Search the Site

Episode Transcript

Stephen J. DUBNER: Hello, John List?

John LIST: Stephen Dubner, how are you doing, man?

DUBNER: I’m great! How are you doing?

LIST: Great! Great!

John List is an economist at the University of Chicago. And he’s a family man – five kids.

DUBNER: Now… do you… do you ever have to bribe any of your kids? I’m just curious, John.

LIST: Every now and then I have to incentivize them. I don’t call it bribing. I call it incentivizing them. My kids refuse to eat seafood. So this comes directly from my wife who claims to have gotten a bad piece of fish when she was a third grader, but I’m a big seafood lover, and we were in the Bahamas about six months ago and I thought of an incentive scheme for my kids that included a large sum of money if they would eat fish for seven consecutive days.

DUBNER: And how’d that work out?

LIST: Three of the five kids it worked for. They collected the money. And I was hoping then that they would acquire a taste for seafood, that they would say, “You know, Dad, that whitefish is actually pretty good.” Or, “You know, one time we had crab legs”. But no, zero of five for the long run, the minute we touched ground back here in the states…the kids have not touched fish since, even the three of five who were eating the fish down in the Bahamas.

DUBNER: How much did they get from you?

LIST: I better talk about that off air. Put it this way, a lot of money.

DUBNER: So three of them at least played you pretty well, though.

LIST: Well, I don’t know about played me. They responded to the incentives, they ate the fish.

DUBNER: But your incentive in giving them that big cash bounty was not just to get them to eat it for the week, presumably.

LIST: Absolutely. I wanted to modify their long-term behavior. And it’s hard to… it’s just a lesson that it’s really hard to change habits.

*      *      *

So is it possible to bribe kids to eat the food you want them to eat? And even if it’s possible, is it desirable? That’s what we’re talking about on today’s show — but not just about bribing and eating; we’ll talk about how bribing incentives may work with all kinds of behaviors. Now, you just heard John List describe his failure to bribe – excuse me – to incentivize his kids to eat something they didn’t want to eat. That was his experience as a parent. So you might think that, as an economist, John List also doesn’t believe that incentives can change the way kids eat. But you’d be wrong. List recently ran a field experiment in Chicago, along with another economist, Anya Samek.

Anya SAMEK: Okay, so what we’re trying to do is understand what kinds of small behavioral nudges or education we can use to actually improve children’s food choices.

That’s Samek. She’s at the University of Wisconsin-Madison.

SAMEK: And we’re using this after-school program called the Kids Café. And there are 24 of these in the city of Chicago and surrounding area. Kids come there, they receive some help on their homework, and then they’re given free food. And what we do is we randomize these different Kids Cafés. And when I say randomize I mean we randomly put different Kids Cafes into different treatments. We have one treatment where we tell the Kids Café just carry on as you are, but at the end of the food-choice we’re going to give kids a second choice, it’s going to be which dessert to have. And they can either have the healthy dessert or the less healthy one.

DUBNER: And what are the two desserts you’re offering, the healthy and the less healthy?

SAMEK: So the healthier dessert — this is actually a good question — we thought about this for a long time, and the healthy dessert is a cookie…I’m sorry.

DUBNER: Wishful thinking.

SAMEK: Yeah, yeah, yeah the unhealthy dessert is a cookie…

DUBNER: What kind of cookie?

SAMEK: We really were quite careful and we chose a low-sugar, whole-wheat cookie. And the reason we did that…

DUBNER: Oh, come on, and that’s the good one?

SAMEK: We didn’t feel comfortable going in and giving 1,500 kids who were at risk for obesity a really high-sugar, delicious cookie.

DUBNER: Okay. Alright.

SAMEK: The healthy dessert then, we again ran into a problem, so we really wanted to use fresh fruit. And unfortunately because of the logistics of how this food is delivered, which is actually another big problem with the way that we deliver food to kids in these school lunchroom settings is that we couldn’t get fresh fruit there so we had to use dried fruit.

DUBNER: Which has sugar concentration problems of its own that fresh fruit doesn’t have, yeah?

SAMEK: Yes, exactly. So one thing that we can’t do is say that for this particular 1,500 kids we really improved their nutrition because all we did was we observed their choices. And all we needed to do was believe that most kids were going to choose the cookie and that’s the less healthy option, and that’s exactly what happened.

DUBNER: Okay so all else equal, you’ve got 1,500 kids roughly. They have their meal and then they do homework and maybe hang out and play a little bit. And then at the end they’re having this snack, this dessert with a choice, yes?

SAMEK: Yeah.

DUBNER: Okay, and just describe to me how this is, how the experiment itself is set up. How is the choice set constructed?

SAMEK: So we announce to kids that they get a choice of this dessert. And we tell them they can only choose one item and they have to eat it there in the cafeteria. And then we come around and we have these trays in which we have a large number of cookies and a large number of fruit and then we just have kids choose one. All of these kids, we know all of their names, we record exactly what they’ve selected, and that’s our control group.

DUBNER: Okay, and you’re doing this for a few days, a few weeks to establish a baseline, how does that work?

SAMEK: We come in twice and then we have kids make this choice, and what we find is that less than 20 percent of kids are choosing healthy. And then we come in for a period of five more days in which we now have these treatment schools actually receive incentives or education. And what we find there, we come in, we read all this information about the food pyramid, so we tell kids that look you really should choose the fruit, fruit’s really good for you, we have this campaign, it’s called Eat Strong, so we tell them you should eat strong. You’re going to be strong on the playground, you’re going to learn more at school, it’s going to be very good for you, this is really good for your health, the USDA recommends that you eat more fruit. And we show on the food pyramid. We have all these posters, and we walk around and show these posters with the food pyramid. And kids don’t improve their food choice at all.

DUBNER: Okay, so teaching them doesn’t improve their choices, or it doesn’t change their choices. Now before we declare that a dead end, do we know that this mode of education was a good one, or maybe…Like when you’re telling me that, if you tell me that if I eat this little cup of raisins and dried apricots that I’m going to be a stud on the playground and a superstar in the classroom, I’m just going to say, ‘Anya, you seem nice, but I don’t believe you. That doesn’t sound compelling to me.’ Do we know if the education is actually considered legitimate by them?

SAMEK: Well, it’s the education that you would administer if you went on the USDA website and pulled off their informational materials.

DUBNER: Right, I know that, but what I’m saying maybe the USDA isn’t so good at this. I realize I’m throwing a stupid thing in the middle of your smart thing. But that’s the way I think.

SAMEK: No, so obviously we could be using the wrong education, that’s true. So we can’t say much about whether it’s the right education or not. But what we can say is that we give them this education, we tell them about how healthy fruit is and 8 out of 10 kids still choose the cookie.

DUBNER: Okay, alright, so that’s not working so well. What else do you have up your sleeve?

SAMEK: So we come in and we have these incentives that we thought, we didn’t even know what to do with these incentives because they were a little bit lame, so we just had these pens, we had these rubber bracelets and then we have these tiny plastic trophies that just said “I ate strong.” And we told kids if you choose the fruit and you eat it, you get to pick out one of these prizes. And now we have 8 out of 10 kids who are choosing the fruit.

DUBNER: Oh my god, for plastic trophies, and pens and rubber bracelets.

SAMEK: That’s right. So these things are very popular. They’re very cheap so they cost us less than a dime each. And kids are choosing healthy and they don’t have any education about why they should be choosing healthy, at least from us. But now we’ve really improved food choice.

DUBNER: Now, how do we know this is not just a novelty, that it’s the first time that they get offered the incentives, or does it not even matter? Is it just a matter of kind of switching the habit or the preference?

SAMEK: Well it actually works every time. So we come in five times and every time we have these really high rates of selection of the fruit.

DUBNER: And do they actually eat the fruit then? Or do they just —

SAMEK: They have to eat it to get the toy. They have to eat it.

DUBNER: And did you check their cheeks and so on like in prison to make sure they swallowed or they’re sticking it under the table, or it doesn’t go that far?

SAMEK: We were actually pretty careful. So this study took a lot of undergraduates from the University of Chicago who were walking around and monitoring these kids.

DUBNER: Okay, so what did the researchers take away from this experiment?

SAMEK: Overall all of these studies really show, first of all, that you can make a small change in a school setting, in a food setting, for kids and have it have a big impact on choices. It shows that education is not enough. We actually do find when we combine education with incentives, that that has the strongest long-term effect, which I didn’t address earlier, but that’s an important point, that in addition to education we need to really give kids a moment where they can make a choice. And that’s the moment where we can provide a nudge, where incentives can act as a great nudge for that.

DUBNER: If I’m a restaurant, I want to sell a lot of food, I want to make the money that I’m able to make. If I’m the government, I think especially as I’m paying more and more for people’s health care long-term as the government, I want to keep people healthier and therefore I have an interest in getting people to eat more nutritiously. So I can put some pressure on restaurants to either serve more nutritious food, serve less food, or help them somehow come up with a scheme to at least educate people or make them a little bit more likely to eat more nutritious food. So I can have calorie counts, or maybe even subsidize healthier food choices. That seems to be fairly viable. Yeah? Do you see evidence of that kind of thing going on and working well?

SAMEK: One project we have that is very effective, it’s a small nonprofit grocery store on the south side of Chicago called Louie’s Groceries. And we’ve gone in and we’ve been running studies in which we give adults incentives for choosing at least five fresh fruits and vegetables in their cart. And we’re actually giving them a dollar for every five servings of fresh fruits and vegetables they buy in a shopping trip. And we’re comparing that to another treatment in which adults are just receiving some educational information about why it’s important to eat fruits and vegetables. And there we see almost the same results that we’re seeing with these kids. So in a real environment where adults are making decisions about purchasing, incentives really work.

DUBNER: So the takeaways here seem to be that fruit must be subsidized whether dried or fresh, yes?

SAMEK: Right.

DUBNER: And that kids can be bribed successfully.

SAMEK: Yeah, well adults can be bribed successfully, too.

Coming up on Freakonomics Radio: so we can incentivize choices that benefit society, at least in some cases – but why should we have to?

LIST: You know, I think that this is one of the most important issues that humanity faces, is making this tradeoff between doing something costly now that will benefit me or humanity in the future. Things like staying in school, things like saving for retirement, things like adopting green technologies.

Also: if you are not a subscriber to this podcast – well, please become one. It’s free. It’s easy. You can sign up at iTunes – and then you’ll get the next episode in your sleep.

*      *      *

We heard Anya Samek describe the field experiment she worked on with John List.

LIST: I’m John List, and I’m a professor at the University of Chicago’s economics department.

What, exactly, does John List do?

LIST: And what I do is I go out into the real world and try to change it for the better using field experiments.

DUBNER: That sounds pretty simple. Can you give me an example?

LIST: Very simple. So let’s talk about the simple example that we’re talking about today. What we’re after is we’re trying to explain or describe first of all the consumption choices of the underprivileged children in America. And once we see the types of foods that they’re consuming, which is not a great thing. You know, if they have a choice between a cookie and a bowl of fruit, they choose the cookie. So our idea is to go out…

DUBNER: Now, John, can I just interrupt you for once second there? Is it fair to say that this is the kind of poor choice, relatively poor choice that an underprivileged kid makes or is that the same choice do you know that a mediocre-ly-privileged adult would make as well. In other words is this a function of childhood or…

LIST: Absolutely.

DUBNER: I’m just curious to know what you know about food choice in children versus adults generally.

LIST: Yeah, so I think that this is a general problem, but you have the problem in an even stronger way amongst the underprivileged who have less resources to use to purchase and consume healthier foods. So when I say underprivileged, I think this is our experimental testing ground, but nevertheless you do have the same problem throughout society. You have kids, you know, running an obesity rate of about one in five across America. And in urban settings the obesity rate is a little bit higher than that. But the general point here about all of this is that you have many problems where what you do now affects what happens later, and usually we choose the easier decision or the easier action now. You think about savings for retirement, you think about getting doctor check-ups, you think about going to school, you think about engaging in risky behaviors, you think about adopting green technologies for our houses. In all of these cases, we usually choose the bad action. And that action is to do what’s best for us now to the detriment of the future, to the detriment of our future self. And nutritional choices right now are just one of these elements that we face in society where we need kids to recognize the choice that you make now will critically affect your outcome in the future.

There’s an idea, quite prevalent in our society, that if we can only teach people that they are making poor decisions now that will adversely affect them later – well, they’ll make better decisions now. But the experiment that John List and Anya Samek did in Chicago, trying to get kids to eat healthier snacks, showed that an educational message didn’t work at all, at least not in that setting. So… maybe it’s reasonable to think that educational messaging isn’t as potent as we think.

LIST: Right, I think at just about every walk of life we have messages like get out the vote, it’s your duty to vote. We had messaging in the States on smoking for decades. And now with a combination of changing the prices of smoking, and what I mean by that is increasing the tax rates on purchases of cigarettes, along with messaging, those together formed a very important duo to curb smoking in the United States. But when you look at other countries, just go to Europe, you can see that smoking is alive and well in Europe. And they surely have the same information that we have on the detrimental effects of smoking. But I think in every walk of life you have people saying stay in school, don’t do drugs, you know, messaging has a really hard time working, and I think by and large because the recipients of those messages is not a demander of that information. And if you’re not a demander, you will have very little use for that information. You know, people say we should be green, we’re ruining the planet because of carbon emissions, global warming. But again, that’s a problem where if we curb our consumption of carbon now, that hurts us, that hurts our economic growth, and we’re not going to see the advantages for 50, 100, 150 or 200 years. People don’t want to make that tradeoff between now and then.

DUBNER: Do you think that policymakers and other, you know, incentive creators are starting to get the message that educational messaging especially when its got a sort of moral tone to it is not very effective, or do you see people in the academic and research realms where you live who understand this quite fully just look out at those policymakers and say, ‘Oh boy, they really don’t get it yet’?

LIST: Right, I sort of see, I sort of see both. On the one side, you have policymakers who don’t get it. On the other side, you have policymakers who really do get it, but that’s really the only tool at their disposal. You see, using messages is cheap. You have government agencies… right now I’m working with the U.K. government to try to convince people to pay their taxes. You have many people who have not paid their income taxes in the U.K. And the U.K. government sends out letters to them every year telling them please pay your taxes. So they ask us, ‘Can you run some field experiments with us to help convince people to pay their taxes’? And we propose all kinds of different incentive schemes — you know, fines, jail time, all these fun things that economists dream about, and the policymakers say we can’t do that. We can’t change law. But what we can do is we can add those two cool sentences that you used back in your 1998 study, and we can use those sentences, which are on moral suasion to try to get people to pay their taxes. So, you see, they’re smart enough to know this might not be a great tool…

DUBNER: It’s the best they have.

LIST: Yeah, and it’s not very costly, and that’s important to them.

DUBNER: So, John, is the takeaway message of your study that it’s better to bribe kids to eat healthy food than to just teach at them?

LIST: I think the message would be you want to do both. I think you do not want to use messaging alone. If you do use messaging you should combine it with an incentive, because that will allow you to convince kids to make better choices, but it will also yield better consumption choices. I think we need to understand that this is a two-part problem, not only the choice but the consumption. And what we find is that messaging along with incentives give you that outcome, or that set of outcomes that you want.

Now take what you learned from this study, which is that incentives work in terms of food consumption, and generalize it for me as much as you can, not only out of the experimental realm, but out of the childhood realm.

LIST: Right absolutely. You know, I think that this is one of the most important issues that humanity faces, is making this tradeoff between doing something costly now that will benefit me or humanity in the future. Things like staying in school, things like saving for retirement, things like adopting green technologies. We have to convince people right now to make the right choice for, in many case themselves, and also society, and I believe that strong dosage of incentives combined with tasteful messaging will allow us to get to that point, where we have people making the right choices now that will yield the right outcomes in the future. So when you think about all of our major problems, it’s a tradeoff through time. And you have kids who would rather skip school and go have fun when they’re 16 years old than stay in a boring algebra class. It’s a very simple point of calculus for them. They don’t see that the future labor market returns are 12 percent for every year that they stay in school. When you’re 16 years old you don’t even think about what’s going to happen when you’re 25 much less when you’re 18. And we need to combine the correct basket of incentives to align the kids’ choices with what society wants them to choose.

DUBNER: Okay, so that’s all well and good to say, and even if one believes it, and I personally tend to pretty much believe what you’re saying, what do you do? What is that basket of incentives for a 16-year-old who’s teetering on the edge of dropping out or not going to school?

LIST: Yeah, I think it is both. It depends on what realm we’re in. If we want to talk about education I think the solution is changing the prices to education. What I mean by that is taking some of the future rewards that we as a society reap from the kids staying in school and not committing crime and going to jail, taking some of those rewards and giving them to the kids now, or giving them to the parents now to incentivize the parent to make sure that the child stays in school, takes school seriously and goes on and gets the right education to help society and the future. That might be different than our food choices, for example, because we’re not talking about 16-year-olds now in our food experiment, we’re talking about six, seven, and eight-year-olds. Incenting a six, seven and eight-year-old with dollars doesn’t make any sense, you know, you have to choose, again, the right incentives. We chose toys, things like rubber ducks really turn a seven-year-old on. So I think in each case there is not a silver bullet, but nevertheless in each case in the long run we need education because, like I said, habits are what keep you going. And I think the education and information change beliefs but only over generations. I think now you have my kids who are coming home and telling me, ‘Papa, you better not smoke that cigar because it’s really bad to smoke.’ When I went home and saw my mom smoking I never thought it was a problem, smoking. But now our kids they’re actually programmed to think that smoking is a really bad thing. That has taken place over generations, and that’s were I think information and education works over the long run when we change habits. But you need initial incentives to change the motivations of our kids now. We just don’t want two generations of lost children because we couldn’t change their habits, we need to change the motivations now, and I think incentives will do that.

Read full Transcript