3 Takeaways - The Latest Findings in Behavioral Science and How They Can Make Your Life Better: Harvard Professor & Nudge Co-Author Cass Sunstein (#54)

Episode Date: August 17, 2021

Cass Sunstein is a professor at Harvard and one of the founders of behavioral science. He has written two books with Nobel Laureates, including Nudge with Richard Thaler. We are unknowingly influenced by our environments, by how choices are framed, by the people we are with, and a myriad of seemingly insignificant factors. Learn the latest findings in behavioral science and how they can be used by everyone - governments, businesses and individuals - to make people's lives better.

Welcome to the Three Takeaways podcast, which features short, memorable conversations with the world's best thinkers, business leaders, writers, politicians, scientists, and other newsmakers. Each episode ends with the three key takeaways that person has learned over their lives and their careers. And now your host and board member of schools at Harvard, Princeton, and Columbia, Lynn Thoman. Hi, everyone. It's Lynn Thoman. Welcome to another episode. Today, I'm excited to be with Cass Sunstein. He's a professor at Harvard and one of the founders of behavioral science. He has recently written books with two Nobel laureates, Noise with Danny Kahneman and Nudge, the final edition with Richard Thaler. We all think that we're rational, but the science says otherwise. We're unknowingly influenced by our environments, by the people we're with,

by how choices are framed, and by a myriad of seemingly insignificant factors. As a result, people make mistakes in areas ranging from savings to healthcare, driving, energy conservation, and even eating. I'm excited to learn the latest findings in behavioral science and how they can be used by everyone, governments, businesses, and individuals to make people's lives better. Cass, welcome, and thanks so much for our conversation today. Thanks to you. Great to be here. Cass, I very much enjoyed both your book, Noise, written with Danny Kahneman and Nudge, written with Dick Thaler. Many of the most important moments in our lives rely on the judgment of others.

For example, we expect doctors to correctly diagnose illnesses and judges to make fair rulings. But there's a massive flaw in human judgment that you and Danny Kahneman discovered and called noise. Noise is invisible. For noise, you don't know what the cause is. It's random. And another kind of error that you talk about is bias. Bias is consistent error, where noise is the opposite. It's scattered and random. Cassiope said, and I quote, bias has a

kind of charisma. It's a little like the person on the screen you can't take your eyes off of, Elvis Presley. Noise is more like the character in a screen you can't take your eyes off of, Elvis Presley. Noise is more like the character in a murder mystery to whom you never pay attention, but who turns out to be the killer, unquote. Can you give some examples with doctors and judges of bias and noise? Yes, thank you for that. So let's start with a scale, shall we? And then we'll talk about doctors and judges. Some scales are biased in the sense that they show you consistently as heavier than you actually are. That's my scale. It's a bias scale.

Some scales are biased in the sense that they show you as less heavy than you actually are. That's not my scale. Systematic error is bias. And with respect to unrealistic optimism or discrimination on the basis of race, we see bias. Noise is very different. It's scattered. So if you go on a scale that shows you as a little heavier than you are half the time and a little lighter than you are half the time, that's a noisy scale. If you have an employer who is sometimes discriminating against women and sometimes discriminating against men, that's a noisy employer. So let's talk about noise in medicine, shall we? That if you have doctors who are inside themselves, susceptible, let's say,

to whether it's sunny outside or really cold, or susceptible to whether their favorite sports team won the day before, or affected by whether they're really tired because it's late in the day, or really energetic because it's early in the day, that would be a noisy doctor. We actually have evidence that doctors are noisy, that early in the day, doctors might treat patients very differently from how they do in the afternoon when they're tired. You can also see noisy hospitals, and this is the more serious problem, or noisy medical professions. When you go to some doctors and

they'll say, go home, I think you don't have a problem. We're going to engage in watchful waiting. And other doctors presented with the same person with the same symptoms will say, I'm worried about you. We're going to do a battery of tests. In the first case, you might not catch serious illnesses. In the second case, you might get overdiagnosis, and that's a big problem. In the criminal justice system, we also see noise in the sense that some judges are very severe and some judges are very lenient, which means that life for many people convicted of crime is really a lottery. The most important moment in their experience of the criminal justice system is not the argument, but is the selection of the particular judge who is presiding over their case. And that's really unfair.

Cass, I wish we could spend more time on noise. But for anybody who's interested, we have a conversation with your co-author and Nobel laureate, Danny Kahneman, on noise. It's episode 27 of Three Takeaways. Cass, let's pivot and talk about your work with Dick Thaler on what you call nudges. Nudges are so powerful that since you discovered them, they've been used by governments, businesses, and individuals. What's a nudge? A nudge is an intervention that preserves freedom of choice, but steers you in a particular direction. We might think of some nudges as involving architecture. So if you are in a store and there are a bunch of products that are the first thing you see, and there are a bunch of others

that are hidden in the corner, you are being nudged to get the ones that are the first you see. If you join a workplace and you are automatically enrolled, let's say, in a savings plan, but you can opt out, you are being nudged to be in that savings plan. If you buy a product and it says there are this many calories associated with it, if it's food, you are being nudged. If you buy an automobile and there's a label on it that tells you something about greenhouse gases and fuel savings or not, because it's a gas guzzler, that will be a nudge. So you can see from these examples, I think, that some nudges are educative. They tell you something. The GPS is an educative nudge. Others are architectural. They make one thing easy and another thing a little harder.

And those are often the most effective nudges. Can you give examples of where you see nudges? Would you see them, for example, in the American judicial system? Sure. In ordinary life, you see nudges in your car. So if it makes a beep, beep, beep as you're about to hit something, that's a nudge. If it makes a little noise as you're being nudged maybe for better, maybe for worse. In the legal system, there's a lot of nudging. There are mandatory disclosure requirements that the law imposes. Those are nudges. I was involved in a bunch of them when I worked for President Obama. And also, if you enter into a contract with someone, there are a bunch of terms

that kind of kick in if you don't specifically say what you want the term to be. So about the time of performance, let's suppose you have rodents, forgive the example, you have rodents in your house and you hire someone to come take them away and they come in three years, that's probably breach of contract because the law kicks in a term about reasonable time or performance, which three years will exceed. So much of the law of contract consists of nudges that specify what happens if the contracting parties haven't said what they want, and it preserves freedom as nudges do. Think of a GPS. You can say, I want to go my beautiful route or my familiar route, even though the GPS is giving you the faster route.

There's nothing mandatory about the GPS. And the same is true of much of contract law. Do nudges apply to something that's so seemingly polarizing and intractable as the problems between Republicans and Democrats? Sure. You can see on social media certain nudges that can cut across ideological lines and maybe affect Democrats and Republicans at the same time. So it might be that on Twitter or Facebook, there's a mandatory disclosure. Twitter does this for some false statements. It doesn't take them down. It

nudges people to look at other information sources. Or it might even tell you, as Facebook sometimes does, false before you see the relevant thing. That's a nudge. It's not coercive. Nobody's being fined. But people are being nudged to have a sense of reality. And how did social nudges work, for example, on panels of federal judges where you have Republicans and Democrats? Social nudges, let's just think that each of us is nudged by others. So if we're at a restaurant and our friends are ordering a ton of food, we might well be affected to order a ton of food. If we're at a restaurant and our friends are being, let's say, abstemious and bypassing everything except the little main course, we might be influenced by that as well. Maybe not,

but we might be. On the federal judiciary, the data shows, and this is spectacularly surprising, I think, that the political affiliation of the appointing president is a pretty good predictor of how judges are going to vote in ideologically contested cases. That's not the surprising finding. So if a Trump appointee is in an environmental case, the fact that it's a Trump appointee will be predictive compared to if it's an Obama nominee. You're not amazed to hear that. But often on three-judge panels, a better predictor of how a judge will vote than the political affiliation of the appointing president is the political affiliation of the president who appointed the two other judges on the panel. So the stunning finding is if you want to ask how a judge is

going to vote, don't ask, did a Republican or Democrat appoint that judge? Ask instead, did a Democrat or Republican appoint the two other judges on the panel? That will predict how that judge will vote. And that's because the data shows a Democratic appointee is going to be pretty conservative when sitting with two Republican appointees. And a Republican appointee is completely the same, looks pretty liberal when sitting with two Democratic appointees. That shows social influence as a nudge. That is so surprising. And social nudges apply everywhere. They apply in meetings. Can you give some more examples?

Sure. Suppose you're in a meeting and the question is whether to hire someone. Let's call her Mary. And let's suppose that more than half the people in the room think that Mary should be hired. Mary's good. And let's suppose that the first speaker in a group of, let's say, seven is in the minority of three who thinks that Tom should be hired and not Mary. So the first speaker says, Tom's better than Mary. Let's go for him. The second speaker, let's suppose just by the line of speakers that was arranged by seniority or something, says, I agree, Tom. And now you've got two votes for Tom. And then the third speaker is either influenced by the first two to shift from Mary to Tom, or is one of the minority of three that's been in favor of Tom. Now you have three votes for Tom. It may well be that the four have been nudged to support

Tom, either because they are swayed by the information conveyed by the view of the first three speakers, or because they don't want to look like an idiot in front of their peers. And the result is that this group ends up hiring Tom, even though the majority actually favored Mary. And let's just say the majority on this one is right and Mary's better. And that happens in groups all the time where who is the first speaker turns out to be determinative often of what decision is made. And this can be on very high stakes matters like whether to close down an operation or whom to hire as the next boss

or what to do even in government. I find that just fascinating. And the ways that people can make better decisions in meetings, what you call decision hygiene, are so important. I'm sorry we don't have more time to talk about the decision hygiene, but people can read both of your books, Noise with Danny Kahneman and Nudge with Dick Thaler, to find out more about them. But here we're talking more about nudge. Can you give some examples of successful nudges, say for savings or for health or organ donation? I'll tell you the one that, the only one that I have a hard time not crying when I discuss, and it's something I had some involvement in when I was in the Obama administration.

So here's the program. Poor children get preschool meals, lunches and breakfasts. It's a program that has bipartisan support. It's for poor kids. They get to eat nutritious meals. But, and here's the problem, a lot of families don't sign their kids up. We don't know exactly why. It might be because the form is confusing. It might be because they're really busy, struggling with economic needs. It might be they're scared, they get a form from the government. It might be they're suspicious, they get a form from the

government. But a lot of them don't sign up. So the policy change, the nudge, is that if the school or the locality knows that kid is eligible, it just has data demonstrating that the child is poor and eligible, It's automatic enrollment. They don't have to take advantage of it. The parents or the kids even can say, I don't want your meal. I want my peanut butter sandwich. But I don't want to eat today. But it's automatic enrollment in the program.

The latest data suggested that 15 million kids are benefiting from what's called direct certification or something similar, which means 15 million poor American children are getting free school meals just because of a modest switch from you have to sign up to you're automatically enrolled and you can sign out if that's what you want to do. And while 15 million is a number, if we put on the screen, let's say, some very small subset of those kids, it's possible there wouldn't be a dry eye in the house. Those kids, their lives are getting much better. Now, that is an illustration of something involving food and hunger and nutrition shifting from opt-in to opt-out.

But that general kind of switch tends to be very powerful. In Switzerland, I'll give just one more example. There's a recent policy change by which electric utility providers are automatically enrolling people by government mandate in cleaner energy sources, but the households and businesses can opt out. So if you're having solar or wind and you think that's more expensive than let's say coal, you can switch. And this is a nudge. It's not a mandate. It's not a tax. It's not a subsidy. It's just switching the default. And the consequence is that not only households, but small businesses, but medium-sized businesses. Not only households and small businesses and medium-sized businesses, here's the surprise, but large businesses are using green energy a ton more.

And the reason I say that's a surprise is that big businesses are profit motivated. That's what they're trying to do, make money. And they have complete freedom to use a cheaper energy source if they want. But the nudge works. They don't switch. We don't know exactly why. It might be a matter of conscience. It might be the shareholders maybe want them to stick with green energy. But the result is that in Switzerland, there's a lot less greenhouse gases being emitted. And the really cool finding is that the benefits of automatic enrollment in terms of pollution averted crush the costs. The costs aren't trivial. The greener energy sources in Switzerland are more expensive, but compared to the benefits, they are trivial. The health and

other benefits are much higher. What are some of the elements of nudges? You've just talked about changing the default option and not having to make people opt in. What are the other elements of successful nudges? Well, it's pretty good if you want success as well as if you want to be ethical, if the nudge fits with people's interests and values. There are some mean or savage nudges out there where people are automatically enrolled in terms, let's say online, that they don't have real visibility into, like hidden fees, or they get some feature of a product that they have no need for, but it costs them money, and they don't have real visibility into that. So that's not consistent with people's interests. In the case of Switzerland and

greenhouse gases, the effectiveness, I think, is that being enrolled in green energy is something the relevant population actually feels good about. If you nudge people, let's say, to give money to their least favorite political candidate or their least favorite charity, let's say it's a charity that they think is not worthy of the term charity, they're probably going to opt out. If you inform people, let's say that some product has chocolate as a way of discouraging them from consuming the product, at least in my case, that will not be an effective. The disclosure, this warning, this product has chocolate it helps me get where I want to go, or whether instead they'll think this is some imposition by someone who cares about themselves, not me, or who cares about some values that are uncongenial. You and Dick Thaler invented the term choice architecture. What is it, and can you give us some examples?

I'll tell you the story that Dick and I, when we were doing the book, had lunch all the time at a little restaurant in Chicago. And Dick read a book by Don Norman about architecture, really, the design of everyday things. And he was extremely excited about it. And he said, there's architecture in everyday things, but actually the effect on choices is much broader. He said, there's choice architecture. And he said, I think this is a central idea in our book, which we were writing. And as was typical with both of us, when one of us had an idea, the other was like a prosecuting attorney interrogating to see if the idea would be sustained. And I said, choice architecture, what do you mean exactly? And it was actually developed into a discussion

by which the idea got, I think, developed over lunch. And the basic idea is if you go into, let's say, a store, it will have not only an architecture in the familiar sense, but it will have a choice architecture. It might be that the diet soft drinks are at the front, and that is intended maybe or an accidental way of getting people to buy diet soft drinks. It might be the candies at the front, and that might be architecture, choice architecture, knowing that people are more likely to get things that are, let's say, right near the checkout counter, and thinking that impulse purchases are fun for consumers and may be very profitable for the store. If you go on a website, there will be a choice

architecture there, meaning that the design will be powerfully influential on what people end up choosing, even if it's quite subtle. So it might be that there's a bright red color that data shows attracts the eye and makes people click on the relevant thing. And the relevant thing might be, let's say, in the interest of the consumer or in the social interest or in the economic interest of the seller. If you go in a doctor's office, there's a choice architecture there with design and how the doctor frames options creates a choice architecture. If the doctor starts by saying, let's say in your situation, 90% of people choose this, there's an architecture being created. Or if the doctor says, if it was me, I would do that. That's the nature of an architectural

intervention. It's a nudge. And in both, saying what most people do or what the doctor herself would do, you're not seeing a mandate. The doctor is preserving freedom of choice, but is creating a choice architecture, which influences outcome. If it's hot or cold, and sometimes temperature is used strategically by people who want to manipulate choice, you can affect what people will do. It sounds like there's choice architecture everywhere. You describe cafeterias where if the fruit comes before the cakes, more people have fruit. You describe cases where the way that parents describe educational options to their children influences the children's decision. Is there such a thing as neutral design? No. And that was one of the fundamental points of the first edition of our book. And it's even

more prominent in the final edition of our book. And it's easy to miss. We've learned over the last decade and more that many people rebel against the idea of nudges. Not most, but many, thinking, don't nudge me. Who the heck do you think you are to nudge me? I'm doing fine. And even if I'm not, it's my life, not yours. And I admire and share that instinct, which is why nudges are choice preserving. But there's no such thing as a neutral design. There's no such thing as choices without choice architecture. A website has to have a design. The same is true of a grocery store. Credit card bill has to have a design that's going to nudge you. If you buy, let's say, sunscreen, it has to have a package and it has to have some words on it at least. And the package and the words will be a nudge. It has to have a package and it has to have some words on it at least.

And the package and the words will be a nudge. It has to have a color. So design influencing choice is unavoidable. The idea of a world without nudges is unimaginable, not because it would be a bad world, but because it's like saying two plus two equals five. That can't happen. So, for example, in the United States where tipping is common and restaurants increasingly put suggested tip amounts at the bottom of their bills, if they put three choices with very high rates, is that a nudge? Yes. It's called an anchor. If you ask people, how many nations are there in the United Nations? And then you ask them, before you answer that question, I want you to tell me the

last three digits of your social security number. There's a good chance that their estimate of the number of nations in the United Nations will be influenced by the last three digits of their social security number. That's to suggest a number that is given to people when they're trying to generate their own number. The first will influence the second, and that can be used in a self-interested way. So if there's a default tip where the easy thing to click, let's say, is 20%, and that's all you got. But then you can enter your own number if you want. You might, because it's kind of a bother to enter your own number, just stick with the 20%, which in some places and times is pretty high. Or you might end up at 15%,

where you otherwise would be at 10% because you were anchored on 20%. So some number given for tipping or basically for anything, like a home that you're thinking of buying or a car that you're thinking of buying, the initial number is an anchor, and it's an anchor as nudge. There are some very large, sophisticated companies like Facebook that use behavioral science. How does Facebook, for example, use behavioral science to influence behavior? I've actually worked a bit with Facebook on some of these issues, and I admire some of the things Facebook is doing with the use of behavioral science and some of the things I admire less, let's say. Facebook is just full of nudges. And that, as noted, is inevitable.

I think it's on your birthday. I think this is true now. It has been true. You are nudged to set up charitable donations. And it's really easy. So they have a social good project that is designed to help people to give to charity. They don't have to. They don't want to. It's just a nudge. But it makes setting up charities and helping you nudge your friends to give to the charity you like really easy. So it both is a nudge from Facebook and it is an enabler of nudges from you. And I think this is terrific. They've generated a lot of money for charities through choice architecture. In emergency circumstances, as I recall, if there's some terrible natural disaster in an area, I believe Facebook has a bunch of nudges to help people respond by sharing information about what's happening and about how

to reduce relevant risks. For voting, there are nudges in the form of, at one point, Facebook made it really easy for you to tell your friends that you voted. And as we were discussing, that is a social nudge. In the context of COVID-19, Facebook has had a lot of nudges to help people obtain information, and I believe to promote both vaccination and mask wearing through information provision. Those are nudges. Your newsfeed is full of nudges, meaning what appears first on your newsfeed is influenced what you're going to click on. And each of us has a designed newsfeed, which is reflective of our own patterns. that means it should fit pretty well with our tastes and not someone else's tastes. And it should also encourage clicking behavior,

which has an upside, but also a downside. Cass, before I ask for your three key takeaways, is there anything else you'd like to discuss that you haven't already touched upon? Yeah, I think the main theme of the final edition of Nudge, he said, was a combination of resolution and sadness. The key thing, the final edition of Nudge is Nudge for good. And each of us has an opportunity, like today, to do something that might be small, won't change the world, but it will change someone's world. And that is the tacit theme of this book. It's been an extraordinarily important book. Cass, what are the three key takeaways you'd like to leave the audience with today?

The first is a poem by Baudelaire, and the poem is called Get Drunk. I say this as a person who doesn't drink much at all. The poem is not about liquor necessarily. It says get drunk with wine, with virtue, or with poetry as you choose. And the theme, which is my first takeaway, is find something that makes you the equivalent of drunk. It might be an idea. It might be a project. It might be an engagement with a family member. It might be a romance. It might be wine. Get drunk. That's number one. That is wonderful. Number two is something I've thought of relatively recently. I'm in government now, and there are a lot of interactions and occasionally conflicts. And the second idea is that everyone is the hero of their own life. And that is a big notation that I think should be in capital letters and large fonts on our view screen. think they've been inattentive to something or mean or something negative, keep in mind that

they're the hero of their own life. That is, they have one life to live. It's theirs. And from their own perspective, chances are, and their perspective should be honored, a negative read of what they did is unfair in the extreme. So when I say everyone is the hero of their own life, I mean, think of other people as people who have every bit as much of an entitlement to be treated with respect and kindness and generosity as one oneself is. And that makes, I think, every day a little more relaxing and a little kinder. The third is consistent with what we've been discussing, which is the immense importance of personal agency, both for ourselves and for everyone else. So nudging insists on maintaining

freedom of choice. And one thing I've learned in the 11 or so years since the book was published, and maybe coming up on 10 in the several days since the final edition was published, is the universal human need for a sense of personal agency, which might mean ownership of your own choice, or which might mean rebellion against foreclosing choice. And of course, I'm for mandatory seatbelt laws. I even think in some places, mandatory mask wearing is a really good idea. But personal agency is a need built into our species, and it's good to keep that in mind. Thank you, Cass. This has been terrific. Thank you for all your work to make the world a better place. And thank you for your current work in government. Thank you. Greatly enjoyed it. Behavioral science is such an important topic. If you're interested, Three Takeaways has several guests on different aspects of behavioral science, including Nobel laureate Danny Kahneman, author of Thinking Fast and Slow, and co-author with

Cass Sunstein of Noise. Danny talks about errors in decision making and noise in episode 27. Bob Cialdini, author of Influence, which has sold over 5 million copies and which is Warren Buffett's number one recommended business book. Talks about how to get people to say yes in episode 42. Bob is called the godfather of influence. Iris Bonet, the former academic dean of the Harvard Kennedy School, talks about unconscious bias and what works to de-bias how we live, learn, and work in episode 49. And lastly, Katie Milkman, a University of Pennsylvania professor, shares on how to successfully change ourselves using evidence-based methods that work from behavioral science. That is episode 41.

Thank you, and I hope you enjoy all the episodes. If you enjoyed today's episode and would like to receive the show notes or get new fresh weekly episodes, be sure to sign up for our newsletter at 3takeaways.com or follow us on Instagram, Twitter, and Facebook. Note that 3takeaways.com is with the number three. Three is not spelled out.

See you soon at 3takeaways.com.