 First, just a few words about the book because this was a wider project and which has just come to completion. The book is going to be published in a month's time by Oxford University Press. And I must say it has been a fantastic experience, this book, because the group with whom I worked with the purpose of producing this book was just a great group. And we will have a sample of, we could say, the chapters because we have three contributors here in addition to myself that are going to speak about their theme. So the book is in fact centered around three main themes. One is marriage, because there are five papers that really concern marriage. And you know, there are two types of paper in the book. They are kind of review papers. And for marriage, there is a review paper on intrahausal bargaining in poor countries by Jean-Marie Ballon and Roberta Ziparo. And then there are more focused paper on different aspects. You will see that with Sarah Law, who will talk on behalf of her work with Nathan Nunn. And there are work by Dominique van der Waal in this respect too, who work with Sylville Lambert and Paola Villa. And there is also a paper by Rebekah Tonton and one by Selim Guleski. The second theme is outside option, you know, the impact of outside option on the status of women and about, you know, the impact on early pregnancy, gender gaps. And here we have a series of paper by Marcus Goldstein, by Shetil Jarvatten and co-author Ricardo Osman, Jim Ferong, Xavier Ginet and co-authors, et cetera. And the third theme is laws and cultural norms, which has six chapters. And the paper I'm going to present in a sense of a review article on that topic. And we have a paper by Sivan Anderson and Dibrageret, another one by Bijou Rao, one by Nancy Kian, one by John Bowen and finally a fascinating paper on the case of abortion in El Salvador with what's presently happening with the state intervention by a team of Latin American people and Jocelyn Viterna. So there are some sociologists in the team, but I must say it's still dominated by economists, there is no doubt about that. So no, so much for the books, so no, I turn to my own presentation and so I'm going to report about a chapter that has been written by Julia Camillotti, who has been my PhD student in Emmanuel O'Reol and myself. Okay, so what is that we are trying to do here? We are reviewing the literature on what we call social engineering, which means the attempt to change oppressive social norms. And in doing that we propose a theoretical, not a single theoretical framework, but we try to think with a link to theory to put some order in the empirical literature, which is not huge by the way. And so the idea is essentially no, you have seen since quite a time a number of international conventions starting with the convention to eliminate all forms of discrimination against women by the United Nations in 79, followed by another one called the UN fact sheet on harmful practices affecting the health of women and children, 95, which listed a number of oppressive social norm and practices, including female genital cutting, early marriage, a son preference, female infanticide, and others. And so we want to assess that to say, is it really useful, should we expect an impact for this kind of approach that through a new law, and in this case an international law, try to influence practices, and if yes, what are the underlying assumptions? And then we try to think, and you will see that there is a clear orientation in the paper regarding this. We are trying to look at other methods to try to change social norm and practices. One is changing preference and the second one is, in fact, a change in the macro technical environment that can be sometimes deliberately influenced by state authorities. Now coming first, I come to the law. The law I think is very important from the very start to distinguish between two effects of the law, the expressive effect of the law or the deterrent effect of the law. And first I'll focus on the expressive effect. The expressive effect, in fact, is assuming that the law can act as a focal point in a coordination game. And in a sense that you are in a game where there are multiple equilibria, Nash equilibria, in fact in pure strategies, and the law is supposed to change the focal point by driving attention to another way of changing or behaving in this case. And so, just look at the present game just to make things clear. So you have a game of female genital mutilation where the choice is between of two parental couples, your cut or circumcise your daughter or you don't do it. And there is another couple. This can be generalized to in person, there is no problem. So I assume that mu is denoting the benefit of coordination. And minus theta v, which is what you get when you cut your daughter and the other one does not, so this is the first row, the second column, is in fact you can conceive, is the cost of following the practice, theta can be conceived of as the health risk, for instance, that you incur by circumcising your daughter, and v is an amplifying factor. And in fact what we have in mind here is that the amplifying factor can be influenced by policy. I'll come back to that. Now, obviously as you all know, here the problem is that you have two Nash equilibria, both reject the practice of both cut their daughters, and you have a problem of equilibrium selection. The focal point, just say it at one point, you have a custom and the custom is a focal point, but if you had a law that emphasise, let's say the benefit or the need to stop her cutting practice, the attention could be shifted to this new law and the practice could stop. This suppose, of course, that we have homogeneous preference at the first thing, and that people don't like the practice, they would just like to change it. The problem is the expectation against, about what the other are doing are pessimistic and so they still do something that they don't like to do, so you have a law it could change, it could change the practice. So, if you know, try to elaborate a little bit on this concept of equilibrium selection and you say, yeah, but maybe there's not only the focal point, but just think about the risk dominance strategy in the way Asani and Zeltan have seen it and these people have an uncertainty about what the other will do and they try to avoid risk, which mean that they don't look only at the benefit that they have if they both cooperate in rejecting the practice, but they look also at the deviation they thought if the other doesn't do it, then you know it has been shown to an evolutionary argument that if the number of repeated periods is large enough, people will select a risk dominance strategy and the whole question is, does it coincide or not with the Pareto efficient equilibrium. In the famous and celebrated Stack Hunt game, it does not, but it happens that in a game of that kind, it does. So in a sense what you would, if you believe in this risk dominant strategy equilibrium selection process, you would expect that the practice should disappear. People should reject the practice because this is the risk dominant strategy and it happens to coincide with the Pareto efficient equilibrium outcome. No, what I want to draw attention to is what about the literature, the empirical literature, what does it does? And in fact, this is an important question because many international organizations, including the UN and NGO, have been driven by this approach and in fact have sometimes explicitly referred to it. And in fact, we have a few studies, I won't show it on the slide because I don't have time, but there is a paper first by Belmar and co-authors in Journal of Development Economics where they find that for West Africa, almost 90% of the variation in support, and yes, support, let's say, for female genital mutilation is explained by individual and household characteristics, which is not what you expect in a game of coordination, where you would explain that what matters is what you ought to do. And in fact, they find no, and that in the communities where the practice of female genital mutilation is more prevailing, then the importance of individual and household characteristics is more important. So this is troubling for this thesis. Then there is a paper on Sudan by Epheson and the author. We just have this very simple argument. Let's say, if you believe in this kind of game, then you should expect to have villages where the practice exists and others where it does not exist. Like this, you should find a discontinuity in the distribution of the practice across villages. And what she finds is that there absolutely no sign of discontinuity. The distribution is completely smooth. Then you could say, yeah, but maybe, you know, this is an argument. The main argument behind this, why do you catch a daughter is the marriage market. I want to enhance the value of my daughter for the marriage market. And so maybe the village community doesn't coincide with the marriage pool, but they find that most of the marriage take place within the village community so that does not apply. They bring out a number of other evidence with a number of implicit association tests, which shows that even the attitude towards the practice does not significantly vary among those who cut their daughter and those who do not. And this is subject to a different kind of interpretation. But I want to maybe give an emphasis to a study done by Julia Camillotti, my PhD student that she did in fall in Senegal, because I feel that she's bringing interesting aspect into light. What did she do? She studied the impact of an intervention of an NGO called Tostan in Senegal. For this intervention, by the way, Tostan got the price of the best NGO, the Hillary Clinton plus of the best NGO in the world for the year, some five or six years a year, two articles from Page of New York Times for the work they did in Senegal. And I wanted to look more carefully into that or difficult to convince them because they say, we know that we have been successful. In fact, all funders come to us to support other actions of us. But still we succeeded in doing it. And what did we find? First, I want to say one thing. Is that something interesting in Tostan is that the process of changing the norm is based on first invoking and referring explicitly to the law. And there has been a new law enacted by the Senegalese government parliament according to which the practice is banned and there are even punishment fees to pay fines, even prison. So it's going a little bit about the expressive thing. But there is a law and they just invoke the law and the International Convention explained that to the women and even the health risks go to that extent. And then second, they asked all the village community to come to a point where they make a public preparation. And that's an interesting article. So here we are at the point of general reading with the 100 villages being represented in which they solely declare we are officially abandoning the practice of female genital mutilation. OK, so they have enlisted more than 100 villages which did go through this public declaration system. So what is the result? What did she find? Essentially, that the intervention of the NGO in this case does not have no impact on the practice or hardly anyone. The second one is that the intervention has even generated perverse effect in the form of a decrease in the age of cutting for the girls. In fact, the idea is to make it more secret. And we all know that the health risk increases if the age at which the girls are cut is lower. Third result is that there is no evidence at all that public declaration had any impact and worked as a coordination and commitment device. OK, so those are the results. There are some others, but they all go into this direction that take us to question this game. But of course, and this is one of the main point we argue in the paper is we say, look, probably you shouldn't throw away the coordination game so easily because there is a very, very simplistic assumption being in this game is that you have homogenous preference. So what you want to understand is what happens if you have heterogenous preferences. And in fact, this is something that is easy to do. The social norm approach based on coordination of incentive is essentially saying that your utility depends on the number of other people, the proportion P of the other people in the community that follow the practice. And the more do it, the more you want to do like they do. But there is a cost. And here you find this V-amplifying factor, which is a function of, I would call the environment E, capital E. That can be a law. That can be a media. That can be the Western donors that could pressure for you to abandon. And that create a rising hostility towards the practice. No theta i becomes a idiosyncratic coefficient measuring the aversion vis-a-vis the customary practice for individual i. And it is distributed in many possible ways. And so we want to just to ask the question, suppose that no, the utility is defined this way rather than the way it was defined in the previous game. What can you say? And you know, unfortunately, I have no time to go into the details and the possible equilibria. But you can see that from that graph, without me going into the detail, that in fact, there are various kinds of solution possible, equilibrium outcome. And it all depends on the way theta is distributed. For instance, we have a situation where many people have a lower version or a strong aversion vis-a-vis the customary practice. Then in this case, you have a unique Nash equilibrium denoted by point A, which is stable because it imposed ability to restrict the number of different equilibria. So in this case, in fact, the practice cannot disappear. There is an interior equilibrium that is quite likely. But if you have many people who have a moderate aversion towards the practice and a few people with strong or weak aversion, then you find that you have two possible equilibria, one in which zero people follow the custom or a large number of the people follow the custom. And you can have even the two situations either everybody follows the custom, this is this corner solution, or nobody follows it. And this is a situation that you can get also in which the curve is so low that in fact, you're sure that everybody will abandon the custom. We also examined the case of a horizontal uniform distribution in which case this curve will be linear and then you have also unique equilibrium, which can be either everybody follows the custom or not. So in a sense, what we learned from here is that this framework is richer than it appears to be. Even though we have to agree that there is not much evidence in the literature from so far for this kind of practices that people are so much influenced by the behavior of the others, which of course would mean that you need alternative approaches if this is the case. Now, of course, an obvious little alternative to that is that you abandon this kind of game, coordination game and you adopt, for instance, a rival claimant's game. In a rival claimant's game, it means that what someone gets, the other doesn't get and so you have a game that can enable you to study the situation of bargaining between the spouses, for instance. No, of course that is the case. It's rather easy to show that the former law can act by powerfully or possibly by changing the threat point of the victimized part of the couple, in this case, the woman and could have an effect on the practice. We have, in fact, proposed in another paper of mine with Zaki Wahaj and Alderchev, a more sophisticated scheme in which customary authority exists, which is deciding about the custom. So instead of saying the custom is there, it is fixed, we say, no, no, it is chosen by the customary authority in repeated judgments that it is making when the case is submitted to him. And what we find is, in fact, that enacting a law that is favorable to women may have no impact, may have an impact, or may have a perverse effect. By perverse effect, I mean that it could worsen the situation of the women. There are various arguments for that. I just tell you what I find is the more convincing argument is that the judge have their own utility preference and the trade-off between following the law, which is their duty, or following their own intrinsic preferences for the customary practice. And the more the law departs from the customary practice, the more you're going to lose a number of formal judges who are going to follow their own preference for the custom done following strictly the law when they have a judgment submitted to him, even in the formal court. Okay, and so we have this result that I think is interesting. Note, by the way, that it is converging with a result obtained in a paper recently published by Asimoglou and Jackson, but which have a completely different setup for norm because for them, the law is reflecting the prevailing norms, where yes, we assume that the state could come with a law that embodies a practice that is at variance with the prevailing norm. But what is interesting is Asimoglou and Jackson is that they find the same result as we do that too radical a law, because it would reduce whistleblowing, would in fact go against the interest of the victimized party. So the whole issue of how radical should be a law I think is an extremely important one. Note, there are many, there are quite a number of empirical literature findings about this practice, you know, this bargaining argument has strong support in the literature, but also we find papers that shows the perverse effect of the law, like the famous paper by Genico and Sivan Anderson, who showed that in India, more pro-women succession law has had the effect of increasing the suicide rates of both the men and the women, which they explained by the fact that when you enhance the woman's bargaining power, you create more conflict within the household. And this is, of course, bad because it creates a lot of quarrels, et cetera, et cetera. Okay, no, let me go rather quickly and I want, I have no time to talk about changing preference, but I want to say a word about the macro environment. Here, the idea, if we come back to my previous, my previous slide, I can go very quickly on that. Yes, the idea is to increase E. And the idea is, how can the macro environment increase the cost of following the norm? And does it have an effect? And in fact, I can go rather straightforwardly towards the literature here because the argument is rather clear. Here we have beautiful paper. For instance, we have a series of beautiful paper on the practice of food binding for women in China. And these practices have persisted for a thousand years since the Sun dynasty came at a peak during the Ming dynasty, 15th to 17th century, collapsed in the 1911 and 12th revolution. No, there is a debate about what caused that sudden abandonment of the food binding practice. The most well-known guy called Mackey's sociologist, they say it's a campaigning, you know, women campaigning. But there is another explanation that I found extremely well argued by economics who have shown that in fact, it is a change in the macro environment in the following sense. That before what was dominated in China was home-based textile artisanal production and which demands that the women be sedentary and so have a restricted mobility and that when the industrial textile production came to the countryside in China, mobility of the women became very important. Of if you want to say this thing otherwise, the cost of following the customary practice of restricting mobility of the women became higher. And so you see that in the area where the industrial textile production was appearing suddenly, the practice has collapsed in a few years. So that's a beautiful illustration. There are other studies, I won't mention, done on Africa, et cetera, which show that when there is a rainfall shock or livestock collapse, et cetera, the age of marriage is changing. But one of the most important paper is Concelling Western Europe by Voye Glander and Vaud, published in 2013 in American Economic Review, and they tried to explain late marriage in Europe. Why did it come at the time when it came, which means essentially 15th, 16th century? And essentially the argument is a black played argument. They say this was the time where there was the black play, so there was a lot of death people because there was a lot of vacant agricultural land. What you saw is the development of animal husbandry. And who were involved in this occupation, women? And so women suddenly, there was a high demand for women to take care of the animals following the black play. But there is something that they show in the contracts in this period is that as soon as you are married, as a woman or pregnant, the contract will be terminated. So an absolute condition for being involved in the in fact yearly renewable contract with an employer, what to remain single? And so they explained this way, the late marriage. Now what I want to add here is that, you know what is very interesting is that, and I think it's not yet in the literature very much, that what you have is interlinked norms. Like you have interlinked games, but here interlinked games depicting norms in the following sense. That norms can be in relationship of complementarity or at least be compatible with each other. And it can be that a change in the macro environment change the cost efficiency of one norm but not the other. And so a beautiful example is just that one, especially in Northwestern Europe. And I have a paper with Catherine Guichangé that is elaborating on that. That shows what? That shows that in fact in Northwestern Europe, it has been the case that when employment opportunities came for women, in fact the status of women change and their mobility became much more liberalized. But in other parts of Europe, Eastern Europe and Southern Europe, no change happened. Why? Because the mobility restricted norms and practices vis-à-vis women didn't change and remain sticky. And so women could not go and seize up the new employment opportunity and the norm didn't change. So it's interesting to know why in some part of Europe it changed or didn't change. And this is something that we discuss in my paper with Catherine. No, the last point and then conclude on that since I have zero minutes, I have to go in less than a minute is the role of culture. Is that something that is usually overlooked is that social norms are part of a local culture that gives a symbolic meaning to the norm itself. And because of that and because this culture are not centered on individuals, but on group, they are group-centered culture, it means that when you touch a norm, you are touching on the identity of the people because you are touching on their culture. And this is something that Tostan have experienced. They had completely underestimated in the beginning is that much resistance of the people. In fact, there are even areas where I couldn't enter villages because there was so much opposition to Tostan where the real issue is the following. They say, you're aggressing us. You want to destroy our culture. You want to criminalize a practice of female genital mutilation that is part of our tradition, which is part of the initiation process whereby women become adult women. And so it is Western imperialism. So in a sense, once you are making and what I want and maybe one conclusion here is that when you have this kind of attack on the norm, this is something that is hardly understood in the West when we talk about human rights approach and the necessity to defend human rights is how it is perceived by the local people. And so anything that resembles a direct confrontation with their culture is bound to have backlash and perverse effects. And in that sense, I think that it is much better for a government to have to kind of pro-women employment policy. I don't say they will automatically succeed. I've just said it. But that is the advantage that you avoid a head-on confrontation with the customary authority and with the people. But having said so, I must end up with a warning remark and that's my own experience in Medina Gunas in Southern Senegal is that you never know in the defense of customary culture by local people who feel being aggressed by Western powers how far they are being feeling or believing that or whether they are being instrumentalized by customary authorities which have an interest in maintaining the customary practice and which use the cultural argument to mobilize the people around their reaction against what is perceived as a Western intervention.