Public goods are freely accessible to all members of a given public, each being able to benefit from it without paying for it. The reason standard theory puts forward for this anomaly is that public goods are by their technical character non-excludable. There is no way to exclude a person from access to such a good if it is produced at all. Examples cited include the defence of the realm, the rule of law, clean air or traffic control. If all can have it without contributing to its cost, nobody will contribute and the good will not be produced. This, in a nutshell, is the public goods dilemma, a form of market failure which requires taxation to overcome it. Its solution lies outside the economic calculus; it belongs to politics.

1. Exclusion Cost

Access to a private good is controlled by its producer or owner by a variety of devices ranging from shop counters, safes, walls and fences to measures against theft, robbery, fraud, illicit copying and breach of contract. The cost of these devices and measures is the exclusion cost of the good. Every good is private or public according to whether exclusion cost is or is not incurred in making it available. A public good is distributed freely to all comers from a given public, avoiding the exclusion cost that would keep it private. This saving is the ‘productivity of publicness’.

Given sufficient imagination and clever technology, every good can be excluded at some cost. Arguably, some would be very awkward to exclude, but none is intrinsically ‘non-excludable’, i.e. doomed to be a public good. By the same token, every good, whether private or public, has many more or less imperfect substitutes that may also be private or public. Thus, contrary to received theory, a more general view tells us that while no good is intrinsically public, the higher is its exclusion cost and the more imperfect are its substitutes, the more efficient it is to provide it publicly.

2. Social Preference for Non-Exclusion

There is one type of exclusion cost that is more important by far than all the rest in putting a good in the public category: it is social preference. It is intangible and is only revealed by the choices it inspires. A pure example is a children’s playground. Access to it is excludable at low cost by a fence and a ticket collector at the gate. However, society would suffer deep moral embarrassment if rich children could use the playground but poor ones could only watch them from the outside. Therefore real exclusion cost would be unbearably high, and children’s playgrounds are provided as public goods.

There are other, less pure but quantitatively far more important examples. One is free universal education Most countries provide it to age 16, some to university degree level. In this case, technical-logistical exclusion cost would be quite low (indeed, in a broad sense negative as exclusion would permit student selection, and that would in turn lower production cost), but social ethics would not tolerate the exclusion of poor, dumb and sub-scholarship standard pupils. With education becoming a public good affording free access, the share of public goods in the national product expands vastly. Organising health care in the form of a free-access public good on the pattern of the British National Health Service expands the domain of public goods even further and multiplies the gravity of the public goods dilemma.

However, it is perverse to argue that this is a true case of market failure. The dilemma presents itself, not because the market cannot cope, but because society does not choose to entrust the matter to it. It may have quite worthy moral reasons for doing so. But it must not be overlooked that since public goods can be consumed at zero marginal cost, a tendency is created to their chronic overconsumption. This, in turn, involves an encroachment of the public upon the private sector and a cascade of adverse indirect consequences.

3. Free Ride Or Sucker

Received ‘market failure’ theory has a false perspective not only in characterising some goods as intrinsically public rather than made public by social choice reacting to intangible exclusion costs. It also mistakes the public goods dilemma for a version of the prisoners’ dilemma. It then finds that like the prisoners’ dilemma, the public goods dilemma has only a non-cooperative equilibrium solution.

Individuals, unless forced to pay taxes, have two choices with regard to a public good: to contribute or not to contribute to its cost while enjoying its benefit. The non-contributor gets a free ride, the contributor is a sucker. For the standard theory, the conclusion is easy: there will be few or no contributors. The market will fail to produce the public good, particularly if it is indivisible or ‘lumpy’, so that a minimum number of contributors is needed to produce even a single ‘lump’ of it (e.g. if the public good ‘education’ comes in ‘lumps’ no smaller in size than a schoolhouse and teacher)

Consider, however, the would-be free rider who must weigh the attraction of a free ride against the risk that by withholding his contribution, he will cause the total of contributions to fall short of the minimum outlay needed to render the good really ‘public’ freely accessible to all and satisfying the accepted criterion of publicness, namely ‘non-rivalry in consumption’. This criterion means that consumption of it by one person does not reduce the amount available to any other person.

Consider likewise the hesitant sucker who must weigh the opportunity cost of contributing against the chance that his contribution will be the one needed to raise total contributions over the threshold of the minimum required for the ‘lump’ of public good needed to permit access to it by the marginal consumer.

In the face of these two pairs of possible outcomes, neither is the free rider strategy unquestionably the best, nor the sucker strategy unquestionably the worst. Which of the two is the rational choice depends on the subjective probability each potential contributor attaches to others going for the free rider or the sucker choice, as well as the value he attaches to having the public good instead of resorting to private substitutes.

The critical values of these variables depend on a complicated set of factors that cannot be detailed in a brief essay. However, it is intuitively fairly clear that there is nothing foredoomed about public goods in general. Whether a good can be ‘made public’ by voluntary contributions depends on how rational calculation and anticipation of the behaviour of others leads to a division within a group between free riders and suckers. Each of the two possible social roles, the free rider and the sucker, leads to a pair of uncertain alternatives. For the free rider they are the free ride (the best) or failure of the public good (the worst). For the sucker, it is that he contributes like everyone else (the second-best) or that he contributes when some others do not (the third-best). In the standard theories of market failure, the free rider strategy is ‘dominant’—it is always the best whatever anyone else may do. In effect, however, the pair ‘best or worst’ is intrinsically neither superior nor inferior to the pair ‘second-best or third-best’. Rationally, one pair is chosen depending on the probability that one member of the pair rather than the other member will in fact turn out to be the case. The problem becomes simply a case in the theory of risky choices.

Public goods can thus be brought back under the calculus that guides homo oeconomicus. The provision of public goods does not presuppose collective choice that overrules individual ones by the brute force of politics. Those who instinctively mistrust collective choices and trust that reasonable solutions emerge from free individual choices need not feel browbeaten by the ‘market failure’ argument.


*Anthony de Jasay is an Anglo-Hungarian economist living in France. He is the author, a.o., of The State (Oxford, 1985), Social Contract, Free Ride (Oxford 1989) and Against Politics (London,1997). His latest book, Justice and Its Surroundings, was published by Liberty Fund in the summer of 2002.

The State is also available online on this website.

For more articles by Anthony de Jasay, see the Archive.