 Dan Ariely is our next speaker. He's talking the honest truth about dishonesty, how we lie to everyone, especially ourselves. He's the James B. Duke professor of psychology and behavioral economics at Duke University. He's got two New York Times bestsellers, predictably irrational, and the honest truth about dishonesty, and his haiku is this. I have never lied, except for what I just said. No wait, that's not true. Please welcome Dan Ariely. So, I want to talk a little bit about dishonesty, and how many people here have lied at least once this year? How about the last week? I'm not going to ask about the last day and the last hour, but there's a very disturbing study in which they take two people who don't know each other, put them in a room and say, talk to each other for ten minutes, introduce yourself to the other person, and then they put them into separate rooms and say, did you lie to the other person? And almost everybody says no. They say, well, luckily we taped your discussion. Let's play back to you sentence by sentence, and let's get your reaction to each sentence, and on average people admit to lie between two and three times in those ten minutes. So if you think about it, the truth is that we lie a lot, we lie frequently. Lots of it is for to be polite or for social grace and so on. By the way, there's a story in the Bible. God comes to Sarah and says, Sarah, you're going to have a son, and Sarah laughs. She said, how can I have a son when my husband is so old? God said, don't worry. Then he goes to Abraham and said, Abraham, you're going to have a son, and Abraham says, did you tell Sarah? And God said, yes, and Abraham said, and what did Sarah said, and God lies. God said, Sarah said, how could you have a son when she is so old? And the religious scholars have wondered how could God lie, and they came up with a conclusion that it's okay to lie for peace at home. And the realization, of course, is that there are many human values out there, honesty is one of them, but not all human values are compatible all the time. And what do we do with sometimes human values don't fit? Something has to give, and honesty often gives, and we understand that, right? We understand that the right answer to the question, honey, how do I look in that dress? Is not always the perfect truth. So on one hand, we realize we tell lots of lies, and we do it for all kinds of social and other reasons. On the other hand, how many people in here in general think of yourself as wonderful honest people? Right, the same, the same group. And the question is, how can that be? How can it be that on one hand, with lying, we recognize that on the other hand, we feel good about ourselves? What are the mechanisms for lying? So to look at this honesty in general, we have to come up with some method that we could quantify this honesty. And we kind of developed a very simple approach. If you're in the experiment, in one of the many experiments we run, one of the methods look like the following. I would give you a sheet of paper with 20 simple math problems. These are math problems that everybody could solve if you had enough time, but I wouldn't give you enough time. I would pass the sheets along, I would say, wait, I would say turn it over, solve as many as those as you can in five minutes. I'll give you a dollar per question. You have five minutes, go. You would start working as fast as you can. At the end of the five minutes, I would say, please stop, put your pencil down and count how many questions you got correctly. Now take your sheet of paper, go to the back of the room and shred it. Come to the front, tell me how many questions you got correctly and I'll pay you accordingly. People do this, they shred, they come forward, they said they solved six problems on average. I paid them six dollars, they go home. What the people in the experiment don't know is that we played with a shredder. Talking about this honesty, the shredder shred the sides of the page, but the main body of the page remains intact. So when you put your sheet in, you feel the vibration, the noise, everything feels the right way, but we can actually see how many questions people really solve correctly. And what do you think we find? On average people solve four problems and reports to be solving six. And this is, this shift in the average is not due to a few people cheat a lot, it's due to a lot of little cheaters. I'll tell you about lots of experiments today and there's many I would not tell you about. In total, we've run these experiments on about 35,000 people. And from those 35,000 people, we found about 20 who were big cheaters and together they stole about 250 dollars from us. We also found 25,000 little cheaters who stole about 50,000 dollars from us. And I want you to think about that. Sure, there are big cheaters out there, but there are very few of them. But on the other hand, there's lots of little cheaters who could cheat just a little bit all the time and that actually has a tremendous economic impact. Okay, so now we have a method for quantifying dishonesty. Now let's think about what is the theory of dishonesty? The standard theory of dishonesty is the cost-benefit analysis. It says every time you're considering an act, you say to yourself, what do I stand to gain? What do I stand to lose? If you think about stealing something, you say, what's the probability I'll be caught? How much time will I get in prison? Is it worth it? And you do the cost-benefit analysis. Now if that's the case, all we need to do is to do harsh punishments and people would do the cost-benefit analysis and we say not worth it. But is that the case? So first of all, let's think about something like the death penalty. The death penalty is a very harsh penalty. It's hard to think about something worse. And you would think to yourself that if people do the cost-benefit analysis, states that have the death penalty would not have crime. At least not the crime that you can get the death penalty for. But any statistical analysis that we can do doesn't show any difference in crime rate between states that have the death penalty and states that don't, suggesting people don't really do the cost-benefit analysis. You can also ask yourself subjectively, when was the last time you thought about the world this way? You must have gone to some friends for dinner and maybe you went to the bathroom to wash your hands and they have these really new nice towels and you know you don't have nice towels and you had your backpack with you and you could have taken them and gone home. Did you ever think about things like this, right? I suspect that most of you haven't, right? It hasn't crossed your mind. Maybe now. Tonight maybe we'll use a lot of cutlery at the dinner or something, but in general, we don't think about things like this. So it doesn't seem to describe the macro data. It doesn't seem to describe the subjective experience. What about our experiments? Our experiments show the same thing. We play with how much money people stand to gain. We don't see a difference. We stand with how much? What's the probability of being caught? We don't seem a difference. So the economic rational model, while seems appealing, that people do the cost-benefit analysis, doesn't have much evidence in our data or in any other data. So what does happen? Because you could say, oh people are angels, but people are not angel either, right? We do see lots of dishonesty. So we thought maybe what's going on is that people try to maximize two things at the same time. On one hand, we want to look at ourself in the mirror and we want to feel that we're good, honest, wonderful people. That can eager utility. Think of ourself in a good way. On the other hand, we want to benefit from selfishly, from cheating a little bit. And you can say you can't do both. You either look at yourself as an honest person or you benefit from being slightly dishonest. You can't do both. Well, maybe you can. Maybe as long as we cheat just a little bit, we can still think of ourself as good people, right? Maybe there's a barrier that says that if you cheat a lot, you have to think of yourself as a thief. But if you cheat just a little bit, all of a sudden you can think of yourself in stealing good eyes. So for example, how many of you have illegally downloaded material in your computers? Okay, not everybody actually. I guess some people just don't know how to do it. And from the people who have this, this information, how many of you would feel bad if other people knew? Almost, almost, right? How can it be? Now think about something else. How many of you would feel comfortable going and having a meal in a restaurant and living without paying? You would feel comfortable? Good to know. Most of you say you would not feel comfortable. I'm guessing you just... So actually in the last few years, every time I go to eat somewhere, I ask the waiters if they can recommend a way to eat and escape without paying. And I usually say, look, I'm not going to do it. I'm just interested. Like how would you do it if you were going to do it? Sometimes they ask me for my credit card. But usually they give me really good advice. They say, wait for a big group to come in. There's an alley. There's a bathroom. They have suggestions of how to do it. And then I say, how often does it happen? And they say almost never. They say sometimes people forget to pay because they don't pay attention to the credit card. They start working without paying, but they say it almost never happens. Now think about these two cases. Illegal downloads, eating without paying. The probability of being caught is very low in both of them. Illegal downloads, if the music industry catches you, you can't say, oh, I didn't think about it. I'm sorry. I didn't pay attention, which you could do in a restaurant. Nevertheless, it's how we feel about it. It's not about the consequences in the real world. So let's go back to our little experiment and say, how would we now modulate the amount of cheating that happens in our experiment? How do we get people to cheat more? And how do we get people to cheat less? And remember, it's not about the cost-benefit analysis. What else would get people to change the tolerance around cheating? And if you think about it, it's all about thinking of yourself as good people and trying to benefit from dishonesty. It's all about rationalization. It's all about saying, how can we justify this little bit of cheating? And if we could justify a little bit more, then we will cheat more. And if we could justify a little bit less, we would feel less. So let's think about rationalization. What kind of things do you think change rationalization? What kind of things are good input that allow people to rationalize more? Just raise your hand and give me some suggestions. Anonymity. So anonymity is usually the multiple interpretations, but one of them, I can't be caught. Another one is you think of yourself as part of a crowd. Everybody else does it. Let me tell you one experiment about everybody else does it and we'll continue. All are good. Let me just tell you about the experiment at a time and then we'll continue with suggestions. So one of the suggestions was everybody else is doing it. So imagine it's the same experiment. I pass the sheet around, but there are two changes. The first change is that I give you an envelope with all the money for the experiment in advance as if you solve all the problems. And I ask you at the end to pay me back the money you did not make. Change number two is I hire an acting student. That acting student sits in the front row and 30 seconds into the experiment they raise their hand and they say, excuse me, I solved everything. What do I do now? Imagine you're in the experiment. Somebody raised their hand, they solved everything. You know that they're cheating. There's no question. You're still in problem number one. And the experimenter said you solved everything, you're free to go. And you see that person taking their envelope and walking, leaving nothing behind. What would happen to your own morality? Lots more people cheat. But there could be a couple of interpretations. One interpretation is that you're saying, well, in this experiment there's no downside to cheating. Look, somebody cheated in an exaggerated way all the way. Nobody changed him. Nobody said anything. There's no consequences. The other possibility is to say in our society people like us seem to behave like this and this seems to be okay socially. So how do we separate those two explanations? We tried multiple things. One of them was that we changed the outfit of the acting student. And here's the story. We read this experiment at Carnegie Mellon. Everybody was a Carnegie Mellon student. The acting student was a Carnegie Mellon student. He was wearing a Carnegie Mellon outfit. In the second version of the experiment he was wearing a University of Pittsburgh sweatshirt. Now what happens when you're a University of Carnegie Mellon student and you see University of Pittsburgh student cheats? From the cost benefit analysis, you still know that in this experiment people can cheat and get away with it. But from the perspective of feeling okay that people like me are doing it, that doesn't work. Now you say oh it's these other people. They are the cheaters and we don't want to be like them and cheating actually goes down. So it is about everybody else is doing it but people like me rather than everybody in general. People that we identify with. You said they deserve it. They're charging too much. We can go into all kinds of personalization you might have about this. So they're doing something bad to me. So we did another version of this experiment. This was in Boston. We came to people in coffee shops and we say hey, would you do this study for five minutes for five dollars? People said yes. We get in the sheet and we explain to them the task. It was slightly different than the task I described to you. And then we said we'll be back in five minutes. Left came back in five minutes. We said here are your five dollars. Please count it and sign a receipt for five dollars and leave the receipt on the table. But we actually gave them nine dollars. How many Bostonians do you think gave us the money back? Slightly more than 50 percent and you could decide if it's a lot or a little. In the second condition, as the research assistant explained to them the task, here is what you do. He pretends somebody called him on his cell phone. He picked up his phone and he talked for 12 seconds to John about pizza tonight. Hey John, what's up? Pizza tonight? Great. Put the phone back. Went straight to explaining these instructions. Said nothing. Left, nine dollars, here's your five. What happens now? Not only 14 percent of the people gave the money back. Right? People said oh, he did something bad to me. That's okay now. I'm restoring karma to the world by doing that. Very good. So other people are doing it. What else would change rationalization? Just raise your hand. Yeah. That's right. So you're not stealing anything. So when we talk about illegal downloads, there's lots of ways to rationalize it. You could say there's no marginal cost. You could say the musicians really want the music to be heard. I wouldn't buy it otherwise. I'm actually doing something good. There's lots of stories you can tell yourself. What else? Person relationship. Do you mean that if you know the person face to face, you might not want to harm them when you don't know them face to face. You would feel differently about it. Absolutely. That's right. So there's something called the identifiable victim effect, which is if the effect is larger and it hurts people in general, you don't feel the effect. If it's something specific, you feel it. I'll give you one. Oh, yeah. You know, we tried that. We couldn't find evidence for that, but I felt the same intuition. We just didn't find evidence for this. So here's another version of this. There's a little joke that little Johnny comes home from school with a note from the teacher that said that little Johnny stole a pencil from the kid who's sitting next to him. And Johnny's father is furious. He said, Johnny, I can't believe you did it. And you never, never, never steal a pencil from the kid who's sitting next to you. You're grounded for two weeks and just waiting till your mother comes. And beside Johnny, you know very well that if you need a pencil, all you need to do is to say something. You can mention it. You can ask for a pencil and I can bring you dozens of pencils from the office. Now, why is this slightly amusing? Because I think we all recognize that the pencil feels very differently from taking 50 cents for a petty cash box. If you took 50 cents from petty cash box, you would feel like you're a thief. Pencil? Not so much, right? Everybody's doing it. It's there for people to take. In fact, even if you took 50 cents and went to buy a pencil, you would feel yourself that you're a thief. So this idea here is that the distance between what we do and the consequence, the people and so on, is a major aspect of whether we feel comfortable or not. So we've done this in a couple of ways. The first one is we looked at golf. Anybody here plays golf? Okay, a few people. So we did a study about about 12,000 golf players. We asked them many questions, but one question was, imagine the ball fell on the rough. Not a good place. And you really, really, really wanted it to be four inches to the left. But it'd been much better. Would you pick it up and move it by four inches? And people said, heaven forbid. I can't imagine doing that. If you're asking this question, it means you don't understand golf. Nobody I know about, nobody does it. Impossible. Not part of the game. Okay. What about kicking the ball a little bit? Yeah, we do that all the time. What about hitting it with a club? That's easy too. By the way, the easiest one is if you're not looking. So you look up and then you kick it a little bit. That's the easiest. Going back to the standard experiment, we did the same experiment I described to you earlier. We passed the sheets of paper, people shredded them, came to it and said, Mr. Experimenter, I solved X problems, give me X dollars. In another version, they looked at the experimenter and said, Mr. Experiment, I solved X problems, give me X tokens. And we paid them in pieces of plastic. They walked 12 feet to the side and changed them for money. So the difference was when you look into somebody in the eyes and you lied, you lied for something that was not money, but it was going to become money very quickly. What happened? In this experiment, people doubled their cheating. And by the way, this for me is one of the most worrisome experiments that we've created. Because if you think about it as a society, we're becoming distant from a direct relationship at distance from money, cash, credit cards, electronic wallets, cash stock, stock options, derivatives, dealing with people directly, dealing with people over great distances. If it's possible that as these distances increases, people feel better about misbehaving than we really need to worry more and more about lots of people cheating a little bit and still feeling good about themselves. And if you think about everything we've said so far, it's really about conflicts of interest. It's really about the fact that if we want to see reality in a certain way and we have the conditions to rationalize it, we can do it. So imagine, for example, that we create something like the financial industry and Wall Street. And imagine that you're an investor in Wall Street in 2005. And I promise you a $5 million bonus if you could only see mortgage-backed securities as a good product. Now ask yourself whether you couldn't see them as better product than they really are. I'm not saying would you lie, would you say, oh, I know that they are terrible, but I'll pass them to my clients. No. Wouldn't you actually shift your belief to believe that they are better than they are? And what if you have some belief in the market? The market is always right. And what if you said, everybody else is doing it? And what if there were multiple steps removed from money and multiple steps removed from the people that you're dealing with? Now you can see how you can create a system that has inherent conflicts of interest and lots of opportunities for rationalization. And no matter who you put in there, you're going to get a bad outcome. You know we can say, oh, this banker is bad and this banker is bad. The problem is that this system is supporting lots of misbehavior. I'll tell you one other story about conflicts of interest. So as you can probably tell, I was badly burned many years ago. I spent a long time in hospital. And about five years after I left the hospital, I came back for a checkup. And the head of the department finds me. He said, Dan, I'm so happy you came. I have a fantastic new treatment for you. Great. I come to his office and he says that when I shave, I have stubble. I just shave, so not as much. I have stubble on my left side. My right side is burned, so there's no stubble. There's no hair. So what is he proposing? He's going to tattoo the right side of my face so that it equals the left side of my face. And he says, go home, shave tomorrow, and come back. And I drive home and I think to myself, what kind of shave do I want to be symmetrical? The morning shave, the afternoon shadow. Like, you know, this is an important decision. So I come to his office and I say, you know what? I'm not sure I want this. Can you show me some pictures of people you've done it? He said, yes, we've done it for two other people. He can't show me the whole face, but he shows me their cheeks and sure enough it looks like, you know, little black dots. And then I said, what happens when I grow older and my hair becomes white? He said, don't worry, we can laser it out when the time comes. And then I say, you know what? I'm not sure I want this. Not sure it's for me. Then he looks at me and he said, Dan, what's wrong with you? He said, do you enjoy looking non-symmetrical? Do you get some deprived pleasure from looking different? He said, do women feel sorry for you? And, you know, he said of some other things, which never happened. I left his office and I was really baffled. And I went to his deputy and I said, what's going on? And he said, well, you know, we've done this for two patients already and we need a third for an academic paper. Now, here's the thing. This guy was an amazing physician. He passed away 10 years ago, but he was an amazing physician. You see the right side of my right eyebrow? It was burned. There was no eyebrow. And he wanted to operate on it. He wanted to create a new eyebrow. I couldn't care less. But I told him there was another operation I was going into. If he wants to do it at the time of the other operation, that's fine. I'm not doing a special operation. I had about 30 operations, another one for half an eyebrow. It didn't look like a good deal to me. And so I went into an operation from my hand. It took about eight hours, but turns out they couldn't, he couldn't do it. He didn't have enough space around me. So he waited the whole day for them to finish and then worked on me most of the night. He took him about eight hours. It's a very, he took a Doppler machine. He tracked the blood vessels. He created really amazing operation. I, the point is that I can't think of this physician as anything but a kind, wonderful person. But at that moment, he had a conflict of interest. And at that moment, he wanted to see this paper out and less cared about my individual benefit. Okay. So all of this is about conflicts of interest and how we could get people to rationalize more. How can we get people to rationalize less? What do you think would get people to cheat less? So of course we can do the opposite of what we've done so far. But what else could happen? What else could work? Ten Commandments. Ten Commandments. Very good. So there's a little joke. There's a little joke. A guy goes to the rabbi and said, rabbi, you wouldn't believe what happened, but somebody stole my bicycle from synagogue. And the rabbi is appalled. Stealing your bicycle from synagogue? This is terrible. So I'll tell you what you do. Come to synagogue next week and sit in the front row. And as we go over the Ten Commandments, turn around and look at everybody in the audience. And when we get to thou shall not steal, you know who's your thief. The guy's very excited. The rabbi is excited. Comes to synagogue, sits in the front row, turns around during the Ten Commandments. At the end of the service, the rabbi waits for him and says, so did it work? And the guy said, like magic, like a charm. He said, the moment, the moment we got to thou shall not commit adultery. I remember. Do I left my bike? That's what's needed to get applause. Good to know. So where's the experiment in this? As was suggested, we went to UCLA and we asked 500 undergrads to try and recall the Ten Commandments. By the way, none of them could recall all Ten Commandments. Many of them invented new interesting ones. This might be a California effect. We don't know. But then after we asked them to try and recall the Ten Commandments, nobody cheated. And it had nothing to do with whether they were remembered more Commandments or less Commandments. In fact, even when we ask self-declared atheists to swirl the Bible and then we give them a chance to cheat, they cheat less. They don't cheat at all. So it's not about the cost benefit of heaven and hell. It's about reminding people about their own moral fiber. And within an experiment like this with the big insurance company, we got some people to fill their odometer reading and then sign at the bottom. And some people signed first and then filled their odometer reading. And the people who signed first cheated by much less. We don't know if they didn't cheat at all because we didn't check their odometer, but compared to the other group, we know they cheated by 15% less. So this suggests that just priming people with their own morality can do something good for dishonesty. There are two more points I want to make. The first one is about what happens to dishonesty over time. So in these experiments that I've described to you so far, it was a one-shot cheating. People have one time. But sometimes we give people hundreds of opportunities to cheat over time. And one of the things we see is that people cheat a little bit. Cheating a little bit, feeling good, cheating a little bit. And then at some point, they switch and start cheating 100% of the time. And different people switch at different points, but the switching pattern is very common. And we call this the what the hell effect. And if you're a dieter, you know this thing, right? You know that you're kind of on a diet and then you start with a muffin and say, I'm not on a diet anymore. I'll start tomorrow or next Monday. And the issue is that we really think of ourselves in binary terms. We're either dieters or not. We're either honest or not. If you're 72% honest, is it worthwhile to move to 78? No, you might as well enjoy it. So we said, okay, if people have this what the hell effect pattern, how could we cure people? How could you get over it? In religious terms, if people think that they're going to hell, why would they ever behave well? So we thought about Catholicism. Anybody here grew up Catholic? Okay, a few. So this is no offense to the Catholics or the ex-Catholics in the audience. But we went to talk to real Catholics. We went to Italy. We went to talk to Catholic priests. And we asked them, please explain to us the logic of confession. After all, if you can confess and get absolved, wouldn't people cheat more? If all you want is to minimize time in purgatory, wouldn't you cheat on the way to confession? But turns out that's not the case. So we came up with a few theories for how confession might work. One is the cost benefit. You think to your future and you say, I might get caught, but for sure I'll have to talk to the priest and that will be terrible. Not worth it. We find no evidence for that. The second possibility is you come out of confession and you feel good and pure and wonderful and you don't want to destroy that feeling for a few days or a few hours. We have some evidence for that. But what about the what the hell effect? So we tried that as well. In the lab, people cheated a lot, switched to cheating a lot, a little, cheated a little bit, switched to cheating a lot. We gave them a chance for a non-Catholic confession. Write some things you've done wrong, write something that you ask forgiveness, shred those pieces of paper. What happens after both of those? Cheating goes down dramatically. So this is a case in which religion has actually figured some interesting mechanism about opening new pages. And I think it's interesting to ask, how could we integrate the general wisdom opening new pages into secular society? What would it look like if we tried to get our politicians and bankers into a ceremony of that? Should we integrate things when people reclaimed their vows and do other things to basically open new pages? And finally, I want to say something about cultural difference, differences. How many people here grew up in a country that is not the US? Raise your hands. Okay, keep them up. Now, how many of you think that in your country of origin people cheat less than Americans? Less than Americans? Keep your hands up. Canadian? So I grew up in Israel. So the first country I went to test was Israel. We test with the same procedure. How many of you think that in Israel people cheat more than the Americans? This is a much more politically correct group than I expected. Anyway, the Israeli cheat just like the Americans. Francesca Gino, my Italian collaborator, said, come to Italy. We'll show you what the Italians can do. But the Italians cheat just like the Americans. We tried China. We tried Turkey, Germany, Colombia, South America. And so far, we find no differences. How can it be? Anybody who traveled in other places gets the feeling that cheating is very different in different places. How can it be? And here is what we're finding. We find that our experiments are not capturing anything cultural because they're not embedded in culture. They're abstract. They're general. It's the first time people experience them. And because of that, they measure the general ability of people to cheat a little bit and feel good about themselves. But that doesn't mean that culture doesn't work. Culture works on a domain by domain specific. Culture doesn't change a human fabric or backbone. What culture does is you take a domain, illegal download, and say, don't worry about it. Culture can take something like paying a bribe to a policeman and say, don't worry about it. Culture can take something like cheating in school and say, don't worry about it. So culture is important. But culture doesn't change people across the domain. They just change them across the whole domain. They change domain by domain. We did find one difference. When we run these experiments, we either run them at university or in bars. And when we run them in bars, we change the payment such that every four questions people solve correctly, they get their money equivalent to one glass of beer in this establishment. So beer becomes our international currency. So we ran this experiment in a bar where congressional staffers hang out in. And we read it also in a bar where bankers in New York, Wall Street, hang out in. And this was the only time we found a difference. Who do you think cheated more? The bankers or the politicians? Who votes for politicians? Who votes for bankers? Huh. Okay, so slightly more for the politicians. The bankers cheated two to one, twice as much. But there are two things we should consider. One is that this was cheating in money, which is domain of bankers much more than the politicians. And the second thing, these were junior politicians, congressional staffers. Lots of room for growth. Let me just finish with the following. I told you a little bit about Judaism and Catholicism. A guy goes to confession and says, forgive me, father, for I have sinned. The father said, what have you done, my son? And he said, I'm 72 years old and I just had sex with two 25 year old twins. And the priest is appalled. He said, I can't believe this is happening. I want you to say 700 times, help Mary and walk around the church a thousand times and give $1,000 to charity. And how long has it been since your last confession? I've never been to confession. What? You're 72 years old and you've never been to confession. How can that be? I'm Jewish. So if you're Jewish, what are you doing here and why are you telling me? I'm telling everybody.