De Jasay argued that the main justification for the state—that it produces or finances so-called “public goods” (e.g., parks, streets, public security)—is economically flawed. Even if we assume that all individuals are self-interested and even selfish, some of them want some public goods enough that they would be willing to pay something to obtain them, notwithstanding that others would free ride on their contributions. Hence, de Jasay claimed, there is no need for an overall social contract to enforce private contracts and property. Let’s test his arguments and, in the process, review important concepts in the economics of government.

The problem of public goods / In game theory, a pure coordination game is an interaction where the acting parties have a common course of action that each considers optimal: driving on the same side of the road (either left or right, but not both), trading instead of fighting with a just-encountered stranger who is equally armed, and so forth. In such cases, coordination through agreement (contract, whether formal or informal) is generally possible without a superior coordinator because it is in the mutual interest of both parties. Contracts would presumably be the main coordination instrument in a state of nature—that is, in anarchy.

A contract needs to be enforced. Spot contracts where exchange is simultaneous and even some forward contracts (especially when future performance will be simultaneous) are self-enforcing when each of the two parties has an interest in the exchange taking place. Generally accepted conventions play a role: before the modern state, councils of elders, priests, guilds, towns, and merchant jurisdictions helped as enforcers of contracts. Only later did the king, with his superior force, succeed in imposing his own courts and public law.

The hard case lies in interactions represented by a different game-theoretical model, the so-called “prisoner’s dilemma” (PD). A simple PD is an interaction in which it is in the interest of each of two parties to defect (not cooperate) whatever the other does, in the hope of free riding on the cooperation of the latter. Defection is the “dominant strategy” for each player.

As an example of the basic PD problem (and a mnemotechnic device), imagine that the “game” consists in deciding whether you and your neighbor will voluntarily contribute to build a sand barrier that will divert the nearby river and protect your two properties in case of flooding. The best outcome for you is if only your neighbor pays for, or builds, the barrier and you free ride on his contribution. Your payoff from the game is F for “free ride.” The second-best payoff for you is that the barrier is built and both you and your neighbor pay your fair share (B for “both”), because by hypothesis your own benefit will still be higher than your cost. Your third-best outcome is if neither contributes (N for “neither”): you may get flooded, but you avoid paying for the barrier. And the worst that can happen to you is that you must pay the whole cost because the neighbor defects (that is, he doesn’t send the check); you are the sucker (S) because you pay for both your benefit and your neighbor’s. Your neighbor has the same preferences mutatis mutandis. The structure of the PD game in terms of payoffs for each player is thus F > B > N > S (the “greater than” operator means “preferred to”). The free rider problem is the following: Since both you and your neighbor individually prefer to free ride and hope the other will pay, neither of you pay, and this “public good”—which you both want—is not produced.

The mainstream economic definition of a public good refers to a good (or service) that is non-rival in consumption (or “indivisible”) and non-excludable. Non-rivalry means that such a good can be simultaneously consumed by several individuals once it is available to one. The flood barrier or an old-style lighthouse meets this definition; so does, for a larger public, national defense or criminal deterrence. Non-excludability means that consumers cannot be excluded from its benefits once the good is made available to some: free riders who don’t contribute to financing a public good can still enjoy it. Thus, everybody free rides and the good is not produced. Public goods are an instance of PD.

Everybody wants a public good but nobody will voluntary pay for it; hence the necessity of the state to provide it. Given the impossibility of reliable contracts to produce public goods, a broad social contract is required that will force individuals to contribute for their own good according to their own preferences. The social contract will receive unanimous agreement—or so goes the mainstream argument. But in Social Contract, Free Ride (as in many articles), de Jasay argued against both the idea that the production of public goods requires the state and that the state can be seen as the product of an overarching social contract.

Voluntary production / De Jasay denied that the PD structure of preferences necessarily characterizes public goods. Although each self-interested individual naturally prefers to free ride if he can, he may value the public good to such an extent that he would rather pay for it himself than go without it, even if he knows that others may free ride on his voluntary contribution. With this ranking, no individual has a dominant strategy like the non-cooperation strategy of the PD model. This alternative game, which de Jasay called the “straddle” game, is also known as the “chicken game.” In this game, some individuals will volunteer for the role of sucker if they think it’s the only way for them to enjoy the public good. Each individual’s preferences are F > B > S > N, instead of the PD preferences seen above. As indicated by the order of S and N, the players prefer to be suckers rather than be deprived of the public good.

De Jasay offered a proof that this structure of preferences is both a necessary and a sufficient condition for some suckers to probabilistically offer a voluntary contribution or “subscription.” In short, if his utility from the public good is high enough compared to its cost, a potential sucker may offer to subscribe, knowing that the worst that can happen is that the project does not gather enough subscribers and is canceled, in which case he will get his money back. Some readers may find de Jasay’s formal proof inconclusive, if only because it relies on the cardinal utility of a “representative member” of the society under consideration. But the proof does point to the possibility that a sufficiently large group of individuals may want the public good with enough intensity to be willing to pay for it even if they risk that others will free ride on their contributions.

For each individual, volunteering to be a sucker is a probabilistic decision that depends on how many other suckers he thinks will come forward. The more of them there are, the less likely one individual’s contribution will prove decisive for the minimum financing required, and the less likely he will be to volunteer. The fewer volunteers he expects to subscribe, the more likely an individual’s contribution will be decisive for the minimum financing required, and the more likely he will be to volunteer. So, there is some probability that some public goods would be produced in an anarchic society.

This probability will be higher or lower depending on the extent to which expectations are coherent among individuals. It is only in rather paradoxical extremes, de Jasay argued, that we are virtually certain that the public good would not be voluntarily provided. For example, if the probability of everybody subscribing is generally thought to be close to 1, I have little reason to subscribe, and the same for everybody else; so, the subscription for the public good will not raise any money. But even in this last case, perhaps another subscription will be launched, by the same or another business, after it is realized that everybody regrets the demise of the project.

That rational and self-interested individuals may choose to volunteer for the sucker’s role seems to be confirmed by experience. Even in our state-dominated societies, many individuals voluntarily contribute to financing “public goods” by contributing to charities, concert halls, etiquette, and so forth. De Jasay gave the example of “the first sewage systems of new towns in the English Midlands and Lancashire in the Industrial Revolution.”

Mainstream economists counter that even if some public goods were financed and produced in an anarchic society, their provision would be below the “optimal” level. De Jasay recognized—along with other economists in the broad Public Choice or Austrian traditions—that there is no way to know if a level of production is “optimal” or not except by letting free markets balance demand and supply. Transaction costs are real costs, and the state carries its own. The claim that, on free markets, production of something is not optimal is, de Jasay noted, “only on the authority of someone’s say-so.”

It is thanks to coercion that the state makes the provision of public goods certain. But, as suggested, the production of public goods does not necessarily require the surrender of free choice, including the freedom to free ride. And there is no general reason to get angry with free riders, because free riding is inseparable from social interaction. “In the last analysis,” de Jasay wrote, “all arm’s‑length social coexistence and cooperation that is not exchange under contract carries within itself an element of potential abuse by free riding.”

These conclusions would be reinforced if motivations other than self-interestedness are in play. Toward the end of the book, de Jasay imagined an “ethics turnpike” with many exits or forks offering opportunities to contribute to public goods out of moral motivations such as humanist-altruist, prudent-Kantian, and clannish-tribal. But we have seen that, even with a self-interested homo economicus, they can be produced, at least to a certain extent.

How “fairness” produces free riders / The social contract and the state emerge not because of the absence of cooperation in the production of public goods, but from the demand for greater “fairness” in their financing. Fairness is invoked against justice, the latter being derived from (voluntary) contracts. In reality, however, the state will not produce much fairness with its public goods.

De Jasay argued that the public goods problem is of concern not because these goods are non-excludable (which is, anyway, a matter of degree and cost: a street or a park can be excluded with a sufficient number of barriers, toll booths, guards, or electronic equivalents) but because they are de facto not excluded. The essential feature of a public good is “the possibility of benefiting without contributing” (de Jasay’s emphasis). What the state achieves is the “uncoupling of individual benefits from contribution.” A public good becomes whatever the state decides it is. (James Buchanan offered a similar idea: for him, the social contract includes rules that determine what public goods will be.)

An interesting consequence follows: “While the intent of the social contract is to suppress free riding, its actual effect is to open up an altogether new ground on which it thrives with impunity,” de Jasay pointed out. In the state of nature, free riding is limited by the risk that the good will not be available at all, but there is no such built-in check under the state. Once the state can create public goods, everybody’s dominant strategy is to ask for more of those he likes—that is, whose benefits are higher than his diluted tax contribution. Instead of free riding on others’ voluntary contributions, free riders in a society governed by the state exploit coerced taxpayers by obtaining “free” stuff for themselves. Not surprisingly, private goods that are made public by the state—healthcare is a good example—become overcrowded as much as more conventional public goods like streets or roads. The more overcrowded is a public good, the more intense are the political pressures to produce more of it.

This way, the social-contractual state naturally moves toward the “Maximal State” instead of the “Minimal State,” “without being ‘chosen’ in any proper sense of the word, and without anybody in particular being noticeably pleased by it.” Buchanan and the school of constitutional political economy argued that rational (virtual) signatories of the social contract will constrain the state with constitutional rules limiting its domain. De Jasay challenged that idea. His counterargument is that the state created by a social contract is designed to be non-unanimous and to prevent some individuals from vetoing what a future majority may want. In a sense, the social contract can only be unanimous regarding a future discretionary state.

Social Contract, Free Ride also challenges the idea that the quest for fairness underlying the social contract will yield greater fairness in redistribution. The last words of the book are that this system “self-evidently leads to less fairness considered as the measure of immunity of each from the unrestrained will of all, and as the measure of the responsibility of each for the consequences of his own actions.” The state replaces contract and justice by command and arbitrary “fairness.”

Conclusion / Did Anthony de Jasay prove that public goods—in the sense of goods that automatically benefit many people whose exclusion is costly—could be voluntarily financed by self-interested individuals (and voluntary groups) in a stateless society? That their provision doesn’t require depriving free riders of their liberty? That the state is not only redundant but positively harmful? That a social contract cannot justify the state? He certainly offered serious arguments. I find his mathematical demonstrations underwhelming; mathematical economist and independent scholar Daniel Kian McKiernan describes them as “at best heuristic.” De Jasay’s demonstrations in English, though, are good enough to be taken seriously.

I think the jury is out on the larger question of “ordered anarchy”—whether it requires the state, as Buchanan believed, or is defeated by it, as de Jasay thought. The latter certainly showed that the state, even if created by a tacit social contract, is far from being the easy and obvious solution that many think it is.

De Jasay espoused classical liberalism together with what he believed to be its natural anarchist extension. In Social Contract, Free Ride, he defined liberalism as “a broad presumption of deciding individually any matter whose structure lends itself, with roughly comparable convenience, to both individual and collective choice.” We can certainly agree with this rather moderate presumption as a general principle.