In conversation: Dan Ariely

Dan Ariely, 45, is a professor of psychology and behavioural economics at Duke University in Raleigh, N.C., and the author of two engaging bestsellers, Predictably Irrational (2008) and The Upside of Irrationality (2010). In his cheekily named Center of Advanced Hindsight, Ariely has lately been concocting simple math tests that pay subjects for correct answers and which, at times, also allow various ways to cheat. The results, as set out in his new book, The (Honest) Truth About Dishonesty, offer fascinating insights into the factors and limits involved in the human dedication to cheating.

Q: How did this topic come to intrigue you?

A: It all started with Enron, and the question of whether the massive fraud there was caused just by a few bad apples or whether there was something more general and systemic, involving more people who were not villains but who turned a blind eye to things that should have been obvious.

Q: Like error, cheating is intrinsically human. Do you look at it in evolutionary terms?

A: I don’t believe we need to evoke evolution. We do stand to gain from being dishonest—if I gave you a situation in which you could only lose from being dishonest no one would be. We have a motivation and there’s no mystery to it. The sad thing we find is that people cheat a lot. The silver lining is that people don’t cheat enough, in another sense: there are lots of opportunities in our studies for people to cheat much more but people don’t take them. There’s this force that stops us from being as dishonest as we can be. Every time I go to a restaurant these days I ask the owner or the waiter about ways in which I could escape without paying, and they always have good suggestions, involving the bathroom or a back door. But when I ask them how often people do it, the answer is almost never. Sometimes, they say, people leave without paying the bill because they forgot, and when they remember they usually come back. So there is something fantastic about human nature.

Q: Do cheaters prosper?

A: In the short term, absolutely yes. In the long term, it comes out more evenly, with high probabilities of success and of doing very badly. But I think we all need to lie a little bit to live in society, okay? When I was in hospital with third-degree burns after an accident, doctors kept telling me that the next procedure wouldn’t be painful and, of course, it was excruciating. But if they had told me the truth I would have lived in dread.

Q: What did you find to be the optimum situation for cheating—the mix of motive, possible gain and freedom from detection?

A: The first thing you need is motivation in the sense of a force that pushes you to see reality in a certain way—a desired outcome for your team or your finances or your reputation. And the second thing is some kind of grey zone in how you’re going to think of yourself afterwards. This is where rationalization comes in. And what the research shows is that there are lots of things that can change rationalization, and that allows people to be more dishonest.

Q: There must be other forces too, such as not wanting to feel like a fool if people around you are making off with stuff and you’re not.

A: Yes, yes, that’s part of the grey-zone rationalization—everybody else is doing this, so you feel that it’s more okay to do it.

Q: And it must help to have a depersonalized (so to speak) victim—it’s one thing to take pencils from the office and quite another to steal them from a blind man.

A: Or a physical object, especially actual cash. When we put the Cokes and the dollar bills in the dorm fridge, the Cokes disappeared almost immediately, but the money wasn’t touched. And this goes back to the restaurant. Nobody steals from a restaurant and few from a record store, but lots of people download illegal stuff from the Internet. In the first two days my book was out there was one website that set it out as an illegal copy. It was downloaded 20,000 times in two days.

Q: The key element governing cheating, you found, was the “fudge factor.” It’s not the possible punishment or the likelihood of being caught (as current criminological thinking has it) or even the possible gain that matters—it’s the living with yourself afterwards.

A: That’s it. The U.S. states that have the death penalty have the same crime rate as the others. Our level of cheating is not related to long-term consequences. Yet most of our attempts to overcome bad behaviour are about catching it after the fact, and exacting some kind of penalty. We think this will deter people from behaving badly, but it turns out to have no effect. We need to understand that what’s stopping us from acting badly is our internal moral compass, which allows us to cheat a little but usually not much. We need to turn to education and not to deterrence. I talked to somebody who grew up in a crime family, where they told him other people just don’t matter—cheating on taxes, on the insurance company, or customers, or other people outside of the family doesn’t matter. But people inside the family matter a lot; they have to trust each other.

Q: The fudge factor balances the desire for gain and the desire to think well of ourselves. Did you ever quantify it? In the tests you gave that allowed subjects to increase their scores and rewards by cheating, they inflated their results by a consistent 15 per cent or so.

A: I think that’s about right. That’s about what we find, in a very general way. Why this number, I don’t know. I think it’s a very interesting idea that the motivation would be to get just a little bit more, so we can still think of ourselves in the same way we always have. You know, in developmental psychology there’s an idea that as you learn new things you try to force them into your current framework of the world, and it’s not until you can’t really hold anymore in the old framework that you develop a new idea of how the world works. And I think, you know, that 15 per cent is about the [limit] there too.

Q: Your foreign associates—except, you note, Scandinavians and Canadians—say that “Americans are chumps at cheating, come to my country, I’ll show you cheating.” Yet your tests abroad show the same results. Is the 15 per cent fudge factor a universal human constant?

A: Our experiments are not embedded in any culture, they measure the general ability to be dishonest and still feel good about yourself. Cheating does vary dramatically across cultures in terms of what areas of life you are permitted to cheat in. In the U.S. and Canada young people feel very good about downloading illegal stuff from the Internet. In fact, when you talk to them—and I talk to quite a few—they can make you feel they’re fighting for freedom rather than benefiting personally. Much depends on how we define something. Infidelity in France is seen very differently than infidelity here—the culture has looked at this phenomenon and said, “This is just not something we think about in the moral domain.” Cheating in university, like students copying in exams, there are countries where this is merely a poker game between teachers and students, not a moral issue. One measure is to consider how embarrassed would you be if a newspaper published your name? Undergrads downloading illegal stuff wouldn’t care: none of their friends would look differently at them.

Q: As with the deeply internalized taboo against stealing money, traditional codes of behaviour remain powerful. You found that when atheists swear on a Bible they don’t believe in, they still cheat less. Is that less than Christians who do not swear on a Bible before one of your tests?

A: Yes. In our experiments the moment people swore on the Bible, cheating went to zero; otherwise, never zero.

Q: That is clearly connected to another finding: when those signature boxes, such as on a tax filing, where people swear to the truth of their responses, are moved from the bottom to the top of a document, cheating goes way down. How does that work?

A: People basically want to be honest, but our internal moral compass is kind of asleep at the wheel most of the time. But once we are prompted to think more carefully about our honesty, we do take notice and we actually behave accordingly. So if you fill out your tax form with the normal fudge factor and then run into that box at the bottom you’re not going to go back and fix things. But if you sign it up top it’s going to affect your fudging as you go through.

Q: So there are relatively simple things that can be done to better control our cheating hearts, and a good thing, too, you suggest, because in the increasingly cashless and online future our moral compasses will need the help?

A: I think, sadly, that the answer is yes. We’re on a bad trajectory because of our increasing distance from people and from our actions. Closeness is why people don’t steal food from restaurants, but have no problems with other things. For all the benefits of distance, the Internet for example, we also need to recognize its costs. Also, with plagiarism in schools, by the way, we’re really on a dangerous trend. I think it’s becoming worse, and there’s a risk that in the near future it will just become business as usual. And the moment it does, people become the perfect economic actors, because at that point it’s just about cost and benefit. So the idea that we have this internal moral compass is incredibly important. It’s a very precious thing that we have, and we need to think about how we sanction it, how we sustain it, how we develop it, because without that society would really be quite terrible.