Read our COVID-19 research and news.

The root of cooperation? Large groups of people cooperated more when they had a symbolic version of money to exchange with each other.


Does Money Inspire Us to Cooperate?

Money is often called the root of all evil, but it may deserve some credit for making us cooperative—at least when we are in big groups. A study of how a monetary system can change behavior finds that filthy lucre may have been crucial for the evolution of large human populations.

Ask most economists what money is good for and they may throw out the word “fungibility,” which means the quality of being exchangeable. Unlike traditional barter systems, where people trade one useful thing for another, a monetary system uses symbolic tokens that can be traded for anything. That opens the possibility of exchanging goods—say, a cow for a year’s supply of bread—for which a fair swap would otherwise be too complex to figure out.

But beyond its basic utility, money could have other benefits to society. One theory holds that money makes cooperation possible in large groups of strangers, where trust is in short supply. For example, if you give your cow to a stranger in exchange for money, you don’t have to trust that person to keep your bread supply coming all year; you can use the money to buy bread from anyone. For this reason, monetary systems may have been crucial for human urbanization.

To test this idea experimentally, a team led by economist Gabriele Camera of Chapman University in Orange, California, brought 200 people into a room full of computers and asked them to play two different games, both based on the goal of earning points called “units.” Everyone started out with eight of these units, and players were divided into groups of two, four, eight, or 32 people for a series of rounds. In each round, the computer made random pairs within the group. Players knew how many people were in their group, but they didn’t know which of those people was their current partner.

In the first game, one of the players in the pair had the option of spending 6 units to help the other get 12 units. That’s an expensive choice, but if someone returned the favor in a future round, it would double the investment. Of course, if no one pitched in 6 units to help out later on, this player was out of luck. The computer stopped the game after a random number of rounds. At the end, everyone’s scores were tallied up and converted into real cash that the participants took home.

As Camera and his colleagues expected, trust broke down as the group size increased. When just two people played the game, they could count on their gifts being returned when their roles were reversed. And indeed, the pairs of people helped each other out 71% of the time. But in larger groups, the chances of being paired with the same person plummeted—below 5% in the case of groups of 32. And that translated to a breakdown of trust. The frequency of cooperation in the 32-person groups fell to 28%.

The second experiment was exactly like the first but included a virtual form of money called “tokens.” Each player started out with two of them. Players had the additional option to use a token to “buy” help from their partner, instead of just receiving assistance as a unilateral gift. Unlike units, tokens had no intrinsic value and could not be redeemed for anything at the end of the game. Using them was voluntary; the game could be played exactly as before without them.

Nonetheless, the participants adopted this monetary system immediately instead of helping each other through gift-giving. In the two-person groups, using tokens eroded trust, and cooperation dropped by 19%. But with more players, tokens had the opposite effect. In the largest groups, people cooperated nearly twice as often when using the symbolic monetary system, and everyone reaped larger rewards, the team reports online today in the Proceedings of the National Academy of Sciences.

Camera theorizes that money makes cooperation possible when people cannot rely on reputation or kinship. He speculates that money drives cultures toward money-based economies by helping to support larger populations.

The study is convincing evidence that money promotes cooperation among Camera’s research subjects, says Paul Rubin, an economist at Emory University in Atlanta. But he points out that they “were undergraduates, meaning they have lived in a money economy all their lives.” The cooperation-boosting power of money may not hold for people with no exposure to modern monetary systems, he says.

Camera agrees that this is a limitation of the study. “We have given some thought to the possibility of running the same experiment with subject pools that are ‘nonstandard,’ ” he says, such as isolated Amazonian tribes. Despite some “nontrivial logistical issues”—such as importing a version of this computer-based game into the jungle—Camera hopes to test the cooperation-inducing effect in groups who live more like our hunter-gatherer ancestors.