Sunday, February 14

Married people can get along (sometimes)

Economics isn't usually very romantic (just ask my girlfriend), but in the spirit of Valentine's Day I thought I'd share an experimental economics working paper probing how married couples interact with each other. It was released in December by three French economists.

The researchers recruited 100 married couples from the Toulouse area and pulled them into their lab to play a prisoner's dilemma game with both their partner and with other people.

The prisoner's dilemma is a game where you choose to either cooperate or "defect" with a partner. If both people cooperate, you get the best possible payout from the game, but if your partner defects while you try to cooperate, you do really badly. Thus, if you don't trust your partner to cooperate, it makes sense for both people to defect, even though you'll be worse off than if you both cooperated.

For the romantics out there, there's a good-news-bad-news side to the results from the experiment (pictured in the bar graph below, taken directly from the paper). Yes, people were much more likely to cooperate when playing with their spouse than with a stranger. But the bad news is that "cooperation in the prisoner's dilemma is not at its maximum," with roughly one quarter of people choosing to defect when playing with their spouse.


The researchers then broke down the demographic info for people who defected. Old rich men are less likely to cooperate with their wives in the game, while younger, more educated women with children are less likely to cooperate with their husbands. "Thus presence of children can in turn lead to greater egoism in social dilemmas played solely with the partner," the authors conclude.

In other words, I think the implications of the study are that if you want to trust your spouse more and be more cooperative with them, don't have children.

That's kind of depressing. And that's why you should never bring up economics in the context of romance.

2 comments:

  1. "In other words, I think the implications of the study are that if you want to trust your spouse more and be more cooperative with them, don't have children."

    I thought you would have known the difference between a cause and a symptom. Couples who do not have children and fundamentally different from those who do.

    "That's kind of depressing. And that's why you should never bring up economics in the context of romance."

    A does not imply B, but that does not mean B is false. Though I actually agree with your B. If you want to be romantic buy some flowers. Did you know that in Moscow there are 24-hour flower shops everywhere? Maybe someone should do a study about that.

    ReplyDelete
  2. Thanks, you're right about the children -> trust causal link. The study doesn't prove that having children makes you less cooperative and trusting -- it could be that children have that effect, or it could be that partners who tend to trust each other don't have kids. The former still seems like a more logical conclusion than the latter, however, if I had to pick one.

    As for my comment about not bringing up economics in the context of romance, you're taking me too literally. It was more of a joke.

    I do agree that flowers are an easy way to be romantic, but you should be careful because people's demands for flowers can easily be exploited, as I discuss in my most recent post.

    ReplyDelete