Researchers Dropped 17,000 Wallets To Test Honesty In Countries Around The World Scientists used "lost" wallets to test whether people are more likely to be dishonest when they might profit. The results were puzzling — so they put more money in the wallets.
NPR logo

What Dropping 17,000 Wallets Around The Globe Can Teach Us About Honesty

  • Download
  • <iframe src="https://www.npr.org/player/embed/734141432/735191324" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
What Dropping 17,000 Wallets Around The Globe Can Teach Us About Honesty

What Dropping 17,000 Wallets Around The Globe Can Teach Us About Honesty

What Dropping 17,000 Wallets Around The Globe Can Teach Us About Honesty

  • Download
  • <iframe src="https://www.npr.org/player/embed/734141432/735191324" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

Researchers dropped more than 17,000 wallets with varying amounts of money in countries around the world. Here, an example of the wallets that held the most money. Christian Zünd hide caption

toggle caption
Christian Zünd

Researchers dropped more than 17,000 wallets with varying amounts of money in countries around the world. Here, an example of the wallets that held the most money.

Christian Zünd

So picture this: You're a receptionist at, say, a hotel. Someone walks in and says they found a lost wallet but they're in a hurry. They hand it to you. What would you do?

And would that answer be different if it was empty or full of cash?

Those are questions researchers have been exploring; Thursday, they published their findings in the journal Science.

The experiment started small, with a research assistant in Finland turning in a few wallets with different amounts of money. He would walk up to the counter of a big public place, like a bank or a post office.

"Acting as a tourist, he mentioned that he found the wallet outside around the corner, and then he asked the employees to take care of it," says Alain Cohn from the University of Michigan, the study's lead author.

The researchers assumed that putting money in the wallet would make people less likely to return it, because the payoff would be bigger. A poll of 279 "top-performing academic economists" agreed.

But researchers saw the opposite.

"People were more likely to return a wallet when it contained a higher amount of money," Cohn says. "At first we almost couldn't believe it and told him to triple the amount of money in the wallet. But yet again we found the same puzzling finding."

The researchers decided to do the experiment on a much larger scale. They put together a team that dropped off more than 17,000 "lost" wallets in 40 countries over the course of more than two years.

All the wallets were about the same — a small clear case holding a few business cards, a grocery list in the local language, and a key. Some contained no money and some held the equivalent of about $13. Research assistants turned them in at the kinds of places people would typically bring a wallet they found on the ground — police stations, hotels, post offices and theaters.

Such a large operation came with a few headaches, Cohn says. One of the researchers was detained in Kenya for suspicious behavior. And researchers worried that a backpack full of wallets might raise eyebrows when crossing borders.

It's also worth noting that for logistical reasons, most of the wallets were not literally returned to the researchers. After people reported a wallet to its supposed owner over email, they were told that the owner had left town and didn't need the wallet anymore.

As results rolled in from around the world, the researchers kept finding the same result. In 38 out of 40 countries, people were more likely to report receiving wallets with money than those without. And in the other two, the decrease in reporting rates for the wallets with money were not statistically significant.

What if the wallets contained far more money? The researchers did a "big money" test in the U.S., the U.K. and Poland. In that phase of the experiment, the staff dropped wallets containing nearly $100, instead of $13.

Cohn says the results there were even more dramatic. "The highest reporting rate was found in the condition where the wallet included $100," he says. Forty-six percent of wallets with no money were reported, compared with 61% of those with about $13 and 72% of those with nearly $100.

What's behind all this honesty? The researchers suggest two explanations.

First, just basic altruism — the person who reports receiving a lost wallet might care about the feelings of the stranger who lost it.

There's some evidence for that. The same team ran a test where some wallets contained only a key — a thing valuable only to the person who lost it. Those wallets were about 10% more likely to be reported than those with no key.

Caring about strangers doesn't explain everything, though. The researchers think their findings also have a lot to do with how people see themselves — and most people don't want to see themselves as a thief. Cohn says they polled people who said that if there's cash in the wallet, it just feels more like stealing.

And, he says, "the more money wallet contains, the more people say that it would feel like stealing if they do not return the wallet."

Duke University economist Dan Ariely, who studies dishonesty, says this shows material benefits do not necessarily drive people's decisions about whether to be honest.

The study "shows in a very natural, experimental way our decisions about dishonesty are not about a rational cost-benefit analysis but about what we feel comfortable with from a social norm perspective and how much we can rationalize our decisions," Ariely says.

The rates at which people tried to return the wallets varied a lot by country, even though the presence of money in the wallet almost always increased the chances. In Denmark, for example, researchers saw more than 80% of wallets with money reported. Peru saw a little over 10%.

The researchers think wealth could be a factor, but there's a lot more research needed to explain the differences. "Now the problem is that we don't really know whether wealth affects honesty or it's the other way around" — whether honesty contributes to a country's relative wealth, says Cohn.

Countries with higher rates of primary education were also more likely to see high rates of lost wallets being reported.

"What this suggests is that what you learn in school is not just math and reading but also social skills, or just more generally how you treat each other," Cohn adds.

The study's results could help policymakers and businesses that want to figure out what motivates people to act for the good of others, rather than for their own enrichment.

"What our study suggests is that there might be a potential to promote honest behavior, first, by making the harm that your behavior can impose on other people more salient," Cohn says.

Cohn says the results also suggest that to promote honest behavior, businesses or policymakers should make it more difficult for people to deceive themselves that they're being honest when they are actually doing the opposite. For example, by having people sign a statement promising truthfulness before they report their car mileage, rather than after.

And sometimes, honesty does pay. Almost all of the people who reported a lost wallet got to keep the cash.