Nicolaus Tideman

Fundamental tax reform is on the agenda these days. There are a variety of proposals for a flatter federal income tax. There are proposals to replace the federal income tax with a national sales tax. I would like to suggest that discussion of fundamental tax reform should be preceded by a discussion of what constitutes a good system of taxing and spending. When principles of good taxing and spending have been examined, what emerges as attractive is greater reliance on local decision-making and, for federal revenue, an allocation of financial obligation to states rather than direct federal taxation of individuals or corporations.

Two of the most troublesome features of taxation and public spending are that taxation relies on coercion, and that the combination of taxing and public spending generates substantially haphazard redistribution. People must pay their taxes on penalty of having their property confiscated, or of going to jail. And we cannot say that this process is for everyone's good.

There is reason to believe, though we do not have the basis to say for sure, that some people receive much less in benefits from public spending than they pay in taxes. How do we justify the harm that is thereby done to some persons by the combination of taxing and public spending? I would like to address this question by first discussing some possible justifications that might seem attractive, but should be rejected.

• One possible justification is conservatism: Things have always been done this way. Our institutions have evolved to solve social problems that we may not even be aware of. Who knows what terrible things might happen if we did things differently?

  There is an element of validity to the conservative position, but not an adequate basis for making conservatism the primary principle of social organization. If we had adopted conservatism as the primary principle of social organization centuries ago, there would have been no end to slavery, no women's suffrage, nor any of the numerous other changes in social organization in response to improvements in our moral understanding. It would be foolhardy to assert that we will never experience further improvements in our moral understanding that will call for changes in social organization.

• A second possible justification of public spending is majoritarianism: Once we have voted we know the right thing to do. The possibility of majority-rule cycles undermines the simple-minded version of majoritarianism that says that the majority is always right. Still, it would be possible to assert that majorities should have their way when cycles are not observed, and that when they are observed, some device for cutting through cycles can restore the viability of majoritarianism. However, that would give free rein to any majority to exploit any minority. That makes no sense as a theory of justice. One could assert that voters, acting as judges of what is best rather than as self-interested advocates, have a claim to determine in a majoritarian fashion what a collectivity ought to do. But then one still needs a theory of the principles by which the voters ought to judge.
  

• A third possible justification of public spending is egalitarianism: People should be required to provide as much for others as they have themselves. This theory suffers from the difficulties of vagueness and questionable operationality. As much what? There is no way to ensure that all persons will have equal quantities of such valued things as affection, talent or self-confidence. It is possible to imagine a crude egalitarianism of goods, but only by eliminating all incentives to be productive. If equality is traded off against some other goal, then egalitarianism is no longer the first principle.

Rawls manages to retain the primacy of egalitarianism without destroying all incentives to be productive by the maximin rule: Laws should be designed to maximize the well-being of the least advantaged person. This formulation, however, generates such anomalies as the following: If there are three persons, and the choice is either a distribution of $10,000, $12,000 and $50,0000, or a distribution of $10,001, $10,002 and $80,000, the maximin rule will select the latter, despite the fact that it makes the middle income person, who is already relatively poor, even poorer while making the richest person noticeably richer. In Rawls's particular formulation there is a further difficulty. He makes the maximin rule lexically subordinate to the rule that all persons should have the maximum individual liberty that all can have. However, he does not deal with the following problem: My individual liberty will be unnecessarily restricted if I am not allowed to sing you a song in exchange for a haircut (without being taxed), but if I am allowed to do so without being taxed, the maximin rule is not satisfied. If Rawls really does mean to put individual liberty lexically first, no redistributive income taxes would be permitted.

Rawls sets his proposals in the framework of a fourth possible principle of public action, the contractarian suggestion that it is just to coerce people to abide by the rules that they would have agreed to if they had had a chance, before they knew their individual circumstances. One difficulty with this rule is that it is so hard to know what it is that people would have agreed to in these circumstances. Rawls is convinced that people would have agreed to a lexical ordering of maximum individual liberty and the maximin principle. John Harsanyi offers a strong argument for the proposition that they would have agreed to the rules that would maximize total (or, equivalently, average) utility. This has some strange implications. For example, if it is possible to separate individuals into ascetics and sybarites, then the maximization of total utility requires that the ascetics be forced to work very hard, with the resulting output used for the benefit of the sybarites. And if one of your kidneys (or your eyes) would provide me with more utility than it provides you, then the state is justified in extracting it from you and giving it to me. Utilitarianism does not accommodate individual rights.

The disagreements among contractarians about what contractarianism implies provide a hint of a difficulty that Bruce Ackerman elaborates in Social Justice in the Liberal State: Contractarianism offers so little defense against people with power who delude themselves about the undeniability of their beliefs about what people would agree to before they knew their personal circumstances. The person who feels oppressed says "I never would have agreed to these rules before I knew my personal circumstances." Those with power reply, "It's obvious to us that you would have. Stop your griping." The scope for further dialogue is small, and the potential for abuse is great.

Another possible framework for the justification of government action is that those with power are simply elitists: They may believe that their exercise of power is justified by their superior understanding of the nature of the good. While it is not unreasonable to suggest that there are some persons who do have superior understanding of the nature of the good, it is extremely dangerous for people to justify the exercise of political power by self identification as members of the elite who have superior understanding of the nature of the good. It was thinking of that sort that gave the world Stalinism. The general thrust of Western political thinking has been that elitism is simply unacceptable as a justification for the exercise of political power.

On the basis of the arguments given, I reject conservatism, majoritarianism, egalitarianism, contractarianism, and elitism as justifications for coercive taxing and spending. A framework for just social arrangements that does make sense to me is classical liberalism, which asserts that it is just to coerce people to accord others the maximum individual liberty that all can have. This means that people have rights to their bodies, their talents, the products of their labor, and the returns to their savings. Anything produced by human effort belong to the producer, or to the producer's successor in title through gift and exchange.

The principle of maximum individual liberty does not address the question of how the rights to natural opportunities (land, water, ocean fish, minerals, the frequency spectrum, etc.) should be assigned. There are at least three approaches within the classical liberal tradition as to how these returns should be divided.

George Reisman advocates what might be called "conservative classical liberalism." This position is that certainty in property rights is so valuable that one should never ask whether unjustifiable violence was used in establishing the existing pattern of control over natural opportunities.

Every natural opportunity belongs to whoever most recently succeeded in establishing control over it.

Murray Rothbard, by contrast, took what might be called the "homesteading libertarian" position. This is the position that when we can know who first used a natural opportunity, it belongs to that person, or to his or her successor in title through gift and exchange. All thefts from victims with identifiable successors should be undone. When we cannot identify the proper successor of the first user, the a thing belongs to whoever is using it now, unless that person stole it, in which case it properly belongs to whoever brings the thief to justice.

The third position, which might be called "geoliberalism" emerges from the work of Henry George. This is the position that all persons have equal rights to natural opportunities, which rights should be secured by having the public treasury collect the rental value of exclusive access to land and other natural opportunities, with the revenue used for public purposes and guaranteed incomes.

Of these three approaches, I find geoliberalism most attractive. It fits my sense of justice that all persons should have equal rights to natural opportunities. In addition, by providing funding for guaranteed incomes, geoliberalism offers a greater prospect for removing more of the distorting taxes that finance the welfare state.

A possible difficulty with classical liberalism as a justification for government action is that it may justify little if any government action. Taxes intrude upon individual liberty. How is this intrusion to be justified? One way to try to get around this difficulty is by the claim that some expenditures on protecting individual rights are so valuable that anyone would understand that, for these expenditures, the reduction in individual liberty from taxation is less than the addition to individual liberty from the protection of individual rights that is possible with taxation. However, if improving the well-being of all citizens is the justification, then one should ask whether there is adequate justification for even these government actions if they do not have unanimous support.

Supporters of government action might argue that if unanimous support were required, then the problem of selfish individuals holding out for greater shares of the surplus would raise transactions costs to the point where nothing could be done, even if there were things that would make everyone better off. But that justification carries the risk of justifying nearly anything. If those in power have an excuse to ignore what citizens say about what they value, what limits to power are there?

Geoliberalism offers a different way of justifying public spending and the taxation that finances it. If people are perfectly mobile, and if there are enough people who value a local public good to fill the area that benefits from it, then the provision of a local public good will raise land rents by enough to pay for the public good. In collecting the rental value of land, governments will collect enough to pay for any worthwhile local public goods that they provide.

Under the conservative or homesteading libertarian versions of classical liberalism, the possessor of land might say, "This is my land. I didn't ask for these public goods. You have no right to tax me to pay for them." Under geoliberalism, on the other hand, the community can reply, "You have as much right to the use of land and other natural opportunities as anyone else. If you want to exercise your rights in this community, these are the taxes you must pay. If you don't like it, claim your share of natural opportunities somewhere else."

If people are perfectly mobile and have an unlimited range of communities to choose among, then they cannot be exploited and need not tolerate inefficiency. They will move. Competing communities will find that the utility-maximizing public sector equilibrium involves providing all local public goods with benefits greater than costs, financing them by a combination of fees equal to marginal costs and taxes on the land that benefits from differential access to the local public goods. There will be no taxes on labor or capital except to internalize externalities.

If tastes vary so much that a community cannot be filled with people with the same taste for local public goods, then it will not be possible to finance all worthwhile local public goods by the increase in rent that they generate.

The community will either have to leave some worthwhile local public goods unprovided, or finance them from some other source. But in either case the residents are treated fairly if they are free to move elsewhere and choose not to do so.

If people are not perfectly mobile, then minorities can be exploited by being taxed without receiving corresponding benefits. If they could move freely they would, but if they can't, they may find it less painful to accept exploitation by the majority than to pay the costs of moving.

The efficiency of the competitive equilibrium requires that the sharing of the pre-development rental value of land (that part of the rental value of land that is not accounted for by local public goods) and the value of other natural opportunities be unaffected by migration or political competition. If the distribution of the returns to natural opportunities depends on where people live or on political competition, then efficiency will fall because of inefficient location decisions and rent seeking losses. For example, if the returns to oil in Alaska are divided among the people who live in Alaska, then people will move to Alaska inefficiently to share in the bounty. If the distribution of the returns to natural opportunities depends on political competition, then there will be rent-seeking losses from such competition. A non-exploitive and reasonably efficient local public sector is thus possible through competition among communities.

What, then, about public goods that have benefits over greater areas? If only a few localities are involved, one might expect them to negotiate voluntary compacts with reasonable efficiency. "Pork barrel" projects--all projects with benefits for only a small region of the country--should not be paid for with national taxes. The practice leads to the approval of inefficient projects as a result of the politicking of those who benefit, and it unjustly exploits those who do not benefit.

For truly national public goods, other ideas must be explored. One of the major national public goods is defense. In a perfectly just world, everyone would be so respectful of the rights of others, and everyone would feel so safe that no defense spending would be desired. In a less perfect world, many people, but not all, want public defense expenditures. How can they be provided justly?

Some financing of defense expenditures can be provided by a Pigouvian tax on the externality of accumulating capital, which makes a nation a more attractive target of aggression. If the U.S. requires a greater defense budget than Canada, which is larger in area, it is because the greater value of the assets in the U.S. makes the U.S. a more attractive target of aggression. Thus anyone who owns capital might reasonably be charged for the increase in the defense budget that is needed to make other citizens as safe as they would be if that one person's capital were not adding to the attractiveness of nation as a target. It would be interesting to know how much of the defense budget could be covered by such charges. I propose a self-assessed tax of, perhaps, 1% per year on the value of all assets and contractual rights, to pay the costs of defense. The owner assesses the value and pays a corresponding tax, and if anyone wants to buy the asset at the assessed value, it is sold. There could be a personal exemption of perhaps $50,000 per year, and an exemption for personal papers. There could be a local add-on to pay the costs of local police and courts.

An even greater share of the federal budget is used for various programs that provide help for people with special needs--welfare, Medicare, Medicaid, Social Security, unemployment insurance, disaster assistance, etc. Some of these programs (social security, Medicare, and unemployment compensation) are funded in part by payments by prospective beneficiaries. But all incorporate substantial elements of deliberate redistribution.

In a world that adhered to a classical liberal understanding of justice, there would be three sources of funding for people with special needs:

1) insurance that those with special needs had bought for themselves prior to the onset of their special needs,

2) donations from family, friends and other compassionate persons, and

3) local public expenditures that would be non-coercively financed, because anyone who objected to the financing could leave the locality.

In a just would, people would ensure, prior to conceiving children, that any expensive special needs that their children might have would be paid for, either by insurance that the parents would buy, or by voluntary conventions in their communities that the community would pay the costs of such needs.

Perhaps you object that some people would not be able to afford such insurance for all their children. If this is true, it implies one of two things.

• Either the parents of such children are behaving unjustly by conceiving those children, who will either lack what they ought to have or will be able to receive it only by the imposition of unjust taxes.
• Or else the parents have been treated unjustly by having started life without enough resources to provide for the children that they ought to have had the chance to bring into the world. I would guess that sometimes the first is true, sometimes the second, and sometimes both.

Whenever a one-time redistribution is proposed, a reaction of many economists is, "Yeah, right. Why would anyone one believe that it would be only one time?" What would make it reasonable to believe that such a redistribution would be a one-time event is its rationale: The recognition that some persons have had unfairly inadequate starting positions in life, and the determination to end that. If the purpose is achieved, there is no rationale for further redistribution, unless, at some future time, our society attains a new moral insight that implies that further redistribution is required.

With transfers replaced by private insurance and local action, and defense pad for substantially by Pigouvian taxes on capital, not much is left for federal taxes to support. Perhaps some international relations, perhaps some research (though in view of the possibility of a conflict of interest on my part, one should not take my word for it.) If I could have my way, the little that would be needed in national spending would be paid for by voluntary subscription by lower levels of government, as the UN is financed.

Of course, we are very far from public acceptance of geoliberal principles. So the question arises of what might be done within current political understandings to reduce inefficiency and exploitation in taxing and spending. It is hard to know how much more than what is being done might be done. Let me be on the imaginative side.

There is a proposal that tax increases be allowed only with a two-thirds majority of both houses of Congress. That has some merit, but it carries a risk of excessive deficits. It also allows the existing level of taxation to go unquestioned. It would probably be better to have a rule that every spending proposal must be approved by a two-thirds majority of both houses of Congress to be enacted. Maybe three-fourths. If spending is truly worthwhile, then, as Wicksell said, there is a way of financing it that will secure the approval of nearly everyone.

The current trend toward returning functions to the states is a step in the right direction. But it encounters understandable objections that poor states cannot afford to do what they ought to do. Some form of revenue sharing is needed. But it is important to have the right definition of which states are rich and which are poor. The level of well-being in a state is determined in part by the wisdom of its public policies. States should not be penalized for adopting productive policies. Revenue sharing should equalize per capita levels of natural opportunities (mineral revenues, fishing rights, pre-development land rents, etc.) Replacing the personal and corporate income taxes with either a flat income tax or a national sales or value added tax would greatly reduce the excess burden of federal taxes. (Excess burden is roughly proportional to the square of the typical marginal tax rate.) But almost all the gains go to the rich. Perhaps the flat tax could be combined with a guaranteed income.

Even better than a flat tax or a national sales tax, in my view, would be a return to the revenue system that was apparently envisioned by the drafters of the Constitution in 1787. I mean a rule that the federal revenue requirement would be allocated to the states in proportion to their populations. The use of population as the allocation device seems anachronistic in this era of concern for impoverished regions. But the allocation could be corrected by an expenditure that equalized per capita access to natural opportunities.

I see three important virtues in allocating the revenue obligation to the states.

• First, it gets rid of the IRS and its intrusions. (A national sales or value added tax would do most of this, leaving only firms subject to the tax man's scrutiny.)
• Second, it requires the states to compete in seeking ways of raising revenue that do not reduce the productivity of their economies. The competitive equilibrium is to raise as much revenue as possible from charges for exclusive use of land and other natural opportunities, and to tax labor and capital only as a last resort.
• The third virtue is the likelihood that placing the revenue obligation on states rather than on individuals would lead to more careful scrutiny of the value of proposed expenditures.

With only fifty states and each state aware of exactly how much each spending proposal will cost its treasury, I believe that it would be much harder to secure approval for inefficient or special-interest spending.