Erik Wengström is a Professor of Economics at Lund University where he studies how people behave in economic and financial situations. In this conversation, we talk about his study about loss aversion when deciding for others and his recent study on prosociality during the COVID-19 pandemic. Along the way, we also discuss the differences and similarities between economics and psychology.
BJKS Podcast is a podcast about neuroscience, psychology, and anything vaguely related, hosted by Benjamin James Kuper-Smith. New conversations every other Friday. You can find the podcast on all podcasting platforms (e.g., Spotify, Apple/Google Podcasts, etc.).
Timestamps
0:00:04: What is loss aversion?
0:05:45: Start discussing Erik's paper Deciding For Others Reduces Loss Aversion
0:16:01: Generalising decision-making across different contexts
0:20:28: The relationship between economics and psychology
0:30:45: Peer-review and publishing in economics
0:44:14: Start discussing Erik's study on prosociality during the COVID-19 pandemic
0:50:09: Deception in psychology and economics experiments
0:53:39: The risk dictator game
1:03:50: Comparing our COVID study with Erik's study
Podcast links
Website: https://bjks.buzzsprout.com/
Twitter: https://twitter.com/BjksPodcast
Erik's links
Google Scholar: https://scholar.google.de/citations?user=SRPZRHoAAAAJ
Twitter: https://twitter.com/ErikWengstrom
Ben's links
Website: www.bjks.page/
Google Scholar: https://scholar.google.co.uk/citations?user=-nWNfvcAAAAJ
Twitter: https://twitter.com/bjks_tweets
References
Andersson, O., Holm, H. J., Tyran, J. R., & Wengström, E. (2016). Deciding for others reduces loss aversion. Management Science.
Andersson, O., Holm, H. J., Tyran, J. R., & Wengström, E. (2020). Risking Other People's Money: Experimental Evidence on the Role of Incentives and Personality Traits. The Scandinavian Journal of Economics.
Andersson, O., Campos-Mercade, P., Meier, A., & Wengström, E. (2020). Anticipation of COVID-19 vaccines reduces social distancing. Available at SSRN 3765329.
Campos-Mercade, P., Meier, A. N., Schneider, F. H., & Wengström, E. (2021). Prosociality predicts health behaviors during the COVID-19 pandemic. Journal of Public Economics.
Kuper-Smith, B. J., Doppelhofer, L. M., Oganian, Y., Rosenblau, G., Korn, C. W. (2021. Risk perception and optimism during the early stages of the COVID-19 pandemic. PsyArXiv.
Smaldino: my conversation with Paul Smaldino about formal modelling: https://bjks.buzzsprout.com/1390924/7048246-8-paul-smaldino-cubist-chickens-formal-models-and-the-psychology-curriculum
Tversky, A., & Kahneman, D. (1991). Loss aversion in riskless choice: A reference-dependent model. The quarterly journal of economics.
-
[This is an automated transcript with many errors]
Benjamin James Kuper-Smith: [00:00:00] Maybe can we start very basic, um, and explain what loss aversion is?
Erik Wengström: Yeah. I mean, this is, this is not such an easy question, right. As you might think. Uh, I mean, to me, I mean that's, that's, uh, sort of a reaction. Uh, we could think about it as sort of a, I mean, going back to Kaman, of course it's, it's, uh, it's the outcome that is coded as a loss in, you know, relation to some reference point.
fAnd of course, this is the reference point here is kind of the trick, right? What, what is the reference point? Yeah. And, and how to think about that and, and, and so on. And often in experiments, what we do is kind of manipulate this in one way or the other with framing. We give people some money and then we, we say we're gonna take them away.
So, and then we interpret this, or we, we hope, or we kind of, uh, assume that the participants, they frame this as a loss. And [00:01:00] if people then treat those losses differently? I mean, depending on, on how we frame things. I mean, we call that, um, a wicked, I mean, call it loss aversion. If you have a, a sort of a higher reaction to losing, you know, say a dollar, then you have to, to gaining a dollar.
But I mean, the definition is if, if you go into it, you know how, how to define it, you know, you have different types. Some people go in and try to measure, you know, the derivative as, uh, you approach zero from both sides and say that, okay, let's say the derivative of the utility function or the value function on the negative domain is steeper than the derivative on the positive domain.
So you react more to losses, uh, than games
Benjamin James Kuper-Smith: in, how should I say, in some technical way. Like that's, that's how lost version is defined, right? In terms of the slope of the curve. So like one, one thing that I really, so I actually did a study where I compared losses and gains [00:02:00] in, um, a, a prisoner dilemma at first. Um, where the outcome could be losses or gains or combination of them, that kind of thing.
And what I realized, I always talked about loss aversion, but I actually didn't really mean technically what the definition of loss aversion was. And it took me quite a while to realize that, um, I was being very imprecise about what exactly I meant by loss aversion, um, because I didn't really mean it as a difference in slope, but more like a, a step, um, and losses kinda thing.
And then I, like, just afterwards I realized like, oh, that leads to completely different prediction than what I've always said. What I was using with loss aversion. One kind of question I had here almost was, I always found that a lot of, a lot of papers I've read have used this phrase, losses loom larger than gains.
And it seems to me that the rhetoric of that. Has almost, which is, you know, a nice rhetorical technique, [00:03:00] um, but is very imprecise. Do you think that's contributed to that or, I dunno. It just seems to me that everyone just says loss is loo larger than gains and then just do whatever they want to do.
Erik Wengström: Um, I have thought about the exact phrase, but I've noticed the same thing as you experienced that loss aversion is often used in a kind of sloppy and non-PC precise way.
It's actually, I, I read about loss aversion on a Wikipedia page. I don't know if it's still there, but they then exemplified this by, uh, the Asian, uh, so-called, let's called it Asian disease problem. And, um, it's not about loss aversion. So it was, uh, I think it was, uh, perhaps, um, the only time you know that I felt, okay, maybe here I should do an edit and think about this.
This is typical referred to as the reflection effect, right? Where you have different. Uh, risk attitudes on the positive and negative domain. It's not exactly the same thing as loss aversion, but [00:04:00] yeah, it has to do with reference points and you frame the outcomes as losses, but, uh, it's not loss aversion per se that's driving this effect.
So that's another example that people are, you know, uh, a little bit, uh, I think lop being imprecise when it comes to, uh, defining and thinking about, uh, loss aversion.
Benjamin James Kuper-Smith: I mean, so you did a study about loss aversion. Is your definition the original technical definition, or how do you I mean, it's, so it's in the title, right?
So you have to
Erik Wengström: Yeah, I think we are not, so we are also contributing to this slot, perhaps a little bit in this paper. I don't know, but we have, uh, we have different ways. It's not easy and it's, doesn't make it more easy if you, uh, you know, make the, uh, utility function richer or if you have a more complex structure.
I mean, if you really want to, you know, apply. For example, prospect theory with probability weighting, and you might allow for differences in probability weighting on the positive and negative [00:05:00] domain, you might allow for different RA risk attitudes on the positive and negative domain. And then you should also estimate a kink or this kind of loss aversion, uh, aspect.
You, you, you realize, I mean, there are many parameters here to estimate and, uh, uh, we don't go that deep in this paper. So we use a kind of a very, uh, brute force, uh, approach where we just look at the number of risky choices when there are losses involved, and, uh, when there are no only gains. And then we do estimate some sort of structural models, but that then depend on, on, um, on certain assumptions, on functional forms of underlying utility functions.
Benjamin James Kuper-Smith: Okay. Um, can you then, uh, briefly summarize the kind of main findings of your paper deciding for others to reduce loss aversion? I guess the title already says quite a lot. Um,
Erik Wengström: exactly. So I think that, that, that the title says it all or says, uh, the main message here, the take [00:06:00] home messages that we had, this online experiment, uh, in Denmark, we, uh, invited people to, uh, uh, to a webpage where they made choices and these choices were over gamble, so they made this kind of risky choices.
Then we had four different treatments. So we had one treatment where subjects made decisions on their own behalf, basically like a standard experiment. They, uh, make a number of choices and then afterwards we pick one and we pay them according to that, um, decision.
Benjamin James Kuper-Smith: So this is a kind of traditional risky gamma, right?
You have two options and each has two potential outcomes, uh, with
Erik Wengström: Exactly, exactly
Benjamin James Kuper-Smith: different, different magnitudes.
Erik Wengström: Yeah. Uh, and then we have had a hypothetical scenario where basically the same, but they were not paid. And then we had two treatments that involved others. So one treatment [00:07:00] where they only made decisions on behalf of someone else, so they were not paid themselves.
And then we had another treatment where you took a decision that, uh, you kind of shared with another person. So you and this other person got the same outcome.
Benjamin James Kuper-Smith: Mm-hmm.
Erik Wengström: Uh, so that was kind of the main setup. And then we had these kind of gambles, some of them were framed as purely gains, so you could only win money basically, whatever happened.
And then we had some, uh, gambles that then involved losses. So there were some potential losses to be made if you were unlucky. What we then do is basically compare behavior between these treatments and, uh, I mean, first of all, it was quite striking to some degree. I mean, when it, when we only look at those gambles that involved gains where you only can make money, there were completely no difference between treatments, right?
So whether you made decisions on [00:08:00] your own behalf or on, um, together with someone else, or where you shared the outcome, or whether you did it, uh, hypothetically, I mean, it didn't play a role. So basically the number of say choices, uh, uh, estimated risk parameters if we do it in a structural estimation was, uh, very similar.
But if we don't put attention on our, on those, uh, situations where they could potentially lose money, I mean, then we find that people take um, uh, less risk when they make decisions on their own behalf. So when in this kind of standard situation where you are the only one that are bearing the consequences of, of your actions, so that's where the title comes from.
Then as soon as someone else was involved, you know, you take, um, more risk. So that means here that one interpretation is of course done, that there are less loss averse.
Benjamin James Kuper-Smith: Yeah. So I have, [00:09:00] I have a maybe a little like, first, a few slightly technical questions and then Yeah. Um, slightly broader questions. One thing, um, that slightly confuses me is that when I look at figure one, right, you see you have, uh, the two panels.
One is the no loss, one is the loss, uh, conditions, and then you have the, uh, average number of safe choices per, uh, treatment used called it, which is individual hypothetical, both or other. Um, there's, that's figure one. And what I find slightly confusing is, so it's, it's obvious to me to see that there's no difference between the conditions and the no loss conditions, but there are differences in the loss condition.
But what I find slightly confusing is that it seems to me that the, the number of safe choices for individual is for myself is I identical between no loss and loss. So how exactly it seems to me like that, that's basically when I just look at the figure, at least it seems to me like there's no difference [00:10:00] between the loss and the no loss condition for individual.
Erik Wengström: Okay. That's a, that's, um, that's a relevant question. And, and now when you ask it, I remember I've gotten this, when I presented the papers, the paper, uh, and I think this is really, I mean, these are not, the gambles don't have the same expected value. They don't have the same variance. So there is no reason actually to, to, to expect, uh, choices or a number of safe choices to be the same between the treatments.
Benjamin James Kuper-Smith: Mm-hmm.
Erik Wengström: So, um, this just happens, I think, to be the case. So if you, if you look at the details of, of, uh, the choices, uh, or the, the gambles they were choosing between, you know, there could be many reasons. You know, I, it could be more, um, it could be a higher, there could be, you know, this can depend on the payoff structure of the gamble.
So I put it that way.
Benjamin James Kuper-Smith: Yeah. That was basically the other question I had is kind of, so. If you, if you look at the [00:11:00] table one, where you have the different, um, where you, you, you list all the gambles in the study, right? Um, I was also confused that somehow like, you know, I'm a mainly experimental, um, psychologist or neuroscientist, and what surprised me as the first thing is that the gambles weren't kind of kept identical apart from the loss, no loss condition.
So for example, I mean the one thing that I noticed immediately is that in the, so you always have a left gamble and a right gamble, right? And you have to choose which one you want. And for the losses, the left gamble always had a much true, or three times as large difference between the two options as the.
As in the no loss conditions. Um, so in the, let's say in in screen, four heads is minus nine and tails is 40, uh, always. And in the no loss condition, heads is 72 and tails is 86. So [00:12:00] just like the, the outcome of that gamble, you know, there's a much larger difference and that's always the case. I mean, does that partly explain why figure one seems slightly confusing?
Um, that as, as you said, like the expected utility of each gamble isn't the same. Is, is that kind of what you meant?
Erik Wengström: Yeah. So that could be, yeah. Yeah. So, so that's, that could be one of the explanations. You know, why you have more, uh, you would expect there to be more, um. More safe gambles on the no loss or more safe choices on the no loss screens.
Right?
Benjamin James Kuper-Smith: Yeah. But why, uh, just from a like practical perspective now, why not keep, for example, the difference? They're the same and see what I mean? Like, it seems to me like, yeah, if, if, if, if, you know, as a second year or 30th PhD student who doesn't know much, if I would've set up an experiment, I would've kept it actually the same and just subtracted something or something [00:13:00] like that, right?
Erik Wengström: Yeah. I mean, uh, I think you are asking a very good question. And, uh, to be honest, you know, now it has passed more than 10 years since we designed these questions. And I must be totally honest and I don't really recall how we came up with these parameters to be honest. So, okay. I mean, it seems like, um. It seems like, uh, we could have made them a little bit more, as you say, we could probably have made, uh, we could have had some lower payouts perhaps in one of these no loss cases.
Yeah.
Benjamin James Kuper-Smith: Yeah. Okay. So it's lost to the mystery of time. Okay. But then basically those, those two questions I had though kind of ex at least one explains the other kind of the
Erik Wengström: Yeah, I, I, I, I mean if you, if we do the structural estimation, we do find loss aversion on, [00:14:00] on both, uh, the no loss and the loss, uh, domain.
Right. So it's, it's not like there is no underlying loss aversion in the individual treatment, and that's as you into is, is due to, you know, the construction of the table. It doesn't show up in, uh, in the figure.
Benjamin James Kuper-Smith: I see. Okay. Okay. I mean, I guess the broad question is kind of like, why do we not have loss aversion when deciding for others?
And the one thing that maybe confused me almost most is that, um, you have no loss aversion when you're deciding for yourself. And for someone else, somehow would've expected that that condition would've been more like self rather than other than other. So I dunno, yeah. What you think is kind of driving this effect that you observe in the study.
Erik Wengström: Yeah, I mean, these are really good and interesting questions and I, I mean, we don't know, I mean, we speculate here right at the end, and what we [00:15:00] propose here is that as soon as you involve someone else, you kind of, um, put yourself in those shoes more. And if you're doing it alone, so just a mere fact that someone else is involved makes you, you know, uh, start to concern about the other payoff.
And then this is in this case enough to, uh, to shift your behavior, but, uh, exactly what's going on there. I mean, that's a, that's a, that's a very good question. And I would be, I would love to, to to see some young PhD students, you know, looking more into this.
Benjamin James Kuper-Smith: Yeah. I mean, you have, um, I haven't read the other paper, but you have another paper about is it risky choices for others or that people make more risky choices when deciding for others, something like that.
Um, do you know which paper I'm referring to or,
Erik Wengström: yeah, we have another paper, but that's, yeah, that's also on, um. On risk [00:16:00] taking on behalf of others.
Benjamin James Kuper-Smith: I'm just curious whether that might give us some sort of like the two combined, give us some sort of general idea of how people change the decision making once they switch from their own frame to someone else.
Erik Wengström: I mean, I, I, I, you know, first of all, I think when talking about these issues, we have to realize that this is probably, and there is some evidence for it, of course, that this is context dependent. You know, it depends on, on what kind of the framing of the situation. Now we're talking about financial risk, but if you were talking about health or safety and so on, um, you know, if I make decisions on behalf of someone else's kids when I drive, you know, I would probably be more on the, you know, the cautious side rather than on the, you know, risky side.
So you have a different, you know, uh, different effect of going from individual to decision making for others. So I think that's also very, you know, important just to keep at the back of our mind [00:17:00] here when we talk about this issue, that, uh, there are differences between, you know, the settings.
Benjamin James Kuper-Smith: Actually, the, the point about the important about the setting and the context is something I was.
I think talking to her recently to someone in the lab that I kind of got frustrated that a lot of stuff you find out just depends on context. That whatever you find out is true in like one context rather than a general finding. I mean, uh, I just, we talked about it 'cause I was just a bit frustrated. Um, but how do you, I don't know, how do you think about that?
Like it's, because it seems to me that like, it's kind of, I, I find it just a bit frustrating the idea that like, kind of if you have something and it's just like a minuscule finding about some specific social context, and as soon as you change one of 30 different parameters, potentially, potentially, like everything can change about the behavior.
Is that, I know that something that you've, or like when you design a study for example, like I guess you have to take the context into account, right? So how do [00:18:00] you decide which context to do or how to do something that maybe generalize across context or do you not care or.
Erik Wengström: I really do think we should care and uh uh, I think this was also one of the reasons why economists, at least for a long time, you know, when doing experiments, they preferred this kind of neutral frames and so on.
Let's assume that if you have a neutral frame and we just induce the preferences with the, uh, incentives, you know, you get more generalizable findings, you know, compared to a situation where you have, uh, you give cues, you give more, uh, detailed descriptions of things you might give context and so on. So, so that was really, I mean, kind of one of the, uh, hallmarks or one of the, you know, building blocks of experimental economics, you know, to, to move away or to remove one of a lot of these context parts.
And I think that's a sound methodology, but I mean, a sound approach. But on the [00:19:00] other hand, there is always a context as well, right? So yeah, even the neutral context might be a context, you know, in some sense. So, uh, and if the, the, the rest of the world is full of context, you know, how, how solid are those findings then?
But, um, we don't know. I mean, um. Uh, so, so I think it's still, I mean, I trained economist. I still think it's, it's a good starting point, you know, to have this as, you know, somehow context free. Uh, you should have the context, free context, you know, to, to start with. And then we can, you know, start to try to understand how easy it is to, to, um, generalize or extend these findings to, to other situations.
But I, I do degree. I mean, I agree that this is of course, uh, partly frustrating as well, you know, uh, but, um. It's also fascinating in some sense as well. I mean, it, it requires more work. It requires more [00:20:00] data, bigger data, more, more studies, you know, before we can say things with confidence. I think, um, some of the claims that you would've made maybe, you know, 30 years ago based on one small study you, you cannot do today, which I think is a, of course, a sound kind of development.
So it's a, it's a challenge and it's a frustration, but it's also, I think reality.
Benjamin James Kuper-Smith: Yeah, I mean maybe, um, you know, as I said, I'd like to talk about the relationship between economics and psychology or kind of what the different perspectives are. Maybe one thing I've always been slightly confused by is.
For example, your, this, this paper we just mentioned, right? Deciding for others reduces loss aversion. To me, that's a straightforward psychology experiment. I don't understand exactly how economics or economies or what exactly they, why they are interested in this kind of thing. So like what I mean, because to me, this is just about how [00:21:00] people make decisions, which is for me, psychology.
I don't know. I'm just curious, like what's the, for you personally, maybe like what, why are you doing these kind of studies or, or why do you, because you said, you know, you're a trained economist and that kinda stuff. Why does that fall into what you're interested in? If that makes sense?
Erik Wengström: Yeah, no, it's a perfect, uh, it's a fine question.
So, so this project, just to give an example, you know, it started in, in the wake of the financial crisis. So there was a lot of, um, risk taking and, uh, some say excessive risk taking. And, uh, the question was, okay, why? Why is this or there? Many, you know, explanations and, uh, often her, you often hear explanations like incentives, shape risk, attitudes, right?
Or risk taking. So maybe it was just misaligned incentives, uh, in the financial sector [00:22:00] that led to this. So that was one explanation, uh, proposed. But there was also others, you know, that yeah, these are greedy people or unreliable, you know, bankers and, and they are taking too much risk on behalf of their clients, you know?
So with that as a starting point, you know, before you can say anything about the effects of incentives, the basic question is then, okay, you, yeah. What if we remove the incentives? Okay. Would there still be differences? Right? So this is the first step that we try to take there, right? So we remove, we try to remove the incentives, we keep it as simple as possible.
And of course then we are kind of down to psychology in some sense. We are, we are looking at, uh, uh, differences in behavior when. But it's also about, uh. For an economist, you know, very relevant behaviors, right? So this is about risk taking in a financial domain. So, uh, as I said, you know, if you want to make claims, then why do, uh, [00:23:00] someone take more risks than another?
You could think about incentives, but you could also think about these kind of underlying, uh, differences in how we perceive the situations when we make decisions on our own behalf or behalf of our clients or someone else. So we started this research. Program. So this was one paper, which was kind of basic first step, and then we added incentives, you know, to see, okay, what if we provide incentives, which, um, look a little bit like they do in the financial sector.
Uh, will that increase, uh, risk taking on behalf of others, and will it do so for everyone? Or are there some sort of offsetting mitigating effects, you know, and, um, that's what we, we then study with. For example, we find that the degree of social preferences, so people that care more for others are also responding less to, you know, uh, incentives to take risk for others or take excessive risk [00:24:00] for others.
So it's, um, I think, you know, at the bottom of this, um, you have. At the bottom of, of, of the economy, you have humans, right? So it's, it's, it's natural for an economist to try to understand how those, uh, humans act and interact when it comes to these kind of, um, economically important behaviors.
Benjamin James Kuper-Smith: Okay. So the, the kind of ultimate goal is to understand the economy and because humans are basically what make the economy, um, that's why it's a relevant question for you.
Erik Wengström: Yeah, and I also think that's, um, I mean talking about the differences between psychology and economics, I also think that's, um. Also, you know, describing a big difference between, you know, how we set up our problems, how we view our problem in particular, probably how we motivate our problems, exactly.
Benjamin James Kuper-Smith: Our
Erik Wengström: research, [00:25:00] uh, questions, right?
So we start often from, I think more often from, you know, some real world, you know, economic phenomena, you know, that.
Information and knowledge on, and, and hopefully in the end, even things like, you know, policy recommendations and, and uh, and these kind of things. We, we, we have this pressure in some sense to always have a policy recommendation in every paper. And you don't have that, I guess, psych as much?
Benjamin James Kuper-Smith: No, not really.
I mean, what I find interesting is that in a way from, I mean of course when I say economics, what I have in mind is the papers from economics that I've read, which means of course they are by definition those that are much closer to psychology anyway. 'cause that's the kind of stuff I happen to be interested in.
So there's probably like, I mean, there's lots of economics, uh, that's I can identify very clearly as being what I would imagine economics is. Um, so maybe in a way I'm just sampling from, [00:26:00] uh, a non representative sample within economics papers. That's very, um, possible. Um, but one thing I just realized is that when you sit with the policies is that it seems to me that kind of economics in some sense is if you're trying to apply it to policies immediately is almost more applied than psychology.
But I've also read a lot of economics papers that are very theoretical and that seem almost a lot less applied than psychology, almost as if economics is like the, the outside extremes and then psychology's kind of the middle of it. I dunno whether that makes sense, but, um, it just occurred to me as you said that.
Erik Wengström: Yeah, no, I, I also, I mean. Economics is a very broad field, right? Mm-hmm. So I guess you can find, um, both ends of the spectrum here. You know, as you say, you know, you have things that are totally abstract, you know, totally, you know, [00:27:00] stripped of any context and then it'ss pure theory in some sense, right? And then on the other hand, you have really applied stuff that really tries to, to give direct, uh, more policy or test testing different things or different, evaluating different programs and, and policy.
So I think you have the whole spectrum, maybe more than you do in psychology. I don't honestly, I dunno, psychology well enough to, to carry,
Benjamin James Kuper-Smith: I mean, in some sense maybe I'm also pretending psychology's less applied than it is just because I'm not interested in clinical stuff. So, you know, of course if you are, people motivate their psychological interest that way.
Where I, whereas I. Don't. But yeah, it's just,
Erik Wengström: yeah, I don't know. Maybe there is more, you know, kind of interaction within economics than there is within psychology. I dunno. I mean, because here we have the Department of Economics in most behavioral economists and experimental economists, they work in departments of [00:28:00] economics.
So they interact with all kinds of, uh, economists every day from different fields. And, um, which of course means that you have to motivate what you do from also, from their point of view to some degree. Uh, but, um, I don't know. Maybe in psychology you have more this division between the more, let's say theoretical or judgment, decision making, that kind of branch.
And then you have the clinical psychology, which is quite different I would imagine.
Benjamin James Kuper-Smith: Yeah, but maybe, uh, I don't know, maybe the kind of clinical stuff is the equivalent of policy recommendations. Just that you're trying to, I dunno, heal the economy whereas we are trying to heal people or something. I dunno.
Um, but yeah, it also occurred to me because I had one discussion about, um, with Post Morino, he's a, well, he does a lot of different things, but he usually does 'em from a theoretical background of modeling. Um, and we had a kind of discussion about modeling, um, formal modeling [00:29:00] and we kind of were wondering whether psychology has out outsourced its theoretical parts to economics or something.
Erik Wengström: Um-huh.
Benjamin James Kuper-Smith: Because sometimes I really read like papers. I think, oh yeah, this is a really cool paper. And then I realize it's an in an economics journal and then it's just 20 pages of maths or something. And, um. A bit trickier to read for me as someone who doesn't have that training.
Erik Wengström: Mm-hmm.
Benjamin James Kuper-Smith: Yeah. Um, but I mean, you, you do consider yourself an economist and I mean, is the kind of applied policy stuff for you, something that's just for you personally actually?
Is that how you think about it? Or is that more like a kind of distant, kind of like where do you position yourself on the applied to theoretical continuum?
Erik Wengström: Yeah, I guess also, um. I'm more on the applied side. Well, depends, you know, we also have something called [00:30:00] applied uh, microeconomics, which is more like empirical, which is, is is,
Benjamin James Kuper-Smith: I see.
Erik Wengström: Have necessarily to be applied. So, but, uh, uh, as a more experimental behavioral economists, I'm somewhere in between, you know, somehow they applied, uh, economics and the more traditional microeconomics I would say. So some, some of my work is more leaning towards, uh, standard micro, which is more abstract, less applied.
But then on the other hand, I often involve real people and so on, which, uh, and, and, and this try to kind of measure and investigate decisions of real people. So that puts me and a little bit more on the applied side in one sense.
Benjamin James Kuper-Smith: Yeah, I have, um, this is, uh, a question that I think is. Uh, very much a technical specific, not even technical, but a kind of nitpicky question about economics, which is why exactly.
From what I can tell, [00:31:00] economics papers can take years between submission. I mean, like you said, you started yours in 2010, which surprised me because it's published in 2016. Um, and I've seen other economics papers where it said like, we, uh, collected the data then, and then like eight years later the paper came out.
Or I, I mean, once I literally saw on a paper it said like, submit it like, I dunno, 2012 and published 2019 or something. Um, what exactly are you guys doing in the, in your peer review process? And what, I dunno, is there something that, I mean, in psychology, I think we feel like we take a long time to peer review this whole stuff, but is there something we can learn from that or is it, are you just slower with admin or what's happening?
Erik Wengström: Uh, yeah, I think the reports are quite different. First thing, I think sometimes the reviewers are more like, you know, a co-author. Yeah. A reviewer, you know, so you can get, uh, five pages, you know, of comments. And [00:32:00] then in some journals you have four of those, and then, uh, they request more and more data and, uh, new experiments and new analysis and so on.
So it's, um, it's slow and, uh, it's getting faster luckily. So I hope we will be able to, to, to change the norm a little bit. Uh, and then you have this development, you know, and economics. That, uh, papers, I mean, you, in some sense you have two developments, right? So over the last, uh, decades, you know, you've had, uh, the development of more bigger and bigger paper somehow, you know, it's, it's, I just published a paper, you know, I think all in all with appendices and stuff, it's like one 50 pages or something, you, those.
Papers that took three years in the referee. It's basically a book started with, yeah, it's, it's almost like you are back to writing monographs in the PhD [00:33:00] programs in economics, but now it's called the job market paper. So that's supposed to be, you know, uh, 40 pages in and then an appendix of, I dunno, 100 pages.
So you're supposed to basically meet all potential objections within this paper rather than, you know, writing a short follow up mm-hmm. Or a comment and so on. So it's supposed to be almost, uh, uh, it's, it's supposed to answer most of the critical concerns already when the, within the same Yeah. Sort of paper.
So I think that's a different stance, a little bit towards, if you compare it with, with other fields. So, uh, on the one hand we've had that development, so, um. Papers and, and this has been growing. But then there is also, uh, we just over the recent, you know, five years or so, there is an alternative trend to, uh, for, especially from, from journals, you know, to open up for shorter papers.
So you have the leading, uh, journals like American Economic Review now have, [00:34:00] you know, a short version of it, and you have restart. You have the, the, the many of the big journals are now, you know, uh, opening up and actively, you know, inviting shorter papers. So I think that's kind of a reaction to this other, uh, development that all findings does not, uh, uh, do not fit into, you know, this kind of huge paper format.
You could have a nice, you know, result and an important finding, which is perhaps just like 10 pages.
Benjamin James Kuper-Smith: Yeah. I mean, yeah. One thing I was wondering is. So it sounds like the, I should say that in economics the attempt is more to solve not all, but most of the problem, potential problems with an idea within a paper.
Whereas I think in maybe psychology is something you are more likely to just write three more papers and then to kind of take it a step by step. Yeah. I'm wondering, I mean, in a sense I feel like the, [00:35:00] the, the meth, the approach from psychology is maybe psychologically a bit nicer because you have these like smaller wins along the way.
But I'm curious whether maybe the economics is actually almost ironically more efficient. Because once you have a paper, you are more confident that it's good and reliable and you don't have to read four papers that might end up being almost longer than the one paper. Um, just because you've kind of managed to really condense it into one really strong piece.
Um, so that, I don't know. I wonder like whether there might actually be more efficient because you kind of. Uh, there's less, there's fewer papers to read. Um, there's less questioning how reliable the results are and that kind of stuff. Does that make sense or is it just, I don't know.
Erik Wengström: No, I think it's a good comment and I guess there is this trade off, right?
On the one hand, as you say, it's kind of, uh, kind of demotivating to work on, on, on papers for five years, and in the end, maybe you get rejected, you know, if [00:36:00] you, if you most, I guess psychology would say that it's better to, to break it down to smaller pieces. On the other hand, as you say, you know, reading some of these papers is also very nice, you know, if you, if you feel that this is solid and you can, uh, you get answers already within this paper to some of your concerns or questions, you know, in these robustness checks and, and alternative specifications and whatnot that's provided within the paper, rather than having to, you know, uh, look up, um, five other paper that are published, uh, subsequently.
But yeah.
Benjamin James Kuper-Smith: But like one of the reason why I kind of, um, asked the question is because, for example, we have, so this thing I mentioned earlier with gains and losses in the prison dilemma I'm in. So I did that fairly early on in my PhD and it kind of worked out fairly neatly. We basically found, I guess what we call loss avoidance in people's behavior there, and it kind of fit very neatly.
But then I thought, well, princess Lema is just one context, right? Um, or it's [00:37:00] only one of the two by two games. And actually we, I should like, if, if this is a general finding, it should also apply to stag hunt and chicken on, you know, all these other kind of games. So I thought like, okay, if we actually want to make this a general finding, then we repeat the entire thing with all these other games.
And that's, I mean, we did a brief part of that and it seems to kind of work, but. The weird thing is then it seems to me almost like I, now I, if I, if I do the latter thing, I almost don't need the first paper anymore because I've got the general finding. So in a way, us publishing the prison cinema dilemma study first almost is a waste of everyone's time because like as soon as you have the general finding, you just don't need the small thing anymore.
I mean, we're probably gonna end up doing it as two separate papers anyway, um, in part, uh, for some practical reasons and that kinda stuff. But yeah, it is this weird thing sometimes where I feel like you, yeah, the latter paper makes the [00:38:00] form paper almost unnecessary to read for anyone. Um,
Erik Wengström: yeah, maybe, you know, exposed in a sense, but the reason that the second paper came into being was the first paper.
Right?
Benjamin James Kuper-Smith: Exactly. Yeah. Yeah.
Erik Wengström: So in sometimes if you view it, you know, not, uh, in terms of the, you know, broader perspective here, is that, uh, because of the first finding you, uh, got interested in this and you, you were motivated to write this bigger paper and then has its, uh, contribution in this, uh, science, in the process of science as well.
Benjamin James Kuper-Smith: Yeah, exactly. I mean, for me personally, it's a, it was a great learning experience in this whole, like thinking about how can I make the, the, how can I test the findings in the most generalizable way and that kind of stuff. That was like really. Very useful. Uh, I think, um, for me as a scientist, but then I do wonder sometimes, like making it into two papers is kind of, I, you know, you know what I mean?
Like the, uh, [00:39:00] just makes you great. You get more one publication more and
Erik Wengström: slicing it up.
Benjamin James Kuper-Smith: Yeah. But then it, I
Erik Wengström: mean, it's, one thing
Benjamin James Kuper-Smith: is, the other problem is then if we don't do that, we have like a, a paper with like eight experiments or something. And that's, that also doesn't help anyone to have this well long paper that, you know, you, you download it and you just put off reading it because it's so long.
Erik Wengström: Yeah.
Benjamin James Kuper-Smith: Yeah.
Erik Wengström: It's too many variables, too many things to keep in mind.
Benjamin James Kuper-Smith: Yeah, that's why basically
Erik Wengström: when you read it,
Benjamin James Kuper-Smith: that's exactly, that's why I basically said like, okay, we're just gonna do it as two things. 'cause then it's, it's just simpler to do from a practical perspective and work through it. But, um, yeah.
Uh, on papers, on economics, just for a second. Again, these are very nitty, basically you are one, you are one of the first economists I'm talking to, to have to ask all the annoying questions. I've, I've always had
Erik Wengström: Okay.
Benjamin James Kuper-Smith: To
Erik Wengström: defend the hold.
Benjamin James Kuper-Smith: Yeah, exactly. [00:40:00]
Erik Wengström: Field here.
Benjamin James Kuper-Smith: Yeah. Um, like, I mean, these are super critical questions, per se.
I think a lot of it's just differences and norms and that kind of thing, but it's kind of confusing from the outside sometimes. So, uh, what I always found really weird also, I mean, you know, I said I'd like to talk about your paper about, um, prosociality and Cop and, uh, in COVID later, and what I've, for example, there found, uh, what I've seen in, you know, many economics papers, is that basically your introduction is a long abstract almost, where the, so like if you look at the introduction of your, of that paper, something like the first two paragraphs are what I, what we would call an introduction almost in psychology, where you outline the literature, then you have like one.
Paragraph of methods of your paper, then like one paragraph of results and then like four paragraphs of discussion almost. So that, that's always been slightly confusing to me when reading economics papers. Why? I don't know. Why, why? It seems to be have [00:41:00] abstract then along abstract then methods.
Erik Wengström: Yeah, I guess, you know, the abstract is supposed to be.
Something separate from the paper.
Benjamin James Kuper-Smith: Mm-hmm.
Erik Wengström: And visa versa, right? So you should be able to read the paper in a sense, without reading the abstract. Right. So I think that's, uh, why it's structured or why this norm has kind of evolved. Uh, I guess, um, it's clear that this is, and, and also, I don't know, I guess, I mean, this would be interesting to, to, to study.
I mean, do, uh, scientists of different field, they probably read papers very differently, right?
Benjamin James Kuper-Smith: Yeah.
Erik Wengström: Yeah. So what parts in, in, in what to order, what sequence, in what detail, you know, you read. But in essence, you know, in most economics papers, if you read, as you say, you know, if you read through the introduction, you, uh, get a pretty good, you know, uh, uh, grip of what the paper is doing.
You, you know, why this is important, you know, what, what they're contributing with. You know, a little bit about the methods and, you know, [00:42:00] uh, yeah. Typically. You get that quite a Okay. Good overview. Um, I find it frustrating sometimes when reading papers from psychology.
Benjamin James Kuper-Smith: Really? Okay.
Erik Wengström: I, I don't find the results, you know, I don't find the, the, the contributions.
It's just, uh, yeah. You know, goes on and on, but I have to keep looking.
I
Benjamin James Kuper-Smith: see. That's, that's funny because I kind of, um, I mean, not, not, you know, about your paper specific, but like, I've had this like, you know, economics paper where like I read the title, which is a kind of summary of the paper. Then I read the abstract, which is a longer summary, and then I read the introduction and it's just the same thing.
I go like, guys, then I read the results. So like, I've read like the fourth time basically by the time I get to the results. Um,
Erik Wengström: oh, I can see that. Definitely.
Benjamin James Kuper-Smith: Um, so that's why, I don't know, I think, um, for me, always the, or the way I was at least taught or whatever, I think the introduction was maybe you have like one paragraph at the end about what you do, but basically it's just mm-hmm.
Summary of the field or something until then.
Erik Wengström: Mm-hmm. Yeah. [00:43:00] Yeah, I guess that's why I keep looking, you know, because I don't find this lecture I can have to go to actually read the full paper to find out what they're doing.
Benjamin James Kuper-Smith: Yeah. Yeah. I mean that, that, again, that also another kind of the nice thing that when I, you know, read the introduction of, of this COVID papers of yours, uh, paper of yours, it was kind of nice to have like pretty much most of it in the introduction already.
Like in a way it does kind of, it is a kind of intermediate step of resolution between reading the abstract and reading the entire thing. Um, so that, that is then quite nice. It's just always unexpected if you
Erik Wengström: Yeah.
Benjamin James Kuper-Smith: For, for someone not from economics or forgets about that.
Erik Wengström: I guess it's also the same these days.
Often when you go to seminars, I dunno about psychology seminars, but the economic seminars, you, it's almost like the introduction of a paper. You get, you know, the motivation you get, uh, the contribution of what they do and a little bit about the method and then also often, you know, [00:44:00] a preview of the main findings already on the fourth or fifth slide.
Benjamin James Kuper-Smith: Oh yeah. I've, I don't think I've ever done that Yeah. In anything I've presented. So
Erik Wengström: that's,
Benjamin James Kuper-Smith: yeah. Shall we start talking about the COVID paper then, or, I dunno.
Erik Wengström: Yeah, why not? Yeah.
Benjamin James Kuper-Smith: Okay. Um, yeah, maybe as, as with, uh, with your also version paper, um, could you provide like a, I dunno, two, three minute summary or whatever of just like roughly what you did as a starting point for the conversation?
Erik Wengström: So this was a paper about basically social preferences. So it's, uh, we wanted to find out, you know, if the things in a sense, the things with traditional method, um, measure and, and label social preferences, uh, have any predictive, you know, bite or predictive power on people's behavior during this pandemic when it [00:45:00] comes to, in particular social distancing, but also other types of compliance to recommendations and, and, and norms.
So what we did was basically to, to run two, uh, separate surveys with a representative sample of Swedes in, uh, the spring of last year, two year 2000. So during the first wave, basically in Europe. And, um, we did one survey on social preferences where we try to elicit, you know, the degree of prosociality in different ways.
So the main task was kind of a classic experimental or a variation of a classic experimental economics task. Um, kind of a risk dictator game we call it.
Benjamin James Kuper-Smith: I'll ask more about that maybe later, 'cause that that was an interesting choice. Yeah.
Erik Wengström: So, uh, basically trying to measure, you know, how much, uh, how you, how people trade it off, you know, getting some dollars for themself.
Uh, [00:46:00] with, you know, exposing others to risk. Uh, and then we get, you know, an individual measure for each person here, uh, on, on their sort of degree of prosociality. And then about two weeks later, you know, in not only seemingly, or we try to make it look like an unrelated survey, we ask them questions about COVID.
So in particular, to what extent they kind of, um, stay home when they're sick. We avoid, you know, social gatherings. They, um, what we more have, you know, we have, um,
Benjamin James Kuper-Smith: the coughing
Erik Wengström: into the, uh, whether they avoid coughing into the elbow and whether they, or, you know, following by the recommendations, more or less washing hands and avoiding touching, uh, their face and so on.
Uh, and what we do find then is also, then we can sort of, and we actually also, these were all. Kind of hypothetical survey questions to also have [00:47:00] some more, uh, you know, incentivized measures. We ask them actually whether they would like to buy one of these cloth masks and, and we give some, some money or some of the participant money to do that so we can elicit their kind of, uh, willingness to pay for or for, for that mask.
We also look at, um, whether they, at the last screen of the experiment, we provide some links, you know, where they can find information about how they can protect themselves and support the healthcare system and so on. And this is also not, you know, it's not supposed to be part of the experiment. It's just, you know, an end screen.
Hey, thanks for participating. You can, yeah, if you would like to find out more information, you can click on these links so we can actually also measure the, the number of clicks to those links. Um, and, um, yeah, we also had some donations. So some of them, uh, received, you know, 20 euros and then our dollars, and then we were, you know, asked to, [00:48:00] uh, donate part of that into, um.
Unicef, um, fund, you know, the COVID pandemic. So these were also a little bit more, you know, incentivized. We as economies typically like to have, uh, some measures that are not entirely just self-reported and hypothetical. So the, the, the overall thing is done to see, okay, is there any predictive power of these?
Measure of prosociality on people's behavior during the pandemic. And the answer is yes. I mean, it's, it's really consistent across all, basically all the kind of measures almost we look at. You know, you find a, a, a correlation, a strong correlation between what people do in this kind of abstract game, you know, when it comes to exposing others to risk and what they claim to do when they, uh, walk on the streets during the pandemic or when they not to walk on the streets during the pandemic.
So I think that was, um, um, kind of a clean, uh, straightforward [00:49:00] finding. And, and we also had access to, for a sub sample of these, we also had, uh, information about. Or we had, uh, they had participated in a previous experiment of ours where we had, uh, elicited social preferences used as some survey questions.
So we could also find that the, the measure, the prosociality measure two years earlier were still predictive of what they were doing during the pandemic.
Benjamin James Kuper-Smith: Mm-hmm. Yeah. And I think that your summary was kind of, that it's, uh, a, a strong unstable predictor of reward behavior. Right. Something like that, as in Yeah.
Operation not only works now, but also like if you look at it in long term, it seems to kind of
Erik Wengström: over time. Yeah. And across context. I think that's also the thing here, right? Mm-hmm. We talked about context dependence before, and whether, at least when it comes to this personality trait or this personality aspect, it seems to travel across the context from the abstract lab [00:50:00] to, you know, the, the wild, uh, the wilderness of the pandemic.
Benjamin James Kuper-Smith: Okay. Um. So maybe just because this slightly relates more to about the, about psychology and economics. So one big difference, I think there's that in general, psychologists don't mind deceiving their participants. And in economics, I think you're basically not allowed to, uh or otherwise they won't publish in certain journals or it's at least.
Not
Erik Wengström: Yeah,
Benjamin James Kuper-Smith: accepted
Erik Wengström: as much. So, uh, referees, uh, can complain and, and and so on. And this is a long tradition. You know, I think these were traditionally the two of the main, uh, differences between doing an experiment in, in economics and psychology was of course no deception. And also typically incentives, you know, not just hypothe ethical choices.
Benjamin James Kuper-Smith: So based on that, my question is, you kind, you kind of deceived participants, right? Like, I don't know if my question was kind of like, I was curious, I thought like, okay, it's really cool [00:51:00] that like you manage to separate these two surveys and make them appear completely different and that people really think they're just taking it in part as a new study, that kind of thing.
But then I also thought, well don't economists, isn't that exactly what they don't like doing? Isn't that kind of. Was, was that, uh, something you thought about there or, um, also that for the participants in terms of, I dunno whether what the ethics boards things are, but that they kind of didn't know they were taking part really in exactly what they were taking part in.
Is, was that like concern for you or.
Erik Wengström: It's something we discussed, but, uh, since we don't, you know, provide, we don't lie to them or we don't deceive them in that sense, we don't.
Benjamin James Kuper-Smith: Yeah.
Erik Wengström: Uh, we never, I mean the, the, the key thing here that people object to when it comes to deception, if, when, if you tell a story and that's not true, we provide some information and that's not true.
And after the study, they can then the participants can potentially [00:52:00] find out that this was, I was actually deceived or someone was lying to me. And then they'll get frustrated and next time they come to the lab, they won't believe what these experimenters are saying to me. Because last time it wasn't true.
And that's not the case here. We did two surveys. Um, there was a matching between the survey.
Benjamin James Kuper-Smith: Yeah. Two
Erik Wengström: surveys. Right.
Benjamin James Kuper-Smith: Okay.
Erik Wengström: I don't think we provided any,
Benjamin James Kuper-Smith: I dunno how you framed it Exactly. I didn't look at all the materials. Exactly. Yeah. Um, yeah, no, I mean about the deception. That's, uh, I once had it, I think this was like very, one of the very first experiments for which I helped collect data as a psychology, like undergraduate.
Um, I, you know, it was like some attention task that's very boring to do, and you just sit there and like press buttons for like an hour or something and then we, yeah, it's fairly boring to do. Anyway, after the experi was finished, uh, the woman said, well, like, she was also like a. A psychologist student, right In an undergraduate and said like, oh, what was the study actually about?
I was like, no, no. [00:53:00] That was what the study was about. No, no, that that was it. She somehow expected that she was expected that there was something else going on that she was not aware of, uh, which is probably a problem if that's the, I mean, I've only had that once ever, but yeah. Um, if that's part of the expectation of going in to doing that kind of experiment.
Yeah,
Erik Wengström: no, it's, it's true. I mean, if you don't believe, uh, what the experiment is saying or you don't trust them, I mean, then yeah, how can you trust the results in a
Benjamin James Kuper-Smith: sense? Yeah, exactly. Okay, cool. Um, but so, uh, as I mentioned earlier, or yeah, as I mentioned earlier, um, I have a few questions about the risk dictator game.
I think the dictator game is well, and it's well known and, you know, well established and people know what that is, but why? Why this, uh, variant of the dictator game.
Erik Wengström: Okay. So this was in a sense, you know, [00:54:00] trying to mimic a little bit more, you know, the situation and, and that we were trying to predict or explain, because what we did was basically, you know, the dictator could, uh, choose, you know, uh, or we framed it in a sense that let's say both, uh, subjects here, they start with the 10 kroners, right?
Benjamin James Kuper-Smith: Mm-hmm.
Erik Wengström: Or 10, $10. That's it. $10. And then if I'm the dictator, I can take a little bit more. So I could take, uh, I could take, um, 200. Say that, but that would mean, or 20 euros or switching here, let's say we have 10 euro now we've got dollars, euro,
Benjamin James Kuper-Smith: and corner.
Erik Wengström: That's
Benjamin James Kuper-Smith: amazing.
Erik Wengström: Yeah, yeah. Sorry, we had kroners. So, but I'll, I'll keep it, I'll keep it in dollars.
Sorry. Or you could translate to euros. So
yeah,
Erik Wengström: we have euros in or dollars in, let's, let's,
Benjamin James Kuper-Smith: you can also just go with the original, if that's easier, I think, isn't it corner, [00:55:00] what was it exactly? I can't remember. Is it 10 times?
Erik Wengström: It's about 10, yeah. Yeah, exactly. So, so that's, it's, so it's kind of simple. Yeah.
But, uh, yeah. So, uh, in, in a sense, you know, what we could do, what we did was, you know, the dictator could choose, you know, to, to take a little bit more, to get a little bit more. But what would happen to the other person is that that person would then risk losing his or her initial $10. You
Benjamin James Kuper-Smith: mean
Erik Wengström: I
Benjamin James Kuper-Smith: can decide.
Then the other person will lose some of the money.
Erik Wengström: So basically you have $10. I have $10. That's kind of the framing of the initial situation. Then I'm the dictator. I can actually increase my payment. But that would impose a risk for you. So if if if I take more, then you might face a risk of losing your 10 euros.
Benjamin James Kuper-Smith: Yeah.
Erik Wengström: And in essence, I could actually take the [00:56:00] maximum amount. That would mean that you would lose for sure. I mean, not many people do that, but that's one option, you know, or some people do. But that's, that's one option, right?
Benjamin James Kuper-Smith: Yeah.
Erik Wengström: And then we had different, you know, benefits. So in the most extreme case, uh, we had that you could get I think 60.
So, uh, if I, if I, if you were to lose your money for sure, I would get $60. Instead of 10. So there were some substantial, you know, gains for me to, to make. So we thought this is, you know, a little bit resembling what, what's happening typically, or the externality part of going out during socializing during the pandemic is that you expose others to risk.
Uh, I might benefit a little bit from, you know, going, uh, to a restaurant, but I also contribute of course to, to, um, spreading or potentially contributing to spreading the disease, making it crowded and so on. Right? [00:57:00] So that's why we choose this kind of risk framing.
Benjamin James Kuper-Smith: Mm-hmm.
Erik Wengström: Because we thought that was a bit more similar to the type of behavior we face or we were trying to predict
Benjamin James Kuper-Smith: about the dictator game.
Um, so I. Uh, it kind of relates again to the question of context and that kinda stuff. So I once did a study where basically I explained to people, um, the dictate game and then they had to make decisions, uh, for practical purposes. I actually explained it to them rather than them reading the instructions and basically they could ask me always questions.
And the interesting thing is like. At least half of the people then said, oh, where does the money come from? Like, I always, I said like, you are given 10 euros in a traditional dictator game to divide it between you and someone else. And half the people they really were like, yeah, but what, like, where does the money come from?
And they said like, no, it's just given to you. Like you can decide. And what was interesting to me was that to a lot [00:58:00] of peop, just the first is that how many people, uh, not psychology students or anything, in this case they were prison inmates, so maybe not the most representative sample, but um, you know, they were really like, did we work together?
Did we put in equal amounts? Like what exactly was going on? What's the story here? It, and how difficult they often found it to just decide in the abstract. And that, that, that did when I was there, make me wonder like how relevant of a measure it is if from at least a large proportion of people, the situation just seems a bit absurd.
Not absurd, but um. As if you're just lacking key information that would probably otherwise dominate any decision you could make.
Erik Wengström: Yeah, and I am, of course, I, I think it's a valid concern that this is a very artificial situation somehow and often in these kind of situations, we would think about, you know, where does the money come from and what's the right or entitlements to these money and so on, how it's, it's created and, [00:59:00] uh, or generated these money.
So that's, uh, that's a natural concern. I mean, in one sense, our paper is, uh, is approved that captures something right. About human behavior or people's behavior in this game captures something about how they, uh, behave in the real world. Are they,
Benjamin James Kuper-Smith: although it's still self-reported, right? Your, your behavioral measures are still
Erik Wengström: Yeah, except from the ones I measured.
Of course. I mean, the number of clicks on the information and so there are some, but, but, but it's clear that most of it is self-reported.
Benjamin James Kuper-Smith: Mm-hmm. Yeah, I mean, I guess for your studies, maybe slightly different because this link between a abstract decision and the reward is kind of almost what you're investigating, right?
Um, so maybe it's less of a problem there than in regular, but to me it was, I don't know, I guess if it is predictive than maybe it isn't that much of a problem as it might seem, you know, it's just when [01:00:00] I was doing, yeah, when I was just doing that experiment, I just kept wondering like, should I, am I wasting my time here because I'm, I'm asking a task that it's just really weird and Yeah.
Erik Wengström: Yeah. No, I mean, it is a concern. We have, I have another old paper on, on the public goods game, which you probably know of is like the prisoner's dilemma, but with more players and more actions to choose from. And, uh, there, you know, it's in a sense it's, it's so strange for some people. That they don't understand it.
A big fraction or bit confused. What's kind of the, the optimal strategy, um, in this game? So a lot of people perceive it very much like, uh, conditional cooperation is the, the optimal strategy.
Benjamin James Kuper-Smith: Yeah.
Erik Wengström: I mean, if you have social preference as it is, but if we explicitly ask them to just think about their own, you know, earnings or their own payoffs, you know, and, and then we ask them what's the optimal strategy for you?
Or what's the [01:01:00] optimal, uh, action for you if someone else, let's say, contributes a certain amount and then there is dominant strategy in reality, right? You should always just don't contribute at all and freer. But most people, I mean, it's kind of con concept of conditional cooperation in the setting is kind of hard.
Grants almost people miss, or about half of the subjects in that study kind of misinterpret this as, you know, the optimal strategy is to contribute as much as the other does. Right. Which might be, you know, yeah. So, so a lot of people kind of get, or there is always this risk when we, when we have these abstract games as well, that people misunderstand or are confused or have misperceptions about the structure of the underlying game.
You know, descentives of the underlying game.
Benjamin James Kuper-Smith: Yeah. Okay. So that was my question about the risk risk dictator again. Uh, by the way, just as a general comment, uh, I thought it was like [01:02:00] methodologically, I liked that like a lot of stuff about the study, like, you know, large representative sample preregistered, you had several measures of all these things link, like linking the different things without other participants being aware of it.
I dunno, I just thought that was kind of a, I just, I read the paper for like, oh yeah, this is well done.
Erik Wengström: Yeah. Thanks to my Coors.
Benjamin James Kuper-Smith: Um, but now I have a very confused question. Question, which is. Do cloth masks cost $25 in Sweden? Because I just thought, so you, the decision was right, like do you keep, was it like you get $20 or you get a cloth mask for 25?
I just thought, what is that mask made? I
Erik Wengström: think they, I think this was actually, uh, well this was the price in April really online for, for this particular, actually this mask.
Benjamin James Kuper-Smith: I mean the, the one literally that, that you have as a photo at the end in the
Erik Wengström: Yeah, I think so. Yeah.
Benjamin James Kuper-Smith: I mean, it's a okay looking mask,
Erik Wengström: but you know, things [01:03:00] that's, uh, things have been developing since, uh, it comes to the mask.
Right? Yeah. There was clear shortage of, of masks back then. I guess it looks like handmade. This one. I know.
Benjamin James Kuper-Smith: Yeah, exactly.
Erik Wengström: It looks not particularly nice.
Benjamin James Kuper-Smith: Yeah, like 20, I mean, yeah. Okay. I was just, I wasn't entirely sure whether that was something you just made up or whether that was, but that's the extra pro.
Erik Wengström: Yeah, I think this was due to supply and demand.
Benjamin James Kuper-Smith: Yeah.
Erik Wengström: That particular time. Point in time. Yeah.
Benjamin James Kuper-Smith: I mean like in Germany at the time, there was, I remember in early April still, or no, late March at least, I was going shopping without a mask because I didn't have one. It was just, I didn't have one. I didn't know where to buy one.
Uh, it just didn't exist really as a thing.
I don't know. I dunno whether this is like a, this is more like a comment of differences. So I mean, like we did a study about COVID and doing risk perceptions and that kinda stuff, and we wanted to [01:04:00] list link risk perceptions to. Health related behavior similar to yours, just, but we had risk perception rather than cooperative behavior.
Um, but what we had was, um, about these protective measures. We had basically physical dis social of, yeah, physical distancing and production of social contacts. And we had like one overall question about hygiene behavior. Uh, ours wasn't as specific as yours, but we just had huge ceiling effects. I mean, I guess you have it too, but less
Erik Wengström: so everyone was washing their hands all the time, or,
Benjamin James Kuper-Smith: I mean, basically we asked like a question was something like, yeah, how much do you, um, stick to the hygiene recommendations and brackets washing your hands, disinfecting, and then later wear a mask.
And like 70% of people said like 95% or more or something. Um, but. Yeah, I guess I don't really have a question here. Um, I guess yours is
Erik Wengström: just a
Benjamin James Kuper-Smith: different,
Erik Wengström: yeah. I mean, [01:05:00] we have up to, I think, uh, the most extreme ones I think is coughing and sneezing into elbow. Yeah, exactly. 70. I think 70%, but for some of the others are much more spread and then we of course make an index out of all of these, so
Benjamin James Kuper-Smith: yeah.
Yeah. Your
Erik Wengström: four, you have extreme answers of one or two, I mean,
Benjamin James Kuper-Smith: oh, yeah, yeah, yeah. Okay. Yeah. One thing that really surprised me is that wears a mask if symptoms. I mean, okay. Point one is, does not apply at all. That might mean people say, I don't know if symptoms therefore hasn't applied.
Erik Wengström: Yeah, you could also think about this issue because in Sweden, masks wearing was not encouraged at that time.
So one way to read this question or misread this question might be, well, I don't go out if I have symptoms, so I don't wear a mask.
Benjamin James Kuper-Smith: Oh, that's the next,
Erik Wengström: because this is the, this is the, the, the, the official recommendation is of course if we should stay home.
Benjamin James Kuper-Smith: Yeah. Yeah. I, I guess I don't have, um, [01:06:00] I dunno, I've kind of run through my list of questions.
Um.
Erik Wengström: Okay. But I can also mention, I mean, relating to your research, I mean that's, we also measure risk attitudes here in this survey. Right? So we could say something about also, not only that there is, uh, not only that there is a relationship between social preferences or prosociality and health behaviors, but we can also, you know, put a little bit in relation to, uh, other types of preferences.
So is that
in
Benjamin James Kuper-Smith: the paper,
Erik Wengström: the risk
Benjamin James Kuper-Smith: perception?
Erik Wengström: Yeah. I think it's mentioned somewhere in the paper. Must have
Benjamin James Kuper-Smith: missed it then.
Erik Wengström: And it's actually, uh, so we, for example, we find that these social preferences are more important than the risk preferences.
Benjamin James Kuper-Smith: Ah, okay. Or
Erik Wengström: health behaviors.
Benjamin James Kuper-Smith: Yeah. I mean, in our study, which we had the problem that we couldn't do that test because we had such seeding effects that.
You know, everyone just said I, I reduced all my contacts and that one we split up in like different social contexts, like [01:07:00] how many, um, the question we had was like, on how many days per week do you have physical contacts with? And then like your family, friends, uh, strangers, like leisure activities and that kind of stuff.
And Okay. From family there was like a, uh, either every day or not at all, basically. Um, you know, either you live with your family and your children or whatever or not. Uh, but for everything else it was just everyone reported zero. Like, so it was just, we, we basically our, I mean, I haven't, I. Well, by the, by the time this is out there will be a new version of our Preprint.
Uh, but basically from what it seems right now in the final stages of finishing it, is that we just can't make that assessment because everyone says they reduce everything. So, which is good, I guess, for society, but
Erik Wengström: yeah, if, if it's, if it's true, if it were only true, but yeah. Yeah. But also I think, you know, sometimes, you know, I've read people saying also that, okay, social preferences are not [01:08:00] enough, and then people have, don't have strong enough social preferences, so on for, to be able to use this as some sort of tool to rely on this and so on.
And of course, I mean, you're right, we need more restrictions at all. But I also think it's, uh, the fact that we need more restriction is not a good evidence that there are not social preferences. Because I think, I mean, a lot of people, you know, especially young ones, you know, they have, um, um, let's say restricting themselves a lot.
And that's not due to themself being particularly afraid. It's probably mostly because of, you know, contributing to, mm-hmm. To the public good or to, to preventing the spread in society. So I think this is the fact that people for more than a year have, you know, uh.
Been, you know, isolating as much as they have been doing and following the restriction as much as they have been doing is in itself, you know, a sign of some sort of social preferences is still being, you know, important. [01:09:00]
Benjamin James Kuper-Smith: Yeah. The fact that there aren't massive riots all the time. No, exactly. They are some, of course, but it's, it's usually a few hundred in Germany at least.
It's usually a few hundred people maybe or something. Um, it's, it's not
Erik Wengström: exactly, it's, it's not the mass movement, you know, and, and, um,
Benjamin James Kuper-Smith: and for example, like in the uk when I studied, in the uk, there were the, like mass protests against them raising the tuition fees for students where there were like hundreds of thousands of people in the street.
We didn't see that for, for this. Right.
Erik Wengström: No. And now, you know, students have been, um, away from campuses and, and most parties for more than a year, and they're still, you know, uh, not going to the streets. So I think that's in itself in sort of an indication that people are, uh, have some sort of concerns for others.
Benjamin James Kuper-Smith: Yeah. Yeah. Or I don't know how much like the, also the, the fears of getting severe symptoms, because I do know someone, for example, who seems to have gotten pretty, a young fit, healthy person who shouldn't really have gotten anything. Got it. [01:10:00] Pretty bad. Yeah. I mean, it's fine again now, but like it took a while.
Erik Wengström: No, sure. I mean, it's a combination. Yeah. But I also, I mean, honestly, I think that, uh, for young people, you know, uh, it's, it's, it's, um, it can be bad, but most likely not. Right. So I think that the social aspects, uh, dominate
Benjamin James Kuper-Smith: that the main concern is not to kill someone's grammar or something.
Erik Wengström: Yeah.
Benjamin James Kuper-Smith: Yeah. By the way, is this, I mean, I know you have the paper on vaccines and or the, uh, that kind of thing.
Yeah.
Erik Wengström: That's the follow up. It's,
Benjamin James Kuper-Smith: is there, I was just curious in general, like whether you have, are you doing more about this, about COVID and that kinda stuff, or is that kind of more or less finished with those two papers?
Erik Wengström: Yeah, we have a third wave paper coming soon, or, um, maybe, you know, yeah, we are planning to do something on, on, on vaccination as well, and hopefully we can get, uh, uh, registered data from Sweden or on whether people actually vaccinated or not.[01:11:00]
So that project, so what we're gonna do there is basically the plan is to do different, uh, interventions, both incentives, but also more kind of nudge style interventions. Uh, and then we can, uh, link this, or the hope is that we can link this to admin data on whether people actually vaccinate or not.
Benjamin James Kuper-Smith: But could you just ask them so
Erik Wengström: we can get,
Benjamin James Kuper-Smith: I mean, that seems like a question that's easy to ask
Erik Wengström: and well, returning to, returning to the issue before, right?
I mean, we'll of course ask, do this before they vaccinate. So we'll do this as soon as, as we can as soon. We are now pending ethical approval. So as soon as we get an ethical approval, we'll uh, launch this study and then we'll follow up afterwards to see whether they're vaccinated or not.
Benjamin James Kuper-Smith: Okay. Well,
Erik Wengström: and of course one benefit is that, uh, as you said, you know, these were mostly hypothetical, but that would then be, you know, a real decision.
Right.
Benjamin James Kuper-Smith: I can differentiate [01:12:00] between which vaccine people are gonna get and that kind of stuff, because that's, at least in Germany, I dunno exactly what it's like in Sweden, but in Germany, there's this big thing with AstraZeneca being available, not available on and off or
Erik Wengström: something. Yeah. I think, but at the moment, this is gonna be targeting the ones below 65 or 60, depending on how fast, uh, the ethical approval goes.
But, uh, uh, those are at the moment, uh, not offered AstraZeneca vaccine, so
Benjamin James Kuper-Smith: Ah, okay. All right. Yeah, it's only the
Erik Wengström: ones,
Benjamin James Kuper-Smith: and it's just different because I'm officially at a hospital, so I mean, I'm, I, I think I'm higher up than I basically should be because I, I don't have really any clinical contacts or that kinda something, but I'm still at a hospital.
Yeah. So that, but I think for us, they said something like where a hospital, even if you're under 60, you can still get AstraZeneca or something like that. But like, I'm just gonna wait for the people who actually have patient contact to get vaccinated first. Yeah, exactly. Stupid for me to get vaccinated.
Makes sense. [01:13:00] Um,
Erik Wengström: to, to, to give it to more prioritized people first, of course.
Benjamin James Kuper-Smith: Yeah. Uh, but can you, through the, through those records, can you, do they tell like when the people were vaccinated and that kind of stuff so you can really,
Erik Wengström: yeah, so in principle we get the when and uh, where they were vaccinated.
Benjamin James Kuper-Smith: Okay. Well, I hope
Erik Wengström: so. Then we can,
Benjamin James Kuper-Smith: I hope that not too many under 60 people in Sweden listen to the podcast. So otherwise this will ruin your sample. Luckily, I'm not that big yet. It's, it's not a, I don't, I don't, it's, it's common. Yeah, exactly.