Analyzing the research on human cooperation from so many disciplines helps identify some levers that may motivate people to contribute to the collective effort instead of pursuing their own interests at the group’s expense. These levers aren’t equally appropriate for all types of systems; different combinations are better for different activities and populations. With that caveat, here are seven preliminary ideas for building cooperative systems.
Nothing is more important in a cooperative system than communication among participants. When people are able to communicate, they are more empathetic and more trusting, and they can reach solutions more readily than when they don’t talk to one another. Over hundreds of experiments spanning decades, no single factor has had as large an effect on levels of cooperation as the ability to communicate.
Framing and authenticity.
People react differently depending on how situations are framed, but they aren’t stupid. It’s important that the frame fit reality. Framing a practice as collaborative or a system as a community may encourage cooperation for a while, but it won’t last if that claim isn’t believable.
Empathy and solidarity.
For reasons biological and social, the more empathy and solidarity we feel with others, the more likely we are to account for their interests. Similarly, solidarity with a group makes us more likely to sacrifice our interest for that of the collective. The difference between solidarity and discrimination is a slippery slope, though. While it’s impossible to deny the role of team spirit in getting people to cooperate, we do need to be wary of its ability to exclude nonmembers.
The first step in constructing a team or encouraging prosocial behavior is to expand the set about whom participants feel they should be concerned. That’s not difficult; experimental economists such as Bruno Frey and Iris Bohnet have shown that just seeing another participant’s face increases cooperation levels substantially. Empathy for those affected by our actions alters the outcomes we care about, and that, in turn, changes our behavior.
Fairness and morality.
People care about being treated fairly. According to work conducted by Ernst Fehr and his collaborators, “fair” does not mean “equal.” In experiments where some people gained because of skill or luck, the others initially were willing to let them walk away with a much larger share of the gains.
We are also flexible in our ability to accept norms of fairness. Historian Andrea McDowell, for example, showed how mining camps during the California gold rush enacted very different codes for what counted as fair or unfair claim jumping. Though the norms differed among camps, each camp applied its own norms uniformly, and newcomers accepted them. In another groundbreaking study, when anthropologists Rob Boyd and Joe Henrich and economists Sam Bowles, Colin Camerer, Ernst Fehr, and Herbert Gintis conducted games with subjects from 15 tribal societies all over the world, they found enormous differences in conceptions of fairness.
Fair Does Not Mean Equal
We are flexible in our ability to accept norms of fairness. Historian Andrea McDowell, for example, showed that mining camps during the California gold rush enacted very different codes for what counted as fair or unfair claim jumping. Though the norms differed among camps, each camp applied its own norms uniformly, and newcomers accepted them.
To some extent, people care about having a fair share of whatever benefits their cooperation produces. Experiments conducted in market-based societies suggest that most participants tend to follow an equal distribution norm if there’s no reason to deviate from it. If some players deviate too far from the norm, the others will punish them even if by doing so they lose, too. In fact, most people prefer to lose everything than to walk away with too small a share of the gains.
People also care about doing the right thing, whatever that is. Clearly defined values are crucial to cooperation; discussing, explaining, and reinforcing the right or ethical thing to do will increase the degree to which people behave that way. It’s important for a cooperative system to have codes that are predicated less on rules and more on social norms. They must be flexible or plastic enough to adapt to change, and they must be transparent. Letting people in a system see what others are doing reinforces social norms and gets people to comply—not because they fear embarrassment or ostracism but because they want to do what is normal.
Rewards and punishment.
In order to foster cooperation, it is critical to set up systems that appeal to participants’ intrinsic motivations—that is, what they want to do from within—instead of systems based on monitoring people and rewarding or punishing them according to their behavior.
Two facts make this tough to implement. First, intrinsic motivation is in its infancy as an area of inquiry. Second, there is a consistent and stable body of work that tells us that if we add money, things may go worse rather than better. That is, monetary incentives and material rewards can crowd out intrinsic motivations to cooperate or display empathetic behavior. If you are invited to a dinner party, you can bring a gift—flowers, wine, or whatever counts as a friendly gesture. If instead you leave $100 on the table at the end of the meal, you will destroy the atmosphere because you have turned a social interaction into a commercial exchange. This captures the findings of studies in experimental economics and psychology as well as many field studies of the crowding-out phenomenon.
For example, a recent study in Sweden, which has a purely voluntary blood donation system, showed that women’s contributions decreased when they were offered payments. Donating blood is a way for people to signal that they are the kind willing to sacrifice for the good of others; offering money spoiled that effect. To test that hypothesis, the experimenters later permitted donors to give the money they would have received to a foundation that works on children’s health issues. Sure enough, the women’s contributions went back up.
The Power of Intrinsic Motivation
A consistent and stable body of work tells us that monetary incentives and material rewards can crowd out intrinsic motivations to cooperate and display empathetic behavior. A recent study in Sweden, which has a purely voluntary blood donation system, showed that women’s contributions decreased when they were offered payments. But when the experimenters permitted donors to give the money they would have received to a foundation that works on children’s health issues, the women’s contributions increased.
Whenever you design a policy that relies on monetary rewards, you have to assume that it will have side effects on the psychological, social, and moral dimensions of human motivation. A change that would lead to more behavior of the kind you are rewarding, or less of the kind you are punishing, may cause the exact opposite behavior because the effects on the material self-interest vector will be more than canceled out by the effects on the intrinsic motivation vectors. We shouldn’t try to motivate people only by offering them material payoffs; we should also focus on motivating them socially and intellectually by making cooperation social, autonomous, rewarding, and even—if we can swing it—fun.
Reputation and reciprocity.
One extremely important form of cooperation hinges on long-term reciprocity, both direct and indirect. Systems that rely on reciprocity, particularly of the “pay-it-forward” kind, are enormously valuable but easily corrupted. Reputation is the most powerful tool against this. As online systems such as eBay have shown, even anonymous reputation systems—such as “handles” that betray nothing of a person’s real identity—are sufficient to keep people in line.
Systems that harness diverse motivations are more productive than are those built for people who care only about material payoffs. Because we differ from one another, cooperative systems have to be flexible. They also need to recognize that we are sensitive to the costs of cooperating, but the degree of sensitivity can change. It’s possible to create a system that depends on massive self-sacrifice, but it’s extremely tough to sustain it. The fate of the nationalist and communist experiments of the 20th-century in Germany, Russia, and China provide ample evidence of that.