Are Humans Much Less Selfish Than We Thought?

In his famous book The Wealth of Nations, Adam Smith noted that “it is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner” but rather “from their regard to their own interest”. And he argued that, in a free-market economy, individuals pursuing their own interests are “led by an invisible hand” to promote the general interest.

The implication is that it doesn’t matter if people are selfish and only look out for themselves and their family. Given the right incentives, they will act in such a way as to make their fellow citizens better off. (It’s worth noting that Smith wrote an entire other book, The Theory of Moral Sentiments, about how people aren’t always selfish.)

Smith’s argument, which has of course been substantially refined and extended by other scholars, seems to imply that a free-market economy is the best overall system—since it channels people’s selfish motives toward the public good.

But what if people aren’t selfish? In that case, there could be a much greater role for the state in guiding the economy and shaping the distribution of outcomes. This is the argument a number of contemporary economists have made based on research into ‘economic games’. The research involves inviting participants into a lab, or recruiting them online, and then having them play specific games designed by economists, such as the Dictator Game or the Public Goods Game.

In the Dictator Game, there are two players. One player, ‘the Dictator,’ is given some money and must decide how much to keep and how much to transfer to the other player. (He can transfer none if he wants to.) In this incredibly simple game, the optimal strategy is just to keep all the money. However, studies typically find that a sizeable portion of ‘Dictators’ transfer some money to the other player.

In the Public Goods Game, there are several players. Each player is given some money and must decide how much to keep and how much to contribute to a public pot. Any money contributed to the pot is increased by, say, 20% and then redistributed equally among the players. In this slightly more complex game, the optimal strategy is, once again, to keep all the money (although everyone would be better off if they all contributed). However, studies typically find that a sizeable portion of players contribute some money to the pot.

The finding that many players choose to transfer money in the Dictator Game, or to contribute money in the Public Goods Game, has been interpreted as showing that we actually have ‘social preferences’—that we care about the well-being of other random people and about upholding pro-social norms.

Now, we obviously do care about those things a bit. But when was the last time you walked up to a random person and gave them some money out of your wallet? The question is whether these economic games can tell us much about the real world. Do they really have big implications for how we govern society?

A number of studies suggest they don’t. In a recent paper, Lina Koppel and colleagues had participants play the Dictator Game, the Public Goods Game and other economic games. Afterward, they asked participants ‘comprehension questions’ to see whether they’d really understood the games. Remarkably, a large percentage of participants answered these questions incorrectly—almost 25% in the case of the Dictator Game and more than 50% in the case of the Public Goods Game. What’s more, participants who misunderstood the games were more likely to play unselfishly.

This finding is consistent with previous studies, which have found that participants play unselfishly even when explicitly told that the other players are computers, and that they behave more selfishly when you give them more information.

So yes, humans have both selfish and unselfish motives—but we knew that long before economics came along. And we shouldn’t draw strong conclusions from economic games that many people do not understand.