When to Rob a Bank | Steven J. Dubner | Talks at Google

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
MALE SPEAKER: There's a freakishly large number of book sales on "Freakonomics." I mean, 7 million is the public figure. I'm sure it's probably a bit more than that in reality. STEVEN J. DUBNER: Why do you assume that it's more? It could be that we lie. MALE SPEAKER: I think it was a 2014 figure. STEVEN J. DUBNER: It could be that we inflated it by 100% though. [LAUGHTER] MALE SPEAKER: It wouldn't surprise me, I suppose. STEVEN J. DUBNER: But you're right. You're right. [LAUGHTER] MALE SPEAKER: I can see how this is gonna go. The blog itself has had about 3 million downloads on each of its episodes, which is pretty impressive. And of course there is the one thing which all of us Googlers who are looking for things to do once we leave Google. We think about careers to do. And so Steven is going to talk to us about robbing banks today, I think. STEVEN J. DUBNER: Yeah. MALE SPEAKER: Cheers. STEVEN J. DUBNER: All right. Thank you. I appreciate it. Good morning. I apologize for being late, but it is entirely your Queen's fault. The book is called "When to Rob a Bank," which is almost entirely random. So if you really want to know more about how to do that, I can tell you. Although the short answer is "don't." The ROI on bank robbery is terrible. That's what it's really all about. It's actually a story of how the book came to be called what it came to be called. So if we want to get into that later in Q&A, we can, which is how little leverage authors exert with their publishers even when they've sold a few copies of a book. Because that was our third or fourth title. The first few got rejected. So anyway, there is a book called "When to Rob a Bank." It is a book that is a compilation of the best, the greatest hits, of our blog from 10 years. So we published by now about 8,500 blog posts over the course of 10 years, most of which were terrible either in the moment or beyond. They just didn't last. But we went through, and we found about 500 that we actually still enjoyed reading. Then we whittled that down to 132. So that's what's in the book. So I'm gonna talk for a while. Then I'll take questions, because I know every time I speak at a Google event, there are a lot of questions which I really enjoy. So I'm not gonna talk too long right now. But the occasion of the book was the 10 year anniversary of the publication of our first book, "Freakonomics." So this is our fourth book, and for 10 years, I've been working together with Steve Levitt, who's an economist at the University of Chicago. And we've been doing this project together that's had many different forms. The first couple books were, largely but not exclusively, based on academic research he'd done. Our third book was called "Think Like a Freak," which was using everything we'd learned from not only methodology but from getting out in the real world to work with firms and governments to try to solve problems. Because after the first couple books, we were called in by people who thought we'd be useful in that way. Almost always we weren't, but we set out a blueprint for problem solving, essentially. And then this fourth book, which as I said, is a compilation of blog posts marking 10 years of us doing our thing together. So it is inherently a retrospective time. As all anniversaries are. And because we've been doing a long time you start to think-- or I've started to think-- a little bit about legacy, and what it means to do anything for that long a time, and what kind of mark if any one leaves. And so I wanted to share with you, I think, the perfect illustration of the legacy of "Freakonomics" as gleaned last night. When I was here in London I was doing media all day yesterday. It started about 8:00 AM, and it went till about 11:30 PM, right? And the last stop was the 500th BBC visit of the day, and it was "News Night." So does anybody watch "News Night?" I assume not, really. So I'm in the green room at "News Night," which is the room where they put the guests before they go on. And there were a couple of MPs, male MPs who-- one conservative, one liberal, or whatever you call them, Labour-- and [LAUGHTER] They were and they were just chatting very amiably with each other. And then they got ushered in to the set to go do their piece. And then someone else came in, and this was a woman who was also an MP. And she's standing now for the head of Labour Party. So I probably shouldn't name her because of what I'm about to tell you about what she said, although it's not that bad. And I understand there are three women running? So you'd have to actually go back and watch last night's show to narrow it down. So I'm not going to do that work for you. But one of the three women who were standing for Labour, you could tell right away she's a great politician because she looks at us, and she wants to know who we are, what we're doing, very friendly, chatty, sticking out her hand to shake. She has her bike helmet, which I'm sure she brought along with her in the taxi to look good. And she says what are you in for? That's what they always ask you when you're in a green room. And I always say homicide, because that's what they ask in my country when they say what are you in jail for. So then she said, what are you gonna talk about? And I said, I'm an author. I'm gonna talk about the books. She said what books? And I said, I write the "Freakomics" books. And she said oh, oh. And she starts nodding. And she said, yeah. I've loved-- I remember reading "Freakonomics." She said that was the book with the ridiculous names like Shithead that's pronounced "Shuh-teed," right? And I said, yeah, yeah. And this is what the MP who's running for head of Labour remembers about the book, right? She says, oh, yeah, Shithead. I loved Shithead. That was fantastic. So it's not about any of the public policy stuff. It's not about the relationship of abortion and crime. It's not about drug dealing. It's not about anything. It's Shithead. And then furthermore, she says, yeah, I really love how you wrote that when parents give their kids these ridiculous names, they ruin their lives for them. And what I wanted to say, what I always want to say in that situation but never do is that, actually, you've got that exactly wrong. The argument that we made using the data on naming, and parents education, and the outcome of children, and so on was that the name that you give your child doesn't matter at all. All the name is is an indicator of who the parents are. But if your Parents A, and you have a boy named James, or parents B and your observationally equivalent to Parents A, and you name your boy Shithead-- or "Shuh-teed"-- Shithead and James are not going to have any substantially different outcome in their life. That's what the data showed, and yet, most people who read this and remember, remember exactly wrong. So the two things that the potential ahead of Labour Party remembers about our legacy is Shithead, and she got it wrong. So that's my legacy. That's what I'll be remembered for, if anything. And I guess it's deserving. Because let me tell you know how I spend my day. So I know you all are mostly engineers? All engineers? Anybody here not an engineer? Anybody sneak in? Where did you sneak in from? Yeah. Are you in the building? [LAUGHTER] AUDIENCE: So I'm not even in this building. I'm in our sales office, but I managed to sneak in from another door. STEVEN J. DUBNER: OK. That's why you have such better glasses than the engineers. [LAUGHTER] There were a couple others. Are you? Yeah, it's so easy to pick out the non-engineers, I have to say. [LAUGHTER] AUDIENCE: I sneaked in from the same building. I work in marketing. STEVEN J. DUBNER: OK, and there was one more? Yeah. AUDIENCE: I work in editorial. STEVEN J. DUBNER: OK, so a good half a percent's smattering of sales, marketing, and editorial along with engineers. OK, so I know how you all spend your day generally. I couldn't survive doing what you do for half a day. I have no idea how to do what you do, but I generally understand what it is you do. So let me just tell you briefly how I spend my day. I live in New York in the Upper West Side. I have a nice family. I've got a wife. I've got two kids. I've got a little dog. And my office, which is another apartment near where we live-- it's about two blocks away. So in the morning, I get up. I have some coffee. I read some papers, say goodbye to the kids, and I go to my office at about 7:30 AM. And I usually walk the dog along the park, which is very nice. And then once I start my day, all I'm doing is I'm walking with the dog, looking at the park, and I'm thinking about stuff. I'm looking at people. I'm looking at the people interacting on the street, in their cars, and bikes, and pedestrians, and wondering why more people aren't killed on that intersection, because it seems to be a very dangerous one. I'm looking at weather, how people respond to it. I'm just thinking about the way people live their lives and how they make their decisions, stuff like that. Then I get to my office, give the dog some food. And then I just sit, and basically all day long, I just keep doing that. I sit and think. And I read. And I write. And it's entirely inconsequential. And that's I spend my day. So for instance, I like economics a lot. I never really liked the form of economics we study in university so much. It felt, for me at least, overly mathematical, not that interesting, not that useful. It sounded like the brightest people among the professional economists were divided on the most important issues of the day. In other words, we look to them to describe the economy, to actually tell us how the macroeconomy really works. And it turns out that they're wrong more often than they're right, and that the best and brightest argue with each other constantly about the most essential facts of that. So I never really liked that kind of economics, but I like the weirder kind of economics, including that my partner Steve Levitt has worked on for years. I like agricultural economics, for instance. Get some of those journals, like to read those. I grew up on a little farm in upstate New York. And food is one of those things that I marvel at every day. A century ago, it was thought that we had roughly a billion people on Earth. It was thought that there's no way we could support 2 billion. There's no way we can grow that much food for 2 billion people. And then after that, there was no way we can get to three or five billion. Here we are at seven billion, more than enough food for everybody. A lot of people still don't get enough food, but that's almost entirely because of political or economic failures, is not because of agricultural failures. So agricultural economics, to me, are really interesting. How did we get so good at this, right? It's not just economics, obviously, but the way that it's sourced, the way that it's resourced, the way it's distributed. When it goes wrong, like in California right now, what's going on with the water prices or with the water shortages is plainly a product of water being badly, badly, badly mispriced for many years in California. So agricultural economics are fun for me. Like, I wonder why is it that living in New York, I can buy a kiwi fruit and a banana for less than it costs me to buy one apple, which is grown often in New York, whereas the kiwi fruit is grown in New Zealand, the banana's grown somewhere in Latin America. Why is that? So those parts of economics and agriculture I like. And one day, I was reading this journal. And I just saw a simple fact. It was nothing but one simple fact. And I'm curious whether you guys, when you run across this kind of simple fact, if it triggers the same kind of thought process. I'm guessing it does. The fact I saw in a journal said that of the turkeys that are bred for consumption in the United States, that roughly 100% of them, of those birds, are the result of artificial insemination. So I thought, first of all, weird. And I'd never in my life thought about turkey sex at all. I did grow up on a farm. We didn't have turkeys. We had chickens. We had some other animals. I'd never thought about turkey sex, so I see this fact, and I think that's odd. And then you immediately start to think, well, who's doing that, and how is it done? OK. But then I wanted to know why. What does it mean? Does it represent anything interesting worth knowing? So whenever you see a fact, or whenever I see a fact, the next instinct is then to surround it or attack it with data so you can put it in context and see what it means. Is it a blip? Is it part of a trend, or whatnot? So of course then you go to the data on chicken sex. Turkeys and chickens are pretty similar when they're rates for consumption. If turkeys are 100% bred artificially, I assume chickens are too. Then you go to the chicken data, and it turns out that's not the case at all. So it turns out that most chickens breed naturally as God intended, right? But turkeys don't. So why is that? They're relatively similar. So then I went looking for other fowl data, and there just wasn't that much. We don't really eat enough other fowl in the States to have a whole lot of data. So it was really turkeys and chickens, and they were really different. So then I wanted to know more. So you go try to find some other data. Then you try to find people who know why things are the way they are. And that's what you do. You get on the phone. You start writing emails. So here's what the story turned out to be. In the States, particularly but also here and in other countries were where there's a lot of turkey-- so first of all, turkey consumption has risen a lot in the last 40 or 50 years. Now turkey plus poultry plus chicken consumption almost equals red meat consumption, which is a big, big change from the old days. And as Americans began to eat more and more turkey, it became very evident that the preference, the primary preference, was for breast meat, OK? People like white meat of turkeys. Now, I don't. I'm a dark meat person. My personal thought is that people don't really like the white meat that much, but it's a better delivery system for gravy and for mayonnaise. But anyway, they want the white meat. OK. So how did that change things? Well, breeders began to grow, to breed, turkeys with larger breasts, OK? So if you look back at the turkeys that were bred for consumption 50 years ago in the States, they had a very, very, very scrawny breast and big hind quarters. They were like wild turkeys. They bred them to have larger and larger breasts over time, OK? So now the male and female turkeys are both bred to have larger and larger breasts. So you can imagine in the old days, the turkeys had scrawny little chests. But now they're growing, and growing, and growing, and growing like this. So now they get to the point where the turkeys now that are bred for consumption have such large breasts that when they meet in the barnyard or where they meet, and they try to get close to have sex, they can't get close enough to physically have sex, thus the need for an entirely new industry, which is the turkey masturbation and artificially insemination. This is how I spend my days. So then I wrote a piece about this. [LAUGHTER] I wrote a piece about this. I thought was pretty interesting. I maintain it's very interesting. And it's so interesting to me. You try to surround an issue, or surround a problem, and consider all the possible data, all the possible explanations, and then you write it, or you put it out there. And I'm sure this happens with you guys with coding as well. You try to think about all the possible avenues, where it can go, but also how it can be responded to once the public gets hold of it. So one thing that happened with that story afterwards that I never could have anticipated was that there was a small but very, very, very vocal group of people who's just railed against the story on the grounds that we were glorifying turkey rape. And they said that you can call it artificial insemination, but what it really is is removing the rights of living creatures to naturally procreate, and you're raping them in order to make more of them to eat them. So when you put it like that, it does sound pretty bad. Honestly, it was something I hadn't thought about. I did eat less turkey for a month or so as a result of that, but I've since resumed. So like I said, this is how I spend my day. So it's fitting that my legacy from the MP is Shithead and that she got the notion of the way the names move exactly wrong. I'll give you another example of how I spend my day. OK, education. I'm guessing that the median level of education here is somewhere between M and PhD, yeah? So you guys have many, many, many, many years of education under your belts. If you were to look at every form of investment known to humankind, it's hard to make this argument definitively, but I would argue that the single best investment ever in history is education, right? So if you have one hour, or $1, or one brain cell to spend on something, if your primary concern is ROI that's scalable, education is the best investment. So the data for that are pretty overwhelming. If you didn't know it-- I think it would be surprising if you didn't know it-- which is that the higher you go up the scale, the greater the ROI is. So it's the opposite of diminishing returns. You guys with PhDs and M degrees earn a lot more than those with B degrees, which diminishing returns would argue against. So we know that, and it's great. And it's the reason that so much of our public policy tries to encourage people who are in position to set up education systems to get their kids educated and so on. Why we urge them to do that is because it's not only the best investment, it's the best insurance. I mean, if you want to ensure against criminal behavior, against poor health behaviors, against risky behaviors, education is a phenomenally leveraged unit. So when we know that. In the past recent years with the recession, there were a lot of people coming out of college with degrees, with good degrees, not able to find a job. So that began this big debate in the States, and I assume here as well, that oh, maybe it's not worth it. What's the opportunity cost? Maybe it's way too high. You know, Steve Jobs didn't finish college, Mark Zuckerberg, Bill Gates, dadadadada. You can find a handful of really famous dropouts and say maybe it's a really bad idea. So we did a pretty multifaceted analysis of the cost of education and the payoff of education for college. And we came to the same conclusion that all the other scholars have, which is it's still really good. So whenever you have something that's really valuable-- in this case it's expensive and valuable. Those two don't always go together-- you also wonder, well, are there people who want to get the value part without paying the cost? Are there people who are gonna try to cheat or try to steal? So this became a really interesting question, in part because it's really hard to figure out. So whenever you're trying to learn about illicit behavior, you know, drug dealing, prostitution-- all the regular kind of crimes-- embezzlement, bank robbery, and so on, the obvious hard part of getting those data is because it's an illicit activity, it's a hidden activity. Because it's hidden, the data are hard to get. So one thing that we've spent a lot of time doing the past bunch of years is finding ways to get data from those realms of illicit behavior. So there's this amazing fella named Sudhir Venkatesh. Some of you may remember his name from "Freakonomics." Although if the MP remembered only Shithead, I'm doubting you're gonna remember Sudhir Venkatesh's name. But he's an amazing guy who's a sociologist, who at the University of Chicago showed up and ultimately ended up embedding himself in a crack gang and lived with this gang for five or six years. And among the trophies he walked away with was a set of financial records from the gang. So they weren't quite KPMG audit quality financial records. But they were pretty good. And from those data, we were able to write about the economics of the crack gang. Until you have that data that, you're guessing. When you watch movies about drug gangs, they're based usually on one person's story that's usually exaggerated, and magnified, and distorted in many ways. So to try to get data like that, getting data like that is really valuable. Similarly, with Sudhir we did a project a few years later on street prostitution in Chicago. So again, if you're trying to ask a very basic question like, why do people do prostitution? Is it because they actually have no choice whatsoever and they're driven there by desperation? How well does it pay? What are the risks of arrest, of violence, of disease, and so on? But until you can get some data you can answer these questions. So again, in this case, we went out and solicited data, set up a bunch of, basically, grad students with clipboards to sit with prostitutes before and after every trick and fill out a survey. And survey data is often very squishy, as I'm sure you all know. But in this case, it worked out pretty well. We could verify that the data was pretty good. And so from that we were able to learn a lot of things about the cost of different sex acts. We were able to learn what kind of premium a prostitute would demand for having the john not use a condom. We were able to look at the interaction between law enforcement and prostitution, which illustrates really beautifully what's known as the principal agent problem. You guys knows the principal agent problem? Most of you? Some know? The principal agent problem is-- again, keep in mind I'm not an economist. So everything I say that the economics you should check the textbook afterwards. But the principal agent notion in economics argues that when there are two parties, let's say, a principal and agent, who would seem to have their incentives aligned, it's often the case that they're not aligned for reasons that I'll explain right now. So let's say in a police department, you would think everybody in the police department wants to arrest people who are doing crimes, right? That's their mission. But let's say that the mayor Chicago tells the chief of police in Chicago, you know, we're bidding on the Olympics and we really want to clean up the streets. And we need to wipe out the most obvious prostitution, OK? So let's make it a mission to get rid of the most obvious street prostitution. So then what happens? The chief of police tells the cops on the street, you know what? Let's go Arrest a bunch of prostitutes for a few months and just tell them that they have to get off the street for a year or so, or maybe take it inside or whatever. So it would seem to be that's an alignment of incentives. Everybody's getting paid to do the job, which in this case is to get prostitutes off the street. But the way the real world works, you often have a principal agent dilemma, which is that the principal has a clear goal and the agent, who's supposed to be carrying out the principal's wishes, seems to have a clear goal, but then personal incentives come into play. Because there are the personal incentives and the public incentives, or the official incentives. So as it turns out, from the data what we learned about this particular case of the principal agent problem was that a given prostitute in Chicago was more likely to have sex with a cop for free than to be arrested by one. In other words, the cops found it much more to their liking not to arrest the prostitutes and actually to take advantage of them the way the customers did, except, because they had the leverage, for free. So those are the kind of findings, if you want to call that, that you just can't get without data. So we work to get those data. And when it comes to education, and college diplomas, and college degrees, we were interested to know, since they're so valuable and so expensive, how many people cheat? How many people steal them in some way, right? There are a lot of different ways to cheat to get a college degree or a diploma. So we began looking into, and it's turned out to be really hard to get any kind of firsthand data. The best we could do, however, was an FBI agent who had spent years doing this, looking into it. And he was really good. Plus, it helps when you're with the FBI. You can go to the people who have the data and demand that they give you the data, which you and I can't necessarily do. And what was really interesting about this guy is he, like a lot of the people that we write about, a lot of the people that I consider heroes for the way that they try to solve problems, he was motivated by, he was driven by a personal interest. And in his case, this was a guy who was an FBI salaried worker. So he's not making a ton of money. And had two daughters getting ready to go to college. He lived in North Carolina. And he realized that even if he sent them to very good state colleges, like University of North Carolina at Chapel Hill, that it was going to cost-- I don't know-- maybe $60 to $80,000. If you add in everything, maybe $100,000 for the two daughters to get their college degree. And the human in him, not the FBI agent in him, the human in him was just thinking that's a lot of money. I wish I could just buy them a fake diploma and buy them a fake degree. And I would pay $5,000 or $10,000 for that happily if it was good enough to convince everybody that needed convincing. So obviously, that's not what we think of as education. That's what we think as just the outcome of the education. But if you think about it, when you show up to apply for a job, they don't really ask-- except maybe at Google-- what you studied in philosophy class. They just want to know that you have it. And nobody even ever asks for the degree. So he thought if I'm willing to pay that money for fake diplomas for my daughters, and I'm an FBI agent, presumably there are some other people out there who would be willing to do it. So he begin to explore the world of diploma mills and counterfeit shops. So it turns out there are a lot of different kinds of ways to get a fake diploma, or to make a fake diploma, and to market it. And the two basic ones are diploma mills, which would be basically like you could take a correspondence course in, like, golf course maintenance from some university. And you could then parlay that one correspondence course into a bachelor's degree for an extra $500 or $1,000. OK? then So that's the diploma mill form. And then there are counterfeit shops where they can just, basically, print very, very, very good diplomas. And the market being what it is, the price variance is huge. But for like about $5,000 or $6,000 you could get a beautiful indistinguishably inauthentic medical degree from Harvard for about $5,000 or $6,000. OK? So when this agent began to gather these data, arrest these people, find their sources, find their customers, began to measure the customers against the actual number of college graduates, it turned out that approximately 1% of all college degrees awarded in the United States every year were totally fake. So 1%. So you think 1% of anything is not necessarily that much, except when it's 1% of a big number like that for whom this credential is supposed to be kind of life altering, and for which you're supposed to have spent four to eight years and maybe $100,000 on, it's pretty significant. So the bottom line of that statistic is if you work with, let's say 100 people, and how many people are in this room today would you say, Will? [LAUGHTER] So I won't point any fingers, but if you look around, the odds are that someone here is totally piping it in and that their diploma is entirely fictitious. Now, it doesn't mean that that person's actually gonna be a worse employee than the people who went to school for the four years or eight years, but that's what the data say. This is how I spend my days. It's in some ways, I guess, not that different from yours, in some ways fundamentally very different. So I'd love to open it up to questions now and see what you'd like to know. MALE SPEAKER: Please raise your hand if you have a question, and wait for a microphone. AUDIENCE: Thanks for coming. I think it's fascinating, this interaction between the counterfeit diplomas and the ROI on education that you mentioned. Is the ROI in the bit of paper or in the knowledge that you get? STEVEN J. DUBNER: That's the question no one can answer, honestly. If you could come up with a good way to measure that, I'd love to know. There are ways you could imagine measuring it, but no one that I've ever found has actually done that. It's not hard to imagine that the vast majority of the value lies in actual knowledge and skills acquisition, but there are those who argue that a great part of the value in a college education lies in communicative skills, networking skills, and so on. And if you think about the cost of a college education, both in terms of dollars or pounds, and in terms of hours, which are very, very, very substantial. The people who make the argument that a lot fewer people should go to college, people who make that argument, their argument is that the opportunity cost is huge. So let's say you take 10 students in the UK. And in terms of academic ability, you take numbers 4 through 7. 8 through 10, you say they're going to a university. 1 through 3, maybe they're going to trade school or going straight to a job. Let's say we take 4 through 7, and we have to decide what we're going to do with them, or what would be the best outcome for them personally and for society. There are those who would argue that the opportunity cost of college is so huge that if you took two years or four years and the amount of dollars that are spent and invested them in that person, or that person was able to invest him or herself in a small businesses or in other training, that could far, far increase the output. So basically the most nuanced argument that I think is valid is that the way that college is pitched to and positioned for all students is pretty poor for all students, in that there should be a lot more segmentation than there is. That's the bottom line. So I think you could argue that for a lot of people who go to college expecting the skills and knowledge acquisition to be the thing that makes them more successful, it doesn't happen-- absolutely. But we don't know. I've never seen great data on that. Even the data, honestly, on the ROI in college are very hard to come by. Because if you think about it, you can't randomize that. What you'd want to do is you want to take like 10 cities in the UK, and pick 1,000 graduating students from those cities, and randomize them all, right? And maybe just do two groups, maybe do four groups, but you send some to great universities, some just out to get a job, and so on. And then you'd measure the effect of the university itself. But we can't do that. So the best studies that have been done are these natural experiments, one having to do with the military draft during the Vietnam War when there were a bunch of people who would have gone to college but didn't, and then other observationally equivalent people who did go to college. And from that one study and from some other similar ones, the conclusion has been had that the ROI is gigantic. But even then, the answers are not nearly as foolproof as we might wish. AUDIENCE: If the alternatives are going to college or going to war, I guess, yeah. STEVEN J. DUBNER: Agreed, agreed. Yeah. Next, please. AUDIENCE: What were your proposed book titles that got rejected? STEVEN J. DUBNER: Oh, OK. So I'll tell you. It's probably not interesting to anyone but me. And Penelope, who's with the publisher might prefer that I not tell the story, but it's not a big deal. So titles with us have always been a bit of a thing. So our first book, we'd written it, and it was just this heap of unrelated stories about baby names, and crack dealing, and sumo wrestlers, and cheating teachers, and nonsense. And so it was very hard to think of a title. We just thought of it as a book about nothing, but that wouldn't work, exactly. So we and the publisher we're trying to come up with all kinds of things that were horrible. One was "It Ain't Necessarily So," which is a wonderful old song but a bad title. One was "E-Ray Vision," like x-ray vision, but E for economics, which is really quite-- yeah. [LAUGHTER] Right, so even when you get it, it's pretty bad. And then Steve Levitt's sister, whose name is Linda Jines, who had been in publishing and advertising for a bunch of years, we gave her the manuscript and asked her to come up with names. And she came up with "Freakonomics." And here's what Levitt maintains. Levitt maintains that he loved it initially and I hated. I maintained that I thought it was so bad that it might be good, but I wasn't quite sure if it was just so bad that it was purely bad. But anyway, we presented to the publisher, to the American publisher, and they dismissed it entirely. Said it's just too weird, said freak is going to make people think of sex. It's going to be weird sexonomics. That can't work. It's too long. It doesn't mean anything. That's what they said. It doesn't mean anything. And I said, have you ever heard of a portmanteau? People make up words. They even make even make up titles. So I made up a word one time. I think this is the only good word I ever made up. You know you run into someone that you used to date, and they got married afterwards? And you realize you were the last person they dated before they married? So what you call that person? There should be a word for it, right? OK, try this-- penultimore. [LAUGHTER] Right? So I thought "Freakonomics" was pretty good. And it took the publisher-- it was right up to the 11th hour of publications, and they finally said, OK. We'll go with "Freakonomics." Then the book came out, and it did very well. And they immediately claimed total credit for having come up with the title. So that was "Freakonomics." Then we did "Super Freakonomics," which was not that controversial, although the publisher, again, was worried that the estate of Rick James was going to come after us, which they did not. And then we did "Think Like a Freak." That was actually pretty agreed upon. But this one, think about the signal you're trying to send with a title, right? it's interesting. We make the argument that names don't matter, right? But the argument that titles don't matter is much harder to make. And the reason is that the name of a person is not marketing in the same way that a product title is marketing, right? So if you name something Google or even ESPN, there are people who are going to hear it repeatedly before they start to use it and learn whether it's any good for them and so on. But a name is a little different. You hear the name, you meet the person, and you know that person. And name becomes secondary. It doesn't continue to serve a marketing function. So a book title name seems to be important. Again, it's very hard to tell. there's not a lot of A/B testing in this realm. And for this one, the book title we came up with was drawn from one of the posts that's in this book. And it was called "Hurray for High Gas Prices." So we liked that because we're American, and we talk about gas prices all the time. And there was a post about peak oil this, peak oil that, gas price is too high, too low, the gas tax-- whatever. There's a lot I could talk about with gas and oil. But this one posted said high gas taxes are great, because what they help do is capture the cost of this activity that otherwise isn't taxed very well, like water. When water is free for farmers, of course they're gonna use it all and not leave any for downstream. So when gas is cheap, of course people are gonna drive a lot. And then that causes congestion, accident risk, pollution, et cetera, et cetera. And unless you can raise the gas tax, which federal officials in DC won't for a variety of reasons, then people just drive a lot. So the post was called "Hurray for High Gas Prices." So that, our American publisher said, yeah, great. And than the British publisher, excellent publisher, Penguin, got hold of it. And they say, well, gas doesn't mean here what it means there. So it'd have to be "Hurray for High Petrol Prices," which doesn't work at all, or it's gonna sound like it's a book of gastroenterology of some kind. So then they said, can you come up with a different title for the British edition. And so I say yeah, but we don't like the idea because it's not fun to write a book and want it be called different things in different places. But then we came up with a title that we absolutely loved, which I thought would be the perfect title for the book, which is called "We were Only Trying to Help." Right? [LAUGHTER] Isn't that good? Because a lot of the blog posts in this book are basically where we just run an idea of the flag pole. Maybe it's about paying politicians more. Maybe it's about gas prices. Maybe it's about proposing a sex tax. All these different kinds of things, knowing that they're either not fully enactable at all, or even a little bit enactable, or just a question to ask to get a different kind of conversation started. Because my frustration is that, politics being what they are, institutional inertia being what it is, institutional habits being what they are, it's really really, really hard to make any real change. I mean, change in most institutions is two degrees this way, when really if you look at health care, education, sometimes you need like 90 degrees or 180 degrees. So a lot of what we do is just proposing these kind of outlandish ideas. One post was Steve Levitt wrote one saying that terrorism is a concern in the US, even though incidents are very, very, very rare. But we're spending a lot of money, a lot of mindshare on it. So if we're really serious about it, we should try to think of all potential ways that a terrorists could attack. So on the blog, he wrote asking, dear readers, if you were a terrorist, how would you attack? Which seemed like a really good idea at the time. We were only trying to help. But as it turns out, "The New York Times," which was then running our blog, immediately panicked, and shut down comments, and, I think, called the FBI on us. It was brutal. [LAUGHTER] So "We Were Only Trying to Help" was a title that we proposed to the British publisher. And the immediate response was yeah, perfect. And that's why I love our British publisher so much. And then I said to the American publisher, you know what? We got a better title than "Hurray for High Gas Prices." "We Were Only Trying to Help." And our editor said, yeah. That's great. That's awesome. And then like three weeks later, on the same day, I get a call from both publishers. Because the way it works now in publishing, the editorial team, and even the marketing team, and the PR team, no offense, they have zero leverage. It's a little bit like probably around here where the engineers have all the juice, right? So in publishing, sales has all the juice, right? So it's the sales team will take a book, title, and cover out to the book stores, the chains, and Amazons, and so on. And they'll say, this is a new book by these guys. And then one person at the book chain will say, I don't like that. That sounds stupid. Or that sounds like they're being self-flagellating or something. So on the same day, I got a from both publishers that said sorry, "We Were Only Trying to Help" won't work. So then we just had to go looking for a title. And we went looking in the posts that we'd selected for the book. And one post that we liked a lot was this one called "When to Rob a Bank." And the reason we liked it was because it's not about how to rob a bank, which is kind of procedural. It's not about why to rob a bank, which is obvious. It's about when. And that's the kind of question that we like to ask, which is if you're gonna do this, this thing that turns out to be pretty stupid. So it turns out the biggest point of this post is that the ROI on bank robbery is just dreadful. In Britain, you take home about, I think, two and a half or three times the amount of money you get from an average bank robbery in the States, which is interesting. I think it has to do with just more cash on hand. But the rate of arrest, the risk of arrest is about the same. So you're likely to get arrested after just three bank robberies. So even if you take on 10,000 pounds, you get arrested after three times. So for 30,000 quid, you're going to spent a few years in prison. And so the ROI is dreadful. If you really want to rob a bank, mornings are much better than afternoons. And yet, the vast majority of bank robbers work in the afternoons, which indicates clearly they haven't looked up the data. Because if they had, they'd be working mornings. But maybe they just can't get up in the mornings. Maybe they're not morning risers, which is probably why they're having to rob banks in the first place. And the very most interesting part of "When to Rob a Bank" for me was that this was all inspired by the story of a woman in Iowa who was embezzling from the bank where she worked. And it turned out to be a bank owned by her father, or run by her father. So if you're gonna rob a bank, rob it from inside, right? They said the best way to rob a bank is to own one or at least to work at one. But what was interesting about this woman was she had been working at this bank for many years, 15 or 20 years. And she was finally caught because she got so burnt out and exhausted that she had to take some time off. And then they discovered that she'd been keeping two sets of books all along. And she's embezzled, I believe, about $2 million. And this was in the 1960s, so that was a lot. Again, from her father's bank. And as it turns out, the way that she had covered it for all these years was by keeping two sets of books, but by never taking a single vacation day, which is how she got so exhausted that she finally broke down. So after she served her time in prison, her parents took her back. They were very forgiving. And then the FBI or some law enforcement agency said do you want to come help us catch embezzlers? And one of the things that she introduced them to was the notion that one of the best metrics to identify in looking for embezzlement is extraordinary vacation or sick time pay. Because if you're not taking it either when everybody else is taking it, or not taking it all, it's almost certainly because you're doing something with your books that you don't want anybody to see. So that's when to rob a bank. Other questions? Yes, please. Oh, you have a mic? Yeah. AUDIENCE: On the turkey story, the thing I found most astonishing was that the same wasn't true of chickens. Why wasn't the same true of chickens? STEVEN J. DUBNER: So short answer, I don't know. If I had to guess, I would say it's one of two things. So one would be that physiologically, even if we want to breed chickens to have as a large breast as possible, that their breasts don't get large enough to the point of interference. Two is there's some artificial insemination in chickens. So it could be that there is a segment of chicken breeding where the breasts are protrusive to the point that it requires artificial insemination. And I would think that's probably true. Now that you mention it, that's an obvious question that I should ask and find out, because it would be interesting to know. We did a podcast once called "Weird Recycling," which I really liked. It was about three different stories about where people took something that was of no value or even negative value, and reused it in a way that they gave it great positive value. So maybe we didn't even include horse manure. Horse manure was one that used to be that. Horse manure in the olden, olden days was worth something. In cities that were growing, they'd have a lot of horse manure. In the cities, they'd sell it to farmers, but then there was a horse manure glut as cities grew. So London and New York were paralyzed by horse manure because of the economic glut. There was so much of it that they couldn't sell it. But then the car came along. There are many fewer horses. And now horse manure is quite expensive. So that's a case. The other instances, there was spent nuclear fuel. So there are some nuclear engineers who figured out how to make nuclear power by using what we call spent fuel. And then one example was chicken feet, or what they call in the industry chicken paws. So in the States, almost nobody eats the feet of chicken. Do you eat them here at all? So if you go to certain Chinese restaurants-- and they look like this, really clawy. And there's that one spur that's right here. They look very much like a human hand was tortured-- a little bit smaller. And they are definitely an acquired taste. I mean, I guess they're an acquired taste. And very few people in the States eat them, and obviously here. But in China, they're huge. So it used to be that chicken producing firms in the US would throw away billions of chicken feet every year at great cost. And then as they begin to do more global business, it turns out that they would sell a whole lot to Asia. And now big chicken producing companies like Purdue tell me that they would not be profitable were it not for the sale of chicken paws, that that's how close the margins are, and chicken paws put them over the top. That said, in the States and here, the taste is plainly for the white meat of chicken as well. So I'm guessing that what you would see is that there are some chickens that are bred for that, and that among those, there might be artificial insemination. But I don't know enough about bird reproductive-- I should now, but yes, I don't know. That's a good question. Yes, anyone else? MALE SPEAKER: Could you pass the mic over? AUDIENCE: Hi. So why are politicians so resistant to evidence-based policymaking? STEVEN J. DUBNER: Why are they so resistant to what? AUDIENCE: Evidence based policymaking. STEVEN J. DUBNER: Oh. Well, before I join with you in accusing them, let me defend them for a minute. Why is everybody averse to evidence-based decision making? I'll just give you two very quick parallels. I'm just going to guess here. A bunch of people sitting here are Google engineers. I'm guessing probably at least 5% of you have a parent who's a doctor, an MD. raise your hand if you have a parent who's an MD. OK. That's about right. So medicine is fascinating. There's so much be grateful for and thankful for, but it's also very, very frustrating. Because even though there's a lot of science in it, most doctors aren't remotely scientific. And it turns out that most doctors are also terrible with data, that they're unbelievably bad with the most basic sets of probabilities. Many doctors are, at least. And then when they communicate it to their patients, it gets even worse, because it becomes a game of telephone, whereas you have a 2% chance within 10 years of developing a potentially fatal condition. And many people hear that in many different ways, but a lot of people hear it in a way that nobody who knows any statistics would ever embrace. So doctors, I always grew up thinking, were scientific in the way that we think of scientific, meaning the scientific method, and things are tried and proven. There are hypotheses. There's data gathered and so on. And it turns out that most medicine isn't really built that way. And the only medicine that's really build that way these days is pharmaceuticals and some forms of medical interventions, actual treatments. But the vast majority of it is not what you would really call evidence-based. And that's a pretty scary thought at the end of the day. And so we've done stories recently in our podcast, again, where even when there are evidence-based proofs of sorts of treatment or in the realm of medical health care delivery that's known to be efficient, even then, policymakers aren't crazy about adopting it. So there's this one case we did a podcast about in Camden, New Jersey that's a very, very labor intensive, cost intensive intervention, where turns out it in this area but in many geographical areas, roughly 5% of the medical patients are responsible for consuming about 50$ of all health care costs of all health care. So it's a huge, huge, huge problem. You've just got massive concentration among people who are going to the emergency room 100 times a year, just consuming all kinds of cost. So these people came up with the thought what if we, instead of treating things the way we are now, which is terribly, what if we come up with a new plan to invent a social services multifaceted network where we offer medical care but we also offer counseling so people actually understand what drugs they're supposed to take and when, so we actually help them get to the doctors they're supposed to go to. So if they need to go to drug rehab, they're doing that, and so on. So an expensive, hands-on project that ended up being very effective and very cost effective, considering what it saved. So even after this kind of thing is proven, no one in the federal government associated with Medicare and Medicaid, I'm told, really wanted to have anything to do with it. Because the constituencies are just so large and ingrained that they really couldn't make any persuasions. So in health care, evidence-based medicine is still more of a dream. In education, evidence-based curricula are only beginning to come online. So for politicians, I would argue, yeah, they're particularly visible in the way that they fail to embrace evidence, but I don't think they're very unique at all. And why they don't do it is because, I think, that most politicians-- unfortunately, it's a principal agent problem in a way. They're the agents and we're the principals. So we're the voters. We want to problems solved that we think government is best suited to solve. And these are typically long term, complex scenarios with a lot of different players-- transportation, health care, education-- things like this-- geopolitics-- that individuals can't do, that businesses, private firms don't necessarily do that well. So we think, well, this is what governments are gonna do, and that's what we want our politicians to do. And that's what we elect them for. But then once they get in office, the way the system is set up in democracies at least is that they are encouraged to respond to different incentives, which is short term self-interested incentives of getting elected, raising money, consolidating power, getting on the right committee, consolidating more power, raising more money in order to perpetuate their own-- So even though we don't like it-- We hate that. We all hate that. But for me, it's hard to blame them for it, because that's the set of incentives that they walked into. There are a few maybe idealistic people who buck it, but very few people's ideology is really as strong as the incentives that they're responding to. So one solution to this that we've proposed, I think, in this book, in "When to Rob a Bank," and I've talked about this with many politicians. And they always nod and smile. They say, oh yeah. Great, great, great idea. But at the end of the day, they hate it because it's totally possible. The idea would be A, we should probably try at least, we should experiment with paying politicians a lot more. So how many parliamentary districts are there in the UK? AUDIENCE: 625. STEVEN J. DUBNER: 625? So it would be really fun to take that, divide it in two or four, randomize, and raise the sal-- what's an MP get paid? 100,000 a year or something? [LAUGHTER] STEVEN J. DUBNER: OH, really? AUDIENCE: About 80, I think. STEVEN J. DUBNER: 70,000 pounds? AUDIENCE: Yeah. STEVEN J. DUBNER: So $100,000-- 70,000 pounds, we'll say. It would be really fun to take an experiment. So for five years, in two districts, we leave it. In one district, we have it. In one set of districts, rather, we do this. And in one set of districts, we double it. All you're really looking for, initially, is do higher salaries attract a different kind of person to go into politics? So some empirical work that has been done on this already. The answer's yes. So in some place in South America, I believe in Mexico-- in Singapore, they pay politicians a lot. There's a lot of things that some people don't like about Singapore. Most political observers, however, argue that it's easily one of the most efficient, uncorrupted, and practical federal governments out there, one reason being-- not all-- being that they make the job incredibly attractive. So rather than maybe coming to Google for a career, the same person might actually want to go into council election or a parliamentary election. So that's one. You could explore the idea of changing the status of political jobs, just the way we explore changing the status of teachers, which we haven't done yet either, by the way. But that's one idea. But then here's the bigger, bolder one that will never, ever, ever happen, but what I think could really, really work. If the problem is as I described, which is that their incentives are not aligned with ours, because they have short term, self-interested ones, we have long term public-interested ones. Well, what if we try to align them better? And one way to align them better would be to remunerate them based on their achievements, merit-based pay, right? So right now, if think about it, how does a politician build up capitol, or build up value in their careers? By being more visible, by being usually safe, right? So that they're gonna turn their political career into a lobbying, or consulting career, or whatever afterwards. But what we really want them to do is be working on those transportation, and education, and health care problems that we sent them to do. So what if we were to pay them for doing those problems, and pay them a lot? So let's say that there's a Department of Education. And we decide that-- I saw that your British math students, you're even worse than Americans. You're terrible, right? Where do you find engineers in this country? Right? AUDIENCE: [INAUDIBLE] British. STEVEN J. DUBNER: So your math scores-- OK, so let's say that the goal for the next 20 years is for British school math scores to rise 15% or 10% against international averages, whatever. AUDIENCE: Make the tests easier then. STEVEN J. DUBNER: OK, well, that's a good solution. That's a very political solution, certainly. So immediately I know you'd be an excellent politician, because you would have a very cheap, effective solution right out of the gate. So let's say that that's your goal. Then let's set up a project to try to figure out how to get to that goal with measurables with all along the way. And you declare somebody the arbiter of these goals and the measurables. So maybe it's an accounting firm, whoever. We know how to do that. We now how to set up projects with measurables. And let's say that you've got a Secretary of Education and 100 people working on this project who do all kinds of experimentation, who put into effect all kinds of curricula, changes, blah, blah, blah. And then let's say 20 years from now, those goals are met, right? So if you raised the test scores of British schoolchildren by 15% in 10 or 20 years, whatever, what's that worth to society? A lot, a lot. It's worth a lot to individuals too. And so what would be wrong with therefore paying the people who worked on that in politics to make it happen a share of that? So I'd love to write a check for a million pounds or two million pounds to the Secretary of Education as a kind of vested stock option 10 or 20 years from now if their plan worked out. So basically, it's merit pay in the form of stock options for elected and appointed political officials, which, I think, is a no-brainer. I think it works. Because that's the way the whole world works, right? If you do well, you get rewarded. If you do poorly, you're gone. The other thing I think is really important to do in politics that I would steal from football is I would relegate. So every year, I would relegate the bottom 15% of all politicians in every political system just to flush them out the bottom. So that's how I would fix politics, but as you can imagine, we were only trying to help, you know? [LAUGHTER] Are we out of time? MALE SPEAKER: We're out of time. STEVEN J. DUBNER: OK, thank you very much for having me today. I appreciate it. Thank you. [APPLAUSE]
Info
Channel: Talks at Google
Views: 87,787
Rating: 4.7333331 out of 5
Keywords: talks at google, ted talks, inspirational talks, educational talks, When to Rob a Bank, Steven J. Dubner, freakonomics, random acts, horse racing, when to rob a bank book, KFC
Id: V3Mwit66h4w
Channel Id: undefined
Length: 56min 2sec (3362 seconds)
Published: Mon Jun 15 2015
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.