Picture the scene: you are walking back home when you stumble upon a lost wallet, stuffed with $10 bills. What do you do? Do you A: walk on by, B: hand it in, or C: take the money?
To find out whether material self-interest trumps civic duty (or vice versa), researchers from the US and Switzerland planted 17,303 wallets in 350 cities across 40 countries and waited to see who attempted to return the "lost" item.
The results, published in the journal Science, show (shockingly!) people are more likely to return a lost wallet with cash than a lost wallet without. And the more money inside the wallet, the more likely it is to be returned.
Is it proof that we are all driven by pure, undiluted altruism?
Not quite. While altruism may indeed play a part, the researchers say, motivation to return the money is likely linked to a keen desire not to think of ourselves as thieves.
For the study, each wallet (actually a transparent business card case) contained a key, a grocery list, and three identical business cards displaying a fictitious male name and a working email address.
The contents of the grocery list and name varied from country to country to account for regional differences. For example, in the UK names included David Brown, Mark Smith, and Michael Wilson; in the US, Brad O'Brien, Brett Miller, and Connor Baker; while both grocery lists contained milk, bread, pasta, and bananas. In India, the names were Gaurav Kapoor, Govind Malik, and Shekhar Kohli, and the grocery list included rice, not pasta. Some wallets also contained a small wad of cash equivalent to $13.45.
Each wallet was then dropped off at a "societal institution" (think: theater, bank, museum, post office, hotel, and police station) by a member of the research team. An assistant would walk up to an employee, tell them they had found the wallet outside, and hand it to them. They then waited to see who attempted to return the wallet using the email address provided.
The team found substantial variation in terms of civic honesty across countries – an average reporting rate of 14 (China) to 76 percent (Switzerland) – but in all bar two, people were more likely to return the wallet when it contained money. And in Mexico and Peru, the two "anomaly" countries, the difference was minimal and statistically insignificant, the study authors say.
Indeed, including money increased reporting rates from an average of 40 percent (in the no money condition) to 51 percent (in the money condition).
To find out if a bigger monetary incentive changed the odds, the researchers performed a second experiment in the UK, Poland, and US with wallets containing $94.15 (seven times the original money condition). But it only encouraged civic honesty, with reporting rates in these countries increased from an average of 46 percent (no money condition) to 61 percent (money condition) to 72 percent (big money condition).
This flies in the face of what the public – and experts – predict, and might feel pleasantly reassuring to anyone feeling evermore cynical about the world. But while it negates economic models based on rational self-interest that suggest with all else being equal, honest behavior decreases as the material incentives for dishonesty increase, it conforms to psychological models that predict a person will cheat for personal gain unless that behavior demands they negatively update their perception of themselves (in this case, to a person who steals).
It also makes a lot of sense evolutionarily-speaking. As the study authors point out, "Honest behavior is a central feature of economic and social life."
Studies have shown that women consider altruism a more attractive quality in a prospective partner than either good looks or a sense of humor. Altruistic behavior has been documented in the wild in animals like bonobos and chimpanzees. Indeed, scientific research suggests we may be hard-wired for altruism, a character trait that makes it easier to trust one another and co-operate – two behaviors we rely on as a social species that relies on a tight social network for our survival.
So, B might be the right answer after all.