Freakonomics Radio - 282. Could Solving This One Problem Solve All the Others?

Episode Date: April 6, 2017

The biggest problem with humanity is humans themselves. Too often, we make choices — what we eat, how we spend our money and time — that undermine our well-being. An all-star team of academic researchers thinks it has the solution: perfecting the science of behavior change. Will it work?

The one problem that really confronts humanity in the 21st century is humanity itself. That's Angela Duckworth. In other words, the problem with human beings is that they're human beings and that they repeatedly make decisions that undermine their own long-term well-being, even when they know full well that they are eating the wrong thing, that they are spending their money on the wrong thing, and they're spending their time in an unprofitable way. Duckworth is a professor of psychology at the University of Pennsylvania with secondary appointments in the graduate schools of education and business.

I am also the founder and scientific director of Character Lab, a nonprofit dedicated to advancing the science and the practice of character development in kids. You may know Duckworth from her bestselling book called Grit, which we discussed on Freakonomics Radio. That's right. I want to redefine genius, if you will. I want to define genius as greatness that isn't necessarily effortless, but in fact,

greatness that is earned, however you do earn it. That's right. That's exactly right. And that is Katie Milkman, also a professor at Penn, primarily in the business school with a secondary appointment in the medical school. My background is actually that I have a PhD in computer science and business, a joint PhD in those two fields. You may know Milkman from an earlier episode as well,

talking about her research on temptation bundling. So when I talk about temptation bundling, I mean combining a temptation, something like a TV show, a guilty pleasure, something that will pull you into engaging in a behavior with something you know you should do but might struggle to do. For instance? So what if you only let yourself get a pedicure while catching up on overdue emails for work? Or what if you only let yourself go to your very favorite restaurant whose

hamburgers you crave while spending time with a difficult relative who you should see more of? Those would all be examples of temptation bundling. Like her friend and colleague Angela Duckworth, Katie Milkman also believes that we humans are often our own worst enemy, that we make poor decisions that lead to self-sabotaging behavior. A problem that if we fixed it could truly solve every social problem we could think of. Now, there has been plenty of progress in the science of behavior change, as listeners of this program know well. But... Angela and I discovered that we both thought the biggest problem that needed solving was figuring out how to make behavior change stick. So not how to intervene and change behavior once or twice so people would make small decisions in a better way,

or even big decisions in a better way, but so that they would repeatedly make good decisions. So together, Milkman and Duckworth began to dream up a project, a huge project. It would seek to experiment with and understand and codify and eventually distribute to all of humanity the most effective behavior change nudges and incentives. Duckworth and Milkman were themselves responding to a rather large incentive. In 2016, the MacArthur Foundation made an announcement that for the first time they were going to dip into their endowment and award a single $100 million prize to one team to solve one social problem anywhere in the world.

Angela Duckworth had won a MacArthur Fellowship in 2013, the so-called Genius Award. Now, what kind of genius would she be if she didn't go for the $100 million MacArthur Prize? She and Milkman began recruiting fellow academics to join their team and corporate and institutional partners, too. They came up with a name for their project. They called it Making Behavior Change Stick. And they submitted their proposal. I'm just curious to know what you guys both think is your probability of winning. So I'm going to ask on the count of three to blurt out the number. OK, probability and percentage terms on the count of three that you will win. One, two, three. Twenty-five percent. terms on the count of three that you will win. One, two, three. 25%. If it were further, I would worry about the two of you. But I guess if it were exactly the same,

I would be higher. Just because she's that kind of optimistic person. He can see who's more delusional. She's more optimistic than I am. She's brought me up. I thought it was lower for a while, but I've hung out with her enough that I'm now at 25%. Now, let me ask you this. If this grant weren't available, would you two still be trying to do some version of this collaboration? Absolutely. And we will. In fact, we are. Right. I mean, we're working on it. I think it was only

a day or two after we turned in our proposal in late August of 2016, before we actually just started getting to work and fundraising, talking to these scientists about their best ideas for enduring behavior change and also talking to our organizational partners. And whether or not we're picked for the $100 million grant, we've now built this momentum around this group. And I truly think that that is the hardest thing, not even funding it. Funding it will be easy compared to getting this incredible group of people and organizations together and moving in this direction. Since Milkman and Duckworth have already gone to so much trouble, and since they're so ambitious and enthusiastic, we thought it makes sense for Freakonomics Radio to chronicle their journey, whether or not they win the $100 million grant. So today, the first installment of making behavior change stick, and we'll check in with them periodically over the coming months or years. After all, we, like they, have long been interested in the science of behavior change. And now that they're rolling up all the small known victories into one potentially giant leap for humankind, well, what kind of show would we be if we didn't

want to go on that ride? From WNYC Studios, this is Freakonomics Radio, the podcast that explores the hidden side of everything. Here's your host, Stephen Dubner. As most of us know from personal experience, changing even your own behavior can be really hard, whether it's how you take care of your mind and body or your physical environment, how you work or interact with other people, you name it. Even when you do make a change, it can be hard to make it stick. And what if you're trying to change other people's behavior? Over the past few decades, a lot of brain power has been spent trying to develop a science of behavior change. And there has been a lot of success, but these successes are often tightly circumscribed for several reasons. One is that

a lot of the experimental research has been conducted by academics who use college students as their subjects. So the sample size is often small and also not very representative. This is called the WEIRD problem. WEIR Weird standing for subjects who are Western, educated, and from countries that are industrialized, rich, and democratic. Weird. Then there's the fact that a lot of experimental research is too artificial, too unlike the real world. Also, too context-dependent and too low stakes.

How much stock do you really want to put in the decision made by a college student in a one-time transaction in a classroom lab where the reward is some free pizza? Another problem with incentivizing behavior change is that incentives wear off. On day one, you might happily choose to eat kale instead of french fries for a $2 reward. By day 10, you might be willing to pay $2, maybe $20 for some french fries. And there's at least one more big problem with designing incentives to change other people's behavior. The people who are typically in charge of the design are fairly accomplished people. Accomplished people tend to be disciplined and driven and cooperative. So the incentives they design may, in their minds, be perfectly logical,

but the rest of humanity may not be as disciplined and driven and cooperative as them. So even if you can find the right levers to press to produce behavior change in the right measure, in the right circumstances, how can you generalize that and scale it up from the individual to the population level? And how can you make it stick? That is the massive challenge that Angela Duckworth and Katie Milkman have given themselves. So massive that they wanted lots of help. Thinking on this scope and magnitude, it was easy for us to start making phone calls to people we would normally be, I think, maybe a little bit too shy to pester with our ideas. So we started sending emails to all of our academic heroes. And because of the scope, because of the excitement around this prize, everyone said yes. They put together an all-star team of more than two dozen researchers, psychologists and

economists and sociologists, but also people from medicine and computer science and marketing. The team includes four members of the National Academy of Sciences, three MacArthur Fellows and one Nobel laureate, the labor economist Jim Heckman. As Duckworth and Milkman write in their MacArthur proposal, this team of researchers has collectively, quote, developed interventions that meaningfully improve flu shot take-up, gym attendance, retirement savings, charitable giving, medication adherence, hand hygiene, energy efficiency, cancer screening compliance, voter turnout, weight loss, smoking cessation, job choice, GPA, attendance, and classroom conduct.

That said, this project will focus on three major areas. Number one. We're focusing on problems in health. So think smoking cessation, healthy eating, increasing exercise, reducing alcohol consumption. Number two, education. Can we get kids to have better outcomes in school and stick to school? And then finally? And then finally, savings.

Can people make better financial decisions on a daily basis so that they'll have better financial outcomes? The research will not be done in academic isolation, but rather in collaboration with real world firms and institutions. I mean, going out and working with a real organizational partner like New York City Public Schools or CVS Caremark or, you know, Bank of America, one of these organizational partners, and literally working with their customers or their students on experiments to try to change outcomes for the better.

I am curious if the subject line of your email was like a Nigerian email scam, like we have $100 million to the team that can change behavior. What was your actual ask to these people to get them on board? How did you describe the level of participation that you wanted from them and so on? We said to our organizational partners, like you, we care about making people's lives better. And like you, we know how hard it is. One thing that we bring to this is the perspective of science. We bring the scientific method. We bring the power of random assignment placebo-controlled studies. You, of course, bring years of hard-fought in the

trenches wisdom. Maybe there's something that we might discover working in collaboration that neither of us would be able to do apart. One of the places where we're going to start is with exercise. So we have two gym partners in this study, Blink Fitness, which has about 400,000 members, and 24 Hour Fitness, which has about 4 million members. What you do is you are encouraged to sign up. And then when you enroll in this program, you experience a personalized five-minute survey. And you're launched into a five-week program where you're going to be getting reminders to go to the gym on a regular basis. You'll be accumulating rewards points for actually engaging in physical activity.

And you'll be getting texted tips about how you can be more persistent in achieving your goals. That's our baseline group. We also have a control group that we'll compare that to in an A-B test that won't actually experience this intervention. And then finally, we'll have additional groups where we're testing new science, new ideas about how we can build on what we already know can be effective. What we really care about is not the five-week intervention period when people are accumulating points and accruing these rewards that they can get excited about, but rather what happens after that. Knowing what we know about loss aversion, there is some research. Is there not about using loss aversion instead of the reward incentive set up for gym participation? What's

that literature say? Yeah, there is some fantastic research showing that if you're thinking about the carrot versus the stick as a way of motivating behavior change, that you can get more bang for your buck in general by penalizing people for not engaging in good behaviors than you can get more bang for your buck in general by penalizing people for not engaging in good behaviors than you can get by rewarding them. There are some downsides to that. And considering that we're partnering with companies that care very much about maintaining satisfaction, you know, we have some constraints in terms of what practices we're going to deploy to try to change behavior for the better. So we're planning to focus on bonuses or pluses that we're going to deploy to try to change behavior for the better. So we're planning to

focus on bonuses or pluses that we're offering people as opposed to the stick for the reason that we care not only about whether or not people get the most out of this program, but also whether they're happy. And I think, frankly, being happy may have a lot to do with long-term benefits. And talk about the sort of subjects who will be enrolled in this research. You know, one problem, as we all know, is that a lot of academic research draws from a pool of subjects, typically, you know, college undergraduates, that's pretty narrow and compliant and willing to do almost anything for a free slice of pizza. So is your subject pool more universal than that?

Far more. The people we're going to be recruiting are going to be the customers of the companies we work with. So is your subject pool more universal than that? Far more. The people we're going to be recruiting are going to be the customers of the companies we work with. And frankly, the work we're building on has not always looked exclusively at undergraduates. In fact, when the pilot studies had 2,500 people from a Fortune 500 company, and we know that the techniques we're planning to deploy with these gym populations were very effective in that sample. In education, it's important for us to be helping all kids, but it's, I guess,

especially important that we help the kids who need the most help. And so it's for that reason that we have deliberately partnered with large urban districts, the New York City public schools, the Philadelphia schools, and also charter school partners who largely serve students from disadvantaged households. So it's not that poor kids have a different psychology. I think that their circumstances are different, and so their challenges can be different. But if you're trying to understand motivation, you're essentially trying to understand human nature. Let's say you're looking at trying to encourage people to make nutritious grocery

purchases. In your study or suite of studies, how would that be measured? Are you also measuring the less nutritious purchases and so on? How does that work? So we'll have to use loyalty rewards cards. Right. So that would be how we would track food. But it's only going to be at that grocer.

So you're pointing out a limitation. We'll have to also rely to some extent on self-report. And then one of the things that's really amazing about our study and the partners we've collected is that we hope we will be able to track people through multiple different partners. a CVS Caremark shopper, a Whole Foods shopper, and also a Humana health insurance customer. If they agree and are comfortable with us accessing all of those different suites of data, then one possibility would be to not only look at whether or not they're making more nutritious grocery purchases, but we could also see, well, what are they buying at the pharmacy? Are they picking up the chips that they have stopped buying at the grocery store there? And then we can look at their actual health outcomes. Are they visiting the doctor? How does their blood pressure look? And so on. I'm exactly the kind of person that you don't want in your study, I guess, because I think, well, wait a minute, if you're incentivizing

that stuff from those places. So if I know that if I go and get a lot of cauliflower and sunflower seeds and whatnot, then I'm getting kind of a rebate on those. And with that money I save, I can buy like four more Burger King burgers every week. And there's nothing you can do or know about that until I show up at the doctor's office and potentially you can tell that I don't have much cauliflower in my bloodstream. You know, the question that you're asking, Stephen, is actually very deep, because we all know the kind of Diet Coke and Big Mac effect, right? The idea that when you get healthier in one choice, you feel like you're licensed to indulge a little in another. So since I'm ordering the Diet Coke and not the sugary Coke, well, I'll just have an

extra side of fries with that, right? And that's actually not what we want. We certainly don't want to incentivize buying cauliflower and sunflower seeds and then have the person not go to the gym because they feel like they made healthy choice A, why make A and B? In fact, I would say that the holy grail of enduring behavior change is when you change identity, when you're the sort of person who buys cauliflower and sunflower seeds and goes to the gym every day. In fact, economists would call this complementarity across your decisions that if you do one, you know, you're more likely to do the other. In fact, the benefits of the other are enhanced. When we talk to our partners, for example, weight loss and talking to the chief scientific officer

of Weight Watchers on this, you know, he would say that if you look at people who really lose weight over the long term, there's a change in the narrative. They think of themselves as different people. And I would say the same thing as a former classroom teacher. The kids who make it out of poverty and are able to thrive are ones for whom, you know, these are not piecemeal decisions. Should I do my homework tonight

or should I, you know, be on social media? I'm the sort of kid who does my homework no matter what. Of the three categories that you're going to be working in health, personal finance, we'll call it an education, which seems to have the largest gap between low income and middle income? That's a great question. I think in all three of starting college in the United States, but not finishing with their degree. But the statistics are grossly distorted. So for the kids who come from disadvantage, who are first in their family to go to college, who are from underrepresented minority backgrounds or from poor households are much, much worse off. I mean, in health, the disparities and outcomes by income are just enormous. And also there are racial disparities in controlling for income that are enormous. So that's another place where we know we'll have

more upside working with people who have less. And certainly, I mean, in financial decision making. In finance, it's almost like exponentially difficult. Exactly. Because finance is actually about money. Compound interest. Yeah, yeah, yeah. Exactly right. And as soon as you get into debt, things spiral. So all of these are just settings where it matters so much on the bottom end of the spectrum

to get in there and make changes. Coming up on Freakonomics Radio, you may be thinking to yourself, Duckworth and Milkman sound really smart. Just how smart are they? We don't think we're smarter than Aristotle. We don't think we're smarter than Aristotle. However, Aristotle's not here to defend himself, but he would be a great Freakonomics guest. All right. Glad that's settled.

But still, you're pretty smart. More after this. Angela Duckworth and Katie Milkman are both professors at the University of Pennsylvania and co-leaders of an ambitious project called Making Behavior Change Stick, but they are hardly the same person. Duckworth, a former teacher, is a research psychologist who's been studying the ins and outs of grit, which is a quintessentially internal motivation. Milkman, with a background in computer science and business, is now a professor of, quote, operations, information and decisions, heavy on the external forces. And we disagree all the time. I think we disagree daily. Right. You know, for example, incentives. Hourly.

Hourly. Yes, exactly. Minutely. You know, incentives is one focal case of dispute, which, you know, I as a psychologist think to myself, you know, there are reams of research studies in psychology that show that incentives can often backfire and that you can, in fact, crowd out intrinsic motivation. Katie has, well, you have your own evidence to, you know, to counterweight against that. And I think that will be the art of this project. If we can get these independent minded organizations, these independent minded academics who are working on one thing in their way to work on other things with other people who have different perspectives, then, you know, our bet is that something will come out of this that hasn't come out before. I mean, the problem of behavior change goes back to Aristotle, if not before. And so it's not just hubris that makes us think that, you know, we're smarter than Aristotle. We don't think we're smarter than Aristotle. Aristotle's not here to defend himself,

but he would be a great Freakonomics guest, by the way, if you could resurrect him. We've tried to book him over and over again and just had no luck with that. What is exciting to me about your project is that all these motivated, smart, hardworking people are working in silos that rarely intersect with each other. And your project is the first that I've encountered where you're kind of blowing up all the silos and trying to cross-pollinate all this behavior and data among private firms and academic researchers and presumably policymakers ultimately. Is my take too grand or too Pollyannish for what you guys are thinking about? Or is that indeed what you're trying to do? That is at the heart of the design. In fact, you could argue that cross-pollination is the theory of change at every level of this proposal. So not to exclusively focus on health outcomes, but to look in concert with savings and education.

Because if you look at a human life, you know, those are domains, only three of which, of course, you know, it's not a complete list of all the domains in which a human being operates, but why be siloed and only look at one and not the other? So there's interdisciplinarity at the level of the domain of life that we're examining. On our academic team of 27 scientists, there are people who really strenuously disagree with each other in terms of the approaches they take. They publish in different journals. They have different sensibilities. They have different, I guess I would say, taste in a way about, you know, what's important to look at and what should you not care about so much. I think that the idea is one thing and the execution is another. And Katie and I hope to

be world-class thinkers on this project with the other scientists, but we really also need to be world-class doers and to get this massive thing with partners around the country and academics with things that they were already working on before we asked them to be part of this project, to work together, to argue argue together to actually produce stuff. So we're grappling with that every day. But we also knew that from go. We kind of signed up for being not only academics in the traditional sense, but entrepreneurs and startup folks. So so that's what it feels like, I think, to me.

This project, unlike any other that I've ever worked on, does already feel like its own little company that we're building because we're trying to create a product. We want it to be something that people will be excited about and engaged with in a way that normal academic studies don't worry about these challenges so much. You know, I think there's something missing in academia that is not missing in Silicon Valley or Madison Avenue, and that is rapid prototyping and really actually listening to the market with both ears. In other words, you know, academics do tend to think a lot and to read a lot and then to come to these top-down conclusions, which can be true and valid. But what they're not doing is a lot of bottom-up

empirical fact-gathering. But I think that it's crucial, and we've got to get this right if we're successful, is to actually introduce design thinking. In other words, this rapid prototyping, really listening and being consumer-oriented for this whole thing to fly. So it's not enough to have the scientists and not enough to even have these, you know, institutional partners. We actually need designers and product managers and folks who actually tend not to hang around the halls of academe. But that's going to be a central part of our team. Let me ask you a very rude or impolite question. I guess I'm thinking about the size of this grant. $100 million is a lot, especially in academia. It's a lot anywhere.

Your year one costs in your proposal, you estimate about $14 million. And I guess I'm just thinking a little bit about what economists call the principal agent problem, where it looks like everybody's incentives are aligned. But in fact, even if you get this grant and do everything that you propose over six or seven years and it doesn't work or nothing meaningful comes from it, then you still got to kind of run this project and give projects and paydays to all these other researchers. So persuade me that this isn't just a kind of rent-seeking jobs program for you and your friends. I think that the proof is in the pudding we've already produced. So we know a lot of the answer already. goals and gives them reminders and provides incentives for good behavior for the course of a five-week intervention has actually been tested repeatedly and replicated to produce

lasting behavior change over up to a year follow-up. So we know that these elements are going to produce a benefit. And then the question is, how much bigger can we make the lift by taking the insights from this incredible team of 27 scientists, this incredible team of organizational partners that have been thinking about this on the front lines? So I have no doubt that we're going to make a big impact by deploying something that's been tested and proven in randomized controlled trials to change behavior in the long term for the better. The question is just how big will the effect be? If all we did was deploy what we know works, we would have a huge impact by simply scaling this to the large populations we're talking about reaching. But by A-B testing and bringing in these insights from different fields and from different practitioner partners, I think the sky is the limit. You write that among the millions of people we will reach through this project, we aim to reduce

high school and college dropouts, financial insecurity, and premature deaths by 10%, which is a huge number. What magnitude of change would you consider ultimately a success? Imagine you're looking back at this 10 years from now. I'd be happy with 10%, but I'd be more happy with 20%. I mean, I think to answer this question, also to think a little bit about the question you bring up, Stephen, the seemingly cranky question of, you know, maybe this is just a rent-seeking, just a way of padding our own research budgets by an extra several million for six years. I mean, to me, the biggest bet of all is to bet your life on something. And Katie and I looked at each other and we said, this isn't about $100 million. This is about our research careers and our livelihoods. We're

both working astronomical numbers of hours per week on this. We're betting our lives on this work. So yeah, the benefit better be huge because the alternative to working on this is to be working on other things. And if we thought that there would be a better way to spend our time to work on a bunch of small projects, then we would. But we've decided that it's worth the risk, if you will, of the next seven years plus to work really singly on enduring behavior change and not on a motley assortment of little things that are unlikely to actually add up to people doing the right thing over longer periods of time than either of us have studied in the past.

You keep gathering data and you come up with what I assume then are policy recommendations, but I'm not sure they're only policy recommendations because also, obviously, this takes into account private firms and school systems and so on. So does your work necessarily get turned into legislation? Is it happening more in the private sector? Where do you see it ultimately taking, you know, the deepest route? Well, first and foremost, we're literally going to build a piece of technology that is our intervention and that can be ported in different settings to help people achieve different goals on a long-term basis. So that's going to be open source. We'll share the learnings as well through academic papers

and presentations and op-eds and so on so that organizations, whether they're for-profit, not-for-profit, whether or not they're governmental or not, can take our learnings and deploy them or take our technology and re-engineer it for their purposes. So our hope is that it will be widely used by everyone who sees behavior change as a problem. And I think that's going to be everyone working anywhere in the world, frankly, on social problems. Excellent. Angela, would you mind answering the same question? I think that the contributions of this project are going to be the insights. You know, it's not that the technology that we're, it's not like we're building a superconductor or that, you know,

we're going to program something in the way that we send out texts that is better than the way people have sent out. The advances are really, I think, going to be in insights and those we want to make completely transparent. I don't think that those insights are necessarily going to lead immediately to a policy change, but they could. I think one of the trends that I see that really is a part of the movement that we're both excited about is that policymakers and private companies and everyone else are realizing that you have to engineer around human nature. You can't just assume that people will make the rational long-term decision. You have to work with the way people are. And if this project in the course of its run not only produces a tangible product that helps in the short term,

we do hope that it creates some kind of long-term knowledge about the way human beings tick and that we would all benefit by it. And what happens if you don't win? We're going to do it anyway. Yep. And how does the scale change? How much money do you think you could raise to do this? It would be an order of magnitude less. I think in order to do this the way we really want to do it, obviously, we wouldn't have applied for the $100 million if we didn't think that it needed $100 million to do the most efficiently. But I think that we could certainly persuade folks in the philanthropic community that this would be worth millions of dollars. So we may not be able to think about smoking cessation and savings and educational attainment all at the same moment.

It may be more sequential, but we'll do the work anyway. That's for sure. Just a couple days after we spoke with Angela Duckworth and Katie Milkman, the MacArthur Foundation announced eight semifinalists for the $100 million grant. Making behavior change stick was not among them. Most of the semifinalists had much more specific goals, curing river blindness in Africa, for instance, and more traditional goals, improving newborn survival, caring for orphans, bringing specialist medical care to underserved patients. It's hard to argue with the nobility of these efforts or the need, but the Duckworth-Milkman proposal, to me at least, is categorically different, not just in its breadth and ambition,

but in its desire to attack a problem way back at its root. Think about the difference between a medical treatment and a vaccine. Once you're already sick, you're grateful for the treatment, but how much better would it be to have never gotten sick? If behavior change is indeed at the root of all the suboptimal, self-sabotaging decisions that we humans make, wouldn't it make sense to start there? The good news is, true to their word, Duckworth and Milkman are still pushing hard without the $100 million. They've already raised $1 million from the only organization they pitched so far,

the Chan Zuckerberg Initiative. And later this spring, they'll be convening their all-star team of scholars and institutional partners in Philadelphia. Freakonomics Radio will be there, too. And we'll let you know what happens next and next and next. And coming up next week on Freakonomics Radio, a thought experiment. If we had a chance to reboot our civilization, to build new institutions and systems from scratch, what would that look like? After all our trial and error on Earth 1.0, how would Earth 2.0 be different? We'll hear from some of the most esteemed economists in the world. Well, that thought experiment's not so far from how I think about things day to day.

And we ask a Nobel Prize winner what he'd do as chief economist of Earth 2.0. Well, you know, I would turn the job down. What I would do. I'm actually quite hostile to the question. That's next time on Freakonomics Radio. Freakonomics Radio is produced by WNYC Studios and Dubner Productions. This episode is produced by Eliza Lambert. Our staff also includes Shelley Lewis, Greg Rosalski, Christopher Wirth, Stephanie Tam,

Merritt Jacob, Alison Hockenberry, Emma Morgenstern, Harry Huggins, and Brian Gutierrez. You can subscribe to Freakonomics Radio on iTunes, Stitcher, or wherever you get your podcasts. You can also check out our archive at Freakonomics.com, where you can stream or download every episode we've ever made. You can also read the transcripts and find links to the underlying research. You can also find us on Twitter, Facebook, or via email at radio at Freakonomics.com. Thanks for listening.