 Good morning. Good morning to all of you. First of all, I would like to welcome you to the University of Pumperfabra, the young university of Merspargic. It's a very big English writer. On behalf of the director, I welcome you to the 26th opening lecture in UPF Economics and Business. I want to excuse the director for not being here, for some major reasons, as he impeded him to be here. Let me introduce the table. My writer have our distinguished guest today, Professor Ernst Ver from UBS International Center of Economics and Society at the University of Zurich. The title of the lecture looks exciting and I'm sure we'll have a good time here. He will speak about one of these black boxes, preferences. This is interesting for both economics and business. Two sites on my left. I want to introduce Jose Urbieta, CFO of Laboratories Dr. Esteve. This is our sponsor. I thank Laboratories Dr. Esteve for sponsoring this lecture uninterruptedly over all these years. Some of these years have been quite difficult, years recently, but this partnership has been here always. And right on my left, I have Professor Rosmarine Agel, who will introduce our lecturer today. So let me continue now a couple of minutes in Catalan for the audience. This is my first inaugural lecture, as of now. It's a pleasure for me to be here. This is the Auditorium Band. I had been to Moscow, the other Auditorium Band. It's a very important event for our center, for our department and for the Faculty of Business Economics. It has been celebrated without interruption since the year 1990 and for which many conferences have been held, in economics and business. I would like to first thank Professor Vicente Orton, who will take care of this place for the last five years. I believe that all the students, the professors, the staff, the services administration are grateful for their dedication and their effort. And the other students have contributed to create a program that is truly enviable and a challenge for those who are in my charge. Our degrees, as you know, are valued as first in many rankings. There is a lot of demand, the degree of detail, our degrees grow year after year. It is so difficult to enter some of our degrees as it is difficult to enter medicine, as I always say, it is very difficult to enter medicine, but our degrees are so difficult to enter as medicine. We can offer up to almost 900 practices, remuneration of companies, more than 300 international exchanges, so many students who come here as they go out, with all the value that this means. And all of this is a moment to continue to plant direct genes. As I assume a new career, I always say, let's see what we have to do, what we have to assume. We have fulfilled 25 years as an OPF and also this center, this faculty. And so we enter the 26th year and then we leave youth and go to maturity, as we have to do. We have very important challenges. I think the values and the mission of this school are always the same, to promote talent and use the resources we have in the best possible way. I would define it, I would summarize it like this. And I always say to my students, when they take their place in the field of studies, it's a miracle, it's a public university and they offer a license to the most outstanding professors at the world level, in the universities, to outstanding associate professors in the professional universities, a multilingual training. I don't know, it's really good that we have been able to do this. Promote talent and bring the best students and we want to accept it. And we want to see students dedicated to academic work that are still here, that we can offer them the best knowledge, the first line, the frontier, that they can also participate in all kinds of initiatives, associations that are still there, and that they really create an environment where we can feel like a community that we are all fighting for the same. I think that the society, judge and value, the Pompeu, always thinks it's a bit special, but we have to work on it. We think we all benefit from it. The students, mainly, think that this is a special center. But we have to keep working on it. With the demand of rigor, I think we all benefit from it, that the value that the society has of the Pompeu, that it works, and it works very well and hard. The methods of science and innovators, too. Professionals of the first line, as I said, producing research of the first international level and outstanding social professors in professional fields. Personal support, too. The first line of so much information, as the professional supports the science. I think we have an immeasurable person. But all this, I think, is at a really good point, and I'm really proud of it. We have some very good initial questions, I think, to make them. B, now I would like to introduce Professor Rosmarinagel. Professor Rosmarinagel will introduce our lecturer today. And at the end, she will also chair the round of questions. So, remember, you have to do that, too. Okay. Professor Nagar, floor is yours. Okay. It's a great pleasure. It's a great pleasure to introduce Ernst Ver. Ernst Ver is an experimental and behavioral economist, like me. And actually, he is one of the most famous experimenters. And he has won many prizes. He has positions in NYU. And I mean, there's so much to tell that I thought I pick one area. And that is, I want to tell you in what areas in experimental economics and what he does with experiments and behavioral economics in very different areas from a very interdisciplinary perspective. So we met 20 years ago when our field was basically small and unknown. We all knew each other. I was a young assistant professor or postdoc. And Ernst Ver actually, as all the people of his generation, went out as a theorist. And then, with his paper on labor contracts about the question of efficiency wages, he was told, well, why don't you also test this? And he said, well, I can test this. I can test it with an experiment. And that's basically how he started his career as an experimenter. As many others of his colleagues were real theorists of experiments. So he used for many years lab experiments. And so in the 90s we had a great data set of experiments and the big dispute was is game theory right or wrong? And there were those who said game theory is wrong and it's clear from the data and with the same data the theorist said the theory is right. And then at the end of the 90s Ernst Ver together with Klaus Schmidt basically found a way to show how the theory can in fact be right. And this is done with incorporating social preferences in the utility function. And you as first year students have all basically heard about this, maybe without the name because you have studied the ultimatum game. You have studied the ultimatum game with, first with behaviors or online and then you studied the theory. The sub-game perfect equilibrium says you should give nothing but you will get something as a responder and this can be reconciled with the theory of social preferences that you will not be so greedy and give nothing to the others and you will really as a responder be not accepting everything. But Ernst did not stop with just doing experiments he thought maybe humans have a brain and this brain also decides how we do our decisions and therefore he went to the field of, which is now called or he started the field with others which is called Neury Economics and he actually published this book which you can get on amades.com and there you bring together the brain with money and in order to bring together the brain with money you need also machinery very expensive machinery and you put people in a scanner which you might know from the movies and there you can actually bring together decision making together with biological data and you can have many other possibilities to bring basic biology together with decision making. So this has been started at the end of the 90s and the beginning of 2000 it's still a fringy subject as experimental economics was a fringy subject but Ernst did not stop there he thought the brain well that's still not the basis of everything let's go deeper and he went to Geno Economics and he looked at the genes well he's a little bit stuck there he told us just that he started this project in 2007 or 8 even with somebody from biology here in our university but the problem is that now everything gets more expensive and everything gets also more data you have instead of 20 million bits with not bits in the brain what is it called neurons you have now 1 million snipes so the data analysis is almost impossible our typical data analysis so he's still stuck in there but there is more to come that is the big question is what does this all mean and to go into the brain and to be only stuck in the lab there is the question of external validity and here the question is well how can we transport all our findings in the field and have similar questions and we now go on the street or Ernst went to Africa and there he has now a science paper which you can immediately download he has a very interesting science paper where he also applies comes to policy application I mean I cannot tell everything and I will probably give Ernst another world but you can download it and you should download it maybe after the talk that was my Instagram introduction of Ernst Fair thank you Professor Nagel so now Professor Fair the floor is yours microphone so thanks a lot for telling the people the portfolio of my activities so it's not a straight line so I'm still doing everything so and this is what you have to do when you are a scientist when you are a portfolio of projects so some materialize and become successful and others don't and that's just the simple truth about risk taking so to speak scientific endeavours are risky and that's why it's sometimes wise to have many different acts in your portfolio now what I'm doing today is I'm giving a talk on the Gustavus Estis Portandum that almost 40 years ago there was a famous paper published by Stigler and Becker two famous Chicago economists with the title the Gustavus none Estis Portandum you can't quarrel about tastes and that's what when I grew up in academia what I learned so teacher usually writes a utility function on the blackboard or on the PowerPoint slide these days and that's the function that's assumed to be maximized and it's given and if you ask where it comes from I mean we have nothing to say basically where it comes from it's an assumption we assume plausible preferences and then we do our analysis now I found there's always this factory because I thought they have to come from somewhere and we should study that but at the same time we didn't have the tools to study them we didn't know how to measure them we didn't know how to infer to really prove that something has a causal impact on preferences we didn't have the empirical tools to study causality so we liked basically all the ingredients to study preferences and what I'm going to today is I'm showing you that we have now all the ingredients so we can start and this is why I call this the emerging science of preference formation and I give you a short outline at the beginning what I'm going to do I will first talk a little bit about the stable preference assumption in economics how sociologists view this and then I I ask the question how sound is the economics viewpoint and when I say economics you see we are not a monolithic community but there is a prevailing view there's always heterogeneity fortunately and it's through heterogeneity progress occurs because some things that were deemed unimportant suddenly become important because somebody can show that it is important so I talk about this initially and then I give you just examples from my own research and if I have time also I give you an example from somebody else's research and I ask the question whether aggregate asset prices affect risk preferences this is typically ruled out in finance models then the next question I ask is does business culture in banking affect preferences for honesty an important question also and then I ask another question which is inspired by the nuclear catastrophe in Chernobyl namely whether such nuclear catastrophes have an effect on people's preferences and the final question here is which excites me a lot is whether we can deliberately shape people's preferences or our children's preferences for example so is it possible for a human being more or less pro-social more or less prone to take risk more or less impatient I give you my prejudice yes I believe it's possible but we know next to nothing whether and how and so on and but it has tremendous implications for educational policy and it has broad implications if you can shape human personality the deep parameters of human personalities that opens up a whole new avenue of research and also of course of ethical questions so now let me start with a few remarks on the assumption of preference stability when I started my research some five or six years ago I wrote an email to one of my friends from sociology and I thought well sociology I remember from my own studies when I studied economics I had to take a sociology course and when I took that course it was clear that they and that was what I remembered it was clear that they assumed that somehow society shapes the individual individual is not a blank slate I mean it's shaped by by social forces and then I thought well let's send an email and start don't replicate things already solved maybe they have already solved this problem partially so let's start where they have been gone so far and I asked him can you give me literature about the insights that are have been acquired in sociology about how society shapes individuals preferences and here is the answer he gave me now this is no here it is so here is a quotation I like this it's beautiful I think so the assumption that society shapes individuals preferences clearly concerns one of the core pillars of sociology but it is not easy to suggest any literature to you it's almost too fundamental for that like asking economists to suggest some tests on the importance of choice almost all sociologists take it as obvious that individuals preferences are formed by society and that society so to speak exists within persons but we just assume that basically he is saying and we don't know anything about it and so there is a science here that assumes something and is taken for granted but doesn't deliver theories or evidence on it it's just an axiom basically taken as an axiomatic departure now let me contrast this with the famous Becker and Stigler view in economics which was written down in Bustibus known as the Sputandum where they write tastes neither change capriciously nor differ importantly between people on this interpretation one does not argue about tastes for the same reason one does not argue over the Rocky Mountains both are there will be there next year too and are the same to all men pretty strong statement so and well Becker these guys were known for a clear language and this is clear language they are like the Rocky Mountains they are there in hundred years and in a thousand years they are there like the Rocky Mountains and notice they assume that it's not they not just assume preference stability they assume basically how much you need they assume it's all the same now we know in the meantime there has been a lot of research on preferences that this is not true that preferences are the same to all men I mean we have lots of evidence that preferences when you look at risk preferences at time preferences at social preferences or just think about your preferences for chocolate some of the people in the room like it and some don't like it so clearly there is preference heterogeneity now that's falsified but what about the stability okay so preference heterogeneity is falsified there is now compelling evidence in many so I self-servingly cite one of my papers but I could have cited also other papers that there is a lot of preference heterogeneity in all kinds of preference domains now but it's still true that when you study economics that 95 or even 98% of what you do is going through thought experiments or through exercises that assume some form of preference stability for example when you want to investigate what's the impact of tax changes, cost changes, price and information changes you typically assume a utility function that's given okay so all the change all the action is in the changes in the constraints in the budget line in the slope where it is and so on or in the changes in property rights and the contractual environment now there is an implicit assumption here and one could distinguish between a strong and a weak preference assumption stability assumption the weakest, the strong assumption says simply like back to Stigler they're stable the weak assumption says they're stable for the problem under consideration so if I'm looking at how does a tax change affect consumption in Spain then this implicit assumption is they are at least stable for that period that I consider in my analysis okay and that is a reasonable assumption I'm not at all at odds with that assumption that is a very powerful research program and has given us many many insights however I want to argue today that there are other important phenomena that we leave out of our picture if we assume that preference if you always assume that preferences are stable in particular in the long run but also maybe in the short run actually I show you examples today that preferences can be pretty malleable in the short run even okay and well back to Stigler when you read their paper well to read they say basically I have it on my slides but I went over it too quickly they basically say if you if people don't know how to explain a phenomenon then the discussion often box down to assuming some obscure change in preferences and then that explains the phenomenon of course and what they do is kind of a lazy attitude which is best illustrated by the following example imagine you are at the petrol station and you see a guy drinking a liter of petrol now asking yourself as a social scientist how can I explain this well your friend tells it's very easy to explain he likes petrol so phenomenon explained now here you see the challenge here you can explain everything by assuming particular preferences and you have to put some constraints on these explanations otherwise if you don't put constraints on the explanation you can explain everything and therefore you have explained nothing and that was traditionally the main criticism of of targeted towards people with preference based explanations ok so but let me here are a few remarks from my side on the assumption of preference stability so first it's important to recognize it's an assumption it's not a fact and there are so many things that are assumed in all the sciences by the way not just economics there are so many things that are conventional that you can if you are a risk taking entrepreneur you can go against these conventions and sometimes you may be very successful in breaking them because they are sometimes just there because they are there and I'm claiming that this convention now should be broken basically it can be broken in a productive way with a productive research program and so it's basically useful practice has been in the past and the accusation has been when you try one can explain everything if one involves changes in preferences completely agree with that completely agree with that however it's too easy to explain changes in behavior by changes in preferences this is a bad argument if something is easy to explain that should not be an argument against it you see but sometimes you hear this changes in the end the only thing that counts is it empirically true that's the only thing that counts so if something is easy to explain well that's good so that should not be an argument but sometimes you hear it in particular from theoretical corners because the people like difficult like you see they like non-obvious explanations the distance between the assumptions and the conclusions has to be large to be such that you so this is the second argument is really bad but the first argument has validity of course so now what are the problems with that argument here is a problem the problem is that I'm claiming and I'm not the only one who claims that one can explain everything if one involves changes in the environment as an explanation now when did I first encounter this argument so I have some papers with my co-author Klaus Schmidt he's a good game theorist applied game theorist and when I met him the first time he once told me look give me any real world contract and I will design you an extensive form game such that that contract is optimal given the game but what does that say I can explain everything assuming the right moves the right set of moves the right information structure kind of asymmetric information and so on and actually he's not the only one who did this who has this view that you can explain everything without you can leave references untouched assume they are constant but you have so many degrees of freedom that you in fact can explain everything and so on who articulated that in a very explicit form is John Sutton John Sutton was president of the European Economic Association he's a great applied game theorist game theorist also a researcher in the Astral organization has published many important papers and he gave his presidential address in 1990 and I still remember it it was a fascinating talk and what John Sutton said is the following let's quote him consequent deliberation so he's commenting on the literature industrial organization in the 1980s the consequent deliberation of a richer class of game theoretic models has been remarkably successful in one respect given any form of behavior observed in the market we are now quite likely to have on hand at least one model which explains it and notice explains is in quotation marks here in the sense of deriving that form of behavior as the outcome of individually rational decisions new explanations of this kind have been used across the entire range of the subject from predatory pricing to vertical restraints our toolkit has been enormously enhanced and then he goes on and characterizes a research program that escapes so to speak these problems but at the end in the middle of his talk he makes the following comment in explaining everything have we explained nothing what do these models exclude so what I'm saying here is there is no methodologically superior status of explanations based on stable preferences and assuming so to speak that the action is in the change in the constraints versus the other way around we have to treat them on a par there's no methodological there's not one thing superior to the other it was just not we were just not able to put constraints in the past we have not been able to put constraints on preference based explanations such that they lose the arbitrariness so that is the comment that I'm making here and actually today we have everything we need we can measure preferences either in surveys or in experiments, behavioral experiments we have all the causal econometrics we need to find the natural experiments that trigger maybe a preference change that lets us examine a causal effect of a societal change on preferences and we can even design field experiments or lab experiments as you will see in a minute that allows us to examine this question now let me go on to my concrete examples that I'm putting forward here are my examples well first I advertise this paper this is my programmatic paper I wrote four years ago in the economic journal in a special volume it's called tastes, case and culture the influence of society on preferences basically lays out the arguments that I've plus a little bit more than what I have told you already but what I want to do in the rest of my lecture is I want to give you examples here are the examples first I will speak about aggregate asset prices the extent to which they affect traders risk preferences then I talk about banking culture then I talk about general bill and finally about how we can change children's prosociality and the first three are based on my own research the last one is based on research done by Armin Falk at the University of Bonn who has conducted a beautiful field experiment where he shows that yes indeed you can basically make children nicer people by treating them appropriately okay so do aggregate asset price movements affect risk preferences why is this an important question well here is basically why it's important some of you may have heard of Bob Schiller who he won the Nobel Prize a few years ago and this is the graph from his Nobel Prize winning paper and what you see on this graph is the following you see on each of the graphs two lines the broken line is what standard economics what kind of the prevailing paradigm in finance predicts stock prices to be in the long run well that view assumes that stock the price of stock is just a discounted present value of dividends that's paid from the stock and he did the tedious task to compute that historically exposed and so what he finds is when you look at this flat line here the flat broken line that's what asset prices in the U.S. should have been if they would have been rational okay now the actual what you see is that actual asset prices they they show huge volatility they show huge volatility and note this I mean they are they deviate from the broken line sometimes for a decade or more this is not just a short run phenomenon this is really prices are wrong sometimes for years and the question is why so this question has bothered financial economics now since the paper has been published there has been many methodological contributions did he measure it right and so on but in the end I think many commentators have shared the view that stock prices are too volatile to be justified by subsequent dividend payments that was the question in the paper and it and the Nobel Prize committee was also convinced that it was a fundamental contribution to finance and economics anyway so why do we see this huge volatility now there are of course nice stories that you can my nice models that exist models of bubbles that can rationalize volatile stock prices the question is whether this is the whole story and that's the what we did in this paper published actually this year in the American economic review so what did we do well if you want to study the impact of a boom or a bust on preferences you face almost insurmountable obstacles why because I mean so many things simultaneously change when asset prices go up we decided there's a whole many individuals become richer our expectations change our expectations about means and variance may change and on and on and on and the same for a bust so it's almost a hopeless exercise to to compute so to speak from observed data in the asset market about underlying preferences are because you have so many unobservables that this is almost hopeless so we we chose a different strategy and our strategy imports a technology from psychology which is called priming what is priming priming is the activation of certain mental concepts so I can prime a memory for example I can make it salient in your mind priming is about making things salient in your mind okay and that's what we did here so we we rendered booms and busts booms and busts mentally salient in people's minds and as you will see the way we did it keeps everything constant their wealth is constant their beliefs are constant so everything else is constant so if they change their behavior in response to our prime it can't be through a change in constraints it can only be through a change in preferences that's the claim so and here I would argue that if you even the mere priming and you will see it's totally innocuous what we did if even the mere priming of a boom and bust changes people's risk preferences how much more likely it is that a real boom or bust which really gets you in the your mind is really in the grip of a bust if you have assets and they lose value how much more likely it is that a real bust will affect your risk preferences if I can even produce that effect already in the laboratory with a priming technology now here is what we did this is now your end experiment you see this graph it looks stupidly simple so it's time and asset prices and after that we ask a few questions imagine you find yourself in a continuing stock market boom would you buy or sell particular stocks investing gold or other precious metals deposit part of your assets on your savings account etc etc we put people in the mindset of a boom and they're randomly assigned to this condition the other 50% of the people are put in the mindset of a bust it's as simple as that it looks really simple exposed but you people forget how much thought it takes to come up with this ex ante it's like always when the proof is made and then it looks simple but you have to have to speak to come up with this ex ante to say it's just simple anyway so who are the subjects here the subjects are not students the subjects are financial market traders so in syria we have every year financial market fair we're the most the newest technologies financial technologies are presented in computer technologies in financial markets and so on and syria is a financial center so we get lots of the tens of thousands of people who work in the banking sector who come to this financial fair and they are our subjects so these are not so to speak non-specialists these are specialists okay now what do we do after they are primed they do a simple lot they participate in a simple lottery task they get 200 Swiss francs they can keep everything in a safe account that means they go home with 200 Swiss francs or they they can invest any of the 200 francs in a risky asset and the risky asset has a 50% probability of winning per invested Swiss franc 2.5 Swiss francs or zero so basically what you have here is the value of one Swiss franc 25 a 25% return rate on this risky asset which is pretty high so you would just actually this is per se an interesting question why do these guys who earn such a lot of money you see these guys earn 150 plus it's a thousand or 200, 300,000 these guys if you work in the banking sector in Switzerland at least they earn a lot of money so this is peanuts for them so they basically you could make the argument they should put everything in the risky asset which is not what they do typically because humans are not as theories sometimes suggest but the interesting thing here is do they invest more or less in the boomer or the bust condition that's the interesting question and since they are randomly assigned everything is kept the same and notice there is another important feature here the blue, the yellow and the red ball so to make it really clear to the most stupid guy that there is a 50% probability we had these balls in the box we covered the box and then some subjects after the experiment puts his hand in the box and takes out the ball and it's clear that it's 50% there's no doubt about that so you can't say their beliefs are somehow affected by the boom and the bust they cannot believe that the probability to win is 60% because it's so transparent here and so we control for everything we control for beliefs in particular for expectations so the only channel left if there is a behavioral change is preferences we also have an ambiguity task I skipped that for reasons of time and I just show you the results here are the results during the the light bar shows you how much they invest in percent of the 200 in the risky asset and the dark bar is what they in the boom and the dark bar tells you what they invest in the in the bust it's a huge difference it's significant in the risk condition so here it is it seems that risk preferences are affected by the mere priming of a boom or a bust now what's interesting here is what we find is in addition that when we prime a boom or let me put it the other way around when we prime a bust they are more fearful we measure self reported fear and priming a bust puts them into a kind of an emotional state of fear the mere priming imagine what a real bust does and so we we put this fear measure into the regressions and yes it explains part of the effect the large part of the effect so our assumption was well maybe it's the emotion of fear that's triggered by a boom and bust that the emotion of fear varies across the cycle but this fear is of course a correlate of our manipulation we cannot causally interpret it so what we do in addition then is we conduct another experiment in which we threaten to impose an electrical shock on subject's hands while they are in a task where they can invest money and notice the electrical shock is completely unrelated to any financial monetary outcome it's just an emotional trigger that we put them in a state of fear very reliably you can believe me because the shock can occur in any second every second during the trial and if you had a shock it can appear again so it's not that you are freed so basically so you are in this state of fear and now we we use every subject as its own control because in some conditions they get a very weak tickling shock so it's I mean if you are in this experiment you are reminded of the condition in which you are in the weak shock or in the strong threat in the weak threat or the strong threat condition by a reminder shock at the beginning so you get a reminder shock you know in which condition you are okay when it's strong and when it's weak you are in this state of no fear of fear and what we find is that people exogenous fear induces less risk taking and again it can't it can only be a preference effect because it can't be it can't affect beliefs it can't affect all these other multitude of factors that could be prevalent in a real world market now you might say well maybe these were the answers among the financial traders the real experience guys they would not behave like that well it's exactly the other way around the more you have experience in financial market trading the more you are prone to this effect we have also questions where we ask your financial expertise so how much do you understand about financial markets the more you understand about financial markets the more if anything you are prone to this effect so you can't say these are just people who don't understand what they are doing apparently they are I mean the effect is not really big the difference and it's not even significant but if anything it goes in the wrong direction you see you can't argue it's the inexperienced people who don't understand how markets work no it's the experienced people who understand how these markets work who are prone to this effect so this is my first indication the feedback loop from a societal phenomenon aggregate asset prices two preferences and it's easy to see that in the boom if I become less risk taking less risk averse then I take more risks so it amplifies the cycle and the same in the bust this is the first example second example is the following does business culture affect bank employees so we all know there have been lots of banking scandals the collective fines that have been paid by the banking industry go into the multiple billions even for single banks for example if you take city bank over 201 to 2115 it's probably in the range of 20 billion in total if you take UBS if you take Credit Suisse if you take other banks these banks had occurrences where scandals broke out libra scandals for an exchange scandal, tax evasion blah blah blah we all know it so and that led to the to the well here you see Sharon Carville who created the loss for associated generalo on our authorized trading for about 5 billion euros and when he was asked why he did it he said well the culture of the trading room was to make as much money as possible as quickly as possible and so they I mean you could say well this guy has every reason to say this to excuse himself but many other people say this too you see it's not just those who finally go into prison who say this and so even the economists who is not known to be a left-wing journal has a story about the rotten heart of finance in 2012 so there are many commentators who think there is something wrong with the business culture in the industry and so we thought let's test it and it's an important question because if business culture now the question is what is business culture ok what I mean with business culture is these informal unwritten rules it could also be formal rules but the informal rules are typically much more important these informal unwritten rules that guide people's behaviors throughout their everyday throughout their work days and preferences are part of that culture they are not the only thing, beliefs about how others behave are also part of the culture of course so it's preferences and beliefs and the concept of social norms is a complex concept it all plays a role but we are interested in the role of business culture in honesty, to what extent is business culture has business culture an impact on the honesty of bankers and so we had we know many people who are in the banking industry in Switzerland, outside Switzerland we had contacts to German to US banks so we never made clear where we did this study but we had access so the interesting thing is you get access the banks are pretty they want to solve the problem you see they are not just the bad guys they want to solve the problem and solving the problem means wiping out these cases and so they gave us access to their own employees such that we could run an experiment in their company and so how we did this, well the research design here is built on work by Akalov and Trenton who have this idea that each of us basically fills out different social roles actually that's a very sociological concept so as a family father different norms may apply compared to when you are in your soccer club compared to when you are among friends compared to when you are among your work colleagues and so basically because different kinds of preferences are associated with these different roles it may be possible to trigger these roles through priming and then you may be able to have an impact on people's behavior so what we did here is we just primed professional identity in bankers these are all banking employees half of them is assigned to the professional identity prime and half of them is assigned to the leisure prime and again when you see the prime you think it's totally it's so innocuous you can't believe that it has an effect so here is the banker identity prime we tell them well this is a survey on XY I forgot what we told them but then we have this innocuous set of questions that prime their professional identity we ask in which industry do you work at which bank do you work what is your function in this bank how many years have you been working in this bank we just make it salient that you are a bank employee and in the other situation we make it salient we ask you about your leisure activities so it's your private identity and your banker identity so to speak that is primed now the question is do you behave differently after such a prime and before I tell you what the behavioral task let me tell you a little bit about how do we check that we prime the professional identity because I mean it's our assumption that we prime this but you have to prove it in order to publish your paper you have to prove it the word in psychologies you have to do a manipulation check and that's what we did and here is our manipulation check they do a word completion task the word was you can complete it with stock or clock or ochre is broker versus smoker money versus honey so it's it's a very nice word completion task and believe it or not what we find is if we prime the bankers professional identity they say more often stock and broker and money and bond the subjects so we know yes we did prime the professional identity then the question is how do we measure their honesty and there's a there's a beautifully simple game available to measure honesty you can basically play that game at every dinner conversation in a sense when you have somebody who helps you in our case it was you can do it with rolling a die you see why but in our case it was a little bit different so they had to to throw a coin ten times and when heads comes up they get $20 and when heads does not come up they get nothing who observes what they throw nobody so what's your incentive if you throw ten times a coin okay so basically you have an individual incentive to lie but in the aggregate I know what the aggregate outcome is if the population is honest so and I can under very mild assumptions I basically can compute the rate of the fraction of individuals who lied and I can compute the lying rate okay so it's a very nice paradigm because it has this gray zone and you can do ten times well only four let's say six okay so the temptation is here to lie and we played this game actually we played this game not me so my co-authors on this paper they played this game also in a prison with prisoners and what they find is the more the prisoners lie the more severe is their crime for which they are incarcerated this game has been played in India with other people where they are able to measure to what extent milk traders cheat you in India they put water into the milk so that they can sell more milk and what they find is the more they cheat in the game the more they cheat their customers so the task really has external validity from several other studies okay and so we did it with our bankers half of them were primed leisure and with a banker prime and here is the here is what we find so what you see here is the light distribution is the binomial distribution if everybody is honest that should occur at the aggregate level okay now what you also find is there are a few guys here I should say ten times I have ten times hats okay it is unlikely it is kind of above what should be expected and you have here a few guys who shift in this direction but overall this is the most honest distribution I have ever seen these are the bankers so we have done this with many different kinds of subject pools and I tell you the bankers in the control condition they are the most honest guys contrary to widespread belief okay now what we did also is we have this other condition the professional prime condition what we find here look there are quite a few more guys who say ten times hats now and ten times eight is also very popular so eight times hats is also very popular there is a clear shift in the distribution to the right and if you look at aggregate cheating rates then the cheating rate is basically not significantly different from zero and so the number of successful coin flips basically reported is this whereas in the banker prime condition we have many significantly more successful coin flips suggesting that there is a shift in honesty so one journalist said bankers are honest unless they are reminded that they are bankers now we have to keep in mind that I mean it has been used for very polemical arguments this paper but that's also why I point out that the bankers are the most honest population I have found I mean we have no representative banks here we have to be careful it's a banking industry in one country it's unclear to what extent we can generalize but it kind of lines up with what many people believed about what's gone wrong in parts of the banking industry and from a scientific viewpoint the nice thing is that it's very hard to assign the change in honesty to anything else but a preference change so we have several controls I don't go into the details we know that people have a preference for honesty we know that the economic man is not so if I do this experiment you always get many many people who are honest okay many many people so if they would just be money maximizers you would have all people ten times ten but that's not what you observe actually it's surprisingly honest is when you but the surprise is always relative to reference point relative to what you would expect maybe when you have studied economics or finance but it's nevertheless a shift in preferences okay so this is the second study now let me come to an end I make this very short the example of general bill and pro-sociality case the example of general bill I stumbled on this data because some people came to me and told me what they do and they this is a data treasure what was available here the data look as follows so we have data on preference measures from the Ukrainian population these are representative data these preference measures have been taken in 203, 204, 205, 206 and so on the general bill accident was in 1986 and what we know is that in the vicinity of the nuclear power plant there was severe damage also health related damage so between 3 and 5% of the population are physically and in terms of health affected by this accident but other 95 to 97% of the population that's known by now didn't basically get much noticeable radiation fallout they in terms of you could express it in terms of annual background radiation if you live a year longer that's what most of the people experience so they are in terms of being exposed to radiation they are as if they are a year older so they consumed so to speak one additional year or 1.5 additional year of background radiation which is nothing in terms of health consequences it's really nothing and so we look at these people who have not been physically affected we can show that there are several articles and the nice thing from a scientific viewpoint general bill accident is of course bad but as a scientist you are looking for natural experiments and the weather condition randomly distributed the cloud over Holo Kerin and that's what we can exploit we have random variation of random variation radiation fallout and so this random variation helps us to identify the causal impact of radiation on preference measures we have time risk preference measures we have measures about political preferences and what we find is that people who have been exposed to more natural background radiation physically imperceptible background radiation they become more risk averse they become more impatient they become more politically conservative they like the Soviet Union more they like planned economies more and so on so it's pretty pervasive the impact and we ask ourselves why does it happen because it's not a physical effect and here is our preferred story why it happens these are just the measures we are showing concretely this quotation basically tells you implicitly the story from Richard Wilson professor of physics at Harvard University who said the worst disease here in Ukraine is not radiation sickness except for children the physical effects are not easy to measure and actually we know for 95 to 97 percent of the population they are in existent the truth is that the fear of Chernobyl has done much more damage than Chernobyl itself and the story that we tell based on data in the paper shows the people who are more affected cycle let me put it differently people who experience more background radiation they observe more public signals of effectiveness what does a public signal mean well the police is telling you you are not allowed to grow your own letters or whatever in your garden okay you have to swallow pills uh your children have to swallow pills so there are many signs of public effectiveness here that create the impression that you are affected that you might have gotten something and the problem is radiation sickness is something that can break out in 20 years you see it's kind of this uneasy feeling the fear might be something in my body that might be dangerous for me and we believe that it's this fear that generates higher risk aversion higher discounting of the future and more conservative political preferences third example very shortly the final example uh from the Armin Falk he did a randomized controlled trials to take socioeconomically disadvantaged kids randomly assign half of them to a treatment condition I tell in a minute what it is and to a control condition which means nothing is happening and the treatment condition is this you assign a pro-social caretaker to the child the family has to agree uh and the pro-social caretaker is typically maybe something like a 22 year old student who meets the child every week for an afternoon and is here for the child helps with homework helps with homework goes to the cinema may go to the park it's just a caretaker the preferences of the child are the important things so the caretaker tries to to care for the child and he measures social preferences so their pro-sociality before and after and what he finds is after a year of pro-social caretaking experience the children become more altruistic they trust more in strangers and uh they are on average more pro-social and what's very interesting is that in the socioeconomic disadvantaged families on average the mothers are pretty selfish and the children are selfish in the controlled group so you have a strong intergenerational linkage in the treatment group after a year the mother is still selfish but the child is pro-social and to the same degree pro-sociality is at the same level than pro-sociality in other social groups so I find this a very stunning experiment that shows you yes indeed we can do pretty fundamental things when we take it seriously that preferences are something that's malleable that can be changed in the short and in the long run we have additional policy tools available that we have to evaluate and that all together brings me to an end so I hope I have convinced you that preferences are an important object of scientific inquiry that time risk and social preferences were shaped in an important place by social and economic factors factors that are affected by asset prices, business culture, education I didn't go into that ethnic and other conflicts so we have this increasing evidence that ethnic conflict wars and so on have an impact on preferences if preferences are endogenous they can and sometimes probably should be targeted by economic and social policies and so I conclude my presentation with the Gusti Busesti Sputandum we must call about preferences or with President Obama we could say yes we can and it's exciting thank you very much it was a very interesting talk bringing broad aspects of preferences we have now time for questions any question you have one? let's start here I just wanted to ask you you say that the results of your research may suggest that you can have an influence on economic policy by affecting preferences what about the ethical implications of this because somehow you also could like distort preferences in some favor given type of policies are there ethical implications in affecting preferences in some way or the other let me say something on this I have strongly argued that preferences are endogenous now let me qualify that statement if we knew the determinants of preferences for a complete picture we could write down a model that takes all that into account and if we had that model I give you an example let's assume that framing effects how you present the problem shifts preferences if we would know how people respond to shifts in frames we could put that into a model and we would have a more general model that would again have as a basis something stable that could be called preference but it's now a preference that's framing dependent you see in the end once you would have a complete picture once you would have the true structural model behind the determinants of preferences you would again end up having something stable that could be the measure against which to evaluate policies that have these effects on intermediate preferences so this is a fundamental issue here because preferences are always unstable relative to some model and when I say they are not stable what I typically mean is they are not stable relative to our standard models that we use in economics that's very important in this sense the view can be reconciled with Becker and Stigler you see but at a kind of different level then and so and it solves the problem of social welfare because you would then make social welfare judgments based on the true structural underlying preferences that we are not yet able to we don't know yet so it's a long research pro it's basically I believe that the science of preference formation could be one of the exciting next frontiers in the social sciences everything is here what we need but the causal inferences economists are the world champions in the social sciences in the domain of causal inference from naturally occurring data experimental tools to measure preferences everything is there you need just to do it you need to just be entrepreneurial and collect the data and then let's go so there is a huge opportunity here I believe for progress for the next two or three decades other questions? correct me if I'm wrong but you said that banks are willing to change business culture and how can we expect them to change when the possibility of being caught is so low when I'm sorry I'm really nervous but when when there's private laws being covered public expense when even if they get caught if they actually go to jail or if they actually have to get back the money it's not proportional in any way to the amount that they have stolen or may disappear so no very good question so you are of course right I mean we cannot expect them voluntarily and they but banks are not the same bank in 208 is not the same a bank in 208 is not the same as a bank in 2015 for example there were cases in city bank where one of the board members of city bank was raising arguments against some of their business practices what happened at the time he was replaced and with a more opportunistic board member now the public pressure has been extremely important and public outrage was extremely important and it's going to be important to keep on the pressure but at the same time what they also noticed is that in the long run they don't benefit from questionable business practices so for example let's say Liber or let's say some of these scandals like the Liber scandal if you look at the at the scale of the problem in terms of involved people it's sometimes not a big number which doesn't make it better for them but view it from this perspective think of a city any city in Spain with 60,000 people how many potential criminals does this city have think of a company with 60,000 employees how many potential criminals does the company have so we have a compliance problem and and I believe that at the top there is now a lot of insight that they want to solve the compliance problem putting reputation first because they fear the reputational loss you could say this is a self-interested reason yes but it's not wrong to do the right thing for self-interested reasons so I think the reputation loss is important that they incur and it's the reputation loss that finally brings them on a road that is better but yes if the public is not vigilant they may stop their attempts but I also want to point out that it's not easy to implement compliance in a large number of people in a bank with 200,000 employees you know you have a few guys who do the wrong thing and what about the country let me take the bank let me take the wealth management example let's say a bank somewhere in Europe wants to go to Brazil now I have to Latin America my belief is there's a lot of tax evasion down there my belief there's quite a lot of tax evasion and now if you do wealth management you get lots of untaxed money every bank takes untaxed money what do you do in that market it takes quite some courage to say no we don't we go for the the sober money because in the long run we that's the right strategy of course but the temptation for them was pretty big or is pretty big to also take untaxed money and then public opinion sometimes shifts there are these radical shifts in public opinion right now in some Latin American countries let's say there's anti-corruption fights going on which is good but then it may everything may shift it may turn things on its head so I'm thinking the world is more complicated they are not all just bad guys they are following their own interests which in the relevant environment may well be improve reputation of their organizations any more questions okay thank you very much thank you Professor Fer so after the question round I thank you all for coming I think our distinguished guest today Professor Fer I think we learned a lot today thanks again to our sponsor so finally on behalf of the rector of the president of the university I say this in Catalan declaro oficialmente inaugurato al curso 2015-2016 bon dia bon