily manipulated through decisions in the private marketplace (such as by investing in "tax shelters") can cause tremendous complexity for taxpayers, who attempt to reduce what they owe, and for revenue authorities, who attempt to maintain government receipts.

Efficiency means that taxation interferes as little as possible in the choices people make in the private marketplace. The tax law should not induce a businessman to invest in real estate instead of research and development—or vice versa. Further, tax policy should discourage work or investment, as opposed to leisure or consumption, as little as possible. Issues of efficiency arise from the fact that taxes always affect behavior. Taxing an activity (like earning a living) is similar to a price increase. With the tax in place, people will typically buy less of a good—or partake in less of an activity—than they would in the absence of the tax.

The most efficient tax system possible is one that few low-income people would want. That superefficient tax is a head tax, a tax on each person that is not affected by that person's income or by any of the person's characteristics. A head tax would not reduce the incentive to work, save, or invest. The problem with such a tax is that it would take the same amount from a high-income person as from a low-income person. It could even take the entire income of low-income people. Within the realm of what is practical, the goal of efficiency is to minimize the ways that taxes affect people's choices.

A major philosophical issue among economists is whether tax policy should purposefully deviate from efficiency in order to encourage taxpayers to pursue positive economic objectives (such as saving) or to avoid harmful economic activities (such as smoking). Most economists would accept some role for taxation in so steering economic choices, but economists disagree on two important points: how well policymakers can presume to know which objectives we should pursue (is discouraging smoking an infringement on personal freedom?), and the extent of our ability to influence taxpayer choices without unwanted side effects (will subsidies for saving merely reward those with the most discretionary income for saving no more than they would have without a subsidy?).

Fairness, to most people, requires that equally situated taxpayers pay equal taxes ("horizontal equity") and that better-off taxpayers pay more tax ("vertical equity"). Although these objectives seem clear enough, fairness is very much in the eye of the beholder. There is little agreement over how to judge whether two taxpayers are equally situated. For example, one taxpayer might receive income from labor while another receives the same income from inherited wealth. And even if one taxpayer is clearly better off than another, there is little agreement about how much more the better-off person should pay. Most people believe that fairness dictates that taxes be "progressive," meaning that higher-income taxpayers not only pay more, but proportionately more. However, a significant minority takes the position that taxes should be flat, with everyone paying the same proportion of their taxable incomes. Moreover, the idea of vertical equity (i.e., the "proper" amount of progressivity) often directly contradicts another notion of fairness, the "benefit principle." According to this principle those who benefit more from the operations of government should pay more tax.

Revenue sufficiency might seem a fairly obvious criterion of tax policy. Yet the federal government's budget has been in enormous deficit for more than ten years. Part of the reason for the deficit is that revenue sufficiency may conflict with efficiency and with fairness. Economists who believe that income taxes strongly reduce incentives to work or save, and economists who believe that typical families already are unfairly burdened by heavy taxes, might resist tax increases that would move the federal budget toward balance.

Likewise, other objectives of tax policy conflict with one another. High tax rates for upper-income households are inefficient but are judged by some to make the tax system fairer. Intricate legal provisions to prevent tax sheltering and thus make taxes fairer would also make them more complex. Such conflicts among policy objectives are a constant constraint on the making of tax policy.

At the federal level total tax collections have hovered in a narrow range around 19 percent of the gross national product (GNP) since the end of the Korean War (see table 1). The individual income tax has provided just under half of that revenue over the entire period. The corporation income tax was the source of almost a third of total revenue at the beginning of the period but has declined dramatically to under 10 percent today. In mirror image the payroll tax for Social Security began at about 10 percent of total revenue, but it increased sharply to over 37 percent as the elderly population and inflation-adjusted Social Security benefits grew and as the Medicare program was added to the system. The relative contribution of excise taxes (primarily on alcohol, tobacco, gasoline, and telephone services) has declined significantly.

---

TABLE 1
Federal Tax Revenues by Type of Tax—Percents of GNP, Fiscal Years
Year	Individual Income	Corporate Income	Social Security	Excise	Other	Total
1954	8.0%	5.7%	2.0%	2.7%	0.5%	18.9%
1959	7.6%	3.6%	2.4%	2.2%	0.7%	16.5%
1964	7.7%	3.7%	3.5%	2.2%	0.8%	17.9%
1969	9.4%	3.9%	4.2%	1.6%	1.0%	20.1%
1974	8.4%	2.7%	5.3%	1.2%	1.0%	18.6%
1979	8.9%	2.7%	5.7%	0.8%	0.8%	18.9%
1984	8.1%	1.5%	6.5%	1.0%	1.0%	18.1%
1989	8.7%	2.0%	7.0%	0.7%	0.9%	19.3%
1991	8.8%	1.7%	7.2%	0.8%	0.9%	19.4%
SOURCE: Office of Management and Budget. 1991 figure is estimated.

One little-recognized aspect of the development of federal taxes is the gradual decline of revenues other than those earmarked for the Social Security and Medicare programs. Although total federal taxes are a roughly constant percentage of GNP, the Social Security payroll tax has increased significantly while other taxes have been cut in approximately equal measure. The result has been that federal revenues available for programs other than Social Security and Medicare have been squeezed—from about 17 percent of GNP in 1954 to about 12 percent in 1991.

States rely primarily on sales taxes but increasingly on income taxes. Local governments rely most heavily on property taxes. Contrary to what many believe, the explosion in taxation has been in state and local taxes. Unlike federal taxes, state and local taxes have increased significantly—from about 7 percent of GNP in 1954 to about 12 percent in 1991 (see table 2).

TABLE 2
State and Local Tax Revenues by Type of Tax—Percents of GNP
Year	Individual Income	Corporate Income	Property	Sales	Other	Total
1954	0.3%	0.2%	2.6%	1.7%	2.1%	7.0%
1959	0.4%	0.2%	3.0%	2.2%	2.1%	8.0%
1964	0.6%	0.3%	3.3%	2.5%	2.4%	9.1%
1969	1.0%	0.4%	3.4%	3.0%	2.7%	10.4%
1974	1.4%	0.5%	3.3%	3.3%	3.0%	11.5%
1979	1.5%	0.5%	2.6%	3.1%	3.2%	11.0%
1984	1.8%	0.5%	2.6%	3.2%	3.7%	11.8%
1989	2.0%	0.5%	2.7%	3.3%	3.7%	12.1%
SOURCE: Department of Commerce, Bureau of Economic Analysis. Data are not comparable to those in table 1.

Thus, although the level of federal taxes has been relatively constant for nearly thirty years, total taxes have increased because state and local taxes have increased. (The data in tables 1 and 2 are computed on different accounting procedures and years and thus cannot be added together; the general picture that they suggest is, however, accurate.) The increase in state and local taxes has obviously added to the taxpayers' burden and has limited the federal government's ability to cut the federal deficit and to increase spending.

Much of the recent interest in tax policy has focused on the federal individual and corporate income taxes. Advocates of "supply-side economics" (most prominently, Arthur Laffer) believed that income taxes had severely blunted incentives to work, save, and invest and that the income tax burden had become excessive. The income tax became a major issue in the 1980 presidential election, and Congress passed a substantial income tax cut in 1981. It provided for a cumulative across-the-board cut of 23 percent in the income tax rate, phased in over four years, along with significant tax inducements for business investment. In the face of a rapidly rising budget deficit, some of the investment incentives were repealed a year later.

An even more radical tax restructuring was passed in 1986. This new law, like the 1981 law, also significantly reduced income tax rates. It was, however, radically different from the 1981 tax cuts in a more meaningful sense, in that all of the tax rate cuts were "paid for" by the elimination of tax incentives—including the remaining business investment inducements from 1981. This tax "reform" made U.S. corporate and individual income tax rates the lowest in the industrialized world. It simplified the tax law in some respects but also included complicated provisions designed to prevent tax sheltering. The new law also provided significant tax relief for low-income taxpayers, especially families with children. And it transferred about $25 billion a year of the tax burden from the personal to the corporate income tax.

Tax scholars have observed the experience of the eighties closely to learn more about how taxes affect economic choices. While much controversy remains, certain results seem clear. First, as many economists expected, the reduction of tax rates in the eighties apparently did induce greater work effort, especially by married women. In 1988, according to Brookings economists Barry Bosworth and Gary Burtless, men between the ages of 25 and 64 worked 5.2 percent more hours than they would have under the pre1981 tax code; women age 25 to 64 worked 5.8 percent more; and married women worked 8.8 percent more. These increased hours would translate into the equivalent of almost 5 million full-time jobs.

Second, household saving fell in the face of tax rate cuts and substantial targeted tax incentives for saving, strongly suggesting that taxes have a limited impact at best on saving. Studies by economists Steven F. Venti and David Wise have suggested that individual retirement accounts (IRAs) were successful in encouraging new saving, but another study by William Gale and John Karl Scholz indicated that much of the IRA deposits came from households that had already accumulated considerable wealth and could simply transfer it into the tax-favored accounts. And finally, while business investment did increase after the 1981-82 recession (as documented by Harvard's Martin Feldstein), other economists (notably Barry Bosworth of Brookings) argue that this increase came primarily in assets (such as computers) that were not highly favored by the tax law. In fact, investment in equipment increased to a record percentage of GNP after the incentives were repealed in the 1986 tax reform, though Alan Auerbach and Kevin Hassett have argued that it would have increased even more strongly if investment incentives had been continued.

Many economists judge the fairness of the tax system largely on how the tax burden is distributed among different income groups. Further, some economists used the distribution of the tax burden as a major criterion of the success or failure of the tax changes of the eighties. Despite considerable effort and innovative methods, however, estimates of the distribution of the tax burden are still limited by imperfect data and the differing perspectives of investigators.

Between 1980 and 1990 the percentage of income paid in federal tax increased for the 60 percent of families with the lowest incomes taken as a group, and decreased for the 40 percent with the highest incomes (see table 3). The decreases were largest for the families with the very highest incomes. Some economists have used these figures to claim that the changes of the eighties left the tax burden distributed less fairly. In fact, the 1986 tax reform taken by itself had the opposite effect when assessed by this methodology, but was outweighed by the 1981 tax cuts.

TABLE 3
Federal Taxes as a Percentage of Family Income, 1980 and 1990
Quintile of Families Ranked by Income	1980	1990	Change
Lowest	8.4%	9.7%	1.3%
Second	15.7%	16.7%	1.0%
Third	20.0%	20.3%	0.3%
Fourth	23.0%	22.5%	-0.5%
Highest	27.3%	25.8%	-1.5%
Top 10 percent	28.4%	26.4%	-2.0%
Top 5 percent	29.5%	26.7%	-2.8%
Top 1 percent	31.8%	27.2%	-4.6%
ALL	23.3%	23.0%	-0.3%
SOURCE: Congressional Budget Office.

These figures are subject to challenge, however. They assume that half of the corporate income tax is borne by owners of business capital, and half by workers in the form of lower wages, and that the employer's share of the Social Security payroll tax is borne by workers, also through lower wages. If you assume that owners of businesses pay all the corporate income tax and the employer's share of Social Security, the tax burden would rest further up the income ladder. Also, although upper-income families paid a smaller percentage of their income in tax in 1990 than they did in 1980, they received a much larger share of total taxable income by the end of the decade. One reason the taxable income of upper-income families is higher is that changes in the tax law, particularly in 1986, caused many upper-income families to reallocate their portfolios from things like municipal bonds to assets that yield taxable income. But there also is evidence that the distribution of income simply became less equal. The net result is that upper-income families now pay a larger share of the total tax burden.

Tax policy remains controversial, and some economists continue to argue for large-scale revision of the federal tax system. Some convervative economists, such as Charles McLure, and some liberals, such as Alice Rivlin, would like to see a broadbased federal tax on consumption, like the sales taxes imposed by the states or the value-added tax (VAT) widely used in Europe. The proceeds of the tax could be used to increase federal spending, to cut federal income taxes, or to reduce the deficit. Advocates argue that a tax on consumption would encourage saving; opponents claim that such a tax would unfairly burden low-income families.

Many economists, including Princeton's Alan Blinder, believe that the income tax should provide a comprehensive adjustment ("indexation") for inflation to eliminate the inflationary component of interest income and expense, depreciation of business investment, and capital gains. The case of capital gains is the clearest. A block of stock bought for $1,000 in 1978 and sold for $2,000 today would yield a $1,000 taxable capital gain. But the investor would have received no real increase in purchasing power because the price level has roughly doubled since then. The same problem, however, afflicts recipients of interest income. With inflation at 4 percent, only half of the interest on a bond yielding 8 percent is real income to the bondholder, but all of the interest is subject to income tax. An adjustment for inflation would be quite complex; it was considered and rejected for that reason in the debate on the 1986 tax reform. The merit of indexing for inflation might depend on the rate of inflation in the economy. In hyperinflation, indexation is essential. But at low rates of inflation—perhaps as high as 10 percent—the markets can offer higher interest rates to compensate lenders (and penalize borrowers) for the inexact taxation of their interest income (and deductions of business interest expense).

Some economists, including Martin Feldstein, argue for a targeted tax cut for capital gains (the profit from the sale of assets like corporate stock or real estate), or for reinstatement of targeted incentives for household saving (like the deductibility for all taxpayers of contributions to individual retirement accounts, or IRAs). Such initiatives are typically claimed to increase economic growth. Opponents, such as Brookings' Henry Aaron, believe that they would be ineffective and that they would unduly benefit upper-income groups who own the most capital assets and have the most discretionary income to save.

Finally, some economists (such as Robert Shapiro) and policymakers (most notably, Sen. Daniel Patrick Moynihan) believe that the Social Security payroll tax should be cut significantly. They argue that the growth of the payroll tax has been a major contributor to the shift in the tax burden toward lower-income families. Such a tax cut, however, would be intricately tied to the structure of the entire Social Security system and is accordingly extremely controversial.