The Upside of Irrationality: The Unexpected Benefits of Defying Logic at Work and at Home

Interviews & Essays

Who wouldn't want Dan Ariely's job? He's a neuro-whistleblower. He gets to sit around all day and conjure up cool experiments that reveal the evil tricks that our minds play on us. His clever tests measure how much we'll screw ourselves to exact revenge, how we over-value our own creations, how we completely misjudge what will bring us happiness.

Ariely is also the James. B. Duke Professor of Psychology and Behavioral Economics at Duke University, which means he also gets paid to celebrate the imperfection of the species. (But, as he shows us in "The Meaning of Labor," a chapter in his just-published new book, money isn't actually the Big Motivator that capitalist fundamentalists make it out to be.)

Predictably Irrational -- Ariely's first book --was a surprise bestseller when it was published in 2008. It got some terrific notes and reviews; the Times described it as "a far more revolutionary book than its unthreatening manner lets on." And two consecutive Nobel Laureates in Economics -- Daniel McFadden, Class of 2000, and George Akerlof, Class of 2001 -- heaped praise upon it.

Like its predecessor Freakonomics, and some of the work of Malcolm Gladwell, Predictably Irrational introduced the general reader to a largely new world of study that seeks to explain the chronic mistakes we make, the mystifying patterns and chiaroscuro of human behavior.

Ariely's sequel is The Upside of Irrationality: The Unexpected Benefits of Defying Logic at Work and at Home. Its timing is sublime; the last few years have been littered with the consequences of fallibility. Historic bubbles, epic meltdowns, galactic over-confidence -- these recent shocks make it all the more important for us to understand the hard-wired irrationality within. Members of Congressional Panels should keep it next to their night tables, right on top of lobbyist cell phone numbers.

Interviewing Ariely isn't like spending time with a starchy scientist. He is voluble and, discursive, burbling with connections and intellectual cross-hatching: the mind's merry Merlin. Yet he also writes from a place of physical pain. In the introduction to The Upside of Irrationality, he takes us through the emotional consequences of the "devastating accident" that left him with third-degree burns over 70 percent of his body. Ariely's rehabilitation and therapy became a heuristic framework for the rest of his life, bestowing upon him "some unique perspectives on human behavior...questions that I might not have otherwise considered but that because of my particular situation became the focus of my research."

Ariely's post-Cartesian mantra for all of us: "I think screwy, therefore I am. --Adam Hanft

Adam Hanft: How do you describe what it is that you do?

Dan Ariely: I like to think of myself as a social hacker.

AH: But you don't want to steal my identity? Actually, you're welcome to it.

DA: My real motivation is to fix something in the world.

AH: There's a lot to fix, and you do a lot of research in a lot of different areas. Do you get bored easily?

DA: I got this way from spending a long time at the Media Lab at M.I.T. I jump from topic to topic. I look at some other people who spend 30 years studying one thing, and by the end of their career they know it really, really well. That's not what I do. I want to think about how we use that knowledge to improve something. Maybe not to solve it, maybe not to do it perfectly, but at least to move along and make something a little bit better.

AH: Was the success of Predictably Irrational predictable or unpredictable?

DA: I didn't expect it to be so successful. It changed my life in many ways. I got invited to places I wasn't invited to before. I got to meet more people. One of the most interesting things is that sometimes, when I go to conferences or sit on planes, and I sit next to somebody who read my book, they talk to me as if we'd already been talking for 20 hours. That's incredibly rewarding.

AH: The new book feels more intimate than your first book. You talk a lot about your horrific experiences in recovery from your burns.

DA: I got more courageous. I'm writing more personally. I try to describe in more detail how I got to think about some ideas -- "Hey, this is what I thought, and this is why I became interested in this idea" -- and some of the ideas that made me think about particular things are difficult or complex, emotional.

AH: You seem to have a secret fondness for irrationality -- beyond the fact that it made you successful. The book is called The Upside of Irrationality, after all.

DA: We usually pooh-pooh irrationality; we say that it's bad. And of course, there are many things that are bad with irrationality, but there's also something good about it. When you ask people, "Do you want people to be rational?" they say, "Yes. " But then you explain to them what it means. It means that if you left your wallet on your desk and one of your co-workers passed by and they knew they could take it and never get caught, they would take it. It means that nobody will ever tip a waitress in a city that they don't intend to go back to or a restaurant that they don't intend to go back to. That nobody would vote. That nobody would volunteer.

Or imagine a friend who would be perfectly rational, who would always work in their self-interest; there will be no friendship, no morality, no love. Everything will be a cost-benefit analysis. Then ask yourself, would you want a friend like this or would you want a spouse like this?

So the book is not just a praise of irrationality. It's showing lots of the weaknesses and the problems. But it's also showing the good side of some of those.

AH: There seems to have been a little mini-boom in books about behavioral economics since yours was published -- titles like How We Decide, Nudge, and SwayDo you think you started this fascination with the illogical side of things?

DA: I don't know if I can take so much credit. But it's true that some of the people who saw that Predictably Irrational was successful came and talked to me about their books and changed what they were going to write. And there are actually more books in the process now.

AH: Is there a club of you behavioral gurus? Do you go sit around and have a drink, and point out how we'll pay more for a vodka label that has less writing on it?

DA: One of the nice things about social science is that it's about everyday life. So, when I meet my friends -- and lots of my friends work in my area -- there's really no distinction between research and the rest of our lives. We'll talk about why we buy car insurance but not life insurance. Or about decisions regarding work -- do we check email first thing in the morning when we come to the office, or do we decide to be productive for a few hours? So every time we sit together we analyze what we're doing; that's a lot of fun actually.

AH: You write that human beings are more Homer Simpson than Dr. Spock. But even Homer thinks he's capable of thinking straight. Do you think we get pleasure out of seeing ourselves as rational beings? I don't hear many people revel in their illogic.

DA: Yes, because there's something positive about the illusion of control, a belief that we're capable and able. It's actually very helpful. How many people do you think would start restaurants if they actually looked at the cold, hard facts of how many of those are successful?

AH: Millions of people also believed they could afford the houses they bought. But they were wrong. You're fascinated by the mechanisms of mistakes. You write about what you call "human incompatible technologies." We put too much faith in systems that fallible human beings built. Systems like collateralized debt and oil rigs.

DA: The sad thing is that the world runs, to some degree, on people who have high over-confidence. We have not been able to do anything without errors. Think about incarcerating people who didn't do anything, or releasing people who did something, or misdiagnosing cancer. There's not a single thing that we can do well even when we have much more data than we have in offshore drilling or in the stock market.

AH: You worry about over-engineering.

DA: So much is automated with so little room for error. That's why I'm always amazed by the faith we put in leveraging. We make mistakes everywhere we turn. We make mistakes in driving, we make mistakes in typing -- we can't spend ten minutes without making a mistake with something. I once asked professional singers how many times they can sing a song without making any mistake; it's very, very rare that they can go four minutes without an error. Now, what's amazing is that the financial system basically assumed no errors.

AH: And when you're leveraged up, there's no such thing as a small error.

DA: One small mistake can create a huge sequence of consequences. A nice analogy is to think about driving. If you're driving and everybody has a safety distance, then everybody is fine. Even if one car stops, people can go around. But if you drive very short distances without braking distance, one car will stop suddenly and everybody will die.

AH: Do you think we can become less mistake-prone if we read your book, and some others that enlighten us about our sloppy brains?

DA: I think so. I get emails every day from people who tell me those things. I sat next to somebody at a conference, and she just turned to me and said, "Based on what you wrote, I decided to go ahead and have an insulin pump installed." She was a Type 1 diabetic, and she was debating for a long time. It was kind of amazing, because I never wrote anything about diabetes and insulin.

AH: How do you want people to use your books?

DA: The ideal use of the book is as a mirror. Because sometimes it's hard to see the mistakes that we make. But through descriptions of other people, you can actually become more introspective. You see somebody else behaving in a certain situation, and you ask yourself, "What do I do?" That can be incredibly, incredibly helpful. Of course, nobody can change everything about how they behave. But if somebody looks at one behavior -- realizes this is something they don't want to do -- and then introspects about it and changes it, that's fantastic. I hear that all the time.

AH: You write that "Pride of creation and ownership runs deep in human beings" and call it the "Ikea Effect." It made me think of our so-called new economy, and millions of free-lancers flitting around like fruit flies from job to job. Is it possible they aren't sticking around long enough to get any real satisfaction?

DA: That's very interesting. It could be that you're right. I think there's much less sense of accomplishment these days. If you just do a part of a job and you move on, you really have no way to see it through. Did you know that last year there were 200,000 books published in the U.S.? One in three people I think is writing a book. A lot of people I think are trying to go to other outlets that give them a sense of completion or achievement that they don't get in the workplace.

AH: So they've gotten freedom but they've traded off satisfaction.

DA: I think that's highly likely. I haven't collected the data on this, but I think it's a really interesting hypothesis that I suspect is correct.

AH: The Obama campaign used the Ikea Effect, didn't they? 

DA: Absolutely. When you hear people in that campaign talking, they think a lot of their success had to do with this effect; they actually created a system in which people felt they had personal responsibility. The campaign would call them and say, "You volunteered. Now there's a hundred more people you need to call."

AH: How about the Ikea Effect and religion? Don't religions achieve loyalty by encouraging behaviors from prayer to candle-lighting?

DA: Well, you know [LAUGHING]... I like this idea of saying that small rituals are things that you can complete, and make your own in some way.

AH: One of the seminal theories of behavioral economics is that we are not "homo economicus" -- we don't always behave as rational economic actors. An example is what you call "contrafreeloading." You show that human beings don't "always choose to maximize their reward while minimizing their effort" through experiments that show we seek meaning in our work. We'll often work harder for less money if we're more satisfied.

Do you worry that a Machiavellian boss could read your book and figure out how to get people to work harder for less money by appealing to their irrational selves?

DA: Yes. But I think every piece of knowledge can be used for good and for bad. You could say: "Let me get people to work harder for less money." Or you could say "Let me make people happier at their work." The same piece of information can get you both of those. I talk to my students a lot about this. What do I really give them? Do I give them tools to trick people or do I give them tools to improve the state of the world? I tell them that these are important lessons in how to get people to do different stuff, and the question is how do you get them to do stuff that you think is good for them.

AH: What's an example of using your technique for good?

DA: Here is something that we were debating last weekend with some friends. We have all these words for wine -- tannin, acidity, complexity. So when you drink it, you have these words in your head. That makes the wine-drinking process much more intense and interesting. Can we design the same types of words for apples? Or grapes? And if we do that, will people start liking them more? This is kind of a design question.

AH: Sounds a bit like the "libertarian paternalism" that Cass Sunstein, who's now President Obama's Regulatory Czar, writes about. Changing behavior with clever motivational tricks.

DA: I'm actually more extreme than he is. Cass believes in what he and Dick Thaler call "nudges." So when you create a cafeteria, in the beginning you don't offer French fries -- you offer salads. It turns out that it works. But it also turns out, it's not enough. So nudges are useful and helpful, but if you understand how magnificently people can fail, you understand it's only the beginning. Some things you have to be more strict about.

Here is an example. Think about how we're regulating driving. You can't park on the sidewalk. You can't double-park. You can't drive on red. You have to get to a certain age to drive. You have to have a license. You have to have insurance. When you think about how much we regulate driving compared to how much we regulate healthcare, it's amazing. There's a reason. In driving, we can see that people are killing each other, so we're regulating. But in healthcare we don't see the death count so easily.

AH: It's the "Availability Heuristic," right. We need examples we can grab onto to make things real. You talk about that in your book in terms of what you call the "Identifiable Victim Effect" and the American Cancer Society.

DA: In the book I go into length describing the American Cancer Society. They do a tremendous job of implementing this learning.  The use of the word "cancer" itself creates a more powerful emotional imagery than a more scientifically informative name such as "transformed cell abnormality." A loaded word like "survivor" lends an additional charge. We don't use that word in connection with osteoporosis. The American Cancer Society clearly understands how to get people to care.

AH: Politicians seem to have no trouble identifying the Identifiable Victim.

DA: One of the things that happened in the last election was Joe the Plumber. Think about it. We're supposed to make arguments about principles, and what people end up talking about are specific examples. A big part of the healthcare debate was like that, politics by examples rather than by talking about real benefits. Politicians understand that it's a much better rhetorical tool. But the real hope isn't people that understand it and abuse it, it's people who understand human psychology in a deep way and use it for good.

AH: Do you think human psychology is the same from culture to culture?

DA: I think that inherently people are very, very similar, but culture creates a filter by which irrationality can actually exert itself or not. Think about savings. Americans save very little compared to other countries. It's not because we're inherently less rational in this regard. It's because in other countries the parents tell their kids that when you grow up you save 25% of your income, and people take this rule to heart. Now, if you left them to their own accord, they would probably fail in the same way we do. Culture creates sub-rules about how to behave, and they make irrationalities either more or less extreme.

AH: Have you ever done experiments that take different cultures and see how they react to the same stimuli?

DA: Yes, we're doing this more and more. I'll give you an example about cheating. As an Israeli, I thought that the Israelis would cheat more than the Americans, but it turns out they don't. They cheat just the same. My Italian collaborator was sure that the Italians would cheat more than the Americans. They don't. They cheat just the same. My Chinese collaborator thought that the Chinese would cheat more. They don't.

So people respond the same way everywhere. What culture can do, though, is place behaviors in the moral domain. Or not. So for example, in France, having an affair is not a moral question; it's a question of time and money. So a culture basically doesn't make people more inherently moral or less inherently moral, but it does take particular activities and re-categorize them.

AH: Ah, France continues to export its martial infidelity comparative standard. Any other examples of value differences?

DA: We went to bars in Washington, D.C., where Congressional staffers hang out, and we went to bars in New York City where bankers hang out -- and who do you think cheated more?

AH: Congressional staffers.

DA: [LAUGHS] No. Bankers, by about twice as much.

AH: I thought that was too obvious an answer.

DA: Actually, I was sure it would be the Congressional staffers. But I do have to say one thing. The Congressional staffers we had in our study were basically junior politicians. So you can ask what will happen to them as they mature through the system.

AH: Where do you stand on psychology? When you do your experiments, you don't really screen. So you don't think that birth order or toilet training or getting frightened by a horse  can influence our levels of irrationality?

DA: I think it could in principle. I've tried a few things. I haven't found it yet. But I'm kind of an empirical guy. I don't have a deep belief in almost anything. I'm happy to be shown.

AH: So we tend to overstate differences?

DA: Here is one general point about differences. If you look at people, they appear to be tremendously different from each other. But the fact is that people are tremendously similar. It's that the vision system that we've developed basically ignores the similarities and just focuses on the differences. Imagine how useful it will be to have a system that shows you that everybody around you is 99.7% the same.

But we have a visual system that emphasizes the differences. Because of that, people expect to see personality differences all the time, because our appearances are so different.

AH: But we are trained to see differences within our culture, right?

DA: I was in China many years ago with a friend. Not only did all the Chinese look the same to me, because I haven't seen people in China before, but we looked the same to them. We would show them the wrong passports -- my friend has curly hair, different eye color, and he's taller and wider than me. They would look at them and nod and let us pass through passport control. It's a skill that you have to acquire by looking at people, and once you do, you don't see the similarities any more.

AH: Are there some countries that have figured out better than others the best balance of pleasure and discipline leading to increased happiness?

DA: The Italians. Italy has lots of problems, too. But in terms of personal life, taking politics out of it, they have a much healthier balance than we do. They don't move as much. They don't work as hard as we do. They spend more time with friends. They go out more. They take much more vacation. I think all of that really affects happiness.

AH: How about the Israelis? We don't generally think of them as conventionally happy.

DA: The Israelis, of course, have the Israeli-Palestinian problem, and security issues. But if you go to Tel Aviv, as long as you don't open the newspaper, you're not aware of this. Israelis actually have a tremendous quality of life. You go into Tel Aviv at four in the morning, and the city is more active than New York. The food is fantastic. People are social. Rather than walking down the street and not talking to anybody, people will chat with you -- I sit in coffee shops in Tel Aviv, and people just approach me and talk. Lots of people are involved at some startup at some point, or thinking about some idea -- and that's wonderful.

AH: Lets move from countries to businesses.

DA: Starbucks has a genius for understanding human behavior. They turned coffee into a ritual and experience, and managed to differentiate it so we don't think about other coffee as equivalent. I think that in many ways people in the fashion industry understand a lot about human psychology. I think the President actually knows a lot about and understands much of it.

AH: Who doesn't?

DA: The people who are doing online content. I think newspapers basically put themselves out of business.

AH: I know it's a huge subject, but do you have any thoughts on what newspapers should do differently?

DA: I think there are a few possibilities. One is that this escalation to free content is just a fight between newspapers to go to the lowest common denominator. Now everybody is stuck. They've got to create a new platform; we're not all of a sudden going to be willing to pay for it, but maybe there's a newer version of a newspaper that we would.

Another approach is to use the name-your-own-price perspective. More and more people are doing experiments on name-your-own-price. It turns out that people do have a sense of fairness and justice, so if you could get them to understand that content costs something to produce, that might work.

AH: How would we put your Ikea Effect to work here? Would it be to get somebody to help co-create a newspaper?

DA: That's an interesting idea, but it's a longer-term perspective. Imagine if we introduced curriculum in schools where kids learned how to become journalists. They'd see how difficult it is to write well. Then they'd value it more.

AH: You have a great chapter on revenge, and describe some fascinating experiments that have been done. Turns out that when we exact revenge, there's actually increased activity in the parts of our brains associated with reward. It makes us feel good, even if it costs us money to punish someone. It made me think of what's happening in Europe right now. Some countries want to punish the Greeks, even if it hurts the Euro.

DA: Yes, absolutely. That's a nice way to think about it. The Germans basically think that the Greeks and others deserve to suffer, so even if the European Union will suffer, too, it's okay.

AH: Towards the end of your book you write about government not doing enough experimentation, not enough small-scale work that could lead to bigger insights that we can use to solve problems. Do you have an example for Barack Obama?

DA: I'll give you a trivial example. Both Obama and Bush gave money as tax rebates. We're talking somewhere in the vicinity of $150 billion. Not pocket money. They had slightly different strategies of doing it. In the Bush Administration we got a lump sum check, direct deposit; in the Obama approach, we got reduced deductions for a long time from our weekly paycheck. But the reality is that there are many ways to do it. I could give you a check. Direct deposit. I could give you a prepaid debit card with $600 with Obama's smiling picture and the statement "spend the government's money." But we could come up with ten versions like this, and try them out.

AH: It's a good idea. But government takes so long getting to one solution, I can't imagine them ever agreeing on multiple options.

DA: We're not talking about something difficult. I'm not talking about experimenting on No Child Left Behind. I'm not talking about experimenting on healthcare. All of those are really important as well. Even in some really simple stuff, I think we could experiment and get improvement.

AH: Are there any governments around the world that are more original and innovative with trying to achieve better outcomes by experimentation?

DA: Not so much. But there is one example I really like, which is Chile. About 32 years ago, Chile decided to follow the Chicago School of Economics and decided that they don't need to tell people to save, that they would just do the right thing for themselves. Of course, thirty years later they woke up and realized that people don't save anything. So now they've created a really smart system. They basically force people to save 11% of their income, and they can actually manage their savings any way they want. Then when people turn 65, they take groups of 250,000 people, and package them up so insurance companies can create annuities for them.

People do two things badly. They don't save enough, and then they don't buy annuities. This system does it for them; it also solves the market deficiency problem. When people buy annuities, there's a self-selection problem -- a "symmetry of information" issue -- which leads to the bad pricing of annuities. Chile solves it by bunching people and selling them off in big groups of people. So it's a really clever approach.

AH: Take something like carpooling, which we can't get enough people to do. Any quick brainstorms?

DA: Here is a very trivial proposal. Imagine that we gave people stickers with different colors to put on their cars, stickers that would say what day of the week they are carpooling. Now you could look at the people who are parked around you, and see who else is going to carpool on similar days to you.

If you knew who else in your neighborhood and who else in your workplace was carpooling, I think that would be very helpful as a first step. Stickers would give you some social utility by showing other people that you carpool.

AH: Speaking of social utility, you write that a simply apology means a lot. But we're in the midst of an apology epidemic, from CEOs to politicians to golf mega-stars. Are we sorried-out?

DA: Do we really believe what they're saying? That's the first thing. I agree, you shouldn't do it too much. But I think it's more effective than people think it is.

AH: Is the reason we respond to a simple "I'm sorry" that we want other people to let us off the hook when we mess up?

DA: No, I don't think it's about reciprocation. I think it's about really defusing your anger. Think about customer service. Something really bad happens, and you call, and somebody says, "You know what? Your television broke. I'm really sorry. It shouldn't happen; you're right." Many times that's all we want -- something to defuse our anger.

AH: Despite all our irrationality, you're actually an optimist. Or maybe, it's our irrationality that inspires you.

DA: If you believe that people are rational, then you believe that this is the best we can do and the state of the world is in some sense the best that it can be. But if you believe that people are irrational, then it means that the world is like this only because we haven't figured out how to do things better.