26.2: Real Life Prisoner’s Dilemmas

There are many situations like the prisoner’s dilemma in the real world, many situations in which the pursuit of narrow self-interest leads to an outcome that is worse for everyone. Since this pattern is so clearly displayed in the Prisoner’s Dilemma, such situations are often called Prisoner’s Dilemmas as well.

International Relations

For several decades, the United States and the Soviet Union were locked in a nuclear arms race. This cost vast amounts of money and led to a situation where the complete annihilation of the world was a real possibility. Both countries would have preferred mutual disarmament to the spiraling arms race. But like our prisoners, they were trapped in a dilemma.

Each reasoned the same way. Either the other side will keep its nuclear arsenal or it won’t. If they keep their weapons, then we must keep ours (so we won’t be at their mercy). On the other hand, if the other side disarms, it’s still good for us to have nuclear warheads so that they will have to stay in line. A perfectly rational line of thought leads to an outcome that is worse for both parties. The situation is depicted in Figure 26.2.1; the figure below; here the top entry in each cell indicates the outcome for the U.S., and the bottom entry in the cell indicates the outcome for the U.S.S.R.).

Figure \(\PageIndex{1}\): The Arms Race

The general pattern in a prisoner’s dilemma can arise when more than two parties are involved. The above line of reasoning would be essentially the same if there were more nations, say the United States, Russia, and China, engaged in a nuclear arms race. All three countries would be better off if everyone disarmed, but the situation makes this almost impossible. Indeed, the basic structure of the situation would be the same for any number of parties.

Cartels face a similar problem. The countries in OPEC (an oil cartel) reach an agreement to limit their production of oil to keep the price reasonably high. But it is impossible to determine precisely how much oil any country sells, and so each country will have an incentive to cheat.

Why? Because each country can reason as follows. Either most of the other countries will stick to their production quota, or they will cheat. If most cooperate, the price of oil will remain high, and we are better off producing more to bring in extra money. On the other hand, if most of the other countries cheat, we are suckers if we don’t; in this case, we had better get as much oil on the market as fast as possible, before the price drops through the floor.

This leads to a situation where many of the countries are likely to cheat, with the result that the price of oil goes down and they all suffer. Each of the parties would have been better off sticking to their original agreement, but strong pressures to defect are inherent in the situation. Many other international situations, e.g., those involving trade negotiations and world standards for pollution reduction, generate similar difficulties.

Public Goods and Free Riders

Prisoner’s Dilemmas are a danger in any situation involving public goods. The following four features are each necessary (and together jointly sufficient) for something to be a public good:

1. Action by many of the members of a group is required to provide that good; the action of just a few members isn’t enough.
2. If the good is provided, it will be available to everyone in the group, including those who don’t contribute to its production.
3. There is no way (at least no practical way) to prevent noncontributing members from enjoying the benefits of the good.
4. The individual’s contribution is a cost (e.g., in time or money or energy or emotional involvement) to that individual.

National defense is a public good. Many people must do their share (either by serving in the armed forces or by paying taxes to fund the defense budget). But once a country has a defense system, everyone will benefit from it, whether they contributed their share or not. And in many cases, it may not be clear whether a person does their share (they may be able to cheat on their taxes, for example). In fact, there are many things that approximate being public goods.

Energy conservation is also a public good. Many people must do their part if a meaningful amount of water, oil, or gas is to be conserved. If enough people cooperate, they all enjoy the benefits of more and cheaper energy sources. This will be so even if a few people don’t do their share, however, and in many cases, it isn’t possible to determine who does pitch in and who doesn’t.

Other examples of public goods include population control, protecting the environment, preventing inflation, public health programs like immunizations, and public education. Not everyone has to contribute for us to have public goods. We all expect there will be some slackers, but problems arise if there are too many. Public goods tend to generate two sorts of difficulties: the problem of free riders and the problem of assurance.

The Problem of Free Riders

In some cities, you pay for public transportation on the honor system. Provided most people pay, the system will work. But since so many people are involved, it will continue to work if a few people take a free ride at the expense of those who do not. This provides a nice metaphor for those who benefit from something without doing their share to produce it; these free loaders are said to be “free riders” on the efforts of those who do.

Rational individuals may easily conclude that they should take a free ride on the contributions of others because of the following (valid) bit of reasoning:

1. Either enough others will contribute for the good to be produced or else not enough will contribute.
2. If enough others do contribute, the good will be produced (my little input won’t be missed), and so it is in my interest not to contribute.
3. If not enough people contribute, then the good will not be produced (despite the trouble I go to in doing my share), so it is in my interest not to contribute.

Since each person can reason in this way, many are likely to conclude that it is not in their interest to contribute. If so, the good will not be produced (even though it would be to the overall benefit of people to have it). The pattern of reasoning is the same as that in prisoner’s dilemma; cooperation is necessary for the best results, but those who do not want to pay the price can enjoy the benefits of public goods by exploiting the cooperativeness of those who produce them (see Figure 26.2.2).

Figure \(\PageIndex{1}\): Taking a Free Ride

The Assurance Problem

A person may be quite willing to forego a free ride and contribute if she could be assured that enough others would also contribute for the good to be produced. But since she realizes that many people may fail to do their share (perhaps because they want a free ride), she lacks assurance about their actions. This provides an incentive for her not to contribute either.

The assurance problem becomes especially acute if some people are detected cheating. As distrust sets in, assurance plummets, and more people defect. This can trigger a downward spiral, since it will become more obvious that more people are shirking their responsibilities, and the motivations for others to do their share decreases.

The Tragedy of the Commons

We will now consider four social dilemmas, some of which are all too real.

The Commons Goes Down the Drain

In many nineteenth-century English villages, the families shared a modest parcel of land, the Commons, on which they grazed their cattle. They sold milk and butter and occasionally beef, and everyone made a reasonable living. But the more cattle a family had, the more it prospered, so each family had a motivation to add more.

If only one or two families added a couple of cows, it didn’t really damage the Commons. But every family could reason in this way, and they all added more cattle. At some point the Commons reached its carrying capacity; it was overgrazed and much of the grass died. As a result, cattle barely survived, some died, and profits dropped. When each family showed restraint, they all did well. When they each pursued their own interest, the result was worse for all.

Voting

Most of us put a very high value on democracy; the ability to vote in free and open elections is one of the most important features of our country. But democracy only works if a reasonable percentage of eligible voters do so. Turnout is already low, typically less than fifty percent, but if it dropped to ten percent, or three percent, there would be little reason to view elected officials as legitimate representatives of the people.

So, most of us want enough people to vote to maintain our democratic institutions. But it takes time and energy to go to the polls, and in a national election, one person’s vote has virtually no chance of making a difference. Of course, as we have seen several times since Bush v Gore, a few voters can sometimes have a very large impact.

Still, it is probably almost as likely that you will be run over by a steam roller on the way to the polls as it is that you will cast the winning ballot in a presidential election. And you are certainly more likely to be killed in a traffic accident than to determine the winner. So why should you vote? Why not sit back with a beer and take a free ride on the backs of those who take the trouble to go to the polls?

Campaign Costs

Raising the enormous sums of money needed to run for national office requires a great deal of time, and many people find it distasteful. Most candidates would prefer that everyone spent less on political campaigns. But each candidate also realizes that others are likely to spend more than they agree to spend, so there is pressure on everyone to spend more. This escalates, and we end up with our current situation, where big money plays a major role in who gets elected.

Salary Caps

Most owners of professional football teams would prefer a situation in which no team paid vast, multi-million dollar salaries to their players. But efforts to cooperate often collapse, because sooner or later some team will spend a lot on some player who is just too good to pass up. Because the owners realize that they can’t trust themselves, they placed a salary cap on what teams could spend.

The State

Our largest cooperative endeavor is working together to maintain a society and government. Once a state is established, there are various mechanisms that help keep it going. But given our recent examples, we may wonder how a state could be established in the first place, and it is even possible to wonder how it keeps working, once it has been set up.

Of course, some states do break down, but many do not. How is this possible? This was the first collective-action problem to occupy philosophers, and figures like Thomas Hobbes, John Locke, and Jean Jacques Rousseau used it as a starting point for their important discussions of the justification of the state (Locke’s work, for example, was a major influence on the founders of this country).

In a large society like ours, a certain amount of free riding is bound to go undetected. The only way to prevent it would be to set up excessive monitoring and other oppressive measures, and this cure would be worse than the disease. This means that everyone will have numerous opportunities to take a free ride on the efforts of others, and so each person can reason as follows. Either enough people will cooperate to keep the state running (by being good citizens, obeying the laws, paying taxes, etc.), or else not enough will. If enough people do their part, then my little bit won’t be missed, so I should take a free ride in any situation where I can get by with it. And if not enough people do their share, my tiny efforts won’t keep the state running. Either way, I’m better off cutting corners whenever I can.

Closer to Home

Prisoner’s Dilemmas also arise in smaller groups. Consider a commune or large family that shares a big house. All of them would prefer a clean house to a messy one, and if each of them cleaned up the mess they made, the house would be spotless and neat. But the house is large, and if only one person cleans up their mess the house will remain almost as untidy as if they did nothing.

Of course, they could clean up other people’s messes too, but then others are free riding and this is likely to lead to resentment and an eventual refusal to do other’s work for them. So, there is some pressure for each person to leave the mess they make, with the result that the situation is worse for everyone. Prisoner’s dilemmas can occur any time trust is required. If two groups (e.g., Protestants and Catholics in Northern Ireland) are extremely suspicious of one another, they will be motivated to violate ceasefires and other agreements for much the same reasons that our prisoners were motivated to squeal. Trust, of course, is an issue in small groups, including groups of two, so we have worked our way back to the two-person case where we began.