Editor’s note: This article was initially published in The Daily Gazette, Swarthmore’s online, daily newspaper founded in Fall 1996. As of Fall 2018, the DG has merged with The Phoenix. See the about page to read more about the DG.
This is the second interview in the series Research Spotlight, in which I share conversations that I have with faculty regarding their research, their journey within their field, and their field in a broader context.
Professor Syon Bhanot is a behavioral and public economist. He primarily uses field experiments to study decision making around cooperation, environmental conservation, personal finance, and various development topics.
This article is the first part of a three part interview with Professor Bhanot.
10/13/2017
AIDAN REDDY: What is behavioral economics, and what has led to its growth in recent years?
SYON BHANOT: In economics we’ve got the standard model of decision making, centered around economic “rationality.” It’s a bit unfortunate that the term “rational” has been used, because people kind of interpret that to mean their decisions “make sense.” That’s not quite what it means. Economic rationality has a really strict set of definitions. If we can observe people systematically deviating from those, then we can improve the models that economists use to study decision-making by incorporating some small adjustments.
For example, there are lots of decisions you make in your life that affect you now, but also affect you later. A common example is going to the gym, or what you choose to eat. Those are decisions that aren’t just about the now; they’re also about “future you.” So, in the rational model, generally we assume that people weigh the costs and benefits of their decisions and make decisions where the benefits outweigh the costs, and not just these costs and benefits now, but in the grand scheme of their lives.
However, behavioral economists have identified a systematic pattern of present bias in humans — that is, we tend to overweigh the present in our decision making, and we do that systematically. That’s why people very rarely under-eat or over-go to the gym; we generally do the opposite. People are overeating now because it’s fun NOW and the pain comes later, and people are not going to the gym enough now because it’s not fun in the present and the payoffs come later. We tend to have a hard time doing things that are, in our overall best interest, if they have a cost now. Behavioral economics is really about improving the way that we think about the decision making process and building models that are more descriptive of how people actually behave.
I think most people think that one “should” be rational — and that we “should” make rational decisions. The use of the term should is a really important one; it implies a normative judgement about what’s best. Behavioral economists try to not take a normative view of things. That is, we are less interested in studying what people should do or what’s the logical thing to do, but rather, we want to understand what people actually do. It’s a more descriptive model-based approach. The idea here is, “I want to observe people’s behavior. Then I want to build models that accurately reflect people’s observed behavior.”
So, I think that’s kind of what behavioral economics is trying to get at. There’s a lot of pieces of that. This is a very broad summary of a very big and diverse subfield of economics.
A lot of people get behavioral economics confused with things like Freakonomics. Freakonomics is about people responding to incentives, and it fits in the standard model pretty well. But, just because it’s cool and because it is about behavior, people think it must be behavioral economics, but it’s not the same thing. (In fact, the term “behavioral economics” is a bit of a misnomer – a lot of economics as a field is about behavior!)
Behavioral economics is also not the idea that people are dumb, or that people make crazy decisions all of the time. It’s about systematic patterns, otherwise they’re not interesting. If, for example, people are, on average right about inflation, and think that it’s gonna be 2%, and some people think its 3% and others think it’s 1% but on average they’re right, that’s actually not inconsistent with the rational model. But, what is a problem is if there is some systematic deviation, if people are systematically using the nominal interest rate to make financial decisions instead of the real interest rate when a model assumes they use the latter, for example. So for example, people are under-going to the gym consistently. It’s not that some people go too little and some people go too much but on average people go the right amount. I would disagree with that notion and argue that most people don’t go as much as they want to. Most people don’t save enough for retirement. These are systematic patterns, systematic deviations from the rational model.
So it’s not just “fun economics”. It’s about systematic deviation from the rational model of decision making.
Why do I think it’s grown? There’s a lot of psychology behind this, how people’s psychological processes work. There are some sociological examples too, thinking about how the people around you might affect your preferences, and things like that. So, there’s a lot of pieces from other fields. In general, academia is a very siloed place. You go to your PhD program and you get your PhD in that field. You learn basically nothing about other fields. Then, you go on and try to convince people in your little field that you’re great. So, there’s never really an incentive to cross over. That’s been a problem in academia for years, and I think it’s increasingly becoming the case that a lot of global problems, interpersonal challenges that people have, involve inputs from multiple sides. If you look at what’s going on politically — really at any given time in history, but now as much as any other — to understand what’s happening, it’s useful to think about things from a multidisciplinary approach. “Solving” climate change will require a multidisciplinary approach. I think people are starting to realize that the old model of academia where everything is siloed just can’t address modern problems. So, we do need to have the “fusiony”-type subfields. I think behavioral economics partly came out of that. The Nobel Prize in economics was given out in 2002 to Danny Kahneman. He’s a psychologist. He got the Nobel Prize in economics for essentially coming up with some of the building blocks of behavioral economics in his work. The subfield is fundamentally multidisciplinary. I think people like it because it kind of sounds like something you can apply to modern problems. How do you help people make better decisions for themselves? Build good habits? How do we address big challenges at the global level that involve cooperation and prosociality? If we use a rational model, it’s really hard to fully address those things. If everyone’s just a selfish, self-interested person, we’re never going to solve some of these issues. But, if we have an approach where we think people aren’t always like that, and that they systematically deviate from that, how can we motivate changes in behavior that are in line with cooperative prosociality? You need to have a multidisciplinary approach. You can’t be rigid to your one model if it doesn’t describe how people actually behave in the real world.
REDDY: Generally, how would you describe your research?
BHANOT: Broadly speaking, I do what are called field experiments. In behavioral economics, there is some theory, some people who are writing formal mathematical models to address some these questions and bring into the formal mathematical models that already exist in economics some of the tweaks that behavioral economists or psychologists might suggest that we need. For example, in the standard economics model, we’ve got this discounting idea: you discount things that happen in the future in some systematic way. There’s a behavioral economics counter, a slight tweak to that model to incorporate present bias.
There are also researchers working on the empirical/data side, which is collecting data from the real world and running experiments —trying to use data to prove the existence of some of these biases, and the importance to markets of these biases. So, that’s a broad subset. Within that group, there are people who are just gathering big data sets from the real world, there are people who are running lab experiments, kind of like in psychology, bringing students or people in the lab to study how they’re thinking about decisions. I do a little bit of that, but not very much. My favorite approach is the field experiment idea. What that is (it sounds more evil than it is, I promise) is going out in the real world and conducting randomized experiments not with students in the lab, but with real people living their lives, who may not even be aware that the study is being conducted on them.
I work with a lot of different public policy partners. I co-founded the Philadelphia Behavioral Science Initiative (www.phillybsi.org) with a professor at Penn, a graduate student at Princeton, an alum of Swarthmore, and the policy team in the Philadelphia mayor’s office, which is integrating randomized experiments into public policy in the City of Philadelphia. So we have a bunch of experiments that we’ve started with the City where they’re sending letters to people’s homes, and we use behavioral science and behavioral economics to manipulate the content of the letters to see how it affects choices that people make. So I have a number of projects in that domain. I also have a number of projects in the developing world, with the World Bank, the Inter-American Development Bank, the US Agency for International Development, and some other partners working in the developing world. With these organizations, I am using field experiments to study how people who are very poor make decisions, and also how we can use different types of behavioral interventions to help these individuals.
So, my big picture projects are generally with partners in the real world, such as policy organizations. We’ll have a huge sample group, thousands of people who are being randomly assigned into various conditions. Just like a medical doctor would run an experiment on a cancer drug, we’re running experiments on behaviorally-motivated interventions, trying to see how they affect people’s choices. We’re trying to get real-world evidence to see if these concepts that people have theorized about and brought students into the lab to test out apply in the real world. If they don’t, then maybe we need to go back to the drawing board a bit with the theory.
REDDY: What’s a specific project you are working on right now, and in what ways have they involved working with people in other fields?
BHANOT: We did one study with the City of Philadelphia school district, which is just about finished. The City has an annual survey that they send out to parents and teachers, and to students as well, about what’s going on with the school. They ask for feedback about what’s working and what’s not, issues people are having, and so on. It’s a really important aspect of their policymaking process, figuring out what’s actually happening in the schools, because they don’t get to see everything every day. But, they get really low response rates relative to other cities. So, they had a conversation with us and said, “Look, we try to communicate with these different groups. We try to get them engaged to take the survey, but they won’t. Can we use behavioral science to try to motivate people to do this more?” If you get one of these surveys, you’re kind of thinking, “I don’t have time for this now. What does it matter if I do it anyway? I’m just one person.” But, really, if everyone thought that way, then no one would vote, for example. It’s hard to get people to collectively act sometimes. So we are trying to use some behavioral tools to motivate people to complete the surveys, using manipulations to the content in emails and letters sent to parents and teachers. So basically every parent who had a kid in the Philadelphia School District was part of our study. We had different versions of the letters sent to their homes, motivating them to log in and take the survey. So, for the parents, we had some that shared information about how much worse Philadelphia was doing than other cities, basically looking to see if the social pressure aspect of looking what the rates are in other cities might motivate them to try to improve, or just to see what they would do. Maybe it doesn’t motivate them to improve, I don’t know. We have theories on both sides. So this is going to get us some real world evidence on which theory seems to bear out. We also had framing around collective action. So yeah it feels like it’s just you, this one lone voice, but if everyone does it, then we can actually see real change in the way the schools are run. With teachers we did a kind of different intervention, where we gave them a financial incentive to complete the survey because that was allowable for the school district, but we varied whether the incentive was for themselves or for their students. So, it was like you either get money for yourself to spend (at Barnes & Noble), or money to buy stuff for your students (at Office Depot). We also varied whether or not they received an email asking them to consider their obligation to the students, and that part of their role is to be the guardians of their students’ futures. The question is: by making them think more about that identity and role, does that make them more or less responsive to financial incentives, and, in particular, more or less responsive when financial incentives are for the students?
Working with people in other fields is a big part of what behavioral economists are doing. You can’t just be surrounded by economists and do this kind of work; you have to be surrounded by people with other academic backgrounds, and also just by “normal” people in the world (academics are, I would say, “abnormal”). Policy makers are a big point of contact for me because they know what’s going on on the ground. They see things affecting people’s choices all of the time. It’s a lot of iteration, a lot of collaboration, and a lot of conference calls. It’s a lot of coordinating big projects, rather than sitting in your office and theorizing, which has merits, for sure, but it’s just not the kind of work I’m doing. It’s a lot more getting together in groups and trying to solve problems. All of my training is in public policy. Actually, all of my higher ed degrees are in public policy! So that’s what I’m built for, I guess: doing policy work. I’m really only interested in academia to the extent that I feel like it can contribute to public policy. Luckily, I can do that in my job.
In part 2, Professor Bhanot gives his take on how behavioral economics can influence policy and people’s everyday lives.
Featured image courtesy of Carnegie Mellon University.