Wealth and Want
... because democracy alone is not enough to produce widely shared prosperity.
Home Essential Documents Themes All Documents Authors Glossary Links Contact Us



The Morality of Taxation: The Local Case
Speech given at the Jerome Levy Economics Institute of Bard College
November 2, 1995
Nicolaus Tideman
Professor of Economics
Virginia Polytechnic Institute and State University

From a moral perspective, taxation is dubious or worse. We tell our fellow citizens that if they do not pay taxes that we say they owe, their property will be seized or they will be sent to prison. Why do we treat people this way? Is there a justification?

The dubiousness of taxation increases when we consider its origins. Government seems to have originated as roving bandits who learned that total destruction was less profitable than protecting their victims from other bandits and allowing them to keep a fraction of what they produced (Olson, 1993). In time, scheduled partial plunder evolved into taxation. Over the centuries, regimes that started as tyrannies evolved into democracies. The public sector 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. Why do we allow ourselves to do this to one another?

There is a gentler side of taxation that provides some explanation of our tolerance of this coercion. Taxation can be the way that people achieve their common purposes. People may agree to be taxed so that there will be money to pay for public services that they want. From this perspective, taxation may be considered no more than the dues for belonging to a club that provides people with things that they would rather pay their share of than do without. However, to make this "voluntary exchange" theory of taxation relevant, people must be able to choose freely whether or not to "join the club," to be a citizen of the taxing jurisdiction. With all land claimed by some taxing jurisdiction, the choice isn't exactly free.

The problem of morality in taxation is the following:

  • How do we retain the possibility of people pooling their contributions to the cost of services that they agree are worthwhile, while eliminating the possibility of citizens treating their fellow citizens as targets of plunder?
  • What are the limits of obligations that we can justly impose on our fellow citizens?
  • And how do we set up a structure of government that will ensure that these limits are observed?
The turn-of-the-century Swedish economist, Knut Wicksell, had ideas that dealt with some of these questions (Wicksell, 1958 [1896]). Wicksell argued that if a public expenditure is worthwhile, then there must be some allocation among citizens of the taxes that are needed 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 allocations of taxes that were unanimously acceptable before they could pass those programs. In that case, 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 be likely to prevent any program from passing, so he was content to recommend a rule of "near-unanimity," without being specific about what this meant.

While Wicksell's insights are interesting, they do not fully solve the problem of moral taxation, because any departure from unanimity opens the door to exploitation of minorities, and the requirement of more-than-majority approval means that the costs of coalition-building will leave some worthwhile activities unapproved. Still, we would probably have a much more efficient public sector if every public expenditure required two-thirds approval in legislative bodies.

But to make taxation truly voluntary, the option to leave must be viable. If people could move costlessly from one jurisdiction to another, taking all of their belongings with them, then competition among jurisdictions would tend to eliminate oppressive taxation. This would leave only the fees that people were prepared to pay to have public services (Tiebout, 1956).

Of course, moving will always have some costs, so the ideal will not be attainable. But what can be imagined is a system in which all taxes were local taxes. Then people would not have to move nearly as far to escape from taxes that they regarded as oppressive. Higher levels of government would not need to disappear; if the services that they provide are desired, they could be financed by levies on lower levels of government.

Consider the economic equilibrium of such a system. What taxes should one expect to find? If it is very inexpensive to move from one place to another, then the utility that people achieve in any one community cannot be significantly lower that what they would achieve in any other community, and localities will only be able to tax people to the extent that their presence in the community generates net costs to the community. And there are some costs of added population --  greater congestion, perhaps higher costs of fire and police protection, and perhaps other costs as well. But there are also benefits to a community of greater population, arising from the opportunity of all other residents to trade with the new residents. Thus communities would not be able to raise much revenue from income tax or taxes on capital before they would drive residents and investment away. It might seem that there would be no way that localities could finance themselves.

Such a conclusion would be unwarranted, because there is a very significant source of public revenue that can survive when localities compete for mobile residents. This source is land. When people are taxed in proportion to the land they possess, no land moves to another locality where taxes are lower. Thus two questions arise:

  • Would taxes on land be sufficient to finance the public activities that ought to be undertaken, and
  • would such a system be fair?

Consider first the question of adequacy of revenue. There is a theorem in economics, known as the Henry George Theorem, that addresses this question (Arnott and Stiglitz, 1979). One of the simpler versions of this theorem is:

If the following three conditions are met:

1. Public expenditures provide benefits only over a limited area,
2. People can move costlessly, and
3. The number of persons who value a public service as highly as anyone does exceeds the number of persons who can live in the area where the service provides benefits,
then for any public service that is worth at least as much as it costs to those who receive it, the increase in the rental value of land that results from providing the service exceeds the net cost of the service.

The Henry George Theorem is true because people who can move costlessly will bid up the rental value of land to reflect the value of public services that are not available elsewhere. The assumption that the number of bidders exceeds the number of people who can benefit from the public service guarantees that the upward movement of rents will not end until all the benefits of the public service are reflected in these rents. If some people who receive a public service value it more highly than others, then they will receive a surplus in addition to the rent they pay for land, and some worthwhile increments of public services will not add quite enough to rent to pay for themselves (Tideman 1993). But rent increments will go a very long way toward paying for worthwhile local public services.

With some local public services, such as weather reports that are broadcasted by radio at a specified wattage, there is no social cost of having another person within the reception area use the service. With others, such as distribution of water, the extra cost of one additional user will be noticeable, but still a relatively small fraction of the average cost of the service. For public services to provide the greatest possible net value, users must be charged the costs of additional use (what economists call marginal cost). If they are charged less than marginal cost, self-interest will motivate them to use the service wastefully. And if they are charged more than marginal cost, self-interest will motivate them to economize inefficiently on the service.

Thus local public services are financed efficiently when users are required to pay marginal costs, and the component of total cost that is not covered by marginal cost charges is financed by levies on land that collect the increases in rental value of land that result from provision of the service. If all citizens in a locality valued the service equally, then every increment in the level of the service that was worthwhile would have a cost that was less than or equal to the sum of the revenue that could be raised from charges for use equal to marginal cost and the increase in the rental value of land in the community that resulted from the availability of the service. When people value the service differently, the increase in the rental value of land that results from the provision of a service will be less than the value of the service to those who receive it. By a criterion of maximizing the real incomes of residents, the best outcome that covers the full cost of a service will involve a level of service somewhat lower than that for which marginal cost is equal to the sum of marginal benefits, along with charges that exceed somewhat the marginal cost of providing the service to additional persons. Still, in general, the combination of user fees approximately equal to marginal cost and taxes on land to finance the remainder of costs is capable of providing financing for local public services that is adequate and reasonably efficient.

Next consider fairness. The fairness of such financing is the fairness of incremental decisions in an environment in which the initial allocation is fair. (It does not provide special opportunities for disadvantaged persons. They would need to be provided for by insurance that operated independently of the provision of public goods.) A person who is treated fairly in the absence of public goods cannot reasonably complain about being required to pay for a service according to its marginal cost, or about have to pay the value that is added to his land by the availability of a public service.

In the latter case, there is an argument that must be answered. A person may say, "It is true that the provision of this public service adds as much to the rental value of my land as I am being taxed, but that is what the service is worth to someone else who might use my land. It's not worth that to me. In fact, the 'service' reduces my well-being."

Here we have a difficulty. We have no way of identifying the value of a service that is provided to a person without his request. If we express a willingness to respond to such statements, people will have a motive for faking a lack of interest in the service in order to reduce their tax bills. If a person can point to some characteristic of his circumstances that would tend to support the claim that he does not value the service (for example, if the service is concerts in the park and the person is deaf, or if the services is a sewer line and the person just put in a new septic system), then we may have an obligation to compensate the person for raising his taxes to pay for something that he does not value. But when the person can point to no such special circumstances, he should bear the cost. People should understand that one of the risks of moving into a community is that their fellow citizens may decide to provide a local public service that most people value but they do not. At worst, the aggrieved person will have to bear the costs of moving (including the psychological costs). Thus if any compensation is offered to people who have characteristics that suggest that they do not value a public service that raises the value of their land, the compensation that they receive should have an upper limit of the loss in the value of things attached to the land (the septic system) plus the financial and psychological costs of moving. While we need to watch out for fakers, these are real costs, and a public service is only worthwhile if its benefits exceed its full costs, including losses in the value of capital and in the human satisfaction of those who did not want change.

The fairness of such a tax system might also be questioned from another perspective. If people are required to pay the costs of the public services that they use, then who will pay for schools?

There are two possible answers.

  • One possibility is that people will agree on the value of living in a community where children are well educated, whether they have children or not, and they will find the presence of another child in the community to be a benefit for which the cost of educating the child is not too high a price to pay. In this case, educational expenditures add to the value of land throughout the community, and education can be financed efficiently by land taxes, without requiring students or their parents to pay. Those who pay will be receiving benefits that are equal to what they pay.
  • The second possibility is that there will be a consensus that, whether anyone wants to pay or not, education is a birthright of all children. In this case, the birthright can be financed efficiently by a tax on land, since land does not leave town or work less hard when it is taxed.

But is such a tax fair? Consider first a case in which everyone has the same number of children. Then the taxation of land value to finance education is equivalent to the assertion of an equal claim of everybody to land. And there is a good basis for such a claim. No one made the land. The titles to land that we recognize today generally originated in conquest. Our affection for the words of the Declaration of Independence, that "All men are created equal," in conjunction with the recognition that titles to land are privileges that are created by government, should lead us to the implication that we have an obligation to share the benefits of land equally. As Henry George pointed out (1960 [1879], 403-407), the sensible way to assert equal rights to land is not to divide the land equally, but rather to collect the rent of land and use it for public purposes. A tax on land takes for public purposes only what nature, public services, and the growth of communities produce, unlike taxes on labor and capital, which take what people produce and which people may properly resent having taken from them. Thus, at least when everyone has the same number of children, it is fair to finance education by taxing land.

Now consider how the situation changes when people have different numbers of children. Consider two possibilities.

  • First, it is possible that additional children provide a benefit to everyone in society, because the number of persons with whom everyone can interact will be greater. If this benefit is as large as the cost of education, then there is no unfairness in paying for the education of all children from public funds.
  • On the other hand if the birth of an additional child does not provide general benefits as large as the cost of educating the child, or if it generates crowding costs, then people who have more children than average can rightfully be required to pay those costs.

"But wait!" you may say. "Many people will not be able to afford to pay the cost of educating their children. How can you expect them to do so?"

I would like to turn the question around. If people cannot be expected to pay for educating the children that they ought to be able to have, doesn't that mean that there is some fundamental unfairness in the starting conditions? Is it not the combination of past injustice and current unequal access to natural opportunities that makes us reluctant to require people to pay the full costs of having children? In my conception of justice, we have not adequately compensated for past injustice until we have put people in a position where we are content to oblige them to pay the full costs of their choices.

But there is another potential inequality that needs to be addressed. What if different communities have different amounts of land value per person? Here a distinction in sources of land value must be made. If a community has higher land value because it has built itself into a wonderful place, then it should be allowed to keep that value for itself. On the other hand, if a community has a higher-than-average natural endowment per citizen, then it owes something to communities with lower-than-average natural endowment per citizen. A program of payments among communities to equalize natural endowments per citizen will be both efficient and fair. It will be efficient because in the absence of such a program, people would gravitate to the communities with higher-than-average natural endowments (like Alaska) even when it was socially uneconomic for them to go there. It is fair because it accords with an equal right of all to natural opportunities.

Thus just and efficient local taxation is achieved by the combination of public collection of rent, marginal cost charges for public services, and a program of transfers among communities to equalize natural endowment per citizen.

If time were not limited I would develop the argument that higher levels of government can also be financed by the combination of public collection of rent, marginal cost charges, and transfers to equalize per citizen natural endowments (composed of the pre-development rental value of land, royalties for extracted minerals and renewable resources such as fish, the rental value of exclusive use of the frequency spectrum, and other opportunities provided by nature). But that is an equally intricate argument, so I leave it for another day.

I hope to leave you with a hunger for answers to such questions as how the natural opportunities of different communities might be assessed, how much revenue could actually be collected from rent, and what we would need to do to create a society where everyone forswore the pursuit of opportunities to be the roving bandit, who grabs through taxes what others produce.


Arnott, Richard J. and Joseph E. Stiglitz, "Aggregate Land Rents, Expenditure on Public Goods, and Optimal City Size," Quarterly Journal of Economics 93: 1979, 471-500.

George, Henry, Progress and Poverty, New York: The Robert Schalkenbach Foundation, 1960 [first published in 1879].

Olson, Mancur, "Dictatorship, Democracy and Development," American Political Science Review 87: Fall 1993, 567-76.

Tideman, "Integrating Rent and Demand Revelation in the Evaluation and Financing of Services," in Hiroshi Ohta and Jacques-Fran‡ois Thisse (eds.), Does Economic Space Matter? London: Macmillan, 1993, 133-150.

Tiebout, Charles M., "A Pure Theory of Local Public Expenditures," Journal of Political Economy 64: October 1956, 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, 72-118.

To share this page with a friend: right click, choose "send," and add your comments.

Red links have not been visited; .
Green links are pages you've seen
Top of page
Essential Documents
to email this page to a friend: right click, choose "send"
Wealth and Want
... because democracy alone hasn't yet led to a society in which all can prosper