Richard Thaler outlines how principles from behavioral economics can help policymakers — and managers — achieve better outcomes.
Q: Could you explain some of the key ideas in Nudge: nudges, choice architecture, and libertarian paternalism?
"Libertarian paternalism" suggests that these two seemingly contradictory terms can actually define a non-contradictory and attractive policy alternative. By "libertarian" we simply mean respecting people's right to choose, whenever possible. And by "paternalism" we mean caring about the outcomes for people, as judged by themselves. So we would like to create environments where people are more likely to choose things that they, themselves, think are good for them.
Now, the person who is in charge of that choice environment is somebody we call a "choice architect." There are choice architects in virtually every environment. When a professor teaches a course, he is the choice architect. When somebody puts this magazine together, they will decide in what order the articles appear and what illustrations and photos accompany them that may or may not attract people's attention. That's a good example, because people are free to throw the magazine away. They are free to read whatever they want, but the magazine designer will have some influence on which articles they read, and in which order. And what we know is that all kinds of small things, like whether there is an illustration that accompanies an article, will influence whether people read that article. And those small things are what we call "nudges." So a nudge is any small feature of the environment that attracts people's attention and alters their behavior but does so in a way that doesn't compel.
Q: Could you give some examples of where nudges have influence?
Probably the areas that have received the most attention so far are savings and investment. The simplest example of a successful nudge is the default option. A default option is simply what happens if you do nothing. Normally, nothing happens, but sometimes even when you do nothing, something happens. So while I'm sitting here talking to you, if I do nothing on my computer long enough, pretty soon the screen saver will come on. How long it is until that happens was itself a default option that came with my computer that I never changed. What we know is that default options are extremely powerful. Many people just go with the flow and take whatever the default is. That means that the choice architect has immense power by choosing the default, sometimes knowingly and sometimes unknowingly.
A good example is in the area of pension policy. In many 401(k) plans, the default option is not to join. If you are going to join, you have to fill out some paperwork. Some companies have tried the opposite default, which is that you are enrolled unless you fill out some paperwork. We know that speeds enrollment greatly and doesn't really cost anything. Shlomo Benartzi and I have added to that a policy called Save More Tomorrow, where people are invited to join a plan in which they agree to increase their savings contribution every time they get a raise. That's another good example of libertarian paternalistic policy. No one is forced to join it. People sign up of their own free will, but in the first company where we did this we more than tripled savings rates.
Q: How is it that nudges haven't been the norm all along?
They are the norm. We've been nudged forever. Eve and the serpent nudged Adam. Religions have been nudging us for thousands of years. Marketers nudge us. Ads are nudges. We can be nudged for good or for evil. There is a company we talk about in the book that pipes the sweet-smelling aroma of their extremely fattening cinnamon buns out into the hall, which acts a bit like the sirens in The Odyssey in terms of drawing people in.
We don't claim to have invented nudges. What we are suggesting is that in lots of domains, people aren't thinking about them as much as they should be. Marketers do think about them, but economists are often in charge of many aspects of public policy and they could think about them more.
In the book we distinguish between two types of creatures: humans and Econs. Humans are the people we interact with every day, and Econs are these strange creatures only found in economics textbooks that are unemotional, really smart, and never have self-control problems. If you want to design policies that will work, you want to design them for humans unless you live in a world of Econs, which, if you do, you have my sympathy.
A simple example of a nudge in the politics domain is the brilliant strategy of the Republicans to rename the inheritance tax the "death tax." Polls show that a majority of Americans are opposed to the death tax and in favor of an inheritance tax. Econs would know that the death tax and the inheritance tax are the same thing. But humans might be in favor of one and against the other. So if you want to repeal the inheritance tax, it's very smart to call it a death tax.
Q: How do defaults get set poorly in the first place?
The status quo is typically the default. And the choice architect typically doesn't think very carefully about this. Let me give you an example. In most companies now, there is a period of open enrollment, typically in November, where you get to rethink your benefit package. And at least at the University of Chicago, you're required to do this online. Of course, some people are going to forget. And the choice architect has to decide what to do with the people who forget. Now, typically there are two options to consider. I call them "same as last year" and "back to zero." For your healthcare plan, back to zero is pretty harsh. So most places go with "same as last year." But for flexible spending accounts, where you lose that money if you don't spend it, many places say, well, I don't want to presume that people want to put money away since they would lose it if they don't spend it, so we'll make that one "back to zero."
Now, at Chicago I had scheduled a meeting to talk to some of the top administrators at the university about changing the default on the supplemental savings account where you could put in money on top of the usual pension. The default was "back to zero," and I was suggesting changing it to "same as last year." It turned out that that meeting occurred on, just by coincidence, the last day that people had to log on and re-enroll, and none of the people at the meeting had yet done that. And that made my case extremely easy.
Q: In some cases, when there is an effort to switch the default, there's resistance. What can cause that?
Most of the time, you change the default and nobody notices. There can be some special circumstances in which people get their backs up. Maybe a good example of that is organ donation. In the U.S., the default is that unless you do something, your organs will not be made available if you die in an automobile accident or some such. Some of the European countries have the opposite default — what they call presumed consent, so you are presumed to give your consent unless you choose otherwise. There have been attempts to switch to that presumed consent in some countries, most recently in the UK, and that was met with resistance, in particular from the Muslim community. Illinois has adopted my favorite solution to this problem, and it's a neat little compromise which is called mandated choice. When you go to renew your driver's license and get a new picture, which you have to do periodically, they simply ask, do you want to be a donor or not? You must answer that question. They won't hand you your license until you say yes or no. And about half the people say yes. And that, as near as I can tell, has met with exactly zero resistance.
Q: And does that increase the number of donors?
Yes, because it is easy and people are reminded that right now is a good time to do this.
Q: Nudges can appear very small and straightforward, yet, played out on a big scale, they become significant.
I think that's right. Because the manipulations can be small, people think of them as minor. There is a nice example, I think, in some research that we mention in Nudge. If you tell people on their electricity bill how much energy they are using compared to their neighbors — and there's a company called Positive Energy that has been doing this — you can reduce energy utilization by something between 2% and 6%. Now that sounds like not a very big deal. But if you think about it, these messages cost nothing. You have to send out electricity bills anyway. If by changing the information you put on the bill, you reduce energy use by 2% to 6%, why in the world wouldn't we do it? If we can come up with 100 of those ideas, we can have a significant impact on the climate.
Q: People seem to have a hard time making decisions that don't give feedback for a long time. How do nudges address these long-term issues?
The solution to all such problems is to reward short-term behavior. So going back to the previous example, it turns out that if they put a little smiley face on the bill if you're using less power than your neighbors, that helps. In some places, they let owners of hybrids use the carpool lanes on the highway. Maybe you'll get somebody who doesn't really care about the environment but cares about getting somewhere on time to switch from a gasguzzler to a Prius. That's going to help. I don't care what motivations they have if we can get the right outcome.
Q: Who decides what the best outcome is?
Well, there's the question. Clearly it's not Richard Thaler and Cass Sunstein. One criticism I frequently hear is that we think we know what's best for people and this is elitist thinking. I think in many situations, it is pretty easy for the choice architect to have a good idea what choices people would really prefer. I think if we go back to default options in the open enrollment period, do we think that people who forget to enroll are going to want no healthcare? Probably not. So it's not that we think we know what's best for them. It's that we have a pretty good idea of what they would want the default to be if they were the choice architect. Sure, there may be some tough cases, but I think most of us would rather be healthier. We'd rather have our kids be well educated. We would rather not starve in retirement. We would rather not wake up some morning and have our mortgage doubled because of some term that was buried in the fine print. So I think we can make a lot of progress without much controversy. And there will be a few cases where it's harder, and that's what the politicians are elected to deal with.
Q: How do politicians sort out those hard cases amid conflicting interests?
It's difficult. I want to emphasize that we don't envision a larger role for government. Government has to make decisions. They have to nudge. So why not do it effectively and transparently? And if we don't like the way the government is doing it, then throw the bums out and elect somebody who will do a better job.
Q: Could you talk a bit about choice? Some research suggests too much choice can be problematic, but the libertarian aspect of nudging is all about making choice available as often as possible.
I think that can be a problem, and one possible solution is what we call structured choice. Let's go back to 401(k) plans. When they first came in, they only had about half a dozen options. And choices were pretty easy. But then a subset of employees would lobby for the fund that invests in Slavic tech funds, and so on and so forth, and now plans have 100 options. And the research shows that with more options, people are slower to join.
So what's the solution to that? One would be to reduce the number of choices, which we don't like. We prefer to start out with a good default fund and say to somebody that's joining the plan, okay, there's lots of choices here, but if you don't want to make a decision, here's a default investment strategy that some experts think might work for you. It might be what they call a target date fund, so the fund will change your asset allocation as you age. Would you like that, or would you like to make a more active decision? If you say you'd like to make a more active decision, then maybe the second stage would be a choice between aggressive or conservative target date funds. Would you like to pick one of those? No? Okay, here are all 100 funds. Go for it. Have fun. I think that that's the way to handle this. We wouldn't want the government to say there should only be three funds.
I think the nudge philosophy is pragmatic, and it says that in some sense we can get the best of all worlds. We can make it easy for neophytes to get something reasonable. At the same time we can make it possible for sophisticated investors — or those who wrongly think they're sophisticated — to make their own choices, and that's the essence of libertarian paternalism.
Q: Is there research behind this? Do we perform better if we have a sense of freedom and choice?
I'm not sure we can give a satisfactory answer to that, but I think most of us would prefer to live in such a society. I also think it's pragmatic in the sense that maybe I want lots of wine options because I know something about wine, but I want somebody else to help me pick a computer. We don't want a one-size-fits-all solution to any kind of problem.
Q: What can managers take from this work?
Managers are choice architects. They need to structure environments to improve behavior. It shouldn't be surprising that many of the traders who took down companies were doing what they were doing, because they were getting paid lots of money while they were doing well, and then could walk away when it blew up. There is a famous old paper in organizational behavior: "On the Folly of Rewarding A, While Hoping for B." If a manager understands behavioral economics, she will not be able to change her employees into Econs, but she will be able to mitigate some of the biases through clever choice architecture.
Editor's note: The interview with Richard Thaler published in the print edition of Q6 ends here. Additional comments on the development of the discipline of behavioral economics continue below.
Q: The field of behavioral economics has increased in prominence in recent years. What does that suggest about its ability to have an impact in the world?
I'm not sure one follows the other, actually. The Simpsons are very prominent, but don't affect the world. So I think, in some sense, we behavioral economists feel a little bit like the band that has been slogging away for 20 years and then gets discovered as an overnight sensation. There is nothing overnight about behavioral economics. We've been doing it for 30 years. It takes a while for things to catch on. In particular, it took the time to generate the second, third, and fourth generations of people doing the work. Yale is lucky enough to have one of the founders of the field in Bob Shiller as well as some of of the leading next generation scholars such as Nick Barberis. I would say that the status of the field, within the profession, has been growing for the last 15 or 20 years, but the rest of the world is catching up on that, including, gratifyingly, some policy makers, most notably President Obama and David Cameron, the leader of the Conservative Party in the UK.
Q: Do you think that the economic crisis has created a window where people are looking for alternatives, and maybe particularly policy makers are looking for other perspectives on economic theory?
Yes, I think so. But it's not just the financial crisis. In the last 30 years, we've had three gigantic bubbles, first in Japan, then the technology bubble here, and now the real estate bubble and the resulting financial crisis. The very fact that we have had these multi-trillion dollar bubbles and crashes does suggest that one aspect of the efficient market hypothesis, namely that asset prices are equal to their true, intrinsic value, is rightly being questioned.
Q: The efficient market hypothesis has held sway for decades. It was a guiding force behind a lot of deregulation and freeing of markets. Could behavioral economics move into that position as a dominant theory?
I wouldn't say that behavioral economics is ready to become a dominant theory, because it is still based on neo-classical economics. The way behavioral economics really has developed is to say, look, here is a list of ways in which human beings differ from Econs, and that leads to a list of situations in which the predictions of standard economic theory are suspect. And let's look into those. It is not a new, up-from-first-principles, grand theory of the world. And I don't think that's really possible. And the reason is that it's just too complicated. There are lots of ways that humans differ from Econs, and there won't be anything as simple and elegant as standard economic theory. The behavioral alternative is going to be a bit of a mess. But messy and right is better than tidy and wrong, don't you think?
Q: Are there criticisms of behavioral economic theory that you have taken to heart?
Well, I think that the criticism in the early 1980s that I most took to heart was "What happens in markets?" The field started by some of us looking at lab experiments run by psychologists, and the natural question economists ask is, "Okay, this seems to work for small stakes with no markets. What happens when you raise the stakes and allow professionals?" For me that was exactly the motivation for getting interested in behavioral finance. No one can say the stakes aren't high. The surprising thing is that finance has probably been the single most successful branch of behavioral economics over the last 20 years. I say surprising because economists would have expected it was the least likely place to work. Now, the frustrating thing is that I still hear people say, well, this is a field based on small-stakes laboratory experiments, and that's just a misinformed view.
Q: What are the most successful examples of where markets are impacted?
Well, you know, we've talked about them. There are two parts of the efficient market hypothesis. In the survey paper Nick Barberis and I wrote we called this "the price is right" and "the no free lunch." And "the price is right" is the one that I alluded to earlier, that says that asset prices are equal to intrinsic value. The "no free lunch" is that it's hard to beat the market. My take on the financial crisis is that the faith in the "no free lunch" principle should, if anything, be stronger, because strategies that looked like they were beating the market, we realize now, were just very risky. But the faith in the "price is right" aspect of it is weakened.
It's not clear what the implications of that are, since no one has yet developed a better way of allocating capital than capital markets. We tried planned economies and they failed. But that's not to say that there are no lessons for policy makers. For example, I suggested that one thing that agencies like Fannie and Freddie might do would be to raise the down payment requirements in regions of the country where the real estate market looks like it's a bit frothy as measured, say, by the price-to-rental ratio.
Interview conducted and edited by Ted O'Callahan.