Predictabily Irrational | Dan Ariely | Talks at Google

Speaker 0:03 I'd like to introduce, uh, Dan Ariely. He studies, as most of you probably know, behavioral economics, which means looking at the way people actually act in the marketplace as opposed to how they would act if they were behaving completely rationally. He's a professor of behavioral economics at Duke University's Fuqua School of Business. This is a new role as of today. Previously, he was a joint professor at the MIT Media Lab and the MIT Sloan School of Management. His latest book, Predictably Irrational, uh, covers some of his research findings, including those that look at, uh, how people behave in, in everyday situations where they're predictably a- a- uh, operating in irrational ways. So with that, uh, please join me in welcoming Dan Ariely [clapping] .

Dan Ariely 0:48 So, uh, thanks so much. Um, it's, it's nice to be here. I was in your New York office, uh, a few days ago, which was also nice. I'm, uh, having, uh, two weeks of Google. Um, I want to tell you a little bit about, uh, behavioral economics, and if you have questions in the middle, feel free to ask. And, and I, I want to start by telling you why I, I chose this, uh, title of, uh, Predictably Irrational. Uh, because while people sometimes, uh, economists acknowledge that people make mistakes, they think that those mistakes will cancel each other in, in the market. So you make one kind of mistake, I make another kind of mistake, and in the market they will cancel each other out. When-- What I mean by predictably irrational is not that. It's that we have the same type of mistakes that we all do repeatedly, and in fact, in the market they could aggregate. And I think if we think about the subprime mortgage crisis, we can get some indication of how, how these mistakes could actually become worse. Um, but, but irrationality is also something, uh, very

Dan Ariely 1:49 individual and personal that touches us individually and I, and I want to tell you how I, how I started, uh, getting, I guess, interested or introduced to irrationality, and it, it happened to me when I was in hospital, uh, which is a place where you would, uh, see many irrational behaviors if you, uh, ever get to spend some time there. So I was, uh, I was burned in an explosion, uh, many years ago, uh, about seventy percent of my body, and I spent three years, three years in hospital. And, uh, you know, hospital is terrible in many ways. Uh, but the thing that, that was most, uh, difficult in the burn department is the process of removing bandages. So, uh, if you ev-ever had a bandage removed, you must have wondered what's the right strategy. Should you rip it off quickly, right, short duration, but high intensity? Or should you take it off slowly, take a long time, but have a low intensity? Well, for me, this was the most important question I could think about. Um- [laughing] ... it's, it's, it's one thing when you have bandage and you still have skin and just a little bit of glue. It's a completely different thing when you don't have skin and the bandage, bandages adhere to the, to the flesh, and it's becoming incredibly painful to, to take them off. So every day, the nurses would come to my room, they would take me on a little forklift and put me into a bath of water with iodine. They would soak me a little bit to release the, some of the stickiness of the bandages, and then they would start ripping them off. And, and the nurses basically had the belief that the right way to do this is to rip the bandages one after the other as quickly as possible, right? So they would, they would start at my feet, and an hour later they would finish, uh, at, at my head. And I hated this intense moment of ripping that just were just awful, right? I mean, it's, it's very hard to describe how, how painful it is. Thankfully, we don't have good memory for pain in this intensity. Um, and I used to argue with them. You know, argue is kind of a polite word. I used to, to scream and beg and, uh [laughing] ... and request and, and say, "Can we, can we try something else? Can we try and take it off a little slower, maybe take an hour and a half, maybe two hours, but just don't have these incredible sensations?" And, and the nurses told me two things. They told me that they were right, that this approach was the one that would minimize my, my pain, and they also reminded me that the word patient doesn't mean getting involved and disturbing them [laughing] uh, in the process. Uh, it's the same word in Hebrew basically that says sit there quietly and, and just be patient. So, um, three years later, I left the hospital. I started

Dan Ariely 4:21 studying at university, and I learned about the experimental method. I learned that if you have a question sometimes you could bring into the lab, you can manipulate something, and maybe you can find an answer to what you were doing. And I was still very puzzled by this question of if you have an experience, a negative experience you have to give to somebody, what's the optimal way to give it in order to create the least amount of negative effect, right? So, uh, I didn't have much money in the beginning, so I went to a hardware store and I bought a carpenter's vice, and I would bring people to the lab, and I would crunch their fingers a little bit. [laughing] And I would crunch them for longer periods and shorter periods and increasing pain [laughing] and decreasing pain. Uh, you remember the pain suit I developed earlier? Anyway, uh, th- then I would ask them, "So how painful was this?" or "How painful was the previous one?" and "Which one would you rather do if you have to choose one of those to do again?" [laughing] Uh, then like all good academic, uh, projects, I got more funding. I moved to electrical shocks and [laughing] and sounds. I even developed a, a pain suit in, in the media lab which was particularly, particularly interesting. And at the end of this whole process, what I learned was that the nurses were wrong. I learned that I would have had less pain if they would have had a longer treatment with lower intensity. I learned that I would have had less pain if they would have started from my head, which was most painful, and moved slowly toward my feet, which was less, less painful, decreasing trend. And I learned that I would have been better off if they gave me breaks in the middle. The question was-- I went, I talked to them, we had all kinds of discussions. That's a separate story, but the thing that bothered me in a more general way is how could the nurses be wrong? This was something that was quite important. It was important to me, but it was also important to them. This was a, a real big issue. Uh, it was something that they had plenty of experience with, and they had very good intentions. It's not as if they were trying to figure out the, the worst way to de- to remove my bandages. And the question is, how can it be that in high-stake situations with good intention

Dan Ariely 6:22 and plenty of experience, people nevertheless get things wrong? And interestingly enough, one of the, the insights for why they got things wrong came to me when I went to talk to them. Uh, so when I went to talk to the nurses, uh, one of my favorite nurses said, "What about my pain?" I mean hers. She said, you know, "It was very unpleasant to the, to remove the bandages from me." And she said that maybe what she was doing is not willing to listen to me to try out something different because it would have caused her unpleasantness. Now, clearly, it's not the right approach. Uh, they should have reduced my pain and not, uh, her pain was secondary. But, but she basically said that maybe her reluctance to try it out, maybe her reluctance to try something else was that she just did not want to expand this duration in which she was wanting it to be over as well. Anyway, so from then on, I started thinking about all kinds of other cases in which people with good intentions and plenty of experience nevertheless get things wrong and, and I want you to, to give you some examples of those today. Uh, now I'm not going to hurt you.

Speaker 7:28 [laughs]

Dan Ariely 7:31 I should say, I did, I did one study, uh, where I, I hired, uh, this... I, I had a friend who was a very attractive woman who would go ahead and recruit subjects-

Speaker 7:41 [laughs]

Dan Ariely 7:41 ... uh, for these pain, for these pain experiments. And I-- this was not what I told her to do, but this is what she did. She would come to men, and she would say, "Can I hurt you?"

Speaker 7:50 [laughs]

Dan Ariely 7:53 And inevitably, we got a lot of subjects.

Speaker 7:56 [laughs]

Dan Ariely 7:59 So, so to give you some, some ideas about irrationality, I want to start with a couple of visual illusions. So if you look at these tables, and I ask you what's longer, the vertical axis on the table on the left or the horizontal axis on the table on the right, how many people here see the horizontal axis on the right being longer? You really see this one as being longer? Now, you know it's a trick, right? I wouldn't show you if it wasn't a trick. But in fact, it's very, very hard to see this as, as being longer. Now, with visual illusions, I can show you it's a mistake. How can I show you? I can put a couple of lines on the board. I can animate them. And to the extent that you believe me that I didn't shrink them, you can say that, "Oh, yes, these two tables are, are actually the same." But, but interestingly enough, with visual illusions, it doesn't help in the sense that if I take the lines off and you look at this, you could say, "Oh, yes, of course. Now I see that they're exactly identical." But in some sense, it doesn't help. You know they're identical, but it doesn't help you see them as identical. Here's another one of my favorite i- illusions. This is by Dale Purves. What do you see as the color of this brown one? Brown, right? I just said brown. What about this one?

Speaker 9:09 Yellow.

Dan Ariely 9:09 Kind of yellow, orangey. Can, can you see them as the same? It turns out they're identical. Can anybody here see them as, as identical? Right. It's, it's basically impossible. Now, what I can do again with visual illusion is hide the rest of the cube and just leave these two squares. And when I do that, you can see that they're actually, actually identical. But again, it doesn't, it doesn't help because if I take it off, you can't, you can't see them as being the same. It doesn't, it doesn't help, right? So, so we have this system that make us have mistakes in the same way repeatedly, and it doesn't really matter. By the way, I have, um... People always think it's not correct. I have a, a, a printed version. So can I ask you?

Speaker 9:49 [laughs]

Dan Ariely 9:51 So just check. You see these two?

Speaker 9:53 Yeah. I believe you.

Dan Ariely 9:55 Okay, you believe me. Thank you.

Speaker 9:56 [laughs]

Dan Ariely 9:59 What, what is the point of this? The point is that our visual system is maybe the best system we have. We have more of our brain dedicated to vision than to anything else. We're evolutionary designed to do it, and we do more hours of vision a day than we do anything else. Nevertheless, we have these systematic repeatable mistakes in vision, and you will never be able to see these tables without seeing it wrong. You'll never able to see this cube. And while I've cured you of the gorilla illusion in the sense that you will from now on, every time you'll see this clip, you will see the gorilla. By no way have I cured you from the, uh, your ability to ignore information that passes in the middle of your visual field, and you're just not thinking about it at the moment. And if we have such mistakes in vision, which is such a good system, what are the odds that we don't have worse mistakes in other systems, for example, financial decision-making, that are not as practiced, that we don't have evolutionary reason to do, and the odd is it's very, very low. So what I want to show you is some examples of irrational behavior in other domains. Now I'm

Dan Ariely 11:00 offering you, do you want the, the blue pill or the red pill? Do you want to think that you're reasonable and rational and noble, or do you want to realize what kind of mistakes we make? And I'll assume we'll take the, the blue pill. Um, so I want to show you a couple of decision illusions, uh, mistakes people make in a regular way. And to start, I, I want to show you this, um, plot. Uh, Eric Johnson and Dan Goldstein collected this data. It's one of-- It's probably my favorite plot in all of social science. And what you basically see here is the percentage of people in different countries that are willing to give their organs to donation. Okay? And basically, you see two types of countries. You see countries on the right who give a lot. By the way, I rounded the numbers up. Nothing is really a hundred. Could be ninety-nine point six. And you see countries on the left that give relatively little. And the question is, what separates them? What makes some countries give so much and some countries give relatively little? And, and when you ask people this question, they think it has to be something substantial, big. It's maybe religion, or maybe it's culture, or maybe it's how much people care about each other. It has to be something kind of substantial. But if you look at these countries... You see very similar countries. You see pairs of countries that are very similar, but nevertheless behave differently. Sweden is on the right. Denmark, very, very similar countries in many ways, is on the left. The Netherland is on the left. Belgium, again very similar countries in many ways, is on the right. By the way, the Netherlands got to 28% after asking every person in the country, sending them a letter asking them to join the program, right? So, so begging only gets you up to 28%. [laughs] [clears throat] Germany is on the left, Austria is on the right. Depending on your view of European history or culture, and you don't want to tell this to the French and British, but you can either think of France and Britain as either similar culturally or not. [laughs] But, but in terms of organ donation, they're very different. So what, what could explain this, this big difference? Yes. Opt-in versus opt-out. Very good. Now ... What explained it? It turns out it's the form at the DMV, right? And here's what happens. In countries where it says, "Check the box below if you want to participate," what do people do? They don't check and they don't participate. In countries where it says, "Check below if you don't want to participate," what do people do? [laughs] They again don't check, but now they participate. [laughs] Think about the responsibility of the person at the DMV who decided how the form is going to be phrased, right? They had the billions of dollars, the responsibility on their shoulders without them realizing the extent of this thing. Now you might say, "You know what? People just don't care. After all, this is what will happen to me after I die. I can't be bothered to lift a pencil and make a decision because the cost in, in terms of, of effort is, is higher than what I can gain from it. What ... Who cares?" But, but in fact, it's the opposite. It's because we do care. It's because we don't know what will happen to us when we die, and how we will reflect on our families and our funeral, and what kind of third-year student will go and ... medical student will get their hands on our organs. It's because it's so difficult. It's because it's complex. It's because we don't know what we want to do, we just take what advice was suggested to us. Now you might say, "Well, you know, this isn't a one-time decision. There are not many decisions we make one time. Maybe it ha- it ... We don't have any of these problems in repeated decisions." So here's another experiment by Redelmeier and Shafer in which they tried the same thing with physicians. So they went to a big group of physicians and they described to them a patient. You have this patient, this 67-year-old farmer who's been suffering from hip pain. You've been treating this patient for a while. Nothing seemed to be working, so you decided to send this patient to hip replacement. Okay, nothing seems to be working, hip replacement. Now, they haven't had the hip replacement yet, but if you ... they're on the process to do it. They're going to meet a specialist and move forward. And then they told half of the physicians, yesterday when you went through the formulary, you looked at all kind of medication, you realized there was one medication you didn't try out. You forgot to try ibuprofen somehow. What do you do? Ibuprofen or hip replacement? Now you'll be happy to know the majority of the physicians thought that ibuprofen is a, is a better choice. After all, hip replacement is dangerous, expensive, irreversible, and so on. To the other group of the physicians, they make the decision slightly more complex. They basically said yesterday when you went through the formulary, you discovered there were two medications you didn't give. You didn't try ibuprofen and you didn't try piroxicam. What do you do? Ibuprofen, piroxicam, or hip replacement? And basically what happened was that they just made the decision slightly more complex and here is what happened. And I, I'm not saying this is what exactly physicians thought, but they basically said ibuprofen, piroxicam, ibuprofen, pi- ah, I don't know. Send him to hip replacement. [laughs] Now, now the majority picked it. It was enough to make the decision slightly more complex for physicians who are experts, well-paid, practice in this thing, that this actually send them again to the default option. Final study was, was done not too far away from here. Anybody here knows Draeger's? It's a very nice grocery store. Uh, Sheena Iyengar and Mark Lepper did this study in which they got people to taste some jams, okay? And sometimes they set a, a jam tasting booth with six jams and sometimes with 24 jams, and the first thing they asked people was ... Uh, they wanted to see how many people would be, would come and look at the jam tasting booth. And as you would expect, more jams were more exciting. So when people saw 24 jams, more of them approached the jam tasting booth. The next thing they measured was how many jams will people actually taste? Turns out it doesn't matter. People try a jam, a jam and a half. [laughs] A jam and a half. Now I should ... Usually I tell my students, of course nobody tried a jam and a half. Some people tried one, some people tried two. Just the average. Um- [laughs] The most important thing, of course, was how many people bought jams. They gave people a coupon for jam that was good for every jam in the store, not just the six and not the 24, any one in the store. Now who do you think was more likely to, to buy jam? People who saw six jams or 24 jams? Six. Six. Very good. So it turns out that people in the six jam condition, 30% of them bought jam. How many of them do you think bought jam in the 24 jam condition? Five. 10. 10, 15. 3%. Oh. Now basically the issue is that jams are not complex. It's, it's sugar and some fruit. [laughs] But, but it turns out that 24 jams are too much. Now generally speaking, a few people have jam on their shopping list, 2, 3%, and those people were not- Overexcited with jam and they were going to buy jam, and they kept on buying jam. But all the extra people who got excited over jam, where there were few of them, basically lost all of this excitement when this happens. And, you know, I... You can think about what this means for search

Dan Ariely 18:35 and getting a lot of, uh, responses. You had a question? Were the jam selected at random? [clears throat] Six from among the 24? Eh, they were not always selected at random, but they were not more desirable- Okay ... than the 24. Good question. [clears throat] Okay, so what is the point of all this? [clears throat] The point is that we have this view of human nature as what Shakespeare... "What, what a piece of work is man, how noble in reason," and so on. And, and while we see often irrationalities in our bosses and spouses [laughs] Oh, you know Tom is getting married soon? [laughs] Um, it's... When we think about the market in general, uh, people think about people as being rational, and we think about the products that we offer and as, as for rational people. Now, behavioral economics doesn't assume, uh, that people are rational. Instead, our, our model of human being is more like this. [laughs] And in many ways, uh, this is a sad perspective of human nature. It means we're myopic, easily confused. We don't know what, what we're doing. You know, all the people here in the room who are doing email at the same time as listening, you know, we think we can do two things at the same time, only later realizing we really can't. [laughs] Um, so in some sense, it's a very, it's a very sad view of human nature. But, but in a very important way, it's also optimistic. And the way it's optimistic is I like to think about is think about free lunches. And here is the idea. When you... In, in standard economics, there are no free lunches. If you start from the idea that everything is at equilibrium, everything is optimal, what can you do to better yourself? Of nothing. Everything is optimal. If you think about people as having all kinds of mistakes, then you can also say, well, it's a sad thing about human nature, but it also means there's a lot of things we can do to improve things. There's all kind of mechanisms that we can actually get, uh, to build, to

Dan Ariely 20:40 build a bet- a better world. And I'll come back to this later. So next, what I want to do is I want to show you some other examples of irrationality. And the next idea I want to present in slight more detail is the idea that we don't actually have well-defined preferences. So, you know, we wake up every day, we open the closet, we decide what to wear, then we open the refrigerator, we decide what we want to, uh, to eat. Uh, we decide how to drive to work, and we have this feeling that we make these decisions all the time. We choose this over this, and these are kind of reflection of our preferences. We have these internal preferences, and we act on them in, in, in life. What if we don't have such well-defined preferences? What if, in fact, we don't really know what we want and what we don't want? That this belief that we have direct access to our brain to feel our preferences is actually more illusory than real. Let me show you a couple of experiments, then we'll come back to this. So imagine I divide, we divide into two groups, and I ask the group on the left to write three reasons why you love your significant other and ask people on the right to write 10 reasons why you love your significant other. And now I give you some time to write. Don't do it now, but, uh, if, if we're doing the real experiment, you would write it down. And then I would ask you, how much do you love your significant other? How likely are you to stay together in six months, a year, two... Maybe this is California, in one month, in two months, in three months? [laughs] How likely are you to have an affair? I ask you, how, how much do you, do you love that person? Will the fact that I present ask you to think about three reasons and ask you to think about 10 reasons should change how much you love that person? It shouldn't. I didn't tell you anything new about them, right? If I told you, "Hey, I met your significant other. Here's some new piece of information," that would be one thing, but I didn't tell you anything new. But of course, it does matter. Who do you think, um, have greater love to their spouse? People who came up with three reasons or 10 reasons? Three. Three. Why? Why three? They didn't have to think as much. Okay. Didn't have to think as much. [laughs] It's easier to think of two. It's easier. That's right. Because you, you're not gonna run out. Oh, th- very good. [laughs] Are you married? [laughs] I hope your wife can see you on YouTube at some point. [laughs] It, it turns out, it turns out, uh... What's your name? [laughs] It turns out he's exactly correct. People run out. And, and as they struggle to come up with reason number six and number seven, they say, "How much possibly can I love that person if I can't come up with 10 reason?" But the interesting thing is that, you know, our spouse is the person we know the best in the world. Uh, what, what should we know better? And if the process of thinking about our spouse shall dramatically change how much we love them, what, what else can we have trust in? By the way, uh, it turns out the same thing goes for BM- BMWs. I don't know if anybody here has or willing to admit they have a BMW, [laughs] but it's the same thing. You ask people to think about 10 reasons, uh, to get a BMW, and they're less likely to do it. It's also very good in teaching. If I ask my students to think about 15 ways in which my class could be improved, I usually get higher teaching ratings. [laughs] So it is practical, as you can see. [laughs] So,

Dan Ariely 24:17 um, here's another example of the same principle. Imagine I asked you, "How often do you floss?" And I ask you to give me on a scale from zero to nine a day- Well, I asked you on a scale from zero to nine a month. Now, then I gave you a, a number for a hygienist and I said, "Should you call her up," right, "to make an appointment?" What happens is that if you got the top scale, you would answer somewhere on the left, and if you got the bottom scale, unless you don't floss at all, you would answer somewhere more to the right. But then you would go a step further and you would say, "My goodness, I'm only flossing once or twice a day. I'm way below the norm." [laughs] Or in this case, you would say, "My goodness, I'm flossing eight, nine times a month. I'm perfectly, perfectly fine," and therefore be much less likely to call the hygienist. Again, your hygiene practices is something you should know. And this, by the way, works on everything. How many hours of television do you watch a week, and how much do you work, and how good is your social life? It

Dan Ariely 25:18 works on a lot of things. [laughs] It turns out we just don't know, and the scale that we're being given to think about something can dramatically change how we think about it. As a consequence, the scale designers have a big influence on our lives. The next example I want to give you for, from this is about the choices that we encounter. So we often depict examples of choices in this way, some multidimensional space. Now, imagine I gave you an option. I say, "Would you like to have a weekend in Rome, all expenses paid, or would you rather have a weekend in Paris, all expenses paid?" Rome and Paris are different. Different food, different culture, different art. Hard to decide. I mean, some people might have a strong preference, but for many people it's hard to decide. What if I added a choice that nobody wanted? What if I said, "Weekend in Rome, all expenses paid, a weekend in Paris, all expenses paid, or having your car stolen?" [laughs] Now, now this is a not exciting offer. [laughs] Nobody should want it. And of course, it shouldn't change your preferences between Rome and Paris. But what if the option of having your car stolen was not as extreme and in fact it was quite similar to one of the two options? So you have Rome, Paris, or you could have Rome without coffee in the morning. Okay? Now- [laughs] ... if I said, if I say Paris all expenses paid or Rome all expenses paid without the coffee, the coffee doesn't really make a big difference. Yeah. Right? I'm not a coffee drinker. Not the... Well, you could pay for it. It doesn't mean you can't drink coffee, it just means you have to pay for it. It isn't included. [laughs] See, I, I bought a heckler with my, uh, previous comment. [laughs] But the thing is, but the thing is that the contrast, the contrast between Rome with coffee and Rome without coffee makes it clear that Rome with coffee is better than Rome without coffee. [laughs] And interestingly enough, it makes it appear not better only from Rome without coffee, it makes it appear even better than Paris. So when you introduce Rome without coffee to the set Rome, Paris, people start choosing Rome. And if you introduce Paris without coffee, people start choosing, choosing Paris. Now, here's a, here's an, here's another example of this. A, a while ago I went on the, a website for The Economist and they had the following offer. You could get an online subscription for $59, sound quite reasonable. You could get a print subscription for $125, again sounds reasonable. [laughs] Or you could get both for $125. [laughs] Now, independently, the idea of getting both for $125 seems very reasonable. But it makes this middle option seems very strange, right? Why would anybody get this option of getting part of what you could get for the same amount of money? [laughs] So I called up The Economist and I said, you know, "Tell me what, what was going on in your minds when you, when you were doing this." [laughs] And, and they didn't tell me and they took it offline, so I never, I never learned what they were thinking. [laughs] But I did run the experiment and, uh, I, ideally I would have liked them to run with me. I, I printed this off and I gave it to 100 MIT students, and this were the market share. A few people wanted the online, vast majority wanted the combo deal. Nobody, nobody picked the dominated option, right? Good for our admission office, right? We eliminated all the people who... [laughs] So, so since nobody took this offer, I said let's take it away. Why, why would you waste space on an offer that nobody wants? So another 100 students got the same offer but without the middle option. They only got the top and the bottom. Now what happened is that the most popular option became the least popular, the least popular became the most popular. You see, while, while the middle option was useless in the sense that nobody wanted it, it wasn't useless in the sense of helping people figure out what they wanted. This basic option screamed, "Hey, look at me. Compared to me, this one below me is fantastic. In fact-" [laughs] "... in fact, you're getting the web for free relative to me if you look at it this way." And this, this all goes back to the difficulty of figuring out how much we like those things. How much is this package of print and web worth together? I have no

Dan Ariely 29:53 idea. And therefore I rely on cues in the environment i- including irrelevant ones, to help me figure out what I'm willing to pay for things and how much, how much they are worth. And I also did this with, uh, sexual attraction. So oft- often we think that when we meet somebody we know immediately if we're attracted to them or not. It's kind of a basic idea. Maybe that's the, the reason for the, you know, the two-minute dating, uh, speed dating ideas. Um, so I took pictures of MIT students and I kind of paired them up. [laughs] I will not show you the pictures of MIT students. These will be computer images to, to protect their privacy. But I basically took pictures of MIT students. How many people here went to MIT? Okay, very good.

Speaker 30:36 [laughs]

Dan Ariely 30:37 There's a lot of good MIT jokes on, on these things, but we- we'll [laughs]

Speaker 30:40 [laughs]

Dan Ariely 30:41 Um, and, and I pair them up to be kind of, like, generally similar in their attractiveness, okay? So I would take somebody, and let's call him Tom.

Speaker 30:50 [laughs]

Dan Ariely 30:50 And I would take somebody, let's call him Jerry, and I would ask people, "Who, who do you, who do you want to date?" Now, what I would do is with Photoshop, I would make an uglier version of one of them.

Speaker 31:00 [laughs]

Dan Ariely 31:03 So some people would see Tom, Jerry, and ugly Jerry.

Speaker 31:06 [laughs]

Dan Ariely 31:11 And, and, and other people would see, uh, Tom, Jerry, and ugly, ugly Tom. And, and basically-

Speaker 31:16 [laughs]

Dan Ariely 31:18 [coughs] And basically the question is will the same thing happen? And the answer is absolutely yes. The, the ugly similar friend always made the other person m- look better.

Speaker 31:27 [laughs]

Dan Ariely 31:28 So ugly, ugly Jerry made Jerry look the most popular.

Speaker 31:31 [laughs]

Dan Ariely 31:32 Ugly Tom, uh, made Tom, made Tom looks more popular.

Speaker 31:36 [laughs]

Dan Ariely 31:39 Okay. Now, uh ... I want to do one ... Just going to do an experiment where ... Let, let me actually talk to you a little bit about one other, one other topic. So the next topic I want to talk to you about is cheating. Um, so I, I became interested in cheating when, uh, Enron came to the, to the field, and a few things kind of bothered me about this. One is that some of the architects of Enron, while they were in many way evil people who stole billions of dollars from the US economy, um, at least some of them seemed to be upstanding citizens in their community. They were giving money. They were volunteering. The other thing is a lot of people worked at Enron. How could it be that they kind of collected all these evil people to, to work at the same, at the same time? Now, we, we had this belief that it's all about the few bad apples, but what if it's not about the few bad apples? What if there was something about the structure of Enron that allowed otherwise normal people to cheat to that level? If that's the case, then we might as well learn about it and figure out what was it about the structure. So I want to tell you about some experiments, uh, that we did on cheating. So like we usually do, we start with very simple experiments. So I created a sheet of paper with 20 simple math problems that everybody could solve, uh, given enough time, but I didn't give people enough time. So if you were in my experiment, I would pass you the sheets of paper. When the five minutes were over, I would collect them back, and I would pay you by your performance. Turns out the average students we tested solved about four of these problems, and we paid them $4. That's the control condition. Then for another group, we passed the sheets of paper, but when the five minutes were over, we said, "Please shred the piece of paper and tell us how many questions you got correctly." Okay. Now, interestingly enough, people solved seven, uh, problems. Of course, they didn't become smarter, they just cheated. And it wasn't the case as if a few people cheated a lot. It was the case that a lot of people cheated just by a little bit. Now, why do people cheat? In standard economic thinking, cheating is basically a function of how much money do we stand to gain, what's the probability of being caught, and how much we will be in prison. So every time you pass by a gas station, that's the question you should be asking yourself.

Speaker 33:52 [laughs]

Dan Ariely 33:54 Uh, in this, how much money is in the till? What's the chance they'll catch me? How much time will I get in prison? So we, we played with those. How did we play with those? We paid people different amounts per question. Some people got 10 cents per correct question, 25, 50, $1, $2, $5, $10 per question. Turns out it doesn't matter. Still, across all of these quite large range of reward, a lot of people cheated, but just by a little bit. Then we played with the probability of being caught. Some people shredded the whole sheet. Some people shred only half the sheet, so there was some evidence left. Some people shred the whole sheet, left the room, and paid themself from a bowl of money that had over $100. These guys could have really walked out with a lot of money. Nothing changed. Still, a lot of people cheated by just by a little bit. So it wasn't about how much people were standing to gain. It wasn't about the probability of being caught. What was there? So we thought maybe what is going on is that people have kind of an individual fudge factor, that we have ... We all want to look at ourself in the mirror and feel that we're good people, right? So we have some utility from this image of ourselves as good people. At the same time, uh, we want to benefit from cheating, and maybe what we're doing is we're balancing these two goals. We cheat up to the level we would have to update our image of ourselves. So we cheat up to the level that we can still look at ourself in the mirror and feel good about it, and that's kind of our personal fudge factor. Now, how, how would you test something like that? What

Dan Ariely 35:25 you would want to do is you would want to think about ways to shrink the fudge factor, and you would want to create ways to increase the fudge factor. So let me first tell you about the shrinking. We got some people to, uh, come to the lab, and we said, "We have two experiments for you. The first one is the memory experiment. The second one is a simple math experiment." For the memory experiment, we asked half the people to recall 10 books they read in high school, and we asked the other half to recall the Ten Commandments. Then we tempted them with cheating. First of all, you should know that nobody in our sample remembered the Ten Commandments.

Speaker 35:58 [laughs]

Dan Ariely 36:00 And those who wrote 10 things usually invented-

Speaker 36:02 [laughs]

Dan Ariely 36:02 ... uh, always invented a couple.

Speaker 36:03 [laughs]

Dan Ariely 36:05 But interestingly enough, those who were asked to think about the Ten Commandments, nobody cheated. It wasn't as if the more religious people, the people who remembered more of the commandments, cheated less, and the people who remembered fewer of the commandments cheated more. Just contemplating the Ten Commandments eliminated cheating whatsoever. So we said, "You know what? It's very hard to bring the Ten Commandments back to, to school. Let's play with the honor code." So we got people to sign, "I understand that this short survey falls under the MIT honor code." They shredded it. No cheating whatsoever. Now, interestingly enough, MIT doesn't have an honor code.

Speaker 36:41 [laughs]

Speaker 36:45 [laughs]

Dan Ariely 36:47 And this actually brings us to an important point because, uh, we also tested this at Pr- Princeton. Anybody here went to Princeton? Okay. So Princeton, as you know, has a very strong honor code. In fact, they, they drive, they spend a week with the freshmen driving honesty into them.

Speaker 37:01 [laughs]

Dan Ariely 37:02 The first question, is the Princeton experiment working? When you meet the average Princeton student on the streets of Princeton or the average MIT students on the streets of Cambridge, do they cheat very differently? No. They cheat just the same level. Now, you could theorize that these Princeton students started from a worse perspective and this education brought them back-

Speaker 37:21 [laughs]

Dan Ariely 37:22 ... [clears throat] to the same level. But most likely, this week-long attempt to create a culture of honesty just doesn't seem to be working very well. The same, at the same time, when either the MIT students or the Princeton students sign the honor code, nobody cheats. So it wasn't so much about the size of the punishment, which is huge at Princeton and nothing at MIT, it is about contemplating your own morality. So, so we saw this and we went to the IRA, the, the IRS, um, and we said-

Speaker 37:54 [laughs]

Dan Ariely 37:54 Sorry. [laughs] Okay. We went to the IRS and, and we said, "Look, we found out that if people sign at the top, they're less likely to cheat. Now, when people get to the bottom of the tax form, it's kind of too late. They've cheated."

Speaker 38:13 [laughs]

Dan Ariely 38:13 "Why don't we try out with getting people to sign at the top?" And they basically told us they have bigger fish to, to fry than this, uh, particular one.

Speaker 38:20 [laughs]

Dan Ariely 38:22 But we did, we did get an insurance company to work with us. So, so many insurance companies, uh, send people, um, forms in the middle of the year asking, "How many miles did you drive?" And we all have an incentive to reduce that number of mile that we pretend to have driven because our insurance policy is based on it. So some people sign at the top, filled in their miles, some people filled their miles and signed at the bottom. Turns out people who sign at the top drove much more, and

Dan Ariely 38:52 than the people who signed, signed at the bottom. So that was all about decreasing the fudge factor. What about, what about increasing it? So, so I heard this story when I was a kid that, that, uh, somebody comes from school ... What's your name?

Speaker 39:14 David.

Dan Ariely 39:14 David. Perfect. Little David comes from school, uh, with a note from the teacher that says, "David stole a pencil from the kid who was sitting next to him." And his father is furious. Says, "David, I'm so embarrassed and humiliated. How we will tell your mother? This is not how we raised you to steal pencils, and I'm just flabbergasted. And beside, if you need a pencil, just say something. I can bring you a dozen from work."

Speaker 39:35 [laughs]

Dan Ariely 39:41 Now, now what, what is the intuition here?

Speaker 39:46 [laughs]

Dan Ariely 39:47 The intuition is that there's a lot of things we can tell ourselves about pencil. We can say, "Oh, yes, uh, workplace really wants us to take a pencil home because we could get more efficient. We can work at home as well." Or we can say, "Well, even if we give it to our kids, they would leave us alone and we could check our BlackBerry and, and answer a couple of more emails." There's all kinds of stories we can tell about it, and maybe that's an important feature. Maybe taking a pencil is so easy and taking 10 cent from a petty cash box is so much more difficult from a morality perspective because there's a lot of degrees of freedom in the pencil. So I'll tell you about two experiments. The first one is not a great experiment, but still cute. So I walk around MIT and I deposit six packs of Cokes in the common refrigerators, and then I come back and I measure what we technically call as the half-lifetime of Coke.

Speaker 40:36 [laughs]

Dan Ariely 40:37 How, how long it lasts, and you wouldn't be surprised to know it doesn't last very long. And then I compare it to a situation in which I take a plate and I put six $1 bills and put in the refrigerator.

Speaker 40:48 [laughs]

Dan Ariely 40:49 And I come back, I come back and I check how my $1 bills are doing. Now, the students could take a $1 bill, they could go to a vending machine, they could get a Coke plus change. But nobody ever took money. And that's kind of the pencil analogy. Now, it's not a great experiment because you often find Coke in the refrigerator and dollars is not really a common-

Speaker 41:09 [laughs]

Dan Ariely 41:09 ... a common thing, so, so we went, we went back to the simple math experiment, and we designed another condition. So a third of the people got the sheets of paper. When the five minutes were over, we collected it, we paid them. They solved four problems. A third of the people, when they finished, we asked them to shred the piece of paper, and they came to us and said, "Mr. Experimenter, I solved X problems, give me X dollars." We paid them. They claimed to solve seven. The third group, when they finished, they shredded the sheet of paper and they said, "Mr. Experimenter, I solved X problems, give me X tokens." We gave them a token per question, they walked 12 feet to the side, and exchanged it for money. In some sense, we gave them a pencil, but we gave them a pencil for a few seconds. These students doubled their cheating. Doubled their cheating. Now, for me, this is perhaps the most worrisome experiment of all we got because we're moving to be an economy that is not about money. It's about things that are at least one step removed from money. Think about a, a CEO who is backdating their stock options. It's not money, it's stocks. It's not stock, it's stock options. It's not asking for more, you just change the, the date a little bit. Could it be that somebody who could never imagine stealing $100 from a petty cash box could nevertheless very easily backdate their stock options and still feel honest about their overall behavior? I think the answer is absolutely yes. Could it be that when people cheat on PayPal and other things that are steps removed from money, credit cards, it's actually easier for them to be, to feel honest and at the same time cheating? And, and there's one more point about this. If you look at me as an economy, um, I, I ran experiments on about 4 or 5,000 students. And by the way, it's not just MIT. Uh, we, we had even students from Harvard, the, the whole range. Uh, no. [laughing] We had students from MIT, Harvard, UCLA, Princeton, Yale, so, so we had a, a good, a good selection of generally good people. Um, and nevertheless, a lot of them cheated. But if you look at how much money I lost, in my whole sample I had about five students who cheated a lot. Some students who basically went all the way, cheated dramatically. I had thousands of students who cheated just by a little bit, and I lost dramatically more. Now, if you, if you make the analogy to society, and of course I'm making some leaps here, but if we think about blue collar criminal, criminals, it turns out the estimated amount that we lose for arson, thefts from homes, car thefts, all of those things together are about $16 billion a year. Now, that's not a small amount, but let's compare it to some other things. The clothing industry. Do you know what this thing called wardrobing? Have you heard of this? You go to a clothing store, you buy some clothes, you wear them to a party, they get torn, they get smell, they smoke, and you return them for a full refund. The clothing industry estimates they lose $16 billion a year on this one, one practice. Uh, insurance property theft exaggerations are $24 billion a year. And by the way, when we talk to the insurance people, they say it's not as if somebody invents a stolen television, but if they had a 31-inch television that was stolen, it becomes 34, and if it's 34, it becomes 37. And there's kind of a slight exaggeration that everybody's willing to take. And when you look at it this way, we pay a lot of attention to the planned crime. But I think that in society in general, much like in our experiments, probably the most prevalent loss to the economy doesn't come from the planned crime, it come from the little fudging that, that good people do just by a little bit. And for example, I'm willing to, to bet that the amount of money people lose for exaggerated expense reports are much higher than what we lose to the whole kind of blue collar, uh, crime, crime market. Let

Dan Ariely 45:16 me tell you, uh, what I think are the big lessons from, from behavioral economics. The first one is that we have these irrational tendencies, and we have to, we have to recognize them. The second thing is that our intuitions are often wrong. Think back to the c- to example I gave you in the beginning with the organ donation. How many people here feel that if you went to the DMV and the form would have been phrased one way or another, your own decision would have been different about whether to donate your organs or not? Right? It's very easy to say, "Oh, these funny Europeans, of course they will be influenced." [laughing] But having, having the intuition that we ourselves will be influenced is incredibly hard, and this is why these, these tricks are incredibly powerful because they don't play into our intuitions. We don't have any intuitions about them. Now, if we make flawed decisions and we have good, bad intuitions about them, what, what can we do? And the only answer is experiments. The only answer is that we have to really think about what, what we do right and what we do wrong, and think about what assumptions we have about the, these kind of things and how do we, how do we test them in a systematic way. Uh, the next, the next implication is that, um, if, if, if you go, uh, bar hopping, who, who do you want to take with you? [laughing] Right? You want to take the middle guy. So, so you always want to take somebody who is similar but slightly less attractive than you are. [laughing] And, and if somebody else invites you to come with them, you know how they think about you. [laughing] Finally, you know, are we, are we, uh, Superman or are we Homer Simpson? You know, when, when we, when we build the physical world, when we build shoes and wa- bottled water and we build, uh, chairs and all of those things, we, we kind of realize our physical limitations. We understand where we fail, we understand where we're short, where we don't do well, and we build the physical world to, to work with that, right? We don't build chairs for Superman and ... But somehow when we move to the mental domain, when we n- move to things like retirement, annuities, healthcare, we somehow assume that people are perfectly rational. And I think that while behavioral economics is kind of, uh, depressing a little bit about our own nature, the, the big hope is for the people who are designing this world to say, how do we want, how do we want to design it? And if we understood the cognitive limitation in the same way we understand physical limitations, maybe we could build a better world. And let me give you an example of one idea is to think about the subprime mortgage crisis. So I, I also have a position at Boston Federal Reserve Bank, and my job is to argue with the local economists from time to time. And a few years ago, we had a debate about interest-only mortgages. Now, interest-only mortgages mean that every month you can decide just to pay the interest or pay the interest plus some of the principal. From an economic perspective, this is a wonderful thing because you increase flexibility. The person who is borrowing can now decide every month what's the right thing for them to do. In fact, one of my friends in the bank found out that people with higher credit card debts were more likely to take interest-only mortgages. And he said, "These are wonderful, smart people. They're stuck with this bad credit card debt. They're going to take the interest-only mortgage, and then they're going to pay back the credit card." Well, there's a lot of assumptions there of course, but that was his, his idea. Um, but I said, think about how difficult it is to compute what is the optimal amount of mortgage for you to get How would you even go about computing that? It's incredibly hard. You have to think about your current income and future income and future market and your other expenses, and you have to think about... And just incredibly difficult to figure out. So what do people do? Do they figure it out? No, they don't. They go to the bank, they say, "How much would you lend me?" And the bank said, "I'll lend you 30%." They say, "Okay, that's what I'll take." Now, what the bank is telling them they would lend them is not the right question to the right answer. They should ask him how much do I want to borrow, not how much I can borrow, not what the max I can borrow, how much I should be borrowing. It's not about the risk the bank is willing to take, it's how much you as an individual want to take. But because it's so hard to figure out, people just don't. Now we created this interest-only mortgages. Now, if people borrow the same amount, they would have extra flexibility. But if people all of a sudden went and borrowed more, they borrowed the whole amount that they could, they would increase... They would not increase their flexibility in any way and would just expose themselves to much higher risks. In fact, that's what we saw. But I think that if we understood how difficult it is... You know, in, in standard economics, we, we think that people would pursue their best interest, but people would pursue their best interest if they know what it is that they're trying to pursue. And if it's too hard to compute, how are they going to do it? Now imagine, for example, that instead of giving people a mortgage calculator that told them what's the maximum they could borrow, we would give them a mortgage calculator to tell them how much they should be borrowing. Nobody wants to have, have their house foreclosed, but

Dan Ariely 50:28 at the same time, we have to give people tool to figure out how they, how they deal with this. Um, okay. Let me tell you a last thing is, um, we do experiments all the time. Uh, if you want to register to participate in some experiments and have your rationality be tested, you're very welcome. It's predictablyirrational.com, and I want to end by doing a little study with you. Can you help me open the box?

Speaker 51:06 Yeah.

Dan Ariely 51:06 So I'm going to give, uh... I have a few books I'm going to give to some of you. Congratulations.

Speaker 51:10 [laughs]

Dan Ariely 51:15 Congratulations. Can I have one? I promise I'll read it. There's no, there's no promising you. You just, you just crossed the whole line.

Speaker 51:21 [laughs]

Dan Ariely 51:36 I have to give you one, of course.

Speaker 51:37 [laughs]

Speaker 51:40 Thank you.

Dan Ariely 51:42 Congratulations. You want to give up a few? Why don't you go ahead.

Speaker 51:47 [laughs]

Speaker 51:47 Is it random?

Dan Ariely 51:49 Not random, it's planned. My goodness.

Speaker 51:51 [laughs]

Dan Ariely 51:59 Next to you.

Speaker 51:59 [laughs]

Dan Ariely 52:02 That's cruel.

Speaker 52:05 There's still a few left.

Dan Ariely 52:06 Okay. Okay.

Speaker 52:11 [laughs]

Dan Ariely 52:28 Congratulations. Okay. Can you help to pass out everybody? Okay. So now, now here is what I want you to do. There's, there's a few people in the audience, uh, who have a new fantastic book that I particularly like.

Speaker 52:43 [laughs]

Dan Ariely 52:45 And there are quite a few of you who don't have it. And, and what I would like you to do is to, to think about how, how valuable is this book. So imagine that the people who own, who own the book, I would ask you, how much money would you be willing to, uh, would you charge, what's the minimum amount you would charge to give away your book, right? You have, you have this beautiful blue and orange book. How much money would you demand to, to part with that? And imagine I ask the people who don't have the book, uh, how much money would you be willing to spend to get such, such a book? Think about the number for a second. Just think about the dollar number for people who own the book, how much you would need to be compensated, people who don't have the book, how much you would be willing to pay. Okay, assuming everybody has a number. The people who don't own the book, how many of you would be willing to pay less than $2 to get this book?

Speaker 53:40 [laughs]

Dan Ariely 53:46 I'm going home. [laughs]

Speaker 53:48 [laughs]

Dan Ariely 53:48 Okay. You? Okay. So this... Ten, I shouldn't have never done. I should have said 100 to start with. Okay. Now, the people who, who have the book, how many of you are willing to part with it for $2? Now, here's... That's the basic story. The basic story is the moment we give... Are you willing to... You don't have one.

Speaker 54:11 [laughs]

Dan Ariely 54:13 Give me a second. The, the basic story is the following. The moment we give somebody something, they start evaluating more. They th- start thinking about it as more, as more valuable. I gave it randomly. It's not as if I saw the craving in your eyes and gave it to the people who-

Speaker 54:28 [laughs]

Dan Ariely 54:28 ... who wanted it more. But the moment you own something, you think differently on it. And this is why the people who own these books think they're wonderful and fantastic, and the people who don't just don't see the particular, uh, attraction in this. By the way, we just did a study on, on the idea called not invented here. So I don't know if you, if you know this phrase that... So we took a group of people in The New York Times and, and we asked them to evaluate ideas. So these were kind of ideas to solve world problem, ideas to solve hunger, to solve water problems, and so on. And sometimes they evaluated an idea of a weak, uh, kind of a one-line idea, and sometimes we gave them 50 words, and we asked them to generate an idea first and then evaluate it. But we doctored the words such that we knew what they were going to generate. They were going to generate the same idea that the other people just got. It turns out when people generate the idea, they thought they were wonderful. They're willing to donate more money to the cause. They're willing to work harder. And the point is that there is something kind of special about everything that is ours. We are special, of course. We're all special. And the moment something is ours, either given to us, our own ideas, we overvalue it. And like many of these things in behavioral economics, not only do we overvalue it, but we don't understand how much we will overvalue it the moment it will become, become ours. Okay? So, uh, thank you very much. [audience applauds]