tax policy center
publications
HOME | TAX TOPICS | NUMBERS | TAX FACTS | LIBRARY | BRIEFING BOOK | EVENTS | LEGISLATION | PRESS | TAXVOX Blog | About Us help get RSS feed

Advanced Search

by Topic:

by Author:

by Type:

by Date Range:
  From last wks

     

library

And Equal (Tax) Justice for All?

C. Eugene Steuerle

Published: November 01, 2002
Availability:
 PDF |  Printer-Friendly Version

Share:  Share on Facebook Share on Twitter Share on LinkedIn Share on Yahoo Buzz Share on Digg Share on Reddit

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

This essay was originally published in Tax Justice: The Ongoing Debate (2002, Urban Institute Press), edited by Joseph J. Thorndike and Dennis J. Ventry Jr. Click here for more information on this title.

This publication is also available in PDF.


Concepts of equity and fairness are at the heart of tax policy. Political leaders pay homage to these ideals in virtually every sphere of lawmaking and regulation. Citizens, moreover, are keenly sensitive to arguments about fairness in almost every policy debate.

Yet, for all its populist appeal, tax equity is loosely understood and inconsistently applied. This concept comprises at least three distinct dimensions: horizontal, vertical, and individual equity. The very real tension among these components complicates efforts to craft "fair" tax policy, prompting political debates that invoke the rhetoric of equity without engaging its substance. The most vociferous arguments have centered on the uneasy relationship between vertical equity—most commonly manifested in progressive tax and expenditure structures— and the demands of individual equity, in which individuals freely engage in transactions of their own choosing.While support for some application of vertical equity seems clear, determining the appropriate degree of progressivity has proved to be difficult.

Various details further complicate efforts to apply tax equity in a rigorous and productive manner. Progressivity remains a touchstone in debates over fiscal policy, yet it often means one thing when applied to tax and something very different in discussions of spending. Moreover, crafting fair tax policy is further confused by discussions about the appropriate tax base—income, expenditure, or something else—as well as the many adjustments that can be made to any of these bases. An adjusted tax base often represents an attempt to define equality and inequality along some measure of net well-being.

In the face of all this controversy and complexity, many theorists have thrown up their hands. Many economists, in particular, have often dismissed fairness as more an issue of aesthetics than of analysis. A large contingent has not merely focused on questions of efficiency but largely abdicated its role in debates over tax equity. In doing so, however, these researchers have ceded the ground once dominated by their predecessors—from Adam Smith in the 18th century to Richard Musgrave in the 20th century. The abandonment of tax equity is misguided, stemming from a misunderstanding of the contributions that economists are uniquely prepared to make. Even assuming that answers to fairness questions are more political than mathematical, economists have a vital role to play. They may not always be able to pinpoint the "right" answer in terms of fairness, but they can frequently identify approaches that more consistently and efficiently accommodate the public's demand for equity—a demand as legitimate as its demand for other public services.

Ultimately, tax equity—for all its complexity and undeniably political nature—remains too important an issue for economists to simply forfeit on or to ignore when creating solutions to difficult public finance issues.

The Universal Demand for Equity

Nothing is more fundamental to the character of a successful democracy than its citizens' trust in judicial and legislative processes that protect basic human rights and provide equal justice under the law. These rights are powerful notions—ones for which people have freely given of their lives and property. From the Magna Carta to the U.S. Constitution to the French Declaration of the Rights of Man, one can trace continual efforts to claim for each citizen—defined in broader and more inclusive terms over time—his or her fair due. Many declarations and conventions, such as the United Nations' 1948 Universal Declaration of Human Rights, assert that certain rights are derived from the human condition; they are not merely temporary, national, or cultural preferences.

Efforts to assert equal justice extend beyond these lofty documents, reaching into the everyday world of politics and policy. Indeed, it is hard to identify any major governmental action—legislative, regulatory, or symbolic—in which fairness is not a dominant concern. Consider, for instance, the names attached to legislation. How often do words like equity, opportunity, and fairness appear? Consider, also, how even regulatory processes are designed to allow anyone claiming unfair treatment access to an almost judicial consideration of their claim. How often are regulations delayed or even reversed because of such considerations? Furthermore, consider the use of symbols in a public setting. What authority can, as a matter of equity, deny a statue to some excluded group—veterans of a forgotten war, women who served in war as well as men, black war heroes as well as white ones—in a park already cluttered with statues?

Equity's status as a political principle is unique, but it is not always the driving force behind action. Other objectives—efficiency, growth, simplicity—often take precedence. Emergencies demand attention and sometimes require equity shortcuts. Still, other objectives are almost never pursued without due attention to the equitable distribution of burdens and benefits. In formulating education policy, for example, government officials may want to subsidize education to promote a strong and growing society. Educational programs, however, will inevitably feature significant redistributive elements. Officials will offer arguments rooted in equality of opportunity—somehow defined—as a standard for designing the program. "Equal access to education" or some other catchphrase will permeate debates over proposed subsidies.

Not every principle receives such homage. For example, compare simplicity with equity. Simplicity routinely falls by the wayside in tax bills designed to make some final outcome more progressive or more evenly distributed among those in similar circumstances. A tax simplification bill, on the other hand, cannot veer too far toward inequity before losing political support. An arbitrarily applied tax might be easy to administer, and, if simplification were the only goal, a random tax might be the most efficient way to achieve the objective, but such taxes cannot be enacted.

That tax and expenditure laws are subject to demands for fairness and justice may sound strange. These laws, after all, are widely considered some of the principal "honey pots" of policymaking. Tax and expenditure systems are vehicles for much of the logrolling endemic to democratic legislatures. Even legislation historically recognized as "fair" often includes questionable riders attached to buy winning votes. Indeed, many such riders fail to meet a standard of equal justice.

Examine, however, the arguments put forward to support even the poorest legislation. Instead of attacking the equity standard, lobbyists frequently try instead to twist it for their own purposes. Rather than arguing for special treatment, they assert that some source of inequity justifies their demand for compensation. Such groups might suggest, for example, that past Congresses intended for them to receive some benefit and therefore that the law should now be recrafted to their benefit. In turn, wherever existing legislation draws a line, these lobbyists seek to secure their place on the side that receives more expenditures or pays less tax. Their arguments usually feature examples in which some individual or group in an almost identical situation receives more favorable treatment.

In arguments over fairness, the ground constantly shifts underfoot; raise one standard of equity and another will be cited in its stead. Equal treatment of those with equal incomes may be fine, but advocates for veterans, members of groups that have traditionally suffered from past discrimination in the tax or spending laws, persons in poor health, and individuals who simply start out with less opportunity to acquire income will reject equal income as an adequate measure of equality. The equity standard itself is not being rejected, but its application.

The demand for equity and equal treatment is so natural we often take it for granted. C. S. Lewis, a theologian known best for his children's tales, once suggested that, as a matter of natural law, we humans always appeal to some standard of behavior. We justify our behavior, whatever it is, so we can live with our consciences. In dealings between individuals, Lewis asserts, "It looks, in fact, very much as if both parties had in mind some kind of Law or Rule of fair play." In effect,we justify even our worst behavior as a matter of equity (Lewis 1943, 17).1 Similarly, individuals reject arbitrarily being treated differently, whether by a family member, stranger, or an elected representative.

Under a rule of law, all government actions must be lawful.Under such a system, however, lawfulness and justice are synonymous. Thus, the individual demand for justice, identified by Lewis, is enhanced by a societal demand in legal matters. Arbitrariness in the law is unacceptable. The most ignoble client must still be represented. Remember the classic legal joke about the defense for stealing a pot: "I never stole it; it was broken, and, besides, it was worthless in the first place." This kind of argument has at its core an equity claim—that "I" should be exonerated or charged with nothing because that would be the only fair outcome, one way or another. Tax and expenditure laws, too,must always appear to meet justice standards.

Equity, then, is the first and most basic set of principles applied to constitutions and laws. While constitutions and courts require "equal justice under the law," they do not require greater efficiency or simplicity. No other standard reaches the lofty status of equal justice in the affairs of government or the souls of humans.While conflicts abound, they are much more likely to arise over how to apply the principle consistently, how to measure who are equals, and the extent to which compensation or special consideration should be applied to those who are different along some scale of fortune, need, or ability.

Horizontal and Vertical Equity

Even if equity were not the highest of principles in both lawmaking and administration, it undoubtedly plays a dominant role in the debate over the allocation of taxes and expenditures. Beyond a broad level of abstraction, however, equity must be defined in practical ways in order to allocate these budgetary obligations and rights. Equity and equal justice do not mean equality in all things. In matters of the state, these concepts refer to the way that government will treat us, not our starting point. Since no two people are exactly alike or equal in all things, the task of maintaining equity under the law is far from dull. Indeed, the complex political undertaking of applying equity principles to practical situations has occupied philosophers and government officials since civilization began. Disagreements abound, generally centering on questions about who should be treated as equals, who should not, and what to do about those differences.

Here the public finance literature, particularly as explained by Richard and Peggy Musgrave, is extraordinarily useful in the way it distinguishes between horizontal and vertical equity.2 Horizontal equity refers to the treatment of equals, vertical equity to adjustments made among nonequals. If income were the only measure of a person, for example, then horizontal equity demands that two persons with equal incomes be treated as equals. Alternatively, vertical equity is based on the premise that someone with little or no income will have difficulty paying the same amount of income tax as someone who is rich.

The easiest and often most useful way to conceptualize horizontal and vertical equity is to imagine people along a particular scale. A classic scale used in the tax literature (but interestingly enough, less in the expenditure literature) is ability—originally thought of in terms of "faculty" as measured by property, but more recently conceived as being measured by income.3 Thus,we speak of taxing people according to their ability to pay.

Horizontal equity requires that those with equal status—whether measured by ability or some other appropriate scale—should be treated the same. They should pay the same amount of tax and receive the same amount of benefits. As a consequence, those who start out as equals before any governmental action would end up as equals after the government acted. For example, suppose equals are defined by ability, which, in turn, is defined by income. In that case, those starting with the same before-tax income should end up with the same after-tax income.4

Vertical equity, for its part, generally requires that those with less ability be treated favorably relative to those with greater ability. Progressivity is often considered synonymous with vertical equity, but even economists trained in the literature of public finance do not apply the term consistently.

Some theorists have argued that horizontal and vertical equity are different sides of the same coin.5 In other words, the contention that those who have less should therefore pay less and receive more reflects the same concern that those with equal status be treated equally. This belief derives partly from a concept of equity whereby people are placed on a single scale so that the appropriate tax or expenditure is a simple function of what is measured on that scale. For example, if taxes are a positive function of income and income only, then both horizontal and vertical equity fall out of the same functional form.6

Despite this functional relationship under one theoretical (almost mathematical) approach to understanding equity, I assert that horizontal and vertical equity should not be viewed as two corollaries of the same principle. Many people strongly support horizontal equity even though they reject the notion that government must adjust the status of any individual along a particular scale. Such a position is not inconsistent.

Examples abound of the application of horizontal equity to government programs: Those individuals with equal incomes should be made to pay equal income taxes. Consumers who purchase the same items at the same price should pay the same sales tax. People living at equal levels of poverty and having equal need should be entitled—at least within their jurisdiction—to equal amounts of food stamps. Indeed, the importance of horizontal equity reaches beyond the realm of economics. Equal crimes should be made to bear the same punishment. Various chemical companies should all be subject to the same environmental limitations on chemicals they can use for agricultural purposes. Automobile companies should face the same limitations on pollution. Each citizen should have an equal right to vote, and so on.

Note, however, that many of these laws are not intended to redistribute or achieve some vertical equity standard. Some, like pollution controls or sales taxes, might be adopted for different reasons and be regressive in their distributional effect. But the demand for equal justice does not go away. Thus, while it is possible to identify government programs that do not aim at and ever contradict some vertical equity goal, identifying one that does not apply some horizontal equity standard, however imperfectly, is virtually impossible. And when equally situated individuals appear to be treated unequally under one definition, it is often because alternative definitions of equality have cast these individuals as unequal.

From one end of the political spectrum to the other, horizontal equity is a universally accepted principle. There are no sides, no divisions between conservatives or liberals, and no conflicts between advocates of big or small government. If a person can prove that he or she is just like another, then no one will likely challenge his or her case for equal treatment under the law. In some ways, horizontal equity is almost tautological: If people are defined as equals, how can government treat them differently, thereby making them unequal? Horizontal equity is a basic application of the broad societal commitment to equal justice.

When it comes to vertical equity and progressivity, on the other hand, agreement quickly breaks down. Economist Herb Stein made this point during 1959 congressional hearings on tax reform, pointing out that while horizontal equity is the "the first, basic rule of taxation," considerations of vertical equity seem to amount to value judgments, questions of degree, and subjectivity. "If A's income is twice B's," Stein asked rhetorically, "should A's tax be twice B's, or one and one-half times or three times as large? . . . Intuitive standards of equity seem to throw no light on questions of 'How much?'" (1959, 110, 114). Conservative economist Harley Lutz condemned the subjectivity of vertical equity. "There is no just or progressive tax rates scale," he insisted in his Guideposts to a Free Economy. "Every such scale is the product of guesswork and of political and fiscal expediency.And where expediency is the basis of policy, it is easy to lapse into injustice" (1945, 70, 82). Every person has a particular notion of how progressive government should be, how much it should be involved in assessing different amounts of tax on individuals with different means, or helping differentially those with different needs. Whatever the degree of subjectivity, at one level or another the debate over progressivity has often dominated public debate and even led to the toppling of governments. It was a major issue in the debate between socialism and capitalism, while progressive treatment of the poor and impaired is used continually to assess the success of governments in developed and developing nations alike.

The attack on vertical equity often goes too far. In the historical debate over vertical equity, almost no one argued that the poor should pay more than the rich. This consistency implies that even those who argued that progressivity was a fluid, subjective standard at some level still accepted it as a requirement. For example, every example given by Stein involved a larger tax on the richer person than on the poorer person.

When governments try to rectify some real or perceived vertical inequity, some amount of redistribution is almost inevitable. This redistribution may come as a higher tax on a richer person or result in some sort of transfer to the poorer person—in cash, in-kind services, or access to opportunities, such as education. In distributing taxes and expenditures—and regulatory requirements as well—there are usually those who pay and those who receive. In both cases, government shapes the lives of individuals. No matter how necessary or valid such action may be, therefore, it often sparks controversy and conflict. After all, any interference involves costs that, in turn, demand justification.

Vertical Equity versus Individual Equity

Vertical equity connects the rights and obligations of individuals to their ability and well-being. While often applied formally to government finance and expenditure programs, a similar standard applies in the ordinary affairs of the family and community. Within the family, for example, those who can work are expected to contribute more financially than those who cannot. Dependents often have obligations, but fewer than those of working-age adults with greater maturity and ability. Thus, common sense, not just philosophy, leads us to accept vertical equity as a general principle. Vertical equity often competes head-on with the principle of individual equity, which emphasizes each individual's freedom to partake in transactions without interference by third parties, including the government. In general, each voluntary transaction involves an exchange between two individuals, each of whom expects to be better-off as a result. Government interference in that transaction diminishes the gains for at least one individual and often the total gains to be shared. Moreover, government intervention can distort the nature of the transaction, perhaps even deter it altogether.

In practice, governmental efforts to promote vertical equity usually involve taking from an individual on the basis of a transaction, such as the sale of labor, the employment of capital, or the purchase of some good or service. Consequently, tax issues (along with regulation) are among the most likely sources of claims that government has violated individual equity.

At times it is possible to tax in a way that either does not violate the principle of individual equity or at least minimizes the extent of the violation. In the former case, the transaction between the individual and the state is voluntary, and the individual would not pay unless he received a benefit worth the price paid. In the latter case, the benefits of government action are designed to be approximately equal to, or closely related to, the taxes or contributions made.

The public finance literature distinguishes between "benefit" taxation and taxation according to "ability to pay."With benefit taxation, the tax paid to the government is roughly equivalent to a price paid in the market—only in this case the good, or service, is furnished by the government rather than the private sector.7 Highway tolls are a common example. By contrast, applying an "ability-to-pay" standard means individuals pay taxes regardless of their highway use.

Adam Smith, the father of economics, is often accused of confusing benefit and ability-to-pay taxation in his argument that individuals "ought to contribute to the support of the government, as nearly as possible, according to their respective abilities [ability-to-pay taxation]; that is, in proportion to the revenue which they respectively enjoy under the protection of the state [benefit taxation]" (1904, 310). But it is not so clear that he was inconsistent, at least for most government taxes in his day and time. Consider the government's primary activities in the latter part of the 18th century: the defense of the state and its people, the maintenance of order and police protection, the sponsorship of trade and new industry, and the enforcement of contracts. Since the resulting benefits could not be calculated easily on an individual basis, it is easy to argue that the benefits were closely related or even proportional to ability to pay. Hence, for some public goods and services, it is possible for benefit taxation and ability-to-pay taxation to result in the same distribution of the tax burden.8

If purely voluntary, benefit taxation would not appear to violate the principle of individual equity—assuming that the government did not run a monopoly that distorted prices or otherwise restricted individuals' purchasing choices. Many governmental transactions, however, are involuntary, and the value of those public goods and services cannot easily be attributed to individuals. Some activities, like defense, require collective support. As a practical matter, taxpayers are compelled to share in these costs simply to eliminate "free riders"—those who avoid paying but get the benefit anyway.9

Of course, when government engages more in making transfers than providing other public goods, the benefits to transferors are unlikely to equal the involuntary taxes each pays. If there are benefits to the social order, or if most people in society want to act collectively to prevent poverty, for example, then some form of coercion is still required. Individuals have little incentive to contribute voluntarily to such actions, since their individual contributions make little difference. Democracy tries to limit this coercion by requiring that at least a majority of people favor such enforced action, but majorities can still reduce the freedom of minorities.

Mandated benefit taxation violates individual equity, even if individuals receive a benefit from what they put into a system. If I give up a dollar and get back a dollar in some mandated benefit, then I have less freedom than if I am left with my dollar in the first place. For example, considerable debate arises today over the establishment of mandated individual saving accounts inside or outside a social security system. Interestingly, such accounts designed as a carve-out from existing taxes garner support from many libertarians as a move toward individual equity. On the other hand, these same libertarians oppose individual accounts as another form of government interference when they are recommended to be financed from an add-on tax. The difference seems to be that in the case of the add-on, interference is viewed as increasing because the sum of mandated taxes plus contributions to the individual account are higher than the old social security tax.

The rise of social insurance in the 20th century has necessitated new thinking about how to achieve a balance between vertical equity and individual equity in the presence of substantial moral hazard. Social insurance is as much a problem about who should pay as about who should benefit when society decides to provide a minimum level of wellbeing.10 Take the case of preventing poverty in old age. A traditional welfare approach—as opposed to social insurance—would simply grant benefits only to those with low incomes. It is very easy, however, for the old and near-old to drop out of the labor force or give their assets to their children to lower their income and disguise their ability to pay. In addition, two people with equal incomes all their working lives may differ in their saving patterns so that one ends up better-off in old age than the other. Most people would consider it unfair to force the saver to transfer to the nonsaver when both had equal saving ability throughout their working lives. Yet, assistance to the nonsaver would force such a result.

Social insurance attempts to deal with this problem by mandating individual contributions for retirement. At the same time, most social insurance programs try to achieve some redistribution from those with greater lifetime ability to those with lesser lifetime ability. Once government mandates that people partly take care of themselves—for example, by contributing to their own retirement, it is difficult then to come in the back door and phase out benefits entirely when income rises as would a welfare system. A hybrid system—one that involves some redistribution but also mandates that individuals receive some return—is a compromise solution.

Even if a social insurance system gives some insurance back to all contributors, the net transfers it makes to those with lower wages or incomes ensures that some people are not going to get back all the money they paid in taxes.11 The point is simply that social insurance exemplifies an approach that combines a type of benefit taxation—mandated because of a moral hazard problem—and taxation according to ability to pay within the same overall program structure.

In many public debates, advocates will approach equity issues only from the standpoint of vertical equity or individual equity. Think tanks are set up to argue for either more redistributive or more libertarian government. Redistributional policy is sometimes presented as being always good or always bad. Such views reflect a lack of balance. The tension between the two equity principles is healthy. That government programs reflect ability and need is only natural, but restricting individuals' freedom to act is costly.

How Much Progressivity?

The debate over vertical equity has raged at least since the dawn of public finance as an academic specialty. That debate centers on how much progressivity is appropriate, if any. (At this point, the term "progressivity" will be used synonymously with "vertical equity," suspending a discussion of the inconsistent definition of progressivity in the tax and expenditure literatures.) While there is no clear-cut answer, therearestandards by which some rational judgment can be made. One standard might try to assess the relative amounts of sacrifice individuals should bear; another might assess efficiency losses from different alternatives. Once again, balance is key.

Throughout most of the relatively short history of economics as a formal discipline, the progressivity debate has centered mainly on taxation, rather than government expenditures. Economists have repeatedly sought to define the optimal amount of progressivity in terms of the sacrifice that individuals should make.12 The very use of the term "sacrifice" narrowly emphasizes the tax (or cost) rather than expenditure side of the issue. That is, one doesn't usually think of sacrifice in the same breath as accepting a benefit. The basic sacrifice theory is utilitarian and based on the commonsense notion that, at the margin, those who have more resources bear less sacrifice when they give up a dollar—for example, by reducing the amount of caviar for their yacht luncheon—than do those with fewer resources who might be threatened with, say, starvation.

Under utilitarian sacrifice theory, then, well-being is a declining marginal function of ability, income, or wealth. This assumption can be examined along several closely related dimensions. It not only implies that a richer person values a dollar less than a poorer person, but, depending upon the rate of decline in utility, it similarly implies that a person with $100,000 might value an additional $10 only as much as someone with $10,000 would value an additional $1. Such comparisons lead to discussions about equal absolute sacrifice, equal relative sacrifice, and equal marginal sacrifice. Economics students are even taught to toy with various precise mathematical relationships between utility and the consumption made possible by income (for example, that utility or well-being equals the square root of consumption). Under these precise mathematical assumptions, economists can develop formal measures to compare the "utility" sacrificed by different taxpayers.

Although these pedagogical devices are useful, no one can really measure someone else's utility nor assess how much sacrifice anyone else has made, regardless of their starting level of ability, wealth, or income. At the same time, despite its inability to prove that some degree or another of progressivity is ideal, the utilitarian approach still enjoys widespread commonsense appeal.

In the middle of the 20th century, a series of scholars, including Henry Simons,Walter Blum, and Harry Kalven, argued that progressive taxation was probably a good thing, but they suggested that the case for it was "uneasy" (Blum and Kalven 1953). In their view, a progressive system seemed desirable, but its justification was essentially aesthetic, not economic.As Henry Simons offered in a famous commentary: "The case for drastic progression in taxation must be rested on the case against inequality—on the ethical or aesthetic judgment that the prevailing distribution of wealth and income reveals a degree (and/or kind) of inequality which is distinctly evil or unlovely" (1938, 18-19). Simons offered this defense during a period in which the worldwide debate between socialism and some sort of refined capitalism or mixed economy was raging far and wide, a historical context that helped tip the scale in favor of substantial progressivity, whatever its shaky intellectual underpinnings.

Again, even the most adamant arguments about the shakiness of the progressivity principle never held that the poor pay more than the rich. Nor can one find any treatise suggesting that the income or wealth distribution should be more unequal than it is. Even the early benefit-tax theorists often suggested that the poor- or moderate-wage earner was not expected to pay tax, following the notion that many or even most of the nonpropertied class earned only "subsistence wages." The modern variation suggests that those with incomes below the poverty level should not be required to pay income tax. Modern flat-tax advocates, for example, usually allow for some amount of wages to be exempt from tax. In effect, this stipulation means that their proposals do not have flat rates, but a progressive rate schedule with two rates—one zero and one positive.

By the 1970s and 1980s, many trained economists were being taught that equity was no longer relevant to their discipline. Economics, they were told, had little to contribute to the debate, thus dismissing 200 years of substantial contributions from Smith to Musgrave. The new view, however, was only a logical extension of the notion that equity was simply a matter of qualitative judgment, more an art than a science. Economists were encouraged, instead, to focus on efficiency. Much of their training centered on how interference in various market transactions among individuals distorts behavior, so it was here, not in qualitative equity judgments, that they had some relative advantage.

By the 1980s, supply-side economists stretched these arguments into the political arena but with very strong emphasis on how progressive rate schedules would reduce aggregate saving and the labor supply. If equity did not matter, and only efficiency mattered, then taxes should be set so as to minimize these very large distortions. The net result, the argument went, might be much higher growth rates for the economy as a whole.

To understand how taxes might distort more when progressivity is increased, a little background is in order. Distortions result mainly from marginal rather than average tax rates. In the case of an income tax, lower tax rates on the first dollars earned by many taxpayers are likely to have inframarginal effects and should not affect their behavior. Correspondingly, rates at the highest income levels are more likely to be marginal, because they are more likely to apply to the last dollars of income. For example, the decisions of someone with income of $50,000 would not be affected much by whatever rate applied to the first $10,000 of income. That individual might, however, decide to either work less or save less depending on whether a 40 percent or 20 percent rate applied to income of more than $50,000.

The supply-side attack on progressive taxes, therefore, scarcely mentioned equity as an issue and argued mainly for lowering the highest marginal tax rates, which often applied at the top of the income distribution.

The logical extension of supply-side theory, in fact, is the old argument that head taxes are the most efficient form of tax. If everybody is taxed the same merely for existing (assuming that such a tax can be assessed and collected and that incentives for having children are not affected), then taxes would not affect any dollar earned or consumed after the tax was paid because there would be a zero tax rate on all marginal decisions. The simple fact that almost no supply-sider, no matter how extreme, proposes a head tax as a substitute for all taxes indicates that vertical equity issues are a concern to supply-siders after all—just as Stein hinted at earlier by never giving examples where the poor paid more than the rich.13

The new view of equity has been iconoclastic in its sweeping generalizations. For example, even if progressivity is in the eye of the beholder, horizontal equity, or the equal treatment of equals, or equal justice under the law cannot be brushed aside so easily. Thus, equity concerns are not beyond the pale of economic design. Moreover, it is possible to think rigorously about progressive scales rather than imply that all choices along those lines are merely subjective. And, finally, efficiency standards can determine whether equity or progressivity is being advanced efficiently. For example, if a program is designed to help the poor, then such standards can assess whether money is targeted efficiently to meet that end or is wasted on the nonpoor. The potential distortive effects of different levels of progressivity can also be compared.

An uneasy truce now exists between the iconoclasts and most practitioners of public policy and finance. The latter recognize the power of equity principles in the development of policy, but no longer ignore efficiency issues as some—perhaps many—did in the past. The former, despite their claims of indifference to equity principles, are often the first to fall back on them. For example, supply-siders have argued that an across-the-board or proportional tax cut is "fair" (are they merely playing to the politics of equity or do they believe that fairness does matter?). Likewise, many opponents of progressive taxation argue that expenditure programs should be targeted more at the poor; using extreme assumptions about the inapplicability of equity standard, wouldn't random distribution of benefits be just as fair and distort behavior less?

The Inconsistent Measure of Progressivity in Tax and Expenditure Systems

Roughly speaking, a tax or expenditure system is more progressive if it tends to redistribute more wealth from those who are better-off to those who are less fortunate. Beyond that basic understanding, however, progressivity can mean different things to different people. Inconsistencies permeate not just the general press but the academic literature—often with powerful effects on policy development.

One of the most arbitrary distinctions arises from the separate treatment of progressivity in tax and expenditure systems. Even in the most sophisticated newspapers and magazines, tax policy and expenditure policy are likely to be covered differently. Sometimes whole publications are devoted to only taxes or only expenditures. Legislatures separate their expenditure authorization and appropriations committees from their tax committees. Politicians running for office separate their advocacy for lower taxes from their support of more spending, as if the two didn't have to come into some balance.

Textbooks compartmentalize taxes and expenditures as well.Modern public finance texts usually examine the progressivity, horizontal equity, efficiency, and simplicity of tax systems. Their approach to expenditures, however, is very different—often focusing on cost-benefit analysis and how it might be applied to different programs.

Tax progressivity is normally defined relative to a proportional tax system. Take the case of an income tax. If the tax rate is constant, then the same percentage of income is owed to the government by all individuals, no matter what their income. This system is defined as proportional. If tax rates rise as income increases, the system is progressive; if they fall with income, then it is regressive. (Of course, many systems are progressive in some ranges and regressive in others.) Note, however, that higher taxes (not tax rates) are still paid by those with higher levels of income in proportional systems and in many regressive systems.

A different analysis is usually applied to expenditures. In welfare and other income-tested programs, benefits fall as income rises. A system is often said to be regressive when middle-class or rich individuals get more benefits per person than poor individuals. In effect, progressivity here is defined as the granting of more dollars to those with less ability, or income, or other measure of well-being. If a public education or highway system provided the same level of benefit to everyone, for example, then few would claim it was progressive. When an expenditure system (for example, educational grants) gives more to those with higher levels of income, it is often attacked as being regressive.

In the drafting of legislation, congresses and parliaments often display distributional tables and analyze policy changes in the same divided way. Tax cuts and tax increases are usually compared by the percentage change in tax liability, while expenditures are usually compared by how many dollars go to an individual in each income class.

The two measures of progressivity are inconsistent. The tax measure is defined with respect to rates, the expenditure measure with respect to dollars. This inconsistency means that a regressive tax system and a regressive expenditure system can together be progressive. Indeed, when expenditures and taxes are considered together, most government programs—including those drawing resources from proportional and even regressive taxes—end up redistributing from the more to the less wealthy.

To understand this concept, assume a world with only two taxpayers, one with an income of $10,000 and the other with an income of $50,000 (see table 9.1). Imposing a "regressive" tax system in which rates are 20 percent for the first $10,000 of income and 10 percent for any additional income means that the poorer taxpayer pays an average tax rate of 20 percent, while the richer taxpayer pays an average rate of only 12 percent. Assume under a simultaneous "regressive" expenditure system that the taxpayer with $10,000 of income gets $3,000 in benefits and the taxpayer with $50,000 of income receives $5,000 in benefits. In this case, total taxes paid equal total expenditures received, so it is easy to see the net effect of the combined tax and expenditure system. The net gain for the taxpayer with $10,000 of income is $1,000, while the net loss for the taxpayer with $50,000 in income is $1,000. Calculating net taxes (taxes less benefits) or net benefits (benefits less taxes) proves the system's overall progressivity by almost any definition.

The relationship between taxes and expenditures is easier to understand in these systems, where taxes are dedicated to specific outlays. Most individuals consider social security and Medicare progressive, even though taxes and expenditures in each are regressive (based on the inconsistent definitions discussed previously). The tax rate is constant up to a maximum earnings level, then falls to zero. Therefore, taxes are slightly less than proportional. Meanwhile, benefits are larger for those with higher levels of earnings and, in most cases, with higher levels of income. Yet, these systems are intended to be redistributive to those with lower levels of earnings and, based on net taxes or net benefits, are usually meant to be progressive.

Table 9.1 The Inconsistent Measure of Progressivity
  Taxpayer A Taxpayer B Total
Income $10,000 $50,000 -
Taxes $2,000 $6,000 $8,000
Average tax rate under
    so-called "regressive"
    tax system
20% 12% -
Benefits under so-called
    "regressive" benefits
    system
$3,000 $5,000 $8,000
Average benefit rate 30% 10% -
Net taxes (taxes less benefits)
    in overall progressive
    structure
-$1,000 +$1,000 $0
Note: In this example, Taxpayer B is richer than Taxpayer A, pays a lower rate of tax, receives more in benefits, and yet the system still redistributes some of B's wealth to A.

In sum, it is limiting and often misleading to define the progressivity of a tax system independently of what is done with those taxes, or as a corollary, to measure the progressivity of an expenditure system without considering how the necessary revenues are raised.

Determining the Tax (and Expenditure) Base

The theory of equity powerfully influences policymakers. Its practice, however, raises difficult issues.Whether discussing horizontal or vertical equity, it is necessary to define who are equals and who are unequals. The application of equity to a program requires some scale or base—or multiple scales. And even if a simple scale is used, it must be amenable to practical measurement.

Take the idea that equals will be determined according to ability. With what scalar does one measure ability? For many centuries, ability in the field of taxation was measured by property. Before the rise of the middle class, the propertied classes were largely considered to be those "able" to pay; the rest of the population often lived close to a subsistence level. Property, in turn, was largely defined by land. Along with tariffs, the property tax tended to be the primary source of government revenue even as late as the 19th century. In the United States during that period, local property tax collections normally far exceeded state and local taxes or federal tariffs and other federal revenue sources (Brownlee 2000).

But property was limited as a measure of ability. Although the term "human capital" is relatively modern, it has long been clear that some individuals are capable of earning more than others, and that differences in earning power are as important, if not more important, than differences in property ownership. The hoarder might have more property, but the enterprising worker might enjoy much higher levels of both income and consumption.

The rising middle class, in particular, focused attention on income, a measure of ability based on a flow concept, rather than property, which is a stock concept. As merchants, manufacturers, and their workers expanded in numbers and economic significance, skills and human capital became more recognized as major income sources.Yet it was only with the ascent of the corporation and other large organizations that wage payments, profits, and other income could be accounted for with enough accuracy and thoroughness that the income tax could move to center stage in the evolution of tax systems.While there were previous occasional attempts at income taxation, such as the Civil War income tax in the United States, they were still constrained by inadequate income accounting systems.

Consider the situation even in developed industrial countries in the middle of the 19th century. Most people still lived on farms, their income was often in-kind in the form of crop yields, and markets were still likely to involve significant barter without the exchange of money (for example, crop sharing and the exchange of meat, produce, and services among farmers).Whether the farmer kept good books or not, there seldom was anyone on the other side of the ledger whose books could be cross-checked by a tax agent. Even today, net income reported by farmers and sole proprietors is estimated to be underreported by more than one-third in developed countries.

The corporation, on the other hand, seriously needed good income accounting for payments to workers and returns on its activities, even if this need meant creating records not easily hidden from the tax agent. To assess optimal employment and investment patterns, the large business needed to know which of its many enterprises and branches were profitable (yielded net income). It had to keep track of wages paid to its many employees in its many divisions. A large organization would find it most difficult to keep a hidden set of books for tax purposes when so many people are involved; collusion is also harder. Moreover, the wage earners' records of wages received could now be checked against the large organization's records of wages paid, and vice versa. Thus, the rapid advance of the income tax in the 20th century coincided with the development of accounting systems whose records could be tapped for enforcement purposes.

Income had another advantage as well. It could be applied as a measure of ability not just to those who paid taxes, but to those receiving expenditures.Welfare and other means-tested programs in most countries now rely primarily on income as the measure of well-being and as a primary determinant of the amount of subsidy or expenditure that is provided.

Income, however, has never been fully accepted either as an appropriate measure of who are equals for tax purposes, or as the base on which progressivity, if any, should be assessed. Localities and states still retain real estate taxes, as well as personal property taxes levied against such items as automobiles—indicating some tendency to revert back to property as a measure of ability. Estate and inheritance taxes also are levied against property value passing in an estate.

Moreover, wage taxes are also assessed quite widely. In theory, they represent solely a tax on the returns from human capital and work effort. In practice, however, separating wage from capital income is nearly impossible unless there is a formal mechanism to achieve this purpose, as when a corporation accounts for stock earnings separately from the wages of workers. Within the small business, on the other hand, seldom are returns for the business easily separable into capital and labor components; most noncorporate business owners pay wage taxes, such as social security tax, on their capital income as well as on labor income.

Wage taxes, however, are usually associated with some form of social insurance, which ties the tax directly or indirectly to a particular benefit, such as social security, unemployment compensation, or workers' compensation. In those cases, equity tends to be defined within each program as a whole. If redistribution within those programs beyond that normally associated with private insurance did not exist, the taxes could be considered benefit taxes in the form of mandates to purchase insurance for oneself, and the primary equity issue would be whether the mandate itself was fair. However, those programs also redistribute or determine taxes and benefits according to ability. Yet, the redistributive function is often hidden within the insurance function. That fact tends to complicate analyses of whether the programs are fair in the way they treat households in similar circumstances (horizontal equity), provide a fair insurance policy for the premium or tax paid (individual equity), and redistribute to those with greater needs (progressivity).

A major tax debate revived in recent years has centered on the notion that consumption, not income, should represent the principal base for taxation. Separate states within the United States have often assessed excise taxes on the purchases of goods and, sometimes, services. In most countries around the world, a value-added tax (VAT) is a major source of revenue and is designed to allow deductions for investments in such a way that it can be considered a consumption tax. While the VAT often competes with an income tax, it seldom displaces one, at least in developed countries. Recent consumption tax proposals, on the other hand, have offered progressive rather than proportional consumption taxes as a complete substitute for progressive income taxes—often attempting to focus on the appropriate measure or scalar to use rather than on the degree of progressivity.

Among the many equity issues surrounding the consumption tax debate is the question of whether it is fairer to assess tax on individual consumption or earnings. Also of importance is whether two individuals with equal lifetime earnings and inheritances should be taxed the same on a lifetime basis, or whether the one who saves more—and, hence, generates more capital income—should be taxed more. Interestingly, advocates of consumption taxes over all other tax bases do not carry their equity arguments beyond the direct tax system itself. They have yet to explain fully how ability can be measured consistently between tax and expenditure programs and why, if consumption is the right base for explicit income taxation, it isn't the correct base for the implicit taxes used to phase out expenditure benefits. For instance, if all taxes and transfers are assessed on the basis of consumption, then millionaires with low levels of consumption will receive welfare benefits. If these benefits are phased out based on income, however, then we are partially back in an income tax world.

In practice,most governments have tended to use various measures—property, income, wages, or consumption—as a tax base or a base for determining eligibility for expenditures. This multiplicity of bases does not mean that equity does not matter—in that case, almost any base would qualify without reason or rationale. But it does reflect the difficulty of reaching consensus on just who are equals before the law and who are unequal enough to pay more tax or receive more benefits. In the end, only democratic processes are able to resolve those differences.

Adjustments to the Tax Base

Treating equals equally and unequals progressively are the two basic equity principles applied to tax and expenditure policy.Measuring who are equals involves more than deciding which tax base—income, consumption, property, wages, or any other—is adequate.

Even when a base, such as income, is chosen, further adjustments and refinements are usually considered. Seven types of adjustments or sources of disparity will be examined here: income in-kind, potential income or consumption, need, transfers paid and received, prices, household size, and measurement period. Because income is the most prominent measure by which industrial nations assess taxes and eligibility for expenditures, this discussion will primarily center on disparities relative to an income base, although many of the same adjustment issues arise with other bases.

Income In-Kind

Measurement for tax and expenditure purposes almost always uses recorded market transactions where money is the medium of exchange. Important barter transactions are excluded. Perhaps most important, existing measures ignore home production, even though a great many services are provided from the home, and many goods are produced there. This equity issue has come to the fore in recent years because of the movement away from home production and the rise in two-earner couples. One question, for instance, is how to treat one-earner versus two-earner couples with respect to work-related expenses and child care and whether or not those market-related expenses are appropriate adjustments to the measure of net taxable income (or consumption).

Potential Income

Perhaps the most serious defect in using income and almost all other tax bases to determine equality is that they focus not on potential but on actualization. If people with equal ability should be treated the same, then there is no equity reason—although there may be very practical administrative reasons—to more heavily tax the person who works harder or retires later. The problem is most serious in cases where individuals simply avoid recognizing or using their potential. Thus, when it comes to designing welfare and retirement programs, determining the extent to which subsidies and lower taxes to low-income individuals should be allowed is difficult because differences in actual potential are hard to distinguish from differences in realized potential.

Take two individuals each capable of earning $40,000. One works and pays taxes. The other does not work, pays no taxes, and collects benefits. If they have equal potential, then their benefits and taxes should be the same, not different.

Often the law tries to make distinctions in potential by relying on a separate categorical qualification other than income—such as age (old or young) or physical or mental impairment. Even here, however, measures are crude. In recent years, for instance, with the extraordinary growth in the number of individuals who are retired for one-third or more of their adult lives, many have questioned whether someone really has less ability (and, therefore, is more worthy of transfers and less capable of paying tax) simply because he has reached a 62nd birthday.

Unfortunately, potential—whether in the form of income, property, or anything else—cannot be measured well. Few would suggest that individuals should pay tax according to arbitrary assumptions about ability. If we assume naively that all individuals have the same ability, then the appropriate tax on ability is a head tax—an equal tax on all individuals just for existing. Yet, few believe that mere existence measures ability or that billionaires have no more potential for paying tax than anyone else.

Assessing ability among those who work full-time is easier. Differences in wage rates are related to differences in potential. Even here, however, it is clear that some jobs are easier than others, and some persons are glad to earn less in exchange for more leisure on the job.

Need

While many individuals may possess the same earning potential or realize the same amount of income, they do not necessarily have the same level of need. For example, one person may be in poor health and have large medical expenses. Few would argue that someone with $50,000 of income and $25,000 of annual medical expenses has the same ability to pay tax or the same need for government subsidy as someone who has the same income and is identical in all other respects except that he has no medical expenses.

Adjustments for Interpersonal Transfers and Other Taxes

Tax and expenditure systems also make adjustments, albeit inconsistently, for transfers made and received. The core issue is whether to measure household ability before or after transfers are made. For example, the existence of tax breaks for households with nonworking spouses reflects, in part, a view that some of the earner's income is transferred to the spouse, who should be granted his or her own tax-free level of income.

Charitable deductions are allowed partly on the theory that the transferor should be taxed on net income available for consumption, although a strict but unenforceable adherence to consistency would then require that the beneficiaries be taxed on transfers received.

Federal deductions for state and local taxes follow a similar transfer logic: If the tax does not generate services closely related to the amount of tax paid, then those taxes are less like fees and more like transfer payments to the eventual beneficiaries of the services.

Adjustments for Prices

Still another problem arises when $1 of income is really worth a different amount in different jurisdictions. Suppose it costs $20,000 in New York City to achieve the same standard of living that $10,000 will buy in Lexington, Kentucky. Then a person with $20,000 in New York City would not have an equal ability to pay an extra $1 in tax as a person with $20,000 in Lexington.

Adjusting for prices, however, is not easy. In national income accounting, there is no pure way to compare one set of prices with another. Moreover, where there are multiple differences in price, comparisons of income can be done using multiple scales. For example, suppose that rent, food, and clothing all cost twice as much in New York City, but only New York City offers access to Broadway plays. For a person uninterested in those plays, New York City may look expensive relative to Lexington, but for one whose life is consumed by such plays, New York City is a real bargain. Once differences are established on average for one region versus another, moreover, the intraregional differences are often as important or more important than the interregional differences.

Adjustments for Family Size

A difficult issue in both tax and expenditure theory centers on the "household unit." For example, if the goal is to tax all units in equal circumstances equally, how should one-person families be compared with two-person families? One attempt to measure equals, according to family size, is through an "equivalence scale." Despite its esoteric name, most people deal with this type of scale all the time. Perhaps the most familiar application is to poverty.When the government reports a certain number individuals in poverty, it is measuring their income against an equivalence scale. That scale might treat a single individual with $8,000 of income as being in poverty, whereas for a married couple the equivalent standard might be $11,500, and for a married couple with two children it might equal $16,000. Here there are two types of adjustments. Each additional individual in the family is generally treated as costing marginally less to support at a given standard of living, and children are usually treated as costing less than an adult.

Equivalence scales are based on the notion that there are economies of scale in living together. Yet, the application is quite arbitrary. For example, adults living together include students in dormitories, the elderly in old-age homes, unmarried couples living together, friends in shared apartments, and married couples sharing a home. Typically, tax and expenditure systems force equivalence scale adjustments on two adults only if they marry, regardless of whether they live together and achieve the economies of scale or not.

On the other hand, tax systems often treat income as if it is shared among married couples in some split, such as 70-30. In a progressive income tax, such as in the United States, this system allows those with more uneven splits—for example, 90-10—to effectively push more income (in this case, 30 percent rather than 10 percent for the lowearning spouse) into the lower tax brackets. Here a "marriage bonus" is created—thus taxing two married adults with very unequal incomes less than two adults with equivalent incomes living separately. On the other hand, those with 50-50 income splits would pay marriage penalties relative to other couples with equal combined income.

Endowment and the Accounting Period

A final and crucial issue often ignored is that a tax system almost inevitably must arbitrarily choose an accounting period over which to tax or determine expenditures. Such arbitrariness is largely a function of practical administration: The tax system usually latches onto the conventions of financial accounting with its annual focus. Nonetheless, a 5- or 10-year period, or even a lifetime, would represent a different way of measuring who are equals and who are not.

Many years ago it was shown how endowment could be considered as equal to the present value of wages and inheritances received and also equal to the present value of consumption and inheritances made.14 According to this accounting, one might not want to tax capital income during life, since that would penalize a person according to when he or she saved and consumed rather than how much funds were available to consume or transfer over a lifetime. Thus, ignoring transfers, this lifetime perspective provides some justification for a consumption tax over an annual income tax. If one takes transfers into account, however, then this perspective suggests that we could tax those with equal endowments equally if we had either a wage tax backed up by an inheritance tax or a consumption tax backed up by an estate tax. In that last case, however, the tax would be more like an income tax with a lifetime accounting period than a consumption tax.

Conclusion

That equity principles have a powerful influence on policy should not be surprising. Equity is closely associated with justice, and justice is closely aligned with lawmaking.

Many, if not most, public laws represent attempts to improve equity. Even laws that emphasize other concerns, such as efficiency, must pay homage to equity principles. From this lofty ideal, we then turn to details. Different notions of equity compete—for example, vertical equity and redistribution toward the needy compete with individual equity, which asserts people's right to transact freely with others. Equity is not even defined consistently between tax and expenditure systems, so that what is sometimes called regressive tax policy and regressive expenditure policy can still be progressive; this inconsistency is made most apparent when programs with both designated taxes and benefits are considered as a whole.

The base or bases by which to measure who are equals and who are more or less able to pay must be determined; this requirement is no easy matter, and much disagreement persists among those who emphasize income, consumption, or some other measure as the base. Finally, in determining who pays taxes and who receives expenditures, possible adjustments for the potential income of the individual, household size, transfers made, lifetime endowment, and many other items must be taken into account.

In the midst of this complexity, it is tempting to conclude that equity must merely be in the eye of beholder and that there is no reason to pay homage to the standard of justice when developing policy. Some economists imply, and others state, that their profession has nothing to say about equity. From this perspective, equity debates are merely over unmeasurable qualitative matters, similar to competing assessments of a work of art.

In some cases, progressivity is really what is being attacked as unworthy of economic analysis; in other cases, almost all aspects of equity are under siege. Economists are told to focus their efforts on the efficiency aspects of government policy and, it would seem, turn the equity debate over to lawyers and advocates.

This chapter reclaims the equity ground on which policymakers instinctively move, and on which economists from Adam Smith to Richard Musgrave quite naturally walked. The problem of public finance cannot be separated from the development of a just society. Political decisionmakers need the best advice on how to create a just society, which means much more is at stake than simply minimizing the inevitable distortions that accompany all tax and expenditure systems. Equity is not only a legitimate field of inquiry for economists, but also a necessary exercise for any would-be policymaker who must balance the benefits and costs of various public actions. Public finance without consideration of equity is like a body without a soul.

The attempt to apply benign neglect to equity involves two types of errors. The more general error is the idea that since no simple standard of equity can be proclaimed universally and applied simply, no standard exists. The more specific misjudgment is that no "scientific" analytical thinking, or at least economic reasoning, can be applied to making equitable choices.

That one cannot necessarily proceed from the particular to the universal, of course, is well-known; this given exposes the logical weakness of the first, more general attack. The presence of hard choices does not mean the absence of viable choices. Suppose a person finds $100. He or she could, perhaps, think of a million good ways to spend the money. He or she could also find a trillion ways to waste the money.

Tax and expenditure choices are merely the public equivalent of these private decisions. If only one choice were feasible or rational, we would have less need for a democracy to sort out choices in a nonviolent manner. Nonetheless, equity principles can be applied usefully even in the midst of disagreement. One example occurred during the tax reform effort at the Treasury in 1984. At that time, those who favored a consumption tax and those who favored an income tax had reached a stalemate. Not knowing which type of tax would eventually be proposed made it difficult to start the decisionmaking process. The solution was simple: Concentrate the initial discussions on those aspects of tax law that were not dependent on the income/consumption debate, such as many itemized deductions, employee benefits, tax credits, and other preferences. In most cases, the equity choices to be made on those issues were the same for a consumption tax as for an income tax.

Thus, suppose a society is undecided over whether two people with equal consumption or equal income should pay the same amount of tax. We can and should agree that any tax system should not unequally tax individuals with both equal consumption and equal income.

Think of two competing principles as two points in space. An infinite number of points represent a compromise between these two points. But there is a subset of points that make up the line stretching between the two original points. The points on this line represent the minimum total distance from the original points and represent an array of compromises that make more sense than the myriad other points floating about in the space. The points not on the line between the two points of principle can then be rejected as being too far from both the principles. That is, compared with these "rejected" points, it is always possible to find points on the line that are closer to at least one principle without being farther from the other.15

The error of the more specific attack, the one more commonly expressed by many economists, proceeds from several sources. Surely economists are trained to understand efficiency. That training, however, no more detracts from their ability to consider equity than does medical training prevent a doctor from examining mind and body. In fact, economists' training prepares them to examine equity issues because their analytical techniques emphasize understanding and quantifying relationships, approaches that can easily be applied to considerations of which individuals are equal and which individuals are not.

Economists' study of efficiency also allows them to spot the inconsistencies in purported theories of equity. For example, one theory espoused in different forms is that government actions must be judged first on what they do for the poor.16 While this principle sounds idealistic, economists recognize that at times it disregards the value of transactions among people, including the nonpoor, that make the parties to the transactions better-off without making anyone else worse off.

This second attack on equity, of course, is concentrated particularly on vertical equity or progressivity itself, where qualitative judgment is a necessity.However, that people care about equity, including progressivity, should be recognized, in the language of economists, as a "revealed preference." If people want to pay for justice—for example, by preventing the poor from starving—then why should this valued service be downgraded relative to other consumption desires? Why should the demand for steel in cars be considered a higher point of inquiry than the demand for poverty relief? Once equity is recognized as having a value which people are willing to sacrifice resources for, then it becomes an issue of efficiency as well.

Examining how different equity goals can be met efficiently requires considerable effort. For example, governments often attempt to target programs toward the poor and must develop least-cost options. In designing many programs aimed at improving equity, the efficiency aspects of both implicit and explicit tax rates need to be examined. And any government consideration of interfering in imperfect markets requires some understanding of both potential equity and efficiency consequences.

Having gone so far to oppose those iconoclasts who would remove equity as a standard, let's go one step further. There is a notion, sometimes taken from a book by Arthur Okun (1975), that equity and efficiency require a "big trade-off." Yet, the analysis here suggests that such a trade-off often is not required, and equity and efficiency often go hand in hand. Consider, in order, individual equity, horizontal equity, and progressivity. Individual equity emphasizes that we are entitled to the rewards of our efforts and from the trades and transactions that we make and thus is closely related to traditional market notions of efficiency. Horizontal equity and efficiency, however, are also linked. The equal tax treatment of different sources of income, for example, often leads to both equity and efficiency gains by simultaneously taxing those with equal incomes equally and removing tax-induced distortions. Finally, the pursuit of progressivity, adequately balanced against claims of individual equity and efficiency, is nothing more than the pursuit of the good society in the end, and a good society is going to be efficient and richer in the broadest sense of that word.

Can the last point be proven? No. From one perspective, it requires a holistic, rather than individualistic, view of humanity—a view that no part of the societal organism is totally independent from the other parts. Of course, the pursuit of progressivity is fraught with costs and dangers, which is why it must always be balanced against other principles.

The final claim that progressivity and efficiency must at least be considered together derives from the demand for equity already discussed. Equity is a service for which individuals reveal a preference, and like all services, it should be provided efficiently.

In sum, despite many sources of complexity, equity principles are the first standard against which policy is assessed and judged. Put simply, our democratic world cannot be otherwise. True, equity might be ignored at certain times and in certain legislation. Certainly, much bad policy today derives from nothing more than an inadequate consideration of the demand for equal justice. And the tension between vertical and individual equity will always remain. Yet, the standard of equity simply cannot be ignored for long or the bounds of inequity pushed too far. Equity will always reassert its rightful place as the first and most basic set of principles applied to constitutions and laws.

NOTES

1. Lewis is basically arguing for a natural law of morality, but his examples constantly seem to reflect some equity standard. See Lewis (1943, 17).

2. See, for example, Musgrave and Musgrave (1976) and Musgrave (1959, 1985, and 1996).

3. Again, see Musgrave (1959, 60-61, 90-115) for a useful summary.

4. In theory, economists will extend this notion to argue that those who start out with equal "utility" or well-being should end up with the same utility after the imposition of the tax.

5. See Musgrave (1959, 160).

6. For a discussion of the dependent vs. independent nature of vertical equity, see Musgrave (1990). See also Musgrave's earlier argument that horizontal and vertical equity are "different sides of the same coin. If there is no specified reason for discriminating among unequals, how can there be a reason for avoiding discrimination among equals?" (1959, 160). See also Kaplow (1989).

7. For example, in 1919, Erik Lindahl (1958) suggested that political processes substitute for the market economy and determine a price for public goods based on marginal benefits received.

8. Richard Musgrave has suggested that Smith "rather ingeniously combined both benefit and ability-to-pay considerations in one dictum" (1996, 344). In another article, Musgrave tried to clarify the meaning of Smith's statement on benefits and burdens. "Smith might have indeed wanted to have it both ways, or he might have been aware (if not stating so explicitly) that the ability and benefit doctrines may be linked via the income elasticity of demand for public goods. His ability-to-pay rule could then be viewed as a prescription for benefit taxation" (1990, 114).

9. For a good textbook discussion of both "free riders" and "moral hazards," see Rosen (1999, 69-70, 206-8).

10. See Steuerle and Bakija (1994, chapter 2). Of course, social insurance has also been used politically as a vague term to justify redistributive policies, whether well designed or not.

11. This passage omits many other issues related to social insurance, such as how much rising levels of transfers from future generations can be used to somehow protect earlier generations and whether mandated contributions are really saved.

12. See contributions to this book by Richard Musgrave, Barbara Fried, and Dennis J.Ventry Jr.

13. It is notable, for instance, that Margaret Thatcher finally lost her position as prime minister of Great Britain not long after proposing a type of head tax as a substitute merely for one small part of the British tax system.

14. See Steuerle (1980). See also Daniel Shaviro's contribution to this book.

15. One can go on with this mathematical analogy when three legitimate principles compete. Three points form a plane, but there is even more space inthree-dimensionalional space outside the interior part of the plane that is formed by connecting the points. By thinking rigorously about the competing principles, one can remove options not lying within the plane.

16. See Rawls (1971, 83), where, among other places, he argues that "social and economic inequalities are to be arranged so that they are . . . to the greatest benefit of the least advantaged." Religions often make claims here as well, as in the case of the "preferential option for the poor" put forward by Roman Catholic bishops. See also Steuerle (2000).

REFERENCES

Blum, Walter J., and Harry Kalven Jr. 1953. The Uneasy Case for Progressive Taxation. Chicago: University of Chicago Press.

Brownlee, W. Elliot. 2000."Historical Perspectives on U.S. Tax Policy toward the Rich." In Does Atlas Shrug? The Economic Consequences of Taxing the Rich, edited by Joel B. Slemrod (29-73).New York and Cambridge: Russell Sage Foundation and Harvard University Press.

Kaplow, Louis. 1989. "Horizontal Equity: Measures in Search of a Principle." National Tax Journal 42 (2, June): 139-54.

Lewis, C. S. 1943. Mere Christianity. New York: Macmillan. (Collier Books paperback edition.)

Lindahl, Erik. 1958."Just Taxation—A Positive Solution." In Classics in the Theory of Public Finance, edited by Richard A. Musgrave and Alan T. Peacock (168-76). London: Macmillan.

Lutz, Harley. 1945. Guideposts to a Free Economy. New York: McGraw-Hill

Musgrave, Richard A. 1959. The Theory of Public Finance. New York:McGraw-Hill.

———.1985."Public Finance and Distributive Justice." In Public Choice, Public Finance, and Public Policy: Essays in Honour of Alan Peacock, edited by David Greenaway and G. K. Shaw (1-14). Oxford: Basil Blackwell Ltd.

———. 1990. "Horizontal Equity, Once More." National Tax Journal 43 (2 June): 113-22.

———. 1996. "Progressive Taxation, Equity, and Tax Design." In Tax Progressivity and Income Inequality, edited by Joel B. Slemrod (341-56). Cambridge: Cambridge University Press.

Musgrave, Richard A., and Peggy B.Musgrave. 1976. Public Finance in Theory and Practice. New York: McGraw-Hill.

Okun, Arthur. 1975. Equality and Efficiency: The Big Tradeoff. Washington, D.C.: The Brookings Institution Press.

Rawls, John. 1971. A Theory of Justice. Cambridge: Harvard University Press, Belknap Press.

Rosen, Harvey S. 1999. Public Finance, 5th ed. Boston:McGraw-Hill.

Simons, Henry C. 1938. Personal Income Taxation: The Definition of Income as a Problem of Fiscal Policy. Chicago: University of Chicago Press.

Smith, Adam. 1904. An Inquiry into the Nature and Causes of the Wealth of Nations, edited by Edwin Cannan. Vol. II. London: Cambridge University Press.

Stein, Herb. 1959. "What's Wrong with the Federal Tax System?" U.S. Congress. House. Committee on Ways and Means. Tax Revision Compendium: Compendium of Papers on Broadening the Tax Base. Vol. 1. Washington, D.C.: U.S. Government Printing Office.

Steuerle, C. Eugene. 1980. "Equity and the Taxation of Wealth Transfers." Tax Notes (September 8): 459-64.

———. 2000. "Social Security and the Preferential Option for the Poor." America (October 23): 8-11.

Steuerle, C. Eugene, and Jon M. Bakija. 1994. Retooling Social Security for the 21st Century: Right and Wrong Approaches to Reform. Washington, D.C.: Urban Institute Press.

This publication is also available in PDF.