The Psychological Cost of Nudging: Julian Jamison

Art by versusthemachines
A conversation with Julian Jamison

Also available on all major distribution platforms, including: Apple PodcastsSpotify, and Google

“So [behavioral science] can be quick, but I wouldn’t push that. To me I wouldn’t want to rest my hat [on the fact that] behavioral science is faster or quicker. I don’t actually think that’s the main advantage. I think the main advantage is sometimes it tells us something that we didn’t know before, and it’s the right way to go about it.”

In today’s episode, we are joined by Julian Jamison, a professor of Economics at the University of Exeter and an affiliate at the Innovations for Poverty Action (IPA) and Jameel Abdul Latif Poverty Action Lab (J-PAL). Prior to this, Julian spent 9 years in the public sector working for the United States government as Section Chief of the Decision-making and Behavioral Studies group and as a Behavioral Economist for the Global INsights Initiative at the World Bank (now known as the Mind, Behavior and Development Unit, or eMBeD). 

Julian holds a B.S and M.S in Mathematics from the California Institute of Technology and a Ph.D. in Game Theory from the Massachusetts Institute of Technology. His academic work focuses on the interaction between individual preferences, decisions, and well-being, and on institutional policies, including explicit welfare tradeoffs. He uses a wide range of methodological approaches, including mathematical theory, lab and field experiments, formal rhetoric, surveys, and large administrative data analytics. Julian’s work has been featured by The New York Times, Foreign Affairs, The Wall Street Journal, National Public Radio, Forbes, Foreign Policy, The Washington Post, Bloomberg, and The Economist

In this episode, we discuss:

  • Julian’s experience working as a behavioral scientist at the World Bank eMBeD unit. 
  • Working in academia and working in industry: pros, cons and lessons.
  • The need to distinguish between behavioral obstacles and behaviorally-informed interventions
  • How the fear of ambiguity makes behavioral science more challenging to adopt within organizations.
  • Why measurement tools are critical in any study.
  • Why the behavioral science we of our decade is different from what has been studied before
  • Julian’s hope for the future of behavioral science: Integrated into our approaches in a way that is complementary rather than a separate field
  • The need for specialization in behavioral science 

Key Quotes

Prices matters, even on non-conventional retail goods
“[A] project we’re doing in Cameroon on the uptake of LARCs (long-acting reversible contraceptives) among young women. I like it because it has a behavioral element. We’re thinking about stigma and biases on both the provider-side and the client-side. But it [also] has an almost neo-classical element where we’re thinking about prices and the effect of prices on this somewhat unusual good. It is not a standard retail good, but we still think prices matter.”

Behavioral science, applied to everything  
“So sometimes whatever your underlying interests are in terms of topic areas, whether it’s development economics or something a bit more stem, health economics, trades [or even] something outside economics, you can almost always bring a behavioral slant to it and that complements rather than substitutes what’s already going on.”

The psychological cost of nudging   
“I think sometimes the people, especially the people doing nudges and pushing for nudges, want to claim that there’s almost zero cost to it because sometimes there’s a zero financial cost. Or its choice architecture. There’s something that’s always there. They’re right, I’m sympathetic to that kind of view. But I think almost always by definition, there is some, at least, psychological cost to doing something else because that’s why it works.

So in that sense I think we should acknowledge, even if we don’t agree in the end, but acknowledge the starting point of some of the criticisms that there is a cost, however small it is. So I’d say it’s something where we’re altering things to make it a little bit harder to change people’s behavior, to change the outcomes”

Distinguishing behavioral barriers and behaviorally-informed interventions
“So you can have a behavioral obstacle and an intervention or policy that’s not behavioral at all that’s regulation or price based or something else. Conversely, you can have issues that you want to deal with something when you’re trying to guide people in a particular direction that are not particularly behavioral. But then have a policy intervention that’s behaviorally informed, or you can have both. But I think distinguishing those two where they’re often confounded now will be another thing that I hope to see more of in the future.”

Behavioral science’s value is in its perspective, not in its speed
“So [behavioral science] can be quick, but I wouldn’t push that. To me I wouldn’t want to rest my hat [on the fact that] behavioral science is faster or quicker. I don’t actually think that’s the main advantage. I think the main advantage is sometimes it tells us something that we didn’t know before, and it’s the right way to go about it.”

Behavioral science might not always be a separate field
“So [behavioral science] is an unusual field. Probably everybody feels that way about their own area. It wasn’t a well defined a field from the role that I was in 20 years ago. Sometimes I wonder if in the future it might also not be a field, but in a very different way. In the future I kind of hope everybody’s doing [behavioral science] to some extent. But it may not exist as a separate field that way. It may be a part of a lot of other things where it does complement them in a lot of ways. 

I think in 5 or 10 years it’s pretty safe to say there will still be something called behavioral science. In 20 or 30 or 40 years, I’m actually not sure. Again not because I think it’s going to disappear, but because it may percolate out. I think that would be a success. I think that would mean we’ve done our job.”

Why specialization is important in applied behavioral science
“But rather than thinking of just adding things to that, maybe go pick something that’s a little bit more left-field and learn that really well. Or maybe that’s already your background and you can bring that. That helps you to stand out a little bit. You don’t have to be doing everything, you’re not going to be expected to that, that doesn’t make sense. But bring something that maybe few other people have and that’s going to be useful skills. 

From revealed preferences to a more holistic model of preference
“I want to be getting into a little bit more normative or welfare analysis— thinking about “what constitutes a mistake?”. What if we depart from the economist view of revealed preference? Behavioral science certainly is extremely suggestive – convincing me that people do make mistakes. That is to say they make choices that are not in line with their own long-term best interest in maximizing their own utility function. So their reveal preference doesn’t always work in that sense.

What do we replace it with and what kind of model? [Should we] take a mix of evolutionary psychology, behavioral science and economic theory and a bit of philosophy thrown in? Survey data on regret and on experiences—how can we sort of pull all these things together to think of a model of mistakes, a model of conflicting preferences where I conflict with my future self. Which preferences do we use for welfare analysis, for policy analysis?

Articles, books and podcasts discussed in the episode:

Julian’s work

Other mentions

Transcript

Jakob: Thank you so much for joining us today Julian, it’s really great to have you. So we would like to speak with you about your take on the field of behavioral science and what trends you foresee in the coming future.

Jakob: But before getting into that, I think many of our listeners would be curious to first learn how you got into behavioral science as a field, what it is that you’ve done at the World Bank before, and what are some of the more exciting projects you currently work on. Can you walk our listeners through some of them?

Julian: Sure, and thanks for having me, it’s great to be here. I guess, I mean I started in behavioral science when I was going to say it wasn’t really a field. That’s not true, because it’s been a field for a long time, for decades now. But when it wasn’t as well developed and certainly for an economist like myself it didn’t really exist as something that one did. My aunt is a psychologist, so I had some exposure to that early on, and as an undergraduate, I did some lab experiments on market behavior.

Julian: But I think one can’t help seeing individuals behave and make choices and make decisions, real humans in the real world without getting interested in something like behavioral science. Because they do all kinds of interesting, and a little bit quirky things that you want to find out more about. So my interest was sparked at that point.

Julian: Then did some game theory, so thinking a lot about individual behavior from different perspectives over time, that was my Ph.D. and game theory and then worked various places after that. But always with this focus on what are people doing and sometimes the policy implications. At the World Bank as you said. So I was there, I was in the behavioral science unit, which was originally GINI, the global insights initiative and then now eMBeD, mind behavior and development unit. We were almost like internal, and I’m still working with them a bit as an outside consultant, and I work on projects there.

Julian: I was going to say almost like an internal consulting team, where we would work with operational units at the bank on projects, often with governments, although not always and bring a behavioral science perspective. Try to bring some behavioral insights to the table at whatever level they needed and try to learn from it ourselves. Try to have a broader takeaway where we could turn it around and make it a bit of a public good for the rest of the bank and more broadly.

Jakob: Fantastic.

Julian: Oh and you asked about current projects, so I’ll just mention one. Obviously happy to talk about more. This is a bank project we’re doing in Cameroon on the take up of LARCs, long-acting reversible contraceptives among young women. So I like it because it has very much a behavioral element. We’re thinking about stigma and biases on both the provider side, and the client side. But also it has a more almost neo-classical element where we’re thinking about prices and what’s the effect of prices and take on this somewhat unusual good. It is not a standard retail good obviously. But we still think prices matter.

Julian: Even then there’s a behavior element because there’s this idea that a zero price might be very different from something else. Marketers have thought this for a while and there’s some recent work on bed nets and subsidies and the development literature that we want to test that again in the context of this unusual but important good. So yeah, that’s a fun one that brings a lot of different aspects together.

Jakob: Fantastic. Thank you for that Julian. So you now teach economics at the University of, and you have to help me out here again, Exeter?

Julian: Exeter, yes.

Jakob: So you know I’m curious, what motivated you to take on this role and how does it differ from your previous role at the World Bank. Then maybe also what aspects of economics do you actually teach?

Julian: Right, so I started my career, my professional career as an academic and then left and worked in the US government at the World Bank. Doing research, but in a sort of policy organization setting. I really enjoyed that, I’m really glad that I did that as part of my career, and I wish in some ways more academics would consider something like that. It’s extremely fulfilling, and you learn a lot.

Julian: But I kind of always had the sense that most likely I would return to academia. I feel like that is a natural home for me. I have just a little bit more freedom on what to work on. Not that anybody was restricting me or telling me what to do. But you’re trying to match the goals of the organization wherever you are. Mostly just internal, you want to be helpful, you want to help out, you want to make it work, and that’s right. That’s how it should be, that’s what I was doing.

Julian: But to have a little bit of freedom to do things that are either slightly more controversial, or a little bit more blue sky or have a pay off that’s 10 years away instead of 18 months away. A little bit more philosophical in my case was my goal. We can get into that a bit later if you want on the future projects. But it’s felt good in that way. Partly I’m taking advantage a little bit best of both worlds. Like I said I am still working with the bank, and I’m still excited about those projects. I like the team there, so I’m getting to continue that, seeing how that goes long term.

Julian: I’m teaching development economics for undergraduates, third year, final year undergraduates here. I’m bringing in of course some behavioral science elements, but a broader perspective on the development literature.

Jakob: Thanks Julian. On that I’d love to ask you a brief follow-up question. Actually I’m just curious in terms of your student group, your student body. How high is the interest in the behavioral science aspects of development economics?

Julian: I think it’s quite high, we’ll see, I just started teaching last week. So I can’t say specifically about this group. But from my experience both talking to some other students here last term, and students of other universities, behavioral science is, as you know this is why we’re doing this and why it’s all set out, it’s become extremely popular. Not with everyone, there’s certainly push back as well. I’m sympathetic to a little bit of that, although obviously I’m a proponent of the field.

Julian: So I think people are excited about it, I think it’s a fun area to get into, and it’s one that you can take a lot of different ways. So sometimes whatever your underlying interests are in terms of topic areas, whether it’s development economics or something a bit more stem, sort of the health economics or even trades. Something else, something outside economics, you can almost always bring a behavioral slant to it and that compliments rather than substitutes what’s already going on. I think for the most part people see that.

Jakob: A segue way into our next question. As you said, interest in applied behavioral science seems to be continuously growing across different counties and sectors. But with that views about what nudging irrationality behavioral economics or behavioral science are have also shifted. So Julian if I was going to ask you, to the average person, what do you think nudging means in 2019. Also, how do you think this is likely to change in the coming years?

Julian: Yeah, good question and I guess I would start by saying and I know you realize this, that nudging and behavioral science are … to me nudging is a subset of behavioral science, but not the whole thing. But anyway they’re intersecting and overlapping, but not identical. I think to answer your question, what does nudging mean to the average person, I would say it’s something that’s a small change, but that’s done with a conscious explicit purpose to alter outcomes. To predictively on average and expectation, obviously not for everybody, alter outcomes.

Julian: So I distinguish that a little bit because I think sometimes the people, especially the people doing nudges and pushing for nudges, want to claim that there’s almost zero cost to it because sometimes there’s a zero financial cost. Or its choice architecture. There’s something that’s always there. They’re right, I’m sympathetic to that kind of view. But I think almost always by definition, there is some, at least, psychological cost to doing something else because that’s why it works.

Julian: So in that sense I think we should acknowledge, even if we don’t agree in the end, but acknowledge the starting point of some of the criticisms that there is a cost, however small it is. So I’d say it’s something where we’re altering things to make it a little bit harder to change people’s behavior, to change the outcomes, but it’s small. So to me, the important part is that it’s small rather than that it leaves all options open or something like that. Although I get the idea of what people are trying to say.

Jakob: Okay, so just as a follow up, the second part of the question. So how do you think this aspect that you just described is likely to change or not to change in the coming years? We’re trying to get a little bit of the trends in the field from your perspective.

Julian: Good question, a bit hard to predict. I actually think in that sense I think that the average person … I liked your question about what does the average person think. I think in some ways they’re a little bit ahead of the field of practitioners. Because I think coming around to the sense of any small change, whether its setting the default option or it’s a small cost or whether it’s something else that that’s how to think of it.

Julian: What’s important is this just shifting things slightly and potentially having a big change in outcomes to me is the most, it’s not right in the sense of morally right. But it’s right in the sense of that’s the consistent view that I think is going to be most productive in thinking about the field. So that’s why I say I think there’s already a little bit of a step in that direction as to where it goes even further in the future. I guess one thing I’d like to see is that it gets applied a little bit of a distinction between behavioral issues. I don’t want to say problems, but choices that are driven due to behavioral insights. Or behavioral factors versus interventions that are behaviorally guided or behaviorally motivated. I think you can have either one without the other or both of them.

Julian: So you can have a behavioral obstacle and an intervention or policy that’s not behavioral at all that’s regulation or price based or something else. Conversely, you can have issues that you want to deal with something when you’re trying to guide people in a particular direction that are not particularly behavioral. But then have a policy intervention that’s behaviorally informed, or you can have both. But I think distinguishing those two where they’re often confounded now will be another thing that I hope to see more of in the future.

Jakob: Interesting so in a way it’s not necessarily going to have to be an either or if I hear you correctly. But an end, and then kind of compare-

Julian: And in particular I think the second one gets a little bit less attention. So the one where it’s not necessarily a behavioral obstacle, it might be something that’s just a traditional issue that we’ve dealt with for a long time. But where the intervention is behaviorally informed, and I think we can, people are doing that as well. But I think we can push a little bit hard on that front, and it’ll help to explicitly view it that way to start coming up with ideas. I’m going a little bit further.

Jakob: Makes sense, got it. Okay, so now I’d like to shift gears to talk about the application of behavioral science, especially in policy making. So obviously the World Bank group has a strong reputation because it applies rigorous and academic approach to policy making. Which can be something that can be very beneficial for clients, counties and governments that request World Bank services.

Julian: But it sometimes also comes with challenges. So at times we have heard that behavioral science is embraced by project leaders at the bank and also in government counterparts because it provides fresh, new and sometimes a bit quicker perspectives than some of the classic economic models have done in the past. But we also hear that units don’t necessarily have the luxury of time and budgets to conduct always complex randomized control trials. Yet they’re still interested enough to apply behavioral science to their projects in a sound manner.

Julian: So what do you think are the biggest challenges for an organization looking to apply behavioral science in an empirical manner, and how can they be best tackled?

Julian: Again a good question. I’ll start by saying that although I’ve done a lot of RCTs, a lot of randomized trials, I think it’s great when you can do them, you learn a lot from it. But I would be the last person to say you always need to do that or it’s always worth it. Sometimes you either don’t have the time or sometimes you don’t need it. Sometimes you can make a good guess. Otherwise, I think my overall response would be, and this is really general theme maybe for the conversation, that I don’t think behavioral science is that different from a lot of other tools in the toolkit.

Julian: It’s maybe a bit new, but it’s not of a different type than the others. So the way that it applies here is that I think it can be done quickly and that’s fine sometimes. It can be done more extensively if you don’t have as good a sense of the answer if you have the time to do that, if you’re focused more on the learning side maybe than we have to get a policy in place side. So I don’t think it’s inherently quicker, although you’re quite right to say that some people want to see that about it. It can be, some of the nudges can be small, they can be easy to undertake, but then price changes can sometimes be small as well, some of the traditional ones.

Julian: So it can be quick, but I wouldn’t push that. To me I wouldn’t want to rest my hat on the reputation on pushing behavioral science because it’s faster or quicker. I don’t actually think that’s the main advantage. I think the main advantage is sometimes it tells us something that we didn’t know before, and it’s the right way to go about it.

Julian: So I’m sympathetic to the view that sometimes they don’t always have time. But I don’t think that’s an argument either against or in favor of behavioral science. I think it’s just that it’s a different way of looking at things. So to that end, what’s the biggest impediment is sometimes that people don’t have expertise. Sometimes that it introduces a bit of uncertainty. To be frank, if you have a way of doing things, even if you know it’s not always right, it gives you an answer, you’re used to it. You don’t have to go to your boss and say well, on the one hand it’s this, on the other hand it’s that. We don’t quite know, and sort of it says like here’s the answer, here’s the prediction, here’s what our model tells us to do, so let’s go and do it.

Julian: If you bring in something that’s just a different way of looking at things, something that’s a little bit more context dependent, that says a little bit more. People are fickle sometimes, then I think just institutionally that’s sometimes a harder sell even though it shouldn’t be. If that’s how the world is, that’s how the world is. It’s complex, it’s complicated, we’re going to get it wrong sometimes and that’s okay, that’s how we learn, that’s how we improve for next time. But it can be a harder sell, so just having a different approach is not always the easiest, but I hope people take the chance a bit more.

Jakob: Really that makes a lot of sense. I hope it’s fair to say that you belong to a group of as I would call groundbreaking researchers that have embraced the topic or have done research in applied behavioral science. Can you share with our listeners how you typically choose teams you are interested in researching about and how you link these to behavioral sciences if you do that, and what tools you use for your research.

Jakob: I think a lot of people are interested in what you do to distinguish good research from bad research, especially in behavioral science. Then what, for the lack of a better word, tricks do you use to translate sometimes really complex academic knowledge that’s always easy to understand for the average person to apply the work without losing any of its depth and rigor?

Julian: Right, so yeah choosing topics, to be honest some of it is as always, it’s opportunities that come up. It’s people you want to work with and find interesting and have conversations with and spend time with. I’ve always found lots of different things that interest me and I suspect that’s true for a lot of the listeners here and a lot of other people. Although I also have met people surprisingly to me who have trouble coming up with research topics or want to know what to work on next and to me it’s usually the opposite problem.

Julian: So to some extent because there’re a lot of interesting things out there, I do try to pick a little bit where do I think I might have some impact. I appreciate you saying, I don’t know how true it is. But it may be somewhat true that starting early on had the good fortune to be thinking about these kinds of things quite a while ago. Hopefully does have some influence and then working at the World Bank where you’re in close contact with governments.

Julian: You feel like you’re making a difference, there’s lots of different ways to do that. But because there are so many interesting questions to me, finding ones where I think it really is going to matter and it’s going to make a difference to people. Especially to maybe some of the more vulnerable people who need help or can’t always speak for themselves, I think that’s a good place to start if you have the opportunity to do it.

Julian: In terms of tools, I haven’t often used field experiments and randomized trials. I’ve used surveys a lot and preference solicitation measures on surveys. I think about methodology. I like to deal into the nitty gritty of measurement issues, which isn’t always the most sexy area. People want to maybe think about the substance and I understand why that’s the case. But I think the measurement area matters and obviously the methods matter both in terms of being rigorous as you said and I can say a couple words about good and bad research. But I think also just sort of exactly how you measure outcomes, for instance, can make a big difference to whether you see an effect or not, to see what kind of effect you see.

Julian: Sometimes people wish that weren’t the case I suspect. So they don’t spend a lot of time always thinking about the measurement. ‘Cause they can take time to get it right or acknowledge like I said the uncertainty. Sometimes just acknowledge if you measure it this way, this is what you get when you measure it a different way something else happens. We’re not really sure which one is right or maybe which one is right depends on exactly what question you’re asking. So that’s a little bit harder to come across as your conclusion, but I think that’s often the case.

Julian: I think that’s how we learn, I think that’s in the end how we do the most good. So good research and bad research I think being explicit about assumptions, being explicit about what your goals are and what your hypotheses are. So for instance, set back a little bit talking about behavioral science more broadly. People have been doing this what’s effectively behavioral science for a long time in marketing and in lots of other areas. That’s to be honest one of the criticisms we get I think a little bit coming in now working out in the field and say well, we’ve been doing that for decades. We know this stuff and that’s true, but I think it hasn’t always been as well spelled out and I’m sure some people will fight back probably actively and point me to examples where it has.

Julian: But they haven’t been maybe as careful. It’s been a bit more intuitive and treating it a bit more as a science. Meaning a science into behavioral science I think is really important. That’s how you make testful predictions, you get things wrong, you learn, you try it a different way. You converge as a field on figuring out what are the patterns here and what do we really believe in versus the things that don’t hold up.

Julian: To make that level of progress to be able to go confidently to a policymaker and say we really think this is going to work in this situation for you right now. You need to have that empirical background and that sense of here’s the structure, here’s the framework, here’s what we tried and what we haven’t in the past. So I think that’s super important for good research that’s going to add even if just one little chink, that’s how we build up the whole edifice overall.

Julian: Let’s see, you asked about translation to applied work from the research. This comes back around a little bit. If the research is done, I won’t say right, but a useful way from the beginning is to really be thinking about the context. Often even the policy question, but whether or not you go that far, embedding it very carefully in the context, especially for behavioral research. As we all know that’s a lot of what’s important that changes things quite a bit.

Julian: So if you started that way, with a particular context, then it’s much easier to go back afterwards. You’re supposed to be able to make policy statements and applied conclusions. But also to have the credence to make those. Maybe you got the relationship, you’ve been talking to people from the beginning. So this is a little bit more than just talk to everybody at the beginning and figure out what questions are important to them. Although there’s an element of that, it’s more to do the good research, even if you’re doing it purely for research purposes, you’re going to start with a particular context. Then that’s going to allow you at the end to be able to make conclusions that are inherently relevant to the context.

Jakob: Got it, thank you Julian. So I want to ask you two more questions that I think our listeners are quite curious about. The first one is more about how to have a career in the field of behavioral science. So since the field is becoming an increasingly appealing career choice for many, especially those people who want to sit at the intersection between various fields as well as between theory and application.

Jakob: However, for that very reason, it is a tough field to prepare well for. So many of our readers and listeners have asked us how they can best prepare for the field. So with this in mind, what skills do you think an applied behavioral scientist will most need in the next 10 years and how can they best prepare for that? Also, how would you distinguish between a behavioral scientist who wants to be a researcher versus someone who wants to do applied work?

Julian: So it’s an unusual field. Probably everybody feels that way about their own area. But probably as I said, it wasn’t so much of a field. It wasn’t as well defined a field at least from the role that I was in, the perspective that I was in maybe 20 years ago. Sometimes I wonder if in the future it might also not be a field, but in a very different way. Before I think people weren’t doing it, in the future I kind of hope everybody’s doing it to some extent. But it may not exist as a separate field that way. It may be a part of a lot of other things where it does compliment them in a lot of ways.

Julian: But that also makes it tricky now to think about where do you place yourself, because it could look quite different. I think in five or 10 years it’s pretty safe to say there will still be something called behavioral science. In 20 or 30 or 40 years, I’m actually not sure. Again not because I think it’s going to disappear, but because it may percolate out. I think that would be a success. I think that would mean we’ve done our job.

Julian: But to be a little more concrete for people on where to begin. I think there’s a little bit of knowing the basics, knowing the core principles and the readings and some of the books and papers that we’re familiar with. There’s some courses now in this and some online courses in others and that’s certainly a good place to start. But then having at least one other thing that you do really well that you bring to the table. Whether that’s a particular knowledge area in finance or health or development or geographics specialty. Or whether it’s a methodological specialty where you’re really good at machine learning and big data. Or you’re good at sociology and qualitative work in addition to quantitative work.

Julian: So having … I think of it almost spatially as this core element that everybody sort of knows to say, you do need that. But rather than thinking of just adding things to that, maybe go pick something that’s a little bit more left field and learn that really well. Or maybe that’s already your background and you can bring that. That helps you to stand out a little bit and that helps you because you are going to be working as part of the team because that’s how things go nowadays. You don’t have to be doing everything, you’re not going to be expected to that, that doesn’t make sense. But bring something that maybe few other people have and that’s going to be useful skills. So think about it in that sort of dual way is what I would suggest.

Jakob: Got it.

Julian: You asked about research versus more sort of the operational side, is that right?

Jakob: Yeah, exactly.

Julian: Well, I think that’s maybe a bit more just what you like doing and for me as a researcher, even when I was working in the policy organizations and enjoying that, I felt like that’s what motivated me. Was to learn things and of course I want them to be helpful things. I want them to be things that people can go and use. Sometimes I’m the one trying to go and use them. But what really motivated me was learning something, even if to be honest I didn’t see it through straightaway.

Julian: In other people I think it’s very much the reverse. So they really want to see it playing out. They want to see it making a difference changing how something is done in the real world and obviously those are both key elements. It’s important to have both. But it’s a bit harder for me to imagine changing yourself from those. Maybe I’m wrong, maybe some people can and maybe some people can do both.

Julian: So I would say sort of be honest with yourself of with which one motivates you and let that guide you. Because there’s lots of openings and both of us know there’s no reason to force yourself to fit into one or the other. Because I don’t think either one is going to be more successful than another way to go.

Jakob: Thank you Julian, I especially liked what you said about distinguishing ones self by having kind of a specialty within the specialty of let’s say behavioral science. How that can be a very value adding element that one can bring to the table to any project or to any team.

Jakob: So as we’re coming towards the end of this chat, we would like to just ask you what short to long-term future you envision for yourself with regards to behavioral science. If there is even that component that you want to explore further down the line in your career, and what types of projects you’re most excited about in the coming years?

Julian: Well it’s good timing. I actually just got off a phone call with a potential future collaborator. ‘Cause I am thinking about that and that was as I mentioned briefly one of the reasons I wanted to return to an academic position. It’s very much related to behavioral science and behavioral policy broadly construed. Whether that’s governments or organizations or family oriented people. I have a young son I think about, I think about his future as well. Or interactions among friends and peer networks.

Julian: So I want to be getting into a little bit more normative or welfare analysis. So thinking about what constitutes a mistake. What if we depart from the economist view of revealed preference and behavioral science certainly is extremely suggestive – convincing to me that people do make mistakes. That is to say they make choices that are not in line with their own long-term best interest in maximizing their own utility function. So their reveal preference doesn’t always work in that sense.

Julian: So what do we replace it with and what kind of model taking a mix of evolutionary psychology, behavioral science of course and economic theory and a bit of philosophy thrown in. Survey data on regret and on experiences, how can we sort of pull all these things together to think of a model of mistakes, a model of conflicting preferences where I conflict with my future self. Which preferences do we use for welfare analysis, for policy analysis.

Julian: Is there some sense of legitimate preferences. There’s not going to be a right answer to that. I’m not going to be able to prove this is or isn’t the mistake that somebody made or these preferences are the right ones and those are not legitimate ones. But I think we have tools at our disposal and a lot of the work in behavioral science leads to this and suggests to this. That then can then feed into the models and say well, this is the set of reasonable outcomes and these other ones are really not too reasonable. We have to be making some very extreme assumptions to get there, which are not plausible.

Julian: Here’s the set of reasonable things for policymakers to go and take ideally a democratically elected government or an organization that’s making a choice in a transparent way from within the set. Or here’s something where in this setting, given this background we might expect people to make these kinds of choices or we might expect people to make what we think of as mistakes and we might guide them a little bit or a more reason to nudge.

Julian: So taking the sort of libertarian paternalism side of things a little bit more seriously, being transparent about it and saying sometimes we are paternalistic. Because sometimes we don’t have a choice or sometimes we do have a choice and we think it’s the right thing to do. But let’s be transparent, let’s do it the best way we can and see where we get.

Julian: So yeah, that’s where I’m headed, it’s kind of a broad agenda I know. I have to start somewhere a little more specific and small scale, but I’m excited about it.

Jakob: Fantastic, well thank you so much for all your insights today Julian. We really like to thank you for your time and wish you all the best for 2019 and beyond.

Julian: Thanks very much, same to you and all the listeners. It’s been a pleasure to chat.

Jakob: Great, thanks.

Read Next

Podcast

Trust in a Technological Age: Kevin Werbach

Blockchain is a concept that is shrouded in myth and misconception. Kevin Werbach is one of the few experts in the world that has both a rigorous understanding of this technology and a vision for how it can make the world a better place.

Podcast

The Future of Banking: Olivier Berthier

The relationship between banks and their customers is constantly evolving. What is the modern bank really selling? We dig into this and more with Olivier Berthier—data scientist, software designer, and CEO of Moneythor.