All Commentary
Friday, February 1, 2002

Prisoners’ Dilemmas and Cooperation

Free Markets Reward Cooperation

Economics is largely about how people cooperate so each can best pursue his or her objectives, whatever they may be. Decentralized market-based economies are wealthier than those based on central direction because markets facilitate the communication of the information and motivation necessary for people to cooperate, while central direction always censors that communication. Even market-based economies sometimes lack markets for important resources and the result is waste and inefficiency due to the absence of cooperation. For example, the pollution problems I discussed in previous columns are caused by the lack of markets in the use of the environment for disposing waste. Without those markets people don’t consider the concerns of others when deciding how much to pollute. Creating artificial markets is the best way of facilitating the cooperation needed to reduce pollution to acceptable levels as cheaply as possible.

But no matter how well markets work, and how extensively they are operating, we remain in a world of scarcity, with people wanting more than they have and often seeing ways to acquire more by behaving in noncooperative ways–ways that create more losses for others than gains for themselves. There is almost always a tension between cooperative and noncooperative behavior, and this column examines that tension in the context of a simple model widely known as the prisoners’ dilemma. By looking at this dilemma we can understand the interaction between politics and economics and the tendency for government action to encroach on market action even though the result is destructive of wealth and, more important, liberty.

To Confess or Not to Confess

The problem I am about to illustrate is very general, but it takes its name from an example involving two prisoners known to have committed a serious crime but who can be convicted only of a relatively minor crime without a confession from at least one of them. The prisoners are separated and each told the same thing: “If you confess to the serious crime, you will receive a sentence of ten years if your accomplice also confesses, but only two years if he doesn’t. But if you refuse to confess to the serious crime you will receive a sentence of 15 years if your accomplice does confess and three years (for the minor crime) if he also refuses.”

The possibilities the prisoners face are shown in the nearby payoff matrix, where the first number in each cell is the sentence Prisoner A receives and the second number the sentence Prisoner B receives. As easily seen, the collective interest of both prisoners is best served when neither confesses, in which case they will serve a total of only 6 years. The worst thing for their collective interest is for both to confess, since this results in their serving a total of 20 years. Yet no matter what each thinks his accomplice will do, the best thing for him to do is confess. For example, if A thinks B is not going to confess, then if he doesn’t confess he gets 3 years, but only 2 years if he does confess. Or if A thinks B is going to confess, then if he doesn’t confess he gets 15 years, but only 10 years if he does confess. And the same situation faces B. So from the perspective of each, the best thing to do is confess even though this leads to the worst outcome from the perspective of both. The action that is individually rational is collectively irrational.

Don’t confess
Don’t Confess
3 years/3 years15 years/2 years
2 years/15 years10 years/10 years

The prisoners’ dilemma is an example of the tension between cooperation and noncooperation that is unavoidable. For example, pollution problems are prisoners’ dilemmas. We would all be better off if everyone cooperated by polluting less. But for each of us it makes more sense not to do so. Regardless of what others do, it doesn’t pay me to reduce my pollution, since the benefit will go primarily to countless others, while I incur all the cost and inconvenience. The reader is encouraged to construct a payoff matrix like the one above showing the payoffs to two individuals from the four different pollution-reduction possibilities (with a higher number representing a higher payoff rather than a longer sentence).* The point of pollution policy is to change the payoffs to eliminate the prisoners’ dilemma by making it pay for each individual to reduce pollution no matter what others do. Creating markets in pollution permits does exactly this, which is not surprising since markets excel at promoting cooperation by eliminating prisoners’ dilemmas.

Consider how eliminating markets destroys cooperation by putting everyone in a huge prisoners’ dilemma. Marxist ideology claimed to substitute the rule “From each according to his ability, to each according to his need” for the “exploitation” of the market. Forgetting that without markets we cannot determine what people’s abilities and needs are, such a rule can be implemented only with brutal force, because without force the rule creates a prisoners’ dilemma: having lots of needs and no ability is the only sensible choice for each person. Whether each person thinks that others will produce to the best of their ability or live off the efforts of others, the rational action is the latter. But obviously this leads to a situation that is collectively irrational, with few abilities being used and few needs being satisfied.

But surely not everyone will shirk since many, probably most, of us want to do our part by being a productive member of society. True, at least initially. But even if only a few people exploit the prisoners’ dilemma initially, soon others will see that they are being treated as suckers by the shirkers. As the number of shirkers increases, it becomes increasingly obvious that cooperation is for suckers, and the productive process unravels. Without markets to eliminate this prisoners’ dilemma by rewarding cooperation, only government force can prevent a breakdown in productivity. But while government can force people to work, it cannot provide either the information or motivation people need to work in their most productive activities and to do so with the creative zeal and cooperative attitude that is characteristic of market activity.

A destructive prisoners’ dilemma is eliminated by the information and motivation provided by markets. But the prisoners’ dilemma comes in many forms and, as a practical matter, markets do not eliminate them all. Because of this, the prisoners’ dilemma is commonly used to justify government and the coercive power it exercises. I will discuss this justification next month.

  • Dwight R. Lee is the O’Neil Professor of Global Markets and Freedom in the Cox School of Business at Southern Methodist University.