RENEE MONTAGNE, HOST:
Researchers have found it over and over that peer pressure works. A company recently decided to apply peer pressure to get their employees to save more money for retirement. Our colleague David Greene spoke with NPR social science correspondent Shankar Vedantam about this experiment.
DAVID GREENE, HOST:
SHANKAR VEDANTAM, BYLINE: Hi, David.
GREENE: So what happened?
VEDANTAM: Well, we know that lots of Americans are not saving enough for retirement, David, and we also know that Americans tend to follow the leader. So various studies have found that people are more likely to conserve energy or to go out and vote when they are told that their neighbors are doing it. Katherine Milkman at the Wharton School at the University of Pennsylvania told me that a large manufacturing company got in touch to ask how they could get more of their employees to sign up for retirement accounts. And the first thing that Milkman and her colleagues thought about was these earlier studies into peer effects. So they decided to tell the employees that lots and lots of their other colleagues had already signed up for the plans. Milkman explained to me why she thought this was going to work.
KATHERINE MILKMAN: It's uncomfortable to be the odd man out or the odd woman out. So if everybody else has an iPhone, then it's sort of uncomfortable for you not to have an iPhone as well. You may feel like a social outcast. And past research had suggested that learning everybody else is doing this, makes you much more interested in doing it yourself.
GREENE: OK, all seems to make sense, I mean, you feel like an outcast if you're not doing what everyone else is doing. Signing up for a retirement plan - people who don't have it see everyone else doing it - they want to do it, right?
VEDANTAM: That's exactly right. Now, the researchers in the company decided they were not just going to implement this plan but actually test to see whether it works. So employees who hadn't yet signed up for retirement plans were divided at random into different groups. Some got just information about the benefits of retirement accounts, and others got the same information, but they also got the peer pressure. They were told, look, 75 percent of your colleagues have already signed up for these retirement plans. When they tallied the results, the company and the researchers got a shock. Here's Milkman again.
MILKMAN: Finding out that the majority of their peers are saving for retirement actually significantly reduces the likelihood that they'll subsequently enroll in the plan. It reduces it by a third, from about a 10 percent response rate to about a 6.3 percent response rate.
GREENE: Shankar, that is kind of crazy.
VEDANTAM: It is crazy, David. And I did a double take when I saw this because I was so certain that this intervention was going to work. The question, of course, is why it didn't work, and the answer is it's not entirely clear. One theory that the researchers have is that peer pressure works when you feel that you have a chance to actually catch up with your neighbors. You feel like your neighbors are within reach. Most employees at this company had already signed up for retirement plans. So it may be that people got the feeling that most of their colleagues were already far ahead of them, and rather than being encouraging, this kind of intervention ended up being demoralizing.
GREENE: Well, for peer pressure to backfire, I mean, that whole concept, I mean, must have some larger implications.
VEDANTAM: It does. I mean, one of the implications is that you can't just run out and apply peer pressure and expect it to work the same way every time. Clearly, peer pressure does work in some situations. But as this study is showing, clearly there times that it's going to backfire. One of the things that Milkman finds is that the backfire effect is strongest among lower paid employees. So it could be these are the employees who are most likely to get demoralized and feel like they're falling behind. There's also, I think, a larger implication from the study, David, which is that when we think about social science research and policy settings, we often have a desire when we learn something about human behavior to run out and implement it in the world, and that's understandable. But I think what the study points to is that if you are not actually studying your outcomes really carefully and systematically like this company and these researchers did, you might end up implementing policies that are not just ineffective but that are actually counterproductive.
GREENE: Do your homework and test things out before you actually implement a policy.
VEDANTAM: That's exactly right, David.
GREENE: Shankar, thanks as always.
VEDANTAM: Thank you, David.
GREENE: That's Shankar Vedantam. He regularly joins us to chat about social science research. You can find him on Twitter at @HiddenBrain. You can find this program at @MorningEdition. Transcript provided by NPR, Copyright NPR.