Shane Shepherd: Standard economic theory views people as rational decision makers, but we know that reality is far different. The new field of behavioral economics focuses on exploring how people actually do behave and the ramifications on consumer choice and financial markets. Can you explain what behavioral economics is and how it works?
David Laibson: Sure. Behavioral economics is part of traditional economics. We are improving on the traditional way of thinking about economic behavior. The traditional, classical economic view is that people are perfectly rational and that they’re also perfectly able to follow through on their good intentions. Behavioral economics psychologizes that and acknowledges that sometimes people are imperfectly rational, and that also, in many cases, people have good intentions but terrible follow-through. The additional psychological realism is our effort to make economics more realistic and more useful.
Shane: A lot of your research focuses on the idea of self-control. That is, in deciding about taking an action now which may have implications both in the immediate present as well as down the road, we’re not good at making that trade-off all the time in the way a rational economic agent might. When I think about self-control, I recall Walter Mischel’s famous “marshmallow test.”
David: Yes, the marshmallow test works the following way. The experimenter tells a five-year-old kid: “Here’s one treat. If you can wait to eat it until I come back, you can have two treats, and if you can’t wait, ring this bell, and I’ll come back immediately, and you can have the one treat.” Many kids can’t wait at all, or can only wait for 15 or 20 seconds. Other kids wait the full 15 minutes until the experimenter coming back.
The kids often want to wait, but have an awfully hard time doing it, so they deploy some strategies that help them wait. Most importantly, they try to divert their attention from the extremely tempting marshmallow by thinking of it as something other than a marshmallow or by playing a game in their heads that takes their focus away from the marshmallow or they literally cover it up or look away. These are the strategies that enable them to wait the 15 minutes and get the double return.
Shane: As we grow up, we employ a lot of the same strategies. Your research documents a framework in which we can think about ways to help with self-control. On the one hand, the approach can be situational or cognitive, and on the other, be self-employed or other-employed. Can you walk us through the framework and describe the different metrics?
David: I think one of the interesting failings of human civilization is the principle that, somehow, we’re going to make better choices by pulling ourselves up by our bootstraps and just deploying more willpower. In the paper I wrote with Angela Duckworth and Katherine Milkman,1 we find that, no, that doesn’t work so well. It turns out that other strategies are better at helping us improve our ability to self-regulate, and those strategies can be put into a space that cuts along two dimensions.
One dimension is whether we’re deploying the strategy or if society or others are deploying it; so that’s what we refer to as self-deployed or other-deployed. The other dimension is whether the strategy is in our own heads or whether the strategy is situational. So, let me give you a few examples.
If we hear “Everyone reuses their hotel towels,” that’s a way of conveying the message that everyone is doing this to try to save water. That would be a cognitive in-your-head strategy deployed by others.
An example of a self-deployed cognitive strategy would be a self-generated deadline. Say, you are trying to get some work done and you tell yourself that if you finish by 10:00pm, then you can relax and watch something on Netflix. This is an internal plan, an internal action goal.
Then we have the situational mechanisms. Maybe the hotel asks guests to decide whether or not to reuse their towels. There is no default; it requires an active choice. This is an example of an other-deployed strategy that is situational.
A situational self-deployed strategy would be a commitment to an external party. Suppose you are trying to complete some work by 10:00pm. A stronger strategy than just telling yourself to do it by 10:00pm is to actually promise someone else, and make that promise ironclad, to deliver the work to them by that time. This type of commitment, using the external party to create credibility for your promise, can be a very powerful way of getting yourself to do what you are trying to do.
We discuss these different strategies in the paper and emphasize that none of them involve having more willpower. Rather, they all involve taking the limited willpower that we have and enabling it to go further.
Shane: Let’s relate this to saving for retirement. A lot of literature shows that many people are nowhere near saving the amount they need in order to provide for a safe, secure retirement. What are some of the strategies that have helped people get closer to that goal?
David: The one strategy that’s not necessarily so obvious, but which is very much part of the ecosystem of savings, is compulsory savings. This is happening in the Social Security System. We essentially mandate that every person in the United States who gets a paycheck puts some of it away for retirement. That’s a paternalistic other-deployed situational strategy. But we’re also using less-strict strategies, such as defaults, in which we default people into investing in a 401(k). If they don’t like it, they can opt out. That would be a second pillar on top of the mandatory Social Security System saving.
We also use a lot of cognitive strategies in which we try to talk people into saving. We tell them how important it is. We tell them how many other people are saving. So we’re deploying all of these cognitive strategies to help people increase their savings. It turns out that these cognitive strategies work a little bit, but they don’t really have a lot of bite.
The strategies that are working best tend to be the situational strategies, either the hard paternalism, as in Social Security, or the softer paternalism in the form of a default, or an active choice, when people are put into a savings system and must opt out. These types of strategies tend to have more bite than simply trying to talk people into saving.
Shane: So you’re saying, even after we’re all grown up, it’s still hard to get people to save their marshmallows for later?
David: Exactly. Though I will say that as we get older and older, self-control problems seem to loom smaller and smaller. So when you’re 5 years old, it’s hopeless, but when you’re 15, you’re on your way, and when you’re 25, you’re making real progress. Then it just slowly gets better after that, all the way to the end.
Shane: One of the other areas that relates to issues of self-control has to do with market downturns. When the market enters a downturn, investors tend to have this sense of panic. They tend to want to pull out and sell to avoid the negative feelings they have about being invested in the market during the downturn, even though they realize perhaps it’s not such a good idea to sell. Could you explore some strategies investors could develop ahead of time to plan for these inevitable downturns so they can respond more appropriately and try to gain self-control over those impulses?
David: Individual investors, as you point out, have a very hard time timing the market. They understand that market timing is difficult, they understand that they tend to do it terribly, and yet, when the moment comes, they tend to panic. So the message is, resist that impulse, sit on your hands. During your cool moments develop the strategy of being passive and not trying to time the market. Then when the panic occurs, and when everyone else is rushing for the exits, that’s the moment to remind yourself of the cognitive strategy you’ve developed, the plan you made during calm moments, to not overreact when the market is melting down.
Shane: Great, thank you. What we’ve learned today is that behavioral economics is really about studying the ways people actually behave, and that we can get useful information from those studies to help people make better decisions for the long run by using self-control. Strategies for enhancing self-control can be either situational, which help us change our environment to have the appearance of more self-control or to make decisions easier for us, or cognitive, which are more of the educational variety. It seems like the situational ones tend to have more bite, although planning and cognitive strategies can also be useful.
David: That’s exactly right. Changing the environment has proven to be much more effective than trying to talk people into changing their behavior.
1Angela Duckworth, Katherine Milkman, and David Laibson. Beyond Willpower: Strategies for Reducing Failures of Self-Control"
Psychological Science in the Public Interest, vol. 19, no. 3 (2018):102–129.