 Okay, well, thanks a lot Milo, and thanks Seis. That's very exciting what's going on, and I look forward to what is going to be produced. It's a big honor also for me to be giving this lecture. So I'm going to talk about market and morality, some work I've been doing with Mathias de Watrypon. We are actually undertaking a very major rewriting of this paper. So it may, you may have to wait maybe two or three weeks before getting the new version of the paper and I apologize for that. Before I start on the paper with Mathias, I would like to have a series of remarks about markets and morality to give the general background on those issues. And you know, as Milo said, you know, feel free to interrupt me and say, maybe it's a keynote, but I want it to be some kind of seminar discussion if you'd like, so don't, don't hesitate. So the paper with Mathias re-revisits a nagging question with is a market weakening or more compass? And that's of course an old question. And you see some big divergence, I will say, between the economist and the public opinion of the politicians and religious leader and I'll come back to that to philosophers who kind of won against the religion of the marketplace. We all know that, and we practice those narratives ourselves, that there are narratives which basically point the finger at the market for immoral behavior. So, you know, if I sell weapons to an autocratic regime, I will say, well, listen, if I don't sell those weapons then somebody else will do that. If I'm a doctor and I sell, you know, I prescribe too many, too much antibiotics in France or opioids in the US, my excuse will be that anyway, somebody else will do it otherwise. So I'm not doing any harm really. Same thing if I bribe an official in the last developed countries, you know, if I don't bribe somebody else who is bribing will get the contract. Athletes who take performance enhancing drugs have the same reasoning and so on and so forth. So that's called a replacement excuse or the replacement logic and actually there's a little bit of evidence pointing in this direction and we need to collect more evidence. But that raises a big issue is that whether such behaviors under intense competition lose our moral overtone and just become part of a cost of doing business. This is very important because if you want to see about if you want to think about on the trust for example, should we create monopolies to avoid the, to avoid having this kind of logic. So that's one thing. Now, before going on, we have to ask what is more? And I think it's fair to say that most economists will think of moral tears being a simple externality. So everybody agrees that polluting is not nice or committing a crime is not nice. It's just an externality. It's about harm or pro-social behavior basically. And this idea, this conception of moralty is probably the most universal, similar across culture or stable across time. I just want to say, it's something I find very appealing but it's very restrictive for at least two reasons. The first is that, and I will come back to that at the end is that lots of experiments have shown our moral behavior is fragile to various narratives and moral wiggle room. So there's an entire literature showing that the very flimsy excuses, situational excuses actually release on oneself of one's more duty. And that's very disturbing somehow that we behave in a moral way but if we have the slightest excuse, we stop doing so. The other thing is that not everybody agrees that moral behavior is about not inflicting harm. So think about, for example, the book of Jonathan Hite on the writer's mind. The idea is moralty is broader than just pro-sociality. It is about not involving no harm to anyone but harm being defined in a broader sense. So there are behaviors which we think, for example, do not involve any harm to anyone. So for example, sexuality between people of the same sex or different races. And nonetheless, for a fraction of the population that violates a convention, the taboos or chained, made, shared, discussed or whatever. And this is very time contingent. It varies a lot across time and across countries. The other thing is that lots of people, including moralty, the sense of duty, the sense of authority, the loyalty to the in-group. So you think about family, workplace, country, religion and so on. And that's viewed as being something very important. We economists tend to say, well, no, I mean, we don't care about authority per se, for example. Authority may be an instrument to achieve a goal, but authority is not a goal per se. The same thing for duty, same thing for loyalty. People have a broader view and low and be all. I mean, this kind of view fluctuates more over time and across cultures, much more than the first view. And of course, the question is then, what is meant by harm? So if I disagree with you, I'm arming you. And that's actually a very complex issue. Now, talking about the market, the economies felt unchown, the world view, if you want, is that incentives are key. Of course, many of us have worked in the last 30 years about warning that low power incentives, so incentives which do not react very much to performance, individual performance may be desirable. Maybe because you have no easy performance measurement, the old Armstrong paper teamwork is a particular case of that. If you have collusion with monitors or capture, if you have repeat interaction, if you have multi-tasking, if you have adverse selection, we have written lots of papers saying, you may not want to have high-powered incentives. So that's well known. Actually, you may even have crowding outs. So crowding out is basically a downward-sloping supply curve. And from my work with Ronald Bilaboo, we know that may arise in two situations. Either the principle is in form and basically high-powered incentives might destroy intrinsic motivation because it conveys something about the task of the person. Or it may be multidimensional signaling what psychologists call the average justification effect, which is that if I behave pro-socially, but I'm being paid for it, then you don't know whether I'm doing it because I'm pro-social or because I'm doing it for the money. Nonetheless, I think it's fair to say that economists, we economists, we think that incentives work. I mean, in most circumstances, we have lots of caveats, but by and large, we want to achieve compliance with contractual incentives, mechanism design and the like. A special case of incentives is markets. And again, the economists have spent, actually, that's much of what we are doing as economists to talk about market failures. Yeah, we talk about anxieties. We talk about internalities, the fact that we may not stand for our own best interests, market power, asymmetric information, inequality, et cetera, et cetera. So again, we have all those caveats and we say, no, no, but market is very important because markets are efficient. So what we must do is actually regulate the market as best as we can. I mean, the social scientists have a slightly different view. I mean, some of that is addressed, of course, in our own research. So for example, psychologists and sociologists will say, material incentives don't always work well. They may undermine intrinsic motivation. They may crowd out such valuable social norms. Okay, so that's something we've taken more and more on board. Same thing legal scholars say that the law is not just a set of incentive and sanction, but it's also expressive in the sand that it's going to reflect societal values. So it's not just about price, you know, punishing you for a crime or something like that. It's also something about society's value. That again, we can account for that in our frameworks. I guess where there's a most disagreement in practice will be with some philosophers. So some of them are very strong distrust of markets or at least some markets. And we may not disagree with them, but maybe, and I should say right away that if you look at the philosophers and they're also very heterogeneous. They're often at the same starting point, often the same title, almost the same title for their book. And maybe even the same conclusion, but they reach that conclusion from very different perspectives. So two extreme case might be Michael Sandel in his best seller, What Money Can Buy, but also Debra Staths at Stanford has written an interesting book on that, but I could also mention is that that understand Michael Wellser and so on. Many philosophers have written books which have a title like this. And that raises a question for us economists. Is economists more into philosophical science? And that's very important because, you know, we believe it or at least we want to believe it and that's important. So let me, before I talk about the table with Matthias, let me make a few more remarks. And the first one, others simplifies very large amount. Our society is something like of a mixture of Smith and Pigu. So on the one hand, we want to have the invisible end of the market to get the market efficiency and it's clear that markets have done wonders in many countries and we see that now with four countries having developed remarkably. At the same time, we also emphasize that markets fail usually, I mean climate is the worst example, but there are many, many other very important example. The markets fail and basically the conception is that you use a market and the state is there to correct market failure. But that raises the issue of why we have a social responsibility and the answer is simply that not only the market fails but the governments fail as well and as badly as the markets. For all the reasons we know, capture, pandering, maybe things happen abroad and we cannot just invade China because China is using coal, right? I mean, this does not make any sense. But also there are transaction costs, there are many other things that raise issues. Now, maybe in defense, I mean, that was, you know, of a profession, you know, I try in economics for the common good to say that the number of what we call repugnant market can be analyzed very simply in terms of very familiar concept like extinities, including image extinities, internalities, non-competitive market, asymmetric information, equality, motivated belief and so on. So in a sense, we have the tools actually to analyze those repugnant market and why is that important? Well, I don't like indignation. I mean, let me explain. I mean, I think moral postures are very unreliable sources of physical inspiration. We have seen too often the past people saying, majority of people saying this behavior is disgusting, it's immoral when you, it's not clear, it's immoral at all. So I don't like moral postures. I don't like indignation. Indignation may being indignant may serve a purpose. So I mean, because it's a warning signal that there might be something wrong. So I see being indignant about something as a warning signal, but that's about it no more because what we need to do is to try to understand why we don't like those markets or why we have concerns about those markets. Should we regain them? Should we private them? What should we do? Should we just let them happen? This is a kind of thing we have to go through and just say, don't content yourself with saying, I don't like that. I don't think it's a scientific posture. We need actually to actually do better. So there are also another issue which I'm not going to touch on is that even if we don't like markets, we have to think about what a counterfactual to the market is. And there are some interesting counterfactuals. There are some awful counterfactuals. If you want to actually, there is always market. The point, my point here is that markets always exist whether we like them or not. So the question is, how do you want to organize them and regulate them? And that's basically an interesting thing. Okay, so I don't want to spend too much time on that. Let me see if there's any question first. And again, feel entirely free to actually interrupt. No? Okay. I guess this one, the presentation of the paper is going to raise more questions and including clarifying questions. So Matthias and I have been working on the market markets. Again, the new version of the paper will be ready only in two or three weeks and I apologize for that. What we ask in a sense is, how is our moral behavior affected by competition? And should we restrict competition in order to get better moral behavior? Now, we IO economies of course are used to thinking about competition as a market structure. The number of firms are substitutability. And that's perfectly fine but we are going to take a broader perspective because we are going to ask, how do firms compete, for example, something that will play a role, for example, is whether prices are flexible or fixed. You're going to tell me in general, in markets prices are flexible and that's true to a large extent, but not always. And if you think about professions, the price of taxi rides is regulated. In many countries, the price of doctors is regulated or not raised in two-sided markets, platforms, apps and franchising apps are constrained to price to charge at zero or bound. I mean, we get lots of service at price zero. So that's basically a fixed price, zero. Franchising environment, same thing the price is often regulated and so on. So there are lots of examples where actually prices are fixed and this is going to make a difference. Something which is also going to make a difference is if you behave immorally, so if you, by the way, I'm going to use acical behavior and more behavior with exactly the same meaning, just to simplify it. Cutting acical corner may help you increase your profits. It's not interesting, I mean, I don't know if anyone who actually behave immorally to reduce is our profit, doesn't make sense, right? So in general, cutting acical corner is about increasing profit. It can either lower costs or boost demand, okay? So for example, cost reduction might be I use fossil fuel and I pollute when I produce or I use child labor in a less developed country. My costs are smaller that way. I'm going to give you examples and actually the examples I gave you earlier are about boosting demand. So it's more likely that I get, I'm selling if I pay a bribe or if I cut various acical corner to boost the demand. So if I presscribe on antibiotics then it's more likely that the patient will come back to me or if I presscribe, oh stop here, it's if I'm engaged in sports and I take performance enhancing drug, it's more likely that of course I'll be invited and I'll get money out of it and so on. So that's one thing. And that's, as we're just going to play a role was actually cutting acical corners are lower costs or boost demand. I mean, you might imagine that competition is going to have bigger effect when it boost demand, right? The social responsibility of stakeholder is important. So we will allow for not only the firm to have acical concern, but also the stakeholders of the firm. So think about investors, think about workers or consumers. And then we will be asking, what about heterogeneity in the industry? So it may be the case that you're in an industry with people with high acical concerns and some people are less acical or something will be interested in as well is whether you have for profit and not for profit corporations in the industry. And we have to ask, do less acical supplier drive out of the market acical one so that might be some kind of crash and slow physical behavior. And similarly, what's going to happen with nonprofits when they're in a market where in which I compete with for profit. So that's a kind of a question we're going to ask. So let me describe the model unless you have questions. Okay. So we are going to consider an oligopoly with N competing firms. This serves a unit demand consumers, continue of unit demand consumers. The firm is going to choose a price PI, but it's also going to choose there will be a second dimension in the choice which will be the model action. So we'll call AI is the action. So higher AI will be a more more action. So AI will belong to some interval zero AI hat. AI hat can be finite or infinite in the number of examples we develop in the paper, it will be finite, but it doesn't matter. But the important thing is that if you have a consumer who comes to you and buys from you and you have chosen the more action AI, then that's going to generate social welfare, WI of AI. So for example, if I'm producing electricity or whatever good, basically I'm going to choose how much pollution I exert. And depending on the level of pollution, what kind of technology I use, then there will be a different social welfare. So if I pollute less, then I will get higher WI. So think of WI as being the externality on society, for example, but there will be other interpretations I will come back to. So the welfare impact WI of AI is going to satisfy, it's going to be an increasing function. It will be concave and with usual properties. Okay, the more action may also affect unit cost. Remember, it might be, for example, that if I employ child labor use fossil fuels, it will be cheaper for me to produce. And therefore my costs will be increasing in the quality of the action. But you could also have the reverse case. So imagine that I behave morally. Then if I have socially responsible investor or socially irresponsible workers, those investors or the workers, they will be willing to accept a lower return or lower wage in order to be able to be associated with my firm. In that case, when I behave more morally, actually reduce my costs. I reduce my costs in that case. But often you have both, right? I mean, in general, so we're just going to assume that C-prime I can be positive or negative, but it's going to be convex. That's the important thing because otherwise we don't get concavity, okay? Now something which is going to play a big role in what follows is the concept of a net price. So I told you that the more action can actually impact demand. Those were the example I gave you on the first slide on the paper with my test. So if I behave morally, then basically I'm reducing the utility of the person I'm interacting with. And it's as if I was increasing the price. So I'm going to call five I of AI. The increase in price relative to, when I increase in the morality of my action, I displease that particular consumer. And therefore it's as if this consumer were paying a higher net price, okay? So that's very natural. And the demand function, my demand function as FMI will depend on my net price and the vector of net prices charged by the other firms P minus I hat, okay? So basically whether I will get the contract will depend on many things but will also depend on the size of the bribe I give to the official and also the size of the bribes which are offered by the other firms. The elasticity of demand DI will depend on the standard IO considerations the number of firms is distributed among products. In general, we can index it by a parameter sigma which is going to be basically an intensity of competition. So it could be the inverse of the transportation cost in the authentic model, for example. So sigma and it could be the inverse but in general it's going to be a parameter which is going to index the intensity of competition. And the estimate demand is defined in the standard way which is minus DDI or DPI or DI or PI. Now the assumption we make is not very strong which is basically that if price is higher then the estimate demand is smaller. I mean, that's standard assumption to make. So if price is outlaw then you get into a region where the estimate demand is very high, okay? So let me get back on the net price. You have, okay, there are three cases you can consider. So the first case is the one I already mentioned which I will call, I mean, you may not like this terminology but I will call socially responsible consumers. So you can, for example, for normalized FI of AI equals AI. But the important thing is that when you behave more morally, they don't like it. They are less likely to purchase, right? So that's the consumers we were talking about but they might be other consumers. So for example, I may be paying bribes in some other country but then for my other product or what I produce with my inputs there would be socially responsible consumer in my country will be upset. So the difference between a socially irresponsible consumer and a socially responsible consumer is that the net price is increasing in the more action in the first case and decreasing with the more action in the second case. So you might think of alpha C as being the, that's just an example. It doesn't have to be that, by the way. But for example, you might internalize the welfare that you create by buying from this firm and alpha C is the internalization of welfare. And then you have, you know, or more economic as I just don't care, the consumers basically care just about the price. Okay. So still no questions? No? Okay. So why is that a more issue? So in the paper, we derive three foundations for demand side benefits. Excuse me of cutting ethical corners meaning that there are irresponsible consumers. The first is the externality. So, you know, if I sell weapons to a dictator, for example, or if I bribe an official or I take a preference and enhancing drugs and I'm working as a people, it could be an internalities of the case of opioids. The case of antibiotics is an anxiety. And then there's something a bit more subtle but basically a product misrepresentation, shrouded attributes and so on. So think about lying or not disclosing fat or sugar content selling financial, toxic financial products without saying that toxic or short termism, income manipulation and so on. I mean, you need to purchase a terminology sometimes but mathematically it's exactly the same as an internality or an externality. Now, the example of a response when consumers will be someone who wants to engage in a fair trade. So I want my coffee to be a fair trade coffee. Okay, so here a little bit more mathematics. Jean, I have a question. Jean. Is it not enough to have the Homo economicus? Why do you need people? Because the price would be lower with all this cutting corner. So why do you need people to like? Because you have two ways in which you can gain from unicycle behavior. The first is what you have in mind right now which is you cut on cost. So you use child labor is cheaper. Okay, but the other thing is that you please the particular person you are dealing with in order to get either business. That's the case of opiates or weapons or whatever. So in that case, it's a demand side benefit of cutting a single color whereas what you had in mind is the cost side benefit of cutting a single color. And you'll see, I mean, that makes a difference in the treatment. I mean, the cost in a sense, the intuition is that cost side, the demand function doesn't really matter for the cost side. Whereas for the demand side, that's a replacement logic that I was talking about at the beginning. If you have very intense competition, then if you don't do it, somebody else will. Okay, so that's very important to actually add more behavior into the demand function as well and not only into the cost function. Maybe it's going to become a bit clearer when I describe the payoff functions. Okay, thank you. Okay, so the payoff function of manager, I, entrepreneur, I, firm, I will be startup profit. So it will be price minus cost. Remember the cost depend on the ethical behavior. It could be increasing or decreasing. And it's going to depend on the price on the demand function, which is going to depend on the net prices. Now, this is a little bit less obvious. So basically the assumption is that the firms have social preferences. That means that they will internalize welfare. And we have to discuss exactly what we mean by that. But alpha I will be the startup social preference parameter and there will be some kind of internalization of welfare. And we are going to assume that supplier I internalize a function W script of p hat and a, okay? Okay, so those are vectors, right? Because it's going to depend on the vector of net prices, if only because it depends on demand. But it's also going to depend on more behaviors by the various suppliers. So examples, so we actually this formalism allows for a number of examples. So one of them is w i of p hat and a is equal to the sum of w i d i or w j d j, right? So basically this is the number of units, the d i is the number of units sold by firm i and w i is societary impact of that. So it could be, for example, this term, this overall term might be total pollution. So, you know, there are d i would choose supply i and then that creates a damage from pollution, w i of a i fair consumer. Or it could be total consumption of opioids, okay? Now, in general, in general, it's not going to be full welfare because in general, you may also have a misallocation cost. So basically the matching, I mean think about an altering model, for example, if one of the producers behave more immorally and attracts more consumers, that means that there will be also, if they are otherwise symmetrical, that there will be basically a misallocation of consumers to firms in this altering model. So there is a misallocation cost which will depend and it should depend only on the vector of net prices because that relates to the misallocation of consumers. Now, it turns out and this is the result on the bottom of the page, we have a result which has, if you look at symmetric equilibria of symmetric oligopolies, then those, the symmetric equilibria when suppliers internalize a sequel welfare are also symmetric equilibria when they internalize full welfare. Why? Because the proof is the one line proof, which is that if you have a symmetric equilibrium, then they won't be any misallocation in a symmetric environment. And therefore, if you deviate, you are going to create a misallocation. So that's going to make deviations even less appealing. And therefore an equilibrium in which there is a sequel welfare entalization is also an equilibrium when there is full welfare entalization. But conversely, you could also look at a narrow RIS equal welfare. So you might say, oh, I'm as a consumer or I'm a supplier, I don't really care about what the other people do. I just care about the morality of my actions. So this term is really a subset of the total term. So I care about my own pollution. It's not a normative point of viewing. I just describe how they perceive the welfare. Now, you need an assumption and that's a very important assumption. So this assumption RIS says that the perception welfare when more action of supplier high changes is proportional to demand. And this is very natural. This is, it's really consequentialist, okay? Because in the end, when there is a change in moral behavior, then there is, it's going to be proportional to demand, right? The impact is going to be proportional to demand. So this is really an assumption about consequentialism. And that's going to be important. Now, this function is going to be decreasing some simply because of concavity, right? And that's, by the way, this assumption is satisfying all the examples we develop in the paper. X19, the entire product misrepresentation, that works. Okay? So it looks like a very strong assumption, but at first it's natural and also it's satisfying all the examples we have developed. Okay. So go back to the startup complain about markets. The startup complaints about markets is that competition is going to erode moral behavior. Okay? Is that correct? Is that correct? And we first say, well, that's not obvious at all. Actually, we have a very general result which tells you that that's not the case. So the question we ask, does an increase in the capacity of demand, either because they are entry of new firm or an increase in the stability? So imagine that the T parameter of the auditing model goes down or Sigma goes up. Does it result in less moral behavior? And the result we get is that it has absolutely no impact whatsoever. Competitive pressure, you know, is something that changes the accuracy of demand. Of individual demand has absolutely no impact on moral choices in equilibrium. If either prices are flexible or they are fixed but moral action affects the cost. So that's a child labor example, but not demand. Okay? So I've, so I'm just going to look at part A which is if you have flexible prices, then the intensity of competition just doesn't matter. And therefore, if you believe that, we should not relax on titraceae. We should not create monopolies to get good behavior. Actually, monopoly or carter doesn't behave better than a very competitive industry. What is intuition? So I'm going to develop it just for flexible prices. So imagine that you have demand side benefits of cutting ethical corners. So basically, when you choose AI smaller, you face a higher demand. Okay? And consider an increase in ASC of demand. Then cutting a cycle corner is going to, and that's really the replacement logic. You know, if you cut a cycle corner, you attract more consumers and that's going to lead to higher increase in market share. It's very tempting, you get higher increase in market share by behaving immorally. So that's exactly in a sense what the replacement logic or the philosophers are saying. At the same time, there is a contributing force which is that if there is more competition, there is a lower market. And if there is a lower market, there is a lower gain actually of gaining market share and therefore there is a lower gain in behaving immorally. And it turns out because of the additivity structure of the net price formula, the two effects perfectly offset each other. Okay? So that really already tells you something about well, you should be careful before saying that markets destroy a cycle behavior. And actually when we will have those assumptions, sometimes we will find it still doesn't have any impact or we find that it's going to destroy more behavior and sometimes we'll find it actually improves more behavior. So the question then is when does competitive pressure matter? The first, I mean, basically this results such as the answer, which is, if you have flexible prices, it's irrelevant. Competitive pressure is irrelevant. So it's a modically an email result. If you have fixed prices, obviously you get rid of the reduced market pay effect of intense competition. Okay? And what we get is that if you are the responsible consumer, so that basically immoral behavior is going to increase demand, then more competition is going to impede more behavior because you do have the replacement logic. If I don't do it, someone else will. And we show that actually more choices are strategy complements for a variety of reasons. The first is the elasticity of effect, which is if the rival's behavior more is that intensifies the competition and therefore that raises a firm's elusive demand and therefore that pushes oneself to actually reduce one small barrier behavior. There are also effects which are actually interesting and which go in the same direction. And let me just mention the social responsibility effect if the other firms behave immorally, then it's very important to take market share away from them. So you're going to behave more, less more yourself actually. So that's part of your social responsibility. Take market share away from rivals while still being more, more than they are. But at the bottom of the page, there's some very important result as well, which is if instead of having irresponsible consumers who have responsible consumers, then that's going to be the mirror image. So more competition is actually going to foster more behavior. For responsible consumers, people who want to engage in fair trade, for example, or something like that, they actually, if you behave morally, then you're going to attract more consumers. And if there is a more competition, the SST is going to be higher. And therefore more competition is going to allow consumers to choose someone who is ethical. So it is just completely reversal of the startup conventional wisdom. This conventional wisdom replacement logic and the philosopher's logic is very much the first result here. But it's confined to fixed prices. And if you have responsible consumer, you get exactly the reverse results. You have questions? I should leave a little bit of time for Q and A I guess, but I'm just going to give you some, Sophie, you have some questions? Yes, Jean, thank you very much. On the flexible price, I had some difficulties reconciling that with the notion of non-nash competition, non-oligopoly, because it seemed to me that if you have fewer firms, then the non-nash competition may induce them to play the bad equilibrium. So this is just a static Nash equilibrium here. So we look at the equilibrium in which they optimize. Now they could collude. I guess that's what you're in mind. So that if it's a repeated game, so each period they charge a price and they choose some more behavior, they might collude on that. And they might reach a monopoly outcome, for example. But this result tells you that the cartel or the monopoly outcome is exactly the same as a competitive outcome in terms of morality. Of course, the prices are lower if you have competition than if you have a monopoly or a cartel. But in terms of morality, it's exactly the same morality in both cases. So you can have as much collusion as you want, you are still going to get the same level of morality. Now you may still want to have antitrust because antitrust will allow you to have lower prices. That's a different ball game. But what I'm saying is that one cannot use just this argument about the replacement logic to basically change antitrust. Thank you. Let me just be very brief and then we can have a chat about what we are doing. So the last, not quite the last thing we do, but something we do is that we have a bunch of results I'm not going to present on symmetric or egopoly. But there is also an asymmetric or egopoly framework for various reasons. One reason is that as I mentioned, people have different basics. So they might differ on their cycle stance. So we capture that by saying people have different alpha i, so they defend internalization of alpha. The other possibility is that they have different corporate forms. So we are actually going to allow both in the same model. So there will be an heterogeneity in terms of alpha i, which is how much internalized social welfare or your conception of social welfare. It could be a sequel welfare. It could be full welfare or it could be narrow welfare. It doesn't matter. But also in terms of corporate form, so you could have for profit and not for profit. We are going to keep, we have two levels of two degrees of heterogeneity. We are going to keep the rest symmetrical. So the demand curve otherwise is same. So it's a symmetrical demand curve. It's a symmetrical cost function. We are going to assume, you remember the relevance result is that when prices are flexible, then market structure is irrelevant. Intensity of competition is completely relevant to moral behavior. Now I told you that's not true if you have fixed prices. Here we are going to look at something which formally is a flexible price, but the factor will be a fixed price in part. So we are going to assume that prices are endogenous. They are not fixed. But at the same time, either because of the difference in ethics or dependent corporate form, then there will be something akin to a fixed price. It's clear for not for profit. Not for profit has to charge a price which is equal to costs. But also when there is heterogeneity in alpha i, the highly ethical suppliers that tend to actually make very low profits that we can show. And you at some point, if they have to break even, then they become not for profit. They are not formally for profit, but they are so ethical that actually they are going to face the break even constraint. And if they have to break even, then they will become de facto, not de euro, but de facto for profit, not for profit. So what we do is to characterize the equilibrium. Again, there is symmetry in demand and cost. It's only in terms of preferences or corporate form that there is an asymmetry. So basically there will be three groups of firms in the industry. I mean, they are not always, there may be some group may be empty, but by and large you have three groups and I've ranked them through increasing morality in equilibrium AI. So the less moral firm will be the for profits which are not constrained. And actually there will be the one with the lowest morality. So I'm going to rank alpha i, alpha one is less than alpha two, less than alpha m, less than alpha n one. Okay, so I rank the alphas that way. And basically the one who was available alpha, alpha one is going to be the one who is going to be, so internalize welfare of the list and is going to be the least moral. And then alpha two and then alpha m. There will be a second group of for profits but those for profit they are more sequel and they are going to be bound by the break event constraint. Okay, otherwise they will go bankrupt. And this group actually even so they are for profit they behave like not for profit. And finally there is the not for profit and notice I haven't written what the alpha is because it doesn't matter. I mean, just maybe I should go back to this. I mean, look at this objective function. If you are not for profit PI is equal to CI so that equal to zero, PI equals CI of AI. So you are just left with this and your alpha doesn't matter as long as it's positive. Okay, so that's why I didn't rank in that diagram I didn't rank the not for profit in terms of AC code behavior. Okay, so that's a characterization. Now the interesting result is this. There is a convergence when sigma grows becomes large. So when there is very intense competition then the for profit will have to mimic the not for profit slow price. So they will charge something close to marginal cost. That will happen anyway with only for profits, right? So you will end up with basically something close to marginal cost pricing. But on the other hand, they are all going to mimic in the limit the least AC code file, the least AC code for profit. So that was one, that's the one by normalization I choose supplier one to be the one with the least morality. So basically the store is of course not for profit they would like to improve software welfare but they can because they are going to lose all their customers. If you have a demand base benefits from cutting AC code corners. So when sigma goes to infinity then basically all firms are going to behave as the same. Okay, and they are going to be very competitive with very low prices but they are also going to be even be able to see clean. And that's of course an issue. And that means actually that if you really want to get the benefit from not for profits, like in the hospital sector or the education sector, you don't want them to be competing with for profits necessarily. Okay, so that will be the thing. Yeah, I think Milo, I guess I should. Yeah, so maybe. Yeah, just a last thing I want to say is that when you look at the examples in the real world for some of them, there is a real events of competition but for some of them, which doesn't mean by way it's more in any sense. But if you have a consequentialist perspective, there is actually importance of competition by their cases in which that's not the case and more bigger room. So it's just an excuse, it's not a logic. So think about bribing an official so the contractor or the bidder then picks a price, which is a bidder and a bribe which means that prices are flexible in that case. And then if they invoke the replacement excuse then that means that it's really more regular. So there's a lot of fun people work to be done on that. And I guess I'm going to stop there and thank you very much of course for doing this. And I look forward to your questions and to the discussion. So thanks a lot, Jean. We have time for a few questions. Don't hesitate to intervene or raise your hand. Good day. Please. Thank you. So, Jean, that was a really nice talk. I was curious about what are some of the policy implications of your work? Is it possible for the planner to directly regulate some level of this action AI or is there a role for information or persuasion in trying to convince consumers to be more socially responsible or perhaps sometimes even less responsible? Yeah, that's a good question indeed. It gets back to the first part of the talk in a sense which is why is there a failure of regulation? You might say, okay, why do we care about social responsibility? The governor is there to regulate another decent carbon price, right? That's what we will be saying, right? And then our social responsibility will become much less important because the governor will be doing the right thing. So basically we are looking at situations and again, if you look at the philosophers complain about economics, those are situations in which the regulation doesn't work well. There are markets for kidneys and they're unregulated or whatever right now. And there are things like that where we feel we have a social responsibility maybe to behave properly. And that's the kind of environment we look at. And I guess many people also they may not say so is that we should have less competition. It was this obsession of economists with competition and one of those who is actually obsessed with competition, this obsession of economists with competition, if you believe in the replacement logic which is a reasonable thing, tells you that probably you should have less competition and let people call you the maybe have a monopoly, a state monopoly or something like that. And what we say is that it's much more complicated. There are a bunch of environments where it's not like everything goes, but there are a bunch of environments where it really doesn't matter. The level of competition should not matter. There are environments which are well defined in which actually people are right in saying competition is going to trigger immoral behavior and there are environments in which actually competition is going to foster for social behaviors. So that's the kind of thing. And the last warning I issued, I don't know how much, I believe in it, but at least from the point of view of the theory, the idea is that if you really want to say elsewhere or education to be done by non-for-profit, you have to pay attention to competition or else you have to regulate well. Because that's an alternative. I tell you, you try to induce more behavior or you do your job as a regulator. I see Marcel is raising the hand. Yeah. Just a quick question regarding child labor. So first case of what seems to please appear in your talk as one of the worst unethical behavior is to use child labor to reduce your costs. Now, child labor is a little bit more complex because if you see it from the point of view of the child or the family or the underdeveloped country, child labor may be welfare enhancing because it takes the kids away from the street. And I remember a report by an ONG, I mean many years ago saying that those actions against, for instance, Nike, which protested in Western countries against Nike using child labor had detrimental effect on the welfare of those children and their families. So in a sense, I mean, we have to be a bit careful in assessing the welfare aspects of behavior from the point of view of child labor in France or Canada, which is not the case if you talk about child labor in very poor countries. Well, that's a good comment. That's why also I want to keep the perception of social welfare very large. I just don't want to, I mean, I still need, we still need some structure in a sense, you know, the two assumptions I gave you on the necessity of demand and on consequentialism, but in general, you want to have a very broad vision of social welfare and that's connected to something which is important is how do we think about morality in those markets and the impact that we have. Let me give you another example that you are familiar with is leakage in environmental behavior. So, you know, I think that by divesting from oil, I'm going to have a big impact. Now, probably my impact will be smaller than I think because the shares are going to be purchased by somebody else or same thing when you talk about exclusion versus best in class and so on. Those are very difficult issues and your child labor, yeah, of course the children are exploited by their parents because they don't decide themselves, it's very, it's very, it's very more but at the same time as you say, they're also, if you stop child labor, they're also some bad consequences. So it's actually much more complicated than you said, than I said, but in the end, you know, again, you know, the question is, you know, I just want to have a broad assumption about what's internalized and that may not be what you or I think is the right social welfare function. That's okay. But of course, when we think about intervening, like, you know, Uday's question earlier, should we, for example, kill competition? Then we need to have our own social welfare function and decide whether child labor is good or bad, right? Socially, independently of what people believe.