The Obviousness of Anarchy: Public Goods
Written by John Hasnas, Associate Professor, Georgetown University, J.D., Ph.D, LL.M
Continued from The Obviousness of Anarchy: Internalizing Externalities
Supporters of government claim that government is necessary to produce “public goods;” goods that are important for human well-being but either cannot be produced or will be under-produced by the market. Public goods are goods that are both non-rivalrous in consumption; that is, its use by one person does not interfere with its use by others, and nonexclusive; that is, if the good is available to one person, it is available to all whether they help produce it or not. Supporters of government argue that such goods cannot be produced without government because, due to the free rider and assurance problems, individuals will not voluntarily contribute the capital necessary for their production.
The free rider problem refers to the fact that because people can enjoy public goods without paying for them, many will withhold their contribution to the goods’ production and attempt to free ride on the contribution of others. The assurance problem refers to the fact that in the absence of some assurance that others will contribute enough to produce the good, people are more likely to regard their own contribution as a waste of money and withhold it. Therefore, government is necessary to ensure the production of important public goods.
The proper response to the argument that government is necessary to produce public goods is: Like what? Like lighthouses? The light they provide is available to all ships and its use by one does not impair its value to others. But wait, lighthouses can be and have been supplied privately.  Like radio and television? A wag I know likes to say that he does something impossible every night by watching commercial television. After all, television signals are non-rivalrous in consumption and nonexclusive. Therefore, they cannot be produced by the market. Like the internet? But wait, that is privately funded also.
Perhaps like police and courts? Theorists frequently argue that police services and courts are public goods that must be supplied by government. With regard to police services, for example, the argument is made that:
Security of person is to a large degree a collective good. . . . [A]n important part of the service provided by public police and systems of criminal justice generally is to deter potential violators from harming people. And this deterrence is an indivisible nonexcludable good to neighbors and visitors. . . . In addition to deterrence, there may be the benefits that follow from incarceration of the thief—namely, incapacitation—benefits that are also indivisible and nonexcludable.
Social order, at least security of persons and possessions, then, is to a considerable degree a collective good. Accordingly, to the degree that this is the case, social order may not be efficiently provided in the absence of a state. 
Similarly, with regard to courts, it is argued that because the existence of definite and widely-known rules of behavior provides a nonexcludable benefit to all, private courts lack an incentive to establish the clear precedents that give rise to rules. Indeed, because clear precedents “would confer an external, an uncompensated benefit, not only on future parties, but also on competing judges, . . . judges might deliberately avoid explaining their results because the demand for their services would be reduced by rules that, by clarifying the meaning of the law, reduce the incidence of disputes.” 
Hence, government courts are necessary for the development of rules of law.
These are perfectly logical theoretical arguments belied only by the facts of reality. The evidence that police services and courts are not public goods is that, like lighthouses, television, and the internet, they have been supplied non-politically for most of human history.
It is true, of course, that if government exists and creates areas of unowned, politically-controlled, property that no private party has an interest in maintaining, police services are likely to be under-produced in these locations. Policing of this “public” property may indeed have to be supplied by the government. However, this is not because police services are a public good that cannot be supplied by the market, but because police services will not be supplied when the market has been suppressed by the government. And although it is certainly true that private police services produce an uncompensated positive externality in that their deterrent effects make even those who have not paid for them more secure, this can hardly be a reason for believing that such services will not be produced. It is actually quite difficult to think of any useful activity that does not produce some uncompensated positive externality. My using deodorant and going about clothed certainly do, but government is not required to pay me to induce me to bathe and dress. Further, it is at least odd to argue that a system of competitive courts will not produce rules of law in a common law legal system in which the rules on which our civilization rests actually arose out of just such a system. 
Like national defense? National defense is perhaps the archetypical public good. The security it provides is both non-rivalrous in consumption and benefits all members of society whether they pay for it or not. Can national defense be adequately supplied without government?
If ‘national defense’ refers to the type of military expenditures associated with contemporary national governments, the answer is an obvious ‘no.’ Once a state becomes invested with the power to expropriate the wealth of its citizenry to provide for national defense, almost any desired expenditure begins to look like a requirement for national defense. Before long propping up Southeast Asian dictators and overthrowing Middle Eastern ones are being characterized as urgent national defense concerns. The fact that there is no non-governmental way to raise sufficient capital to realize this conception of national defense proves nothing about the viability of anarchy, and, in fact, serves as one more argument in favor of markets.
However, if ‘national defense’ refers to only what is strictly necessary to protect the citizens of a nation against outside aggression, I am willing to admit that I do not know the answer to this question. I am not discomforted by this admission, however, because as I said at the outset, the question of national defense is, as a practical matter, a trivial one. No one believes that we can transition from a world of states to anarchy instantaneously. No reasonable anarchist advocates the total dissolution of government tomorrow. Once we turn our attention to the question of how to move incrementally from government to anarchy, it becomes apparent that national defense would be one of the last governmental functions to be de-politicized. If my argument for anarchy is flawed and anarchy is not a viable method of social organization, this will undoubtedly be revealed long before doing away with national defense becomes an issue. On the other hand, to the extent that the gradual transition from government to anarchy is successful, the need for national defense continually lessens.
Consider what it would mean for a nation to seriously undertake a process of de
politicization. Every reduction in the size and scope of government releases more of the creative energy of the population. The economic effects of this are well-known and are currently being demonstrated in China. As economists point out, revolutionary change can be wrought by marginal effects. Even a slow process of liberalization that is sustained over time will produce massively accelerated economic and technological growth. And the increase in freedom and prosperity in this nation would have profound external effects as well. Many of the bravest and most industrious residents of more repressive nations would attempt to immigrate to the liberalizing one, and some other nations would learn by the liberalizing nation’s example and begin to copy its policies.
As the economic and technological gap between the liberalizing nation and the rest of world widens, as the rest of the world becomes more dependent upon the goods and services manufactured and supplied by that nation, and as a greater number of other nations are moved to adopt liberalizing policies themselves, the threat the rest of the world poses to the liberalizing nation decreases. Evidence of this is supplied by the demise of the Soviet Union. Radical regimes and terrorist organizations may constitute a serious and continuing threat, but consider it in historical context. Such a threat is considerably less serious and less expensive to address than the threat of thermonuclear war.
Recall that we are considering the cost only of protecting citizens against aggression, not the cost of foreign adventures or “pre-emptive” warfare. How significant a threat of foreign invasion does the United States currently face? How much of its “national defense” spending is actually devoted to preventing such invasion? After years or decades of continual and sustained reduction in the size of government, how much wider will the economic and technological gap between the prenatal anarchy and the more repressive nations be? How much more sophisticated its defensive technology?
How much more dependent will the repressive nations be on its goods and services? Let a nation begin to tread the path toward anarchy and by the time the question of whether national defense is a public good that must be supplied by government becomes relevant, it is very likely to be moot.
Continued in The Obviousness of Anarchy: Conclusion
 See Ronald H. Coase, The Lighthouse in Economics, 17 JOURNAL OF LAW AND ECONOM ICS 357 (1974).
 CHRISTOPHER W. MORRIS, AN ESSAY ON THE MODERN STATE 60-61 (1998).
 See William M. Landes & Richard A. Posner, Adjudication as a Private Good, 6 JOURNAL OF LEGAL STUDIES 235 (1979).
 For the true intellectuals among my readers who simply cannot accept that facts should be allowed to undermine a perfectly good theoretical model, I refer you to DAVID SCM ID TZ, THE LIM ITS OF GOVERNMENT: AN ESSAY ON THE PUBLIC GOODS ARGUM ENT (1991). Schmidtz explains how the assurance problem can be handled by the assurance contract or money back guarantee and how the free rider problem can be cabined to a relatively small number of cases in which using coercion to produce the public good is ethically questionable.