Published in Russian in Alexander Pozdnyakov (ed.) Order and Chaos in the Development of Socio-Economic Systems, Tomsk: The Institute of Optical Monitoring, 1998, pp. 16-25.

Nicolaus Tideman
Professor of Economics
Virginia Polytechnic Institute and State University

Introduction

Taxation began as extortion by conquerors. Over the centuries, regimes that started as tyrannies have evolved into democracies. The public sector has evolved from an apparatus for implementing the will of despots into a mechanism for carrying out democratic decisions. But public finance continues to rely on the power of tax collectors, developed under early tyrants, to coerce citizen to pay taxes. The wrath that citizens feel toward tax collectors is probably the strongest antagonistic feeling that citizens have toward a governmental institution. Nevertheless, few economists question the need for continued reliance on coercion to collect public revenues in democracies.

How do we justify coercion? Legislators and bureaucrats who devise coercive tax collection rarely bother to address the question. Nor do interest groups that promote expansions of the public sector, with their inevitable accompaniment of higher coercively collected taxes. Thus coercive tax collection has every appearance of a power play against the weak by those who have the strength to get away with it.

We need to reexamine our acceptance of coercion in public finance. If we are going to continue to rely on coercion, we need to develop a consensus about the framework that justifies it.

This paper begins with a definition of justice and a taxonomy of theories that might be used to justify coercion in public finance. I argue that theories based on Conservatism, Contractarianism, Utilitarianism, or Libertarianism all have value in some circumstances, but not for justifying the use of power to extract resources from people. None of these theories are consistent with what we ought to understand justice to be, because they do not accept the equal standing of all persons in defining what is good. A theory of justice that does accept the equal standing of all persons in defining what is good is Liberalism as elucidated by Bruce Ackerman (1980).

To oblige people to pay taxes while satisfying the limitations that Liberalism sets on the use of power, three conditions must be met:

Justice

Begin with what justice is not. Justice is the antithesis of might-makes-right. A powerful person might say, "This is just because I say it is just, and if you voice any disagreement I will silence you," and he might have the power to make his threat stick. However, he would be misusing language.

Justice entails the recognition by those who have power that the use of power ought to be constrained by principles whose relevance does not depend on the views of those with power.

Justice is well represented by the classical figure of the blindfolded goddess, with scales lifted up in one hand and a sword in the other. The scales convey the idea of equal division. The blindfold represents complete impartiality in determinations of what constitutes equal division. The sword says that there will be power to ensure that the determinations of justice are carried out. In other words,

Justice is the evenhanded adjudication of conflicting claims, backed by the threat of force.

To achieve the impartiality that justice requires, we generally insist that the person making a judicial determination have no personal interest in the matter being decided. The maxim that no person should be the judge of his own case is offered repeatedly in classical discussions of justice. But in the most fundamental questions of justice, we have no choice but to abandon this valuable rule. When it comes to determining the foundations of justice, there are no disinterested parties. When a person determines what we must all accord to one another, he determines for himself as well as for others.

Liberalism

The lack of disinterested parties to define justice is confronted in the principles that underlie Liberalism, as set forth by Ackerman (1980, pp. 4- 12). In Ackerman's Liberalism, every person who exercises power is obliged to defend his exercise of power with a consistent set of reasons that do not violate Neutrality, which is the principle that:

No reason [for exercising power] is a good reason if it requires the power holder to assert:

Neutrality blocks the appropriation for oneself of a greater value of natural opportunities than the per capita value of the natural opportunities that one leaves for others. In other words, a Liberal abides by Locke's famous proviso (1960 [1689], pp. 328-29) that there be "as much and as good left in common for others."1 While Ackerman extends Liberal theory to a wide variety of topics in very interesting ways, the Liberal analysis of claims to natural opportunities will be sufficient for discussing the ethics of coercion in public finance.2 The subsections that follow evaluate various alternatives to Liberalism for dealing with the lack of disinterested parties to define justice.

Conservatism

One way to sidestep the lack of disinterested parties to define justice is to identify justice with tradition. To the extent that tradition can be identified unambiguously, it is at least a disinterested source for justice. Conservatives stress the further virtue of tradition, that it has passed the test of evolutionary success. The traditions that we have inherited may have hidden strengths that we do not discern and would lose if we sought to design our own institutions.

As valid as these claims for tradition are, they are not sufficient to make tradition an adequate source for justice. In some times and places, tradition has given us slavery, serfdom, denial of voting rights to women, and any number of other inequalities that we have come to realize we cannot accept. While the Conservative perspective is valuable for the insights it provides into rationales for the status quo, its automatic acceptance of tradition cannot guarantee the evenhandedness that is essential in a theory of justice.

In rejecting tradition as a fundamental source for justice, we must also reject the ultimate authority of laws and constitutions as sources for justice. Justice is the standard to which we hold laws and constitutions; it is not defined by them. Nevertheless, our traditions often embody promises that we have made to one another, which we may need to take into account when we decide how we will get from where we are to where justice tells us we must be.

Contractarianism

A second common way of coping with the lack of disinterested parties is through Contractarianism. This is the axiom that it is just to coerce people to abide by rules to which they would have agreed before they knew how they would be personally affected by the rules.

There are several theories of justice that employ the Contractarian axiom. In The Calculus of Consent (1962), Buchanan and Tullock develop a theory of democratic principles growing out of a "constitutional setting." In this setting people are presumed to reach a consensus on rules with the greatest aggregate value, because they do not know what their roles will be in conflicts to which the rules will be applied.

The most famous modern application of Contractarianism is John Rawls' 1971 book, A Theory of Justice. Rawls argues that behind a "veil of ignorance" regarding one's personal characteristics, a person would want the rules that maximized the well-being of the representative member of the least advantaged class. To most economists, Rawls' specification of what would concern people behind a veil of ignorance sounds completely arbitrary and unfounded. If people are going to be overwhelmingly concerned about losing in the lottery of life, why would they focus on the representative member of the least advantaged class and not on the worst possible individual outcome? Why would people be unwilling to trade a minute loss by the least advantaged class for a substantial gain by the second-least advantaged class?

John Harsanyi (1975) argues that behind Rawls' veil of ignorance, people would not be concerned exclusively with the well-being of the representative member of the least advantaged class. Rather, they would recognize that they had equal chances of being all persons, and therefore, to maximize their expected utility, they would choose the rules that maximized the sum of individual utilities. The Buchanan and Tullock framework is consistent with Harsanyi's claim.

Another theory that applies the Contractarian axiom is Ronald Dworkin's (1981) theory of justice as equality of resources. Dworkin justifies an income tax as the expression of an insurance policy that people would desire before they knew their talents.

The fact that people using the basic idea of Contractarianism could come to conclusions as different as those of Rawls and Harsanyi is a reflection of a limitation of the axiom. Even if one of them has it wrong and the other right, the disagreement suggests that the Contractarian axiom is inevitably too vague and too subject to self-serving or idiosyncratic interpretations to serve as a basis for achieving consensus.3

To put the difficulty with Contractarianism in a somewhat different light, suppose that a person asks, "Why is it just for you to use your power to deprive me of what I want?" The Contractarian exercising power responds, "Because you would have agreed to the rule that deprives you, before you knew what your role in this dispute would be." The deprived person can respond, "No, not me. I never would have agreed to that." To justify the deprivation, the Contractarian must declare that the deprived person is mistaken or deceitful about that to which he would have agreed. Anything that the deprived person says about who he is or what has happened in his life can be declared to be irrelevant to the question of what rules he would actually have favored in a properly constructed contractual setting. The flesh-and-blood person asking for a justification of power goes unanswered.

Contractarianism does have potential value as an organizing principle for seeking to resolve disputes, because sometimes it will be possible to persuade a person involved in a dispute that what he seeks is inconsistent with a rule to which he would have agreed before he knew his personal role in the particular dispute. But Contractarianism is dangerously vague as limit on what people with power accord to themselves. It is too easy to convince oneself that people would have agreed to whatever you like in the hypothetical contractual setting.

Utilitarianism

While Utilitarianism is presented above as a version of Contractarianism, some people might argue for Utilitarianism independently of any such foundations. If the argument simply begins with the proposition that aggregate utility should be maximized, then there is a similar disregard for personhood. A person is then simply a device for generating utility; whether the utility is for herself or for someone else is irrelevant. People matter only to the extent that they are generators of utility. To those seeking explanations of why it is right for power to be used to prevent them from achieving their ends, this is just as dissatisfying as explanations based on hypothetical people in a contractual setting.

The great value of utilitarianism is as an ethical principle for allocating largess that is rightfully one's to allocate. If a person must decide whether to allow A to achieve purpose 1 or B to achieve purpose 2, the judgement that purpose 1 would contribute more to person A's utility than purpose 2 would contribute to person B's utility is an excellent reason for deciding in favor of person A. But justice is not concerned with largess. It is concerned with what is due to persons just from being persons, without having to prove their worthiness any further.

Libertarianism

Utilitarian and Contractarian theories of justice share the feature that they do not acknowledge self-ownership. Under Utilitarianism, persons are inputs into the generation of aggregate utility. Under Contractarianism, persons might have owned themselves in the contractual setting, but there they sold themselves. The rights of real people are gone. Theories in these classes can be contrasted with a class of theories that justify power with direct responses to those who question power, treating them as individuals who own themselves. There are two theories in this class: Liberalism and Libertarianism. The two theories share the idea of self-ownership, the idea that each person is free to decide the purpose of his or her life, provided that the rights of others to do the same are respected.

The fundamental difference between Libertarianism and Liberalism concerns the basis for claims to own things. The Libertarian axiom (Rothbard, 1982, pp. 29-43) is that everything in nature is the property of the first person who transforms it. This contrasts with the Liberal axiom (Ackerman, 1980, pp. 11, 31- 68) that the claims that people make on natural opportunities are respectable only if the person making the claim leaves as much for others as she takes for herself.

When a Libertarian is asked, "Why should you have that opportunity instead of me?" he can reply, within his own theory, "Because I got here first and did something." The Liberal says that any appropriation of natural opportunities that leaves less for those who come later fails to satisfy the condition of evenhandedness that is required for justice.

The Libertarian and Liberal theories both have limitations in terms of efficiency. Acceptance of the Libertarian theory promotes a land rush, in which people waste resources doing the minimal work necessary to establish claims to land, so that they can later sell it to others. (Whoever burns down the most rain forest gets the most land.) This first inefficiency is compounded by a continuing inefficient use of land, arising from the inability of people to manage all that they have claimed. If the Liberal theory is accepted, an opposite inefficiency occurs. There is no incentive for people to seek to discover opportunities hidden in nature, because whatever is discovered belongs equally to all.

While it may be difficult to avoid having these inefficiencies color our evaluations of the theories, we should remember that the pursuit of justice does not guarantee efficiency. It is possible that a commitment to justice will entail accepting some inefficiency.

The basic problem with the justice of Libertarian theory is that it allows the first arrivers to deprive those who come later of any natural opportunities that it suits the first arrivers to claim and transform. To permit this is inconsistent with the evenhandedness that justice requires.

A Libertarian might allege that it would never be possible for the first arrivers to leave as much for all later arrivers as they take for themselves, because there may be an endless stream of generations, and each person would be unable to claim even enough land to stand on. This objection presumes that any claims would be made on the stock of land, rather than on the flow of land services. Under Liberalism, each person may respectably claim the use of an amount of land during her lifetime that leaves as much land rent per person for others who are alive at the same time as the rent of the land that the claimant reserves for herself. People may also respectably reserve more for themselves, if they compensate those who therefore have less. There will be plenty of room for everyone to stand.

Libertarianism is an excellent framework for analyzing justice when there is no scarcity of natural opportunities. But when natural opportunities are scarce it denies the equal rights of those who come later.

Justifications of Coercion in Economics

Cost-Benefit Analysis

The most common way that economists justify coercive governmental actions is through cost-benefit analysis. This is oversimplified Utilitarianism, and it is subject to the same criticism as Utilitarianism. If it were pure Utilitarianism, it could be criticized for treating persons as deserving concern only to the extent that they generate utility. Being oversimplified, cost-benefit analysis treats persons as deserving concern exactly to the extent that they experience financial gains and losses.

Any financial harm to one person can be justified by a financial gain of a greater magnitude to someone else. It is imaginable that people in a constitutional setting would agree to a rule that projects that passed cost-benefit analysis could be financed by coercive taxes, but this is plausible only if all reasonable efforts are made to finance projects by taxes whose incidence is in proportion to the benefits of the project. An suggestion that people can respectable be coerced to finance projects simply because they pass a cost-benefit analysis is a mockery of justice.

Social Welfare Functions

A social welfare function is fancy cost-benefit analysis. If the social welfare function were defined as the sum of the money incomes of the members of society, it would be exactly equivalent, and again subject to the criticism of Utilitarianism. If one accepts Samuelson's (1947, p. 221) invitation to employ a more general welfare function specified by some individual, then that individual becomes a dictator. The evenhandedness required for justice is completely absent.

The Pareto Criterion

The Pareto criterion is closely connected to Conservatism. It declares that there is no way to judge whether one initial allocation is better than another, while any action that benefits at least one person and makes no one worse off (How would we know?) is undeniably good. This implicitly puts questions of the justice of the initial situation beyond consideration. Unless the initial situation is subject to scrutiny, there is no justice.

Public Choice

Economists often analyze public decisions as predictable consequences of self-interest operating in democracy and bureaucracy. This is fine as positive analysis, but not as normative analysis. As normative analysis it is an example of Conservatism and has nothing to do with justice. If one were to declare that an outcome is just simply because it resulted from democratic process or from procedures that are consistent with a constitution, that would eliminate any possibility of scrutinizing the actions of tyrannical majorities or biased constitutions.

Near-Unanimity

Knut Wicksell (1958 [1896]) offered a theory of public finance that comes close to justice. He argued that for every worthwhile public expenditure, there must be an allocation of taxes to finance the expenditure that would make everyone better off. If legislatures were required to achieve unanimity to pass spending programs, then they would have to find an allocation of taxes that was unanimously acceptable before they could pass a spending program. Thus majorities would have no opportunity to exploit minorities, and inefficient proposals would be prevented from passing as well. Wicksell recognized that if complete unanimity were required, strategic holdouts would prevent any program from passing, so he was content to recommend a rule approaching unanimity, without being specific about what this meant. But any departure from unanimity means that there must be a Conservative principle at work to justify coercion, and this opens the door to injustice.

Competing Local Governments

Another economic tradition that comes close to consistency with justice is the Tiebout (1956) tradition, which is concerned primarily with efficiency. In the Tiebout tradition, people sort themselves into communities composed of individuals who have the same preferences for local public expenditures. They are treated justly when they are required to pay taxes to finance public goods, because anyone who is unhappy with his tax bill is free to move to another community.

The principal difficulty with Tiebout's suggestion as a way of achieving justice is that it is developed in the context of a world with an unlimited amount of land. In today's world, people have little opportunity to form communities that express their own tastes for public services, because all land is already claimed by some sovereign nation that specifies the taxes that people who live there must pay.

A Just Framework for Taxation

Consider how Ackerman's principle of Neutrality constrains a person who wishes to levy taxes. How can person say to others, "You must pay these taxes," without violating Neutrality? One possibility is by permitting emigration. If there were unclaimed land with natural opportunities as good as those of existing nations, where those who did not like existing nations could form their own nations, then a right of emigration would make taxation consistent with Neutrality. The person who levied taxes would not be claiming more for himself than he left for others. In fact, the best natural opportunities are fully claimed. Still, a right of emigration goes some of the way toward making taxation just. It prevents at least the worst exploitation of taxpayers.

When the most valuable natural opportunities are fully appropriated, Neutrality requires that any effort to induce people to accept an obligation to pay taxes provide an alternative of leaving the tax-levying jurisdiction while retaining a level of access to natural opportunities that is no less than the per capita level within the tax-levying jurisdiction. "Leaving" could mean emigrating, or it could mean seceding.

The most straightforward way of achieving equal access to natural opportunities for emigrants is to recognize the requirements that Neutrality places on relations among nations or their subunits with respect to claims on natural opportunities. To guarantee equal access to natural opportunities, governments must acknowledge that their claims to natural opportunities, that is, their claims to land and natural resources, are just only if they are not disproportionate to the number of persons making the claim.

If land were uniform in its economic potential and there were no depletable natural resources, the application of such a rule would be straightforward. But land is not economically uniform and there are depletable natural resources, so the problem of equal access is conceptually more difficult.

To consider a simple case first, suppose that there is no disagreement about the relative value of land in different places and that there are no depletable natural resources. Equal access to natural opportunities is satisfied if the ratio of aggregate land rent to population is the same in all jurisdictions. Equal access is also satisfied if there is a set of transfer payments among jurisdictions, such that the ratio of the sum of aggregate land rent and net transfers received to population is the same in all jurisdictions.

The rent of land in these formulas is not calculated on a site-by-site basis, but rather on an aggregate basis, under the assumption that there are no man-made improvements in the territory being evaluated. Thus, in asking whether the United States was claiming more than its share of natural opportunities, one would ask what the rental value of the territory occupied by the U.S. would be if it were completely devoid of human improvements and human habitation, and the only bidders were other nations or their citizens. A similar calculation would be made for every other nation. Of course there can be no guarantee of agreement about such valuations, but agreement on the need to make such estimates and to try to reach consensus on their magnitudes would constitute enormous progress in global justice.

Next, suppose that there are depletable natural resources. This raises two new issues: appropriate depletion and intergenerational equity. Appropriate depletion is depletion on a path that maximizes the present value of net returns from using a resource. Devising a set of incentives that identify and achieve appropriate depletion is not an issue of justice and is therefore outside the scope of this paper.

Intergenerational equity requires that the value of natural opportunities available to the members of all generations be equalized. The value of depletable natural resources can be shared among generations by investing the net proceeds of depletion in capital and paying dividends to all generations. This could be done centrally or by individual nations. In any case, a determination of whether a nation's claim to territory is disproportionate requires consideration of the nation's appropriation of depletable natural resources as well as its appropriation of sustainable rent.

One of the ways in which harmony is promoted by a world order in which nations acknowledge an obligation not to make disproportionate claims on natural opportunities is through the incentives that such an order generates for nations to amalgamate.

Consider the question of whether the Sri Lankan Tamils will have a separate nation. If the Tamils have a separate nation, then in evaluating the proportionality of their territory, one would include the bids that other Sri Lankans would make for the land that they occupy. If the Tamils are part of Sri Lanka, on the other hand, only bids by others would be considered, and these might be considerably lower.

Within any nation, justice requires equality of access to natural opportunities among subdivisions. This is achieved by a similar process. The rental value of any subdivision is evaluated as if the subdivision were completely unimproved and uninhabited. The potential bidders would be the citizens of other subdivisions. Because of the value of proximity to the economic activity in the other subdivisions, the sum of such valuations would generally exceed the rental value of the nation. If subdivisions or their citizens claim rights to extract depletable resources, the opportunity cost of such extractions is added to the sustainable rental value of the land of the subdivision. Transfer payments among subdivisions equalize aggregate rent per person among the subdivisions. This amount is larger than the amount used to equalize rent per person among nations, because of the value that each subdivision receives from proximity to the others.

In the next step, equal access to natural opportunities for each locality in the subdivision is achieved by a similar process of evaluation and transfer payments. There is another increase in rent per person, because of the value that localities receive from proximity to each other's economic activities.

Finally, at the local level, assignments are made of the rental value that each parcel of land would have it were unimproved. Land value per person is yet higher, because of the value that improvements on one site add to surrounding sites, and transfers among individuals equalize this value.

The transfers that have been described provide equal access to natural opportunities, and to the advantages of being near the economic activities of others. The transfers would also tend to promote an acceptance of immigration, because with each immigrant that a jurisdiction accepted, it would gain an added claim on rent.

While the transfers described above are discussed as if they were obligatory, justice actually allows them to be discretionary, in the following sense. If the citizens in a locality share an understanding of justice that entails some pattern of entitlements or transfers other than that described, they are free to implement that pattern among themselves. Anyone who doesn't like it is free to go somewhere else, where he will have an undiminished claim on natural opportunities.

Similarly, the localities within a national subdivision are free to implement a pattern of locality entitlements that differs from the ones described, provided that they allow any locality to "emigrate." Since locations do not move, "emigration" here means secession. But secession is potentially expensive, because of the difference in the way that entitlements are calculated when membership in larger entities changes.

For example, suppose that Toronto wanted to secede from Ontario and become a separate province. As long as Toronto is a part of Ontario, the group that shares the rental value of Toronto (in an unimproved condition) consists of all Ontario localities. If Toronto were to become a separate province, that rental value would be shared among all Canadian provinces, and a much smaller share would come back to Toronto. The opportunity to share such rent would be glue that would tend to bind localities in provinces and provinces in the nation. But if the pressures to separate ever became great enough, the right to do so would be recognized.

If localities are permitted to secede from national subdivisions in place of emigrating, one might ask whether individuals may secede from nations. The answer is, in principle, yes, but it is likely to be impractical for them to do so, except on the fringe of civilization. If an individual who wants to declare himself an independent nation, the share of rent that he can claim is the land value per person in the equalization of rent among nations. But, unless he wants to live on the edge of civilization, he will need land in the midst of other people, where rents will be very high. Furthermore, his neighbors would not be treating him unjustly if they responded to his secession with an economic boycott. Thus it could be expected that people would not propose to secede in the middle of existing jurisdictions.

When all individuals and jurisdictions are free to emigrate and to secede in whatever ways they find practical, and equal access to natural opportunities for all members of any entity is the default option from which the entity is free to depart, then any taxes that are imposed as a condition of continued membership in the entity are consistent with Neutrality. People really can go somewhere else if they don't like it, and the equal access to natural opportunities required by Liberalism is satisfied.

If people were entirely selfish, and if outcomes were dominated by competition among jurisdictions, then in the Liberal world that has been described we would see efficient taxes -- on externalities and on the appropriation of natural opportunities. To the extent that people have feelings of community and of compassion for one another, we would find people agreeing to accept obligations to share the rewards of their efforts with others, though justice could not oblige them to do so.

Notes

1. Whether Locke should be classified as a Liberal is not clear. He seems to have believe that there was enough unclaimed land in North America that no one who enclosed land would be diminishing the opportunity of anyone else to do the same (1960, pp. 334-35). He did not address the issue of justice in claims on natural opportunities when those opportunities are limited.

2. It should be noted that some of the conclusions that I reach from Ackerman's premises conflict with conclusions that he reaches.

3. For a more detailed argument along these lines, see Ackerman (1980), pp. 336-42.

References

Ackerman, Bruce, Social Justice in the Liberal State, New Haven: Yale University Press, 1980.

Buchanan, James and Gordon Tullock, The Calculus of Consent, Ann Arbor: Michigan University Press, 1962.

Dworkin, Ronald, "What is Equality? Part 2: Equality of Resources" Philosophy and Public Affairs 10 (1981), pp. 283-345.

Harsanyi, John, "Can the Maximin Principle Serve as a Basis for Morality? A Critique of John Rawls Theory," American Political Science Review 69 (June 1975), pp. 594-606.

Locke, John, "Second Treatise of Government," in Two Treatises of Government, Cambridge, England: Cambridge University Press, 1960 [1689].

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

Rothbard, Murray, The Ethics of Liberty, Atlantic Highlands: Humanities Press, 1982.

Samuelson, Paul, Foundations of Economic Analysis, Cambridge: Harvard University Press, 1947.

Tiebout, Charles M., "A Pure Theory of Local Public Expenditures," Journal of Political Economy 64 (October 1956), pp. 1145-59.

Wicksell, Knut, Excerpts from Finanztheoretische Untersuchungen [1896], translated as "A New Principle of Just Taxation," in R. A. Musgrave and A. T. Peacock (eds.), Classics in the Theory of Public Finance, London: Macmillan (1958), pp. 72-118.

see also: Coercion Decision Tree