Humans are social beings, and many of the choices we make in our lives happen in a social context, with neighbors, friends and family judging our decisions. How does this social pressure affect the choices we make about what to buy, whether to vote, or what political views we hold? Leonardo Bursztyn, a behavioral economist at the University of Chicago, has focused his research on exploring the complexities of that question.
Bursztyn describes what he and others have learned from real-world experiments on social pressure and behavior in a 2017 article he coauthored in the Annual Review of Economics, “Social Image and Economic Behavior in the Field: Identifying, Understanding, and Shaping Social Pressure.”
This conversation has been edited for length and clarity.
How do you study social pressure?
The approach I’ve been taking is to use field experiments to identify the effects in a natural setting. One type of manipulation I’ve been using is to randomize whether or not someone’s action is visible to others. The first paper I wrote with this approach, we were interested in understanding whether kids felt pressure to avoid making educational investment because they didn’t want to signal they were uncool or that they were nerds. We did an experiment in which high school students in a low-income setting in Los Angeles were offered free access to a $200 prep course for the SAT [a standard college entry examination]. We randomized so that some kids expected their decision to be completely private, and others believed their classmates could find out their decision. You see a big decrease in the likelihood that they’ll take the offer when their decision is public, which would have to be explained by their concern about their classmates finding out.
If kids are avoiding some opportunities in school because of peer pressure, why are they doing it? What type of image are they trying to portray to their classmates? Are they trying to signal that they’re cool, and that’s why they don’t need to study, or are they trying to signal that they’re smart, and that’s why they don’t need to study? If you think there’s a stigma associated with not being smart, that leads to certain policy recommendations. If you think that the stigma is something like “too cool for school, school is lame,” then it leads to different policy implications. In fact, both reasons are important, though for different groups of students. So beyond showing there is an effect, understanding the mechanics, the underlying reasons, is also important.
One of the main lessons from your work is that social pressure is a very complex force. People are part of multiple groups with competing pressures sometimes. How difficult does that make it to draw clear conclusions?
You need to understand who the peer group is, what is driving the peer pressure, what image people are trying to portray, and whether the social pressure people think they face is based on a correct assessment of their social environment. This makes it a very difficult problem, and that also makes it very interesting for a social scientist. People are more and more aware of the importance of social forces. Understanding them, and what to do with them, and thinking of policies associated with them — I think that becomes the new frontier.
What are some other ways that social pressure affects our behavior?
People sometimes have incorrect beliefs about what other people think. In those cases, just by correcting people’s beliefs about others, you can lead to changes in the effect of social pressure. I’m working on a project right now indicating that married men in Saudi Arabia seem to be relatively progressive regarding female labor force participation, but they think that other men are against it. What happens when you teach them that other men are progressive? It may actually change their actions relative to female employment.
I have another project studying the unravelling of political correctness in the US, in some areas, with the rising popularity of Donald Trump. People who were xenophobic believed they were in a small minority, because on TV they never saw people saying those things. They thought they would be judged for saying xenophobic things. Trump’s rise in power served as a way to tell these people that in fact there are more people like them than they thought there were. Those people don’t feel the pressure to avoid this type of language any more — and we see increasing expressions of hate and xenophobia.
Have you tested this by comparing private versus public actions?
Exactly. We asked people to donate in private or in public to anti-immigration organizations. Originally there was a difference: A large share of participants would not donate in public but would donate in private. After the election, they’re no longer embarrassed to donate to the xenophobic organization in public. So we showed that social pressure is very powerful, but at the same time it can be shifted very fast.
Why do we care about our social image?
That’s still an open question. Is it for some instrumental reason, meaning I want to cause a good impression because I’ll get good things out of it — for example, if people think I’m rich, they’re going to treat me better? Or do I care if people think well of me even if they’re not going to treat me better? We call that second case a hedonic social concern.
How can you tell the difference?
One way to think about it is: Someone that you’re never going to interact with in the future, do you still care about how they view you? The answer seems to be yes, which suggests there must be some hedonic concern about social image, something intrinsic that doesn’t have anything to do with anything else. Why we feel that way is an interesting question that will require more work.
How does the importance of social media today change the way that social pressure works? Is the quality of pressure different now, or is it the same, just more intense?
If the norm is to have everything that you do being portrayed on social media, then for sure, social image becomes more important. Many actions are now expected to be shared, and the mere fact of not sharing sends a signal that there’s nothing interesting to share. That creates pressure to share. This is just a conjecture, but we might be getting to the stage where people care more about the image of what they’re doing, how it looks to others, than the actual experience. This may create anxiety. People might look at what others are sharing and feel bad about their own life. They don’t understand that everyone’s trying to make their life look amazing on social media.
How can we use this knowledge to help us be better at nudging?
One way is to vary observability. An action that is private, you can make it public. Some countries use shaming incentives — if you don’t pay your taxes, your name is going to be posted in the newspaper. That’s one type of nudging.
There’s other ways to do it. For example, you can create a leaderboard at a company. Then people can see who the best-ranked people are. If you care about being seen as a high-performing employee, you work hard, not necessarily because of the money, but because you want to be seen as a top person. On the other hand, many high school students avoid effort if there is a public leaderboard because they don’t want to be seen as nerds.
In a later study of SAT courses, you found that the appropriate nudge is different in different contexts, because of different social pressures.
We showed that in the same schools, with kids who take both honors and non-honors classes, in the non-honors classes, people are less likely to sign up for the SAT course in public than in private, suggesting the direction of social pressure is away from signing up. Now when you look at the same group of kids in honors classes, they’re more likely to sign up for the SAT course in public than in private, suggesting that the direction of social pressure is toward signing up. They want to signal to others that they’re doing it. The results suggest that even within the same school, different local norms might exist.
That’s very important. If you think about monetary incentives, dollars are pretty much dollars everywhere. Social pressures are more complex, because it really depends on the norm that exists within the peer group. So it’s important to understand the specific context.
What about the ethical concerns about using social pressure in nudging? You mentioned shaming, which is a powerful kind of social pressure, but not necessarily the sort of force we want to harness.
This is yet another open area for research. How do you think about society’s welfare in the world of social pressure? With shaming, you might be creating a lot of dissatisfaction. How do you weigh this against the positive outcomes of actually increasing tax payments? It becomes very complicated. There’s very little work done on understanding welfare and how to take social image concerns into account when designing policy.