Strong institutions reduce in-group favouritism


People in societies with supportive government services, food security and institutions that meet their basic needs were very likely to follow impartial rules about how to give out money.
Photo: iStock

Ineffective social and political institutions make people more likely to favour their family and own local social group, while good institutions make them more likely to follow impersonal rules that are fair to everyone, suggests a new study in press at the journal Human Nature.

A series of experiments found that people in societies with supportive government services, food security and institutions that meet their basic needs were very likely to follow impartial rules about how to give out money. By contrast, those without effective, reliable institutions showed favouritism toward members of their local community.

“If you don’t have well-functioning governments then you need these kinds of motivations, because then you’re doing what’s best for your group and for your local community,” says Senior Fellow Joseph Henrich (University of British Columbia), the co-principal investigator on the study.

“You’re just trying to survive in a world where there’s no higher-level governmental institutions you can depend on,” says Henrich, a member of the Institutions, Organizations & Growth (IOG) program.

The study, done in Bangladesh, Bolivia, Fiji, China, Iceland and the United States, tested motivations using a game. Researchers gave subjects half a day’s wages in cash and placed them before two cups. They told the subjects that the money placed in one cup would go to an unspecified member of their community or group at the end of the game, while the money in the other cup would go to an outsider.

Played fairly, the game would result in both cups having the same amount of money at the end. Researchers made sure individual players knew that no one could see them cheating while the game was being played. But statistical analysis after the game was over detected whether allocation biases or “cheating” had taken place.

The study found that people from countries with effective institutions followed the rules, while people from countries with poor institutions were biased in favour of community members.

From the government deemed least effective, Bangladesh, to the one deemed most effective, the United States, participants showed a significant decline in favouritism toward their own group. In a Bangladeshi village, the subjects allotted 55.7 per cent of the money to their fellow villagers. At a U.S. church, the congregation ended up with 50.1 per cent of the share.

“In a world with well-functioning institutions, this gets inside of people and actually affects their basic motivations, even when they’re in a situation when no one is watching,” Henrich says.

Henrich says the research ties into the themes of the IOG program, as it demonstrates the complex social and psychological effects of institutions on societies.

— Lindsay Jolivet

Print Friendly
This entry was posted in CIFAR Six. Bookmark the permalink.

« Return to CIFAR Exchange

  1. Brad says:

    Very interesting experiment, but the description was lacking. In what way was there a ‘game’? Was one of the rules that the subject was to distribute the money evenly? How does cheating occur?
    Can you provide a link to the study report or the journal with publication date so I can answer the questions for myself?
    Why the graphic with 3 Canadian flags?
    What is IOG and is it related to CIFAR?

    • Kurt Kleiner says:

      Here’s a link to the paper which details the study design.

      The graphic we ran is of the Canadian Parliament, which we chose as an example of the sort of strong institution referenced in the study. And IOG is the CIFAR program in Institutions, Organizations & Growth.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>