 And in particular, if you're thinking about these things from a policy perspective, they matter to the way policy is going to interact with the economy. Now, much of what I'm going to be saying today is summarized in my chapter with Natalie Bau. It's forthcoming in the Handbook of Family Economics. So if you want to see a fuller description of this, I would urge you to go there. This talk is also a bit unusual and is not necessarily focused so much on my work, but really on the whole field. What I'd like is for the people, particularly PhD students and young assistant professors and old professors too, is to take away from here is that taking the family seriously and incorporating it into your analysis is a very interesting thing to do. And this is a very dynamic field in economics. OK. So you might want to start with some definitions, which I don't particularly love doing, but let me do it nonetheless. So you could have a definition for family. You see that one I've put up here on the slide. It's the smallest group of individuals who see themselves as connected to one another. It fulfills basic human needs, providing for children, regulating sexuality, passing property, and knowledge. And knowledge is also culture between generations. That's more of an anthropological view of family culture. There's a million and one definitions. I've just put the one from the dictionary there. And again, it's the customary beliefs, social forms, material traits of a particular group. But it also is human knowledge, belief, and behavior that's transmitted from one generation to the other. Now, for those of you who think about institutions primarily, you might question whether a family is an institution or whether it's culture. And I don't think this is a very useful distinction in general. And I'm sometimes sorry when the economics literature goes there. We don't distinguish between the two. We think, and here I'm quoting a paper by Shills and co-authors of the family as a set of culturally transmitted norms that influence a broad set of social relationships and that shape patterns of marriage, residents, relatedness, alliance formation, and configure social networks in ways that profoundly influence social incentives and behavior. So I won't get into this whole institutions versus culture. I tend to think of institutions more as formal institutions. And the family, of course, is an informal institution for the most part. There's a lot of variation in traditional family institutions and cultural practices that govern the way the family acts. They govern the form that social unions, such as marriage take place, what legitimizes them, who inherits, who supports parents in their old age, and who is considered a family member. And I'm just going to give you very, very quickly just an idea. This is ancestral family practices but of current population groups. But it's not necessarily their practice, it's rather the ancestral ones. This is a patry locality. And we're going to come back to this later. I just want to show you that the places that are darker is where patry locality, which is when people go and live in the house or community of the parents of the husband, of the newly formed couple of the husband, as you can see, that has some span across the grove. And here, for example, is another practice at a bride price. And again, you can see how it varies across the grove. So the first thing I'm going to do is nothing related to my own research, but it's very interesting. I wanted to give you more macro outcomes in some sense of why the family matters. And what anthropologists do is that they very much distinguish between places where you have a small nuclear family that's not particularly embedded in social networks, in large kinship groups, and those where kinship groups are tight. And one particular tight form or intense form of kinship is that of segmentary lineage and conflict. So here I'm going to tell you a little bit about a really interesting paper, Moscona Nunn and Robinson, about a particular form of kinship, intense kinship relationships, which is a segmentary lineage system and how that relates to conflict. So a segmentary lineage system has unilineal descent. That is, you trace your descent through the father, which is patrilineal, or through the mother, and that's what makes a matrilineal system, as opposed to, say, in a nuclear family, you trace it through both, the father and the mother. Now, a segmentary lineage system has subsets towards segments of a full lineage, of a full lineage, they function as autonomous groups. So those small little circles that you see there, those ovals, those are all different segments of a particular lineage. And you can see that triangle, this is only showing men, so those triangles aren't men only, and you can see they start with one person and then they segment off and they keep on segmenting off, forming a lineage. And what the authors do is to say, there might be a relationship, and this had been proposed already in anthropology, but had not been studied rigorously, between having a segmentary lineage and conflict. Why would that be? Well, in a segmentary lineage, if, say, that individual that you see down on your left-hand side, denoted by one, has a conflict with the individual on your right-hand side, denoted by nine, then what one would do is expect to have an alliance and help from everyone who is in that major segment A. And everyone, and the individual number nine, would have an alliance with everybody in segment section B. So any conflict between one and nine would expand to encompass both major segments aligned against one another. And you can see why you might think that conflict would be broader and take longer to resolve. So the authors study this in Africa where you have segmentary versus non-segmentary lineage. These maps are a little bit hard to distinguish, but the first one on the left is showing you in dark ethnic groups that come from segmentary, the light gray are non-segmentary and the other ones are not in the sample. And then on the right, which looks very much like the one on the left, because it might be hard to distinguish, it has little dots which are the conflict incidents as measured by the ACLED between 97 and 2014. So the authors do a lot of this and I'm just gonna show you two results. One is these partial correlation plots which look at the log of conflict and segmentary lineage conditional on country fixed effects and a whole host of geographic and historical controls and the historical controls have to do with the ethnic group, how kind of the politically centralized it was, et cetera. And as you can see, no matter what type of conflict we're looking at from A through D, you see a positive relationship in these scatter plots between and these partial correlation scatter plots between on the y-axis the number of incidents and the x-axis the segmentary lineage after partialing out the other variables. The other thing that the authors do to really get rid of, to be able to establish more directly the causality, the direction of causality is to look at ethnic groups that are segmentary versus non-segmentary and compare those that lie right on the border. So take the left to be segmentary, the right to be non-segmentary. The question is, do we see a change in conflict as we move from the right towards the left, okay? Do we see a conflict increase because the segmentary group might have more conflicts? So that's a question that they're doing and the reason that they're doing it this way is because it's a boundary regression discontinuity and what they want to argue is that other things would be expected to move smoothly. What would move smoothly? Their historical experience perhaps, geographic conditions, et cetera. So when they do that, they get the following bin scatter plot on the right, you see the distance of the border, the borders given by that dashed line and sorry, that dashed line and then as you go towards the right, you're getting into, within the area I lived in by the segmentary group on the left, the non-segmentary group and as you can see at the border, there's a discontinuity. So that was a pretty neat paper and I think it shows very convincing the relationship between extensive kinship groups or intense kinship groups and conflict. Where else has this kinship intensity raised its head? I'm gonna talk to you about three things, individual psychology, per capita GDP, we're just gonna be a scatter plot and then looking more at causation. So there's several authors that have contributed to looking at and hypothesizing about the relationship between the Western church and individual psychology and the reason that they have and particularly anthropologists is because the Western church very early on started with prohibiting certain types of marriages, levered marriages, marriages where the man married the sisters of his deceased wife. They started to rule those things out, ban those things and then by the early middle ages, they ended up banning marriages even to distant cousins. They wanted to promote it marriage by choice, they encouraged new married couples to set up independent households and by 1500, uniquely in Europe, most of Europe did not have extensive kinship ties but instead they had monogamous, nuclear households, bilateral descent, mother and father and neolocal residents, meaning that you go and you live somewhere where neither your father, neither the bride's parents nor the groom's parents are living. The idea is people who grow up with these type of households as opposed to households with intensive kinship norms might have different psychologies. This was very much popularized in the relatively recent book by Henrik called Weird, Western Educated Industrialized, I forgot our means, Rich and Democratic, there we go. And this, in particular, if you grow up an intensive kinship norm, so you're embedded in these social relationships with many, many people who form your very extended family, that's gonna reward greater conformity, greater in-group loyalty and it would discourage independence, individualism and impersonal motivations for fairness and cooperation. Now, all this sounds very negative and I'm not trying to say that it's bad to have intensive kinship rules, kinship norms, but there are going to be certain downsides. So what would be the positive parts? Things like insurance, for example, that's gonna be much easier to maintain if you're an embedded in a social network than if you're an independent nuclear family. But here the question is that they're looking at as one of psychology, this is a paper by Sholz and all, and what they look at is on the X axis, centuries under the medieval church and on the Y axis, the cousin marriage rate. So what fraction of individuals currently, basically recently, marry their cousins. And marriage to cousins is not as unusual as you might think in some parts, some parts of the world is 50%. So we're talking about first and second cousins and here as you can see that having spent much, many centuries under medieval church, you're much less likely to have cousin marriage. Here they're showing the two things that really matter to them. One is that there is a negative relationship between the cousin marriage rate and a measure of what's called individualistic impersonal psychology which is measures proclivities towards individualism and independence, lower conformity, lower obedience and treating and less likely to treat strangers with the same degree of cooperation and fairness that you would treat your in-group. Okay, and so that relationship is obviously negative and then the same relationship can be shown using the centuries under the medieval churches directly. So this is a psychological relationship. Does it matter? That's something I think is much less clear for economic outcomes. So here's simply a very suggestive graph. This comes from a different paper, there we, oh gosh, and co-authors. And what it's showing you is on the X axis, GDP per capita and this is in logs. And on the right axis, also in logs, the percent of marriages that are between first and second cousins. And as you can see again, there is a fairly strong negative relationship but this is simply a correlation. Now you might ask why might this exist? And one answer, which I like, was given in a paper by Hoff and Senn, a chapter in a book, where they argue that this might hinder mobility. So while these relationships might be very useful in terms of conflict, in terms of providing insurance, when a society is modernizing and you might need a lot of mobility, particularly think about people moving from farms to the cities, then it may not be so positive because you would tend to stay behind as you form part of a social fabric. Either because you yourself want to stay behind or because you face pressure to stay behind. So now getting a little bit more towards causality, I want to tell you about this paper. It's a recent working paper by Gosh, Hwang and Squires, which I really enjoyed reading and actually just incorporated now into the handbook, which has to do with cousin marriage bans in the US. So now we really want to get at the question of causality. What they do is they look at the large variation that exists in the United States towards banning marriages between cousins. This started in the late 1850s in the United States with Kansas and then as you can see on the slide, every decade a few more states joined in. The authors argued that the variation in timing is due to states entering the Union and therefore adopting the more recent legislation and also simply idiosyncratic activism. This wasn't the thing that was particularly important for the US, but some senator might get it into his and it was always a his at this point into his head that this was an important thing to rule out. In any case, what the authors do is to conduct an event study where they compare outcomes for men born in the same state, born in the same decade, and who have high cousin marriage surnames and they compare those to the same type of men, but who have low cousin marriage surnames. So the only way to really measure are you marrying your cousin at this point is to go back into marriage records and look at your wife and have the same surname as you did when she was single and married you. So all these are married, these measures of the surnames comes in the period prior to the reform, prior to 1858, and it's done for the whole US, not at a particular state level. So the authors run the following regression. They look at, sorry, they look at outcomes as a function of state cross decade of birth fixed effects, state cross high cousin marriage surnames, a census year fixed effect, and then what they're really interested in is how, since they're looking at an event study, is how the outcome changes with the decades that pass after the banning in that state of cousin marriages. So let me show you a few results that I found very interesting. The first is, does this legislation have any impact on cousin marriage rates? You'd like to know that it does, otherwise it wouldn't make much of a difference or you would be picking something else up. And as you can see, the decade of the ban is at zero over there and negative one is not shown because those are people who will be born 10 years before the cousin marriage ban and consequently they might be also affected by the ban. And as you can see, there is a drop within the same state for men, and I should be very clear, this is white men, the history for black Americans is completely different here and many change their names after freedom. So this is for white men only, okay. So as you can see, it really goes down, even though we don't really know how much this ban was enforced, it obviously had a bite. The next outcome I'd like to show you from their paper is the impact of these bans on different measures of urbanization and mobility which is what that Hoff and Senn paper was discussing at a more theoretical abstract level. And as you can see here, these are different measures as we go ABCD or 1234, these are different measures of urbanization and mobility. The first one is showing you urbanized people were less likely, about three decades later, to live in places with small populations. The second one is showing you that they're less likely, again, about three decades later to live in farms. The third one is showing you that they're more likely to live in urban settings and the fourth is that when they're picked up in the census, they're more likely, so again, three birth decades later, to be living in a state different from their state of birth. So interstate mobility has gone up. And lastly, they have several measures of income. None of them are exactly the ones that you might think because there's no way to get wages at this point for the US for the most part. Instead, what they have is an income percentile rank where that income percentile rank is based on a decade by decade kind of socioeconomic index based on the years of education needed for you to be in the occupation that you're in. And again, what you see here is that for birth cohorts, some three to four decades after the Banzer Institute, you see much greater occupational mobility or socioeconomic mobility measured this way. And again, I'm just gonna emphasize, this is not showing you that the US has become more socially mobile. It's not showing you that the US has become more urban. All this is always comparing people with the same, with high cousin surnames, relative to low cousin surnames, born same decade, born in the same state. So it's not picking up a general trend. Okay, enough for those, both of those, I think they're very interesting and I spent a lot of time learning about these papers when I was writing the handbook and modifying it right now at the end. Next I would like to pass on to the family's cultural beliefs and practice. So this is where I have worked and so of course I'm the least excited. I never understand people who are excited about their own work two years later. But in any case, this is many more than two years later so I don't find it exciting. So I will go quick on it. So to study the impact of cultural beliefs that the family brings, people have followed what I've called and the epidemiological approach. So what the epidemiological approach does is to exploit the fact that beliefs are much more transportable when you move than are your institutions and economic backgrounds, okay? So what people do is that they study usually second-generated natives. So suppose people have moved to the US and immigrated to the US, you'd be studying the second generation, so the kids who were born in the US, who live in the same geographic area was a city, a town, a commuting zone and very often you still have, you've transported other things maybe other than culture with your parents, maybe you've transported things that were not measuring like wealth or like human capital embedded in your parents or your parents' education. So very often, at least for my own work, I would control for things like similar education level even if it's taking away some of the impact of culture because if you don't control for it, you're not really sure you've gotten rid of the other things that get transported when people cross boundaries. So the idea is again that people, these individuals face the same set of institutions but their choices or economic outcomes can differ in a systematic fashion because their parents transmit it different cultural beliefs and the differences somehow are reflecting their country of ancestry, their country of origin to other parents. So this approach has been used to study many outcomes. My own work was on married women's labor force participation and fertility both on my own and with Alessandra Foley and also Alessina Juliano-Nan have worked in this area. I'm just gonna show you like the classic, classic, I'm gonna show you a scatterplot. On the x-axis, I have how much women used to work in 1950 in your country of ancestry. In 1950, neither the parents were there and certainly these kids were not there. I'm looking at them in 1970 and I'm looking at married women and asking about their labor force participation. Here is the hours worked per week in the US in 1970 given that your parents were born say in Lebanon, in Mexico, Spain, Cuba, et cetera. And this is just a raw scatterplot and a little regression is controlling for your own education, your husband's income, your husband's education to try to get rid of other sources of variation that are coming from culture, although you might think that those things would be affected by culture as well and we find a strong positive relationship. This has also been used to study the relationship between various measures of gender roles or sexism bluntly towards gender gaps in mass scores, incidents of intimate partner violence, a different form of, not gender roles, but a different form of discrimination against women which would be preference towards sons and sex selective abortions. Again, this is all following the epidemiological approach so this is outside of their country. This is looking at people in a different country. The traditions of metrolocality and petrolocality and what we saw before, psychological characteristics, they also apply the epidemiological approach. Okay, so I'm done with that part. Now I wanna tell you about two different cultural practices that have to do with marriage payments. Okay, so payments at the time of marriage were really common throughout the entire world and they typically could be classified as a dowry or as a bride price. And bride price were widespread in sub-Saharan Africa and dowry is still widely practiced in India but also it's becoming more popular in other parts of South Asia. In modern contexts, you would think that bride price and dowry can be thought of marriage clearing transfers and as such, you would expect them and maybe the whole practice to be affected by things like technological change, environmental change policies, et cetera. And what I wanna show you is that the way that the family is organized, its cultural beliefs can have very different effect depending on what the family practices. So here I'm going to summarize a paper by Ashraf Bao, Nun and Voena. And what they do is to look at human capital investments and how that might interact with bride price, particularly when a policy is introduced. So if you ask, well, why do we have certain family institutions, a leading hypothesis is that it helps to compensate for markets that might be missing. On the other hand, even though those markets may no longer be missing, it might take a long time for these cultural practices to disappear. It depends. Now the authors investigate that the reason that a buy price might exist is because it allows imperfectly altruistic parents to capture some of the investments that they make in their daughter by capturing the marriage market, it's part of the marriage market returns to education. So the idea is the bride price is going to incentivize parents to invest more in their child's and their daughter's education if they can get some of the returns via bride price to that investment. So they look at two countries, Indonesia and Zambia. As you can see, what we've graphed here is people are coming from different ethnic groups. They have different, these ethnic groups have different practices, and not all of them come from Ancestriza practice, bride price. So looking at the current population, both in Zambia and in Indonesia, the darker colors are associated with ethnic groups that were more likely to be practicing bride price. So both between Indonesia and Zambia, they're going to distinguish between individuals whose ethnic group practiced bride price and those that did not. And it's very important to note that no ethnic group practiced dowry. First they show that the hypothesis is true in the sense that in both countries, girls receive more education if they come from bride price ethnicities. And they attempt to rule out other explanations like the parents might be wealthier, maybe they have lower fertility of those parents and that's why their wealthier or have lower fertility which makes them de facto wealthier on a per child basis, that's what makes them more educated and they rule those out. And also the link is there, in both countries, girls with more education receive higher bride price. So education is valued, which is what you need. And then they study two school construction programs. The one that is most famous probably is the one in Indonesia in which they built a large number of primary schools between 1973 and 1978. This was Esther's to close paper from a long time back. She studied treated versus untreated cohorts in the same district in Indonesia and found that when these schools were introduced, it had a positive effect on boys education and that this effect was larger in districts with more school construction. That was the point of that paper, it was a long time ago and it was really showing, okay, schools matter, the possibilities for education matter, at least for boys. Subsequent studies, however, did not find any effect on female education. So what Ashraf and her co-authors did was to revisit the setting, but they now distinguish between bride price addition and those without bride price tradition in Indonesia. And again, they compare treated versus untreated cohorts within the same district. And they find that ethnic groups with a bride price tradition did increase the girls rate of primary school completion, whereas the other groups did not. So the negative result was really resulting from the other groups. But here you see that when there's an opportunity to invest more in girls, these parents did. A similar school expansion happened in Zambia, a longer period, maybe a little bit less neat in terms of thinking about how this was randomized. In any case, once again, they compare across districts with different numbers of schools. And what they find is that there is no effect in general of being in the districts with more school for girls. But again, at the district level, if you distinguish between girls from bride price traditions versus non-bride price traditions, you see that the districts with more schooling see increases in schooling for girls that come from a bride price tradition. So your ideas about whether schools are going to make a difference or not really need to take into account what the cultural beliefs of different families are to be able to distinguish its true effects. So now I want to tell you about a different practice which is that of dowry and how it interacts with some preference in particular in the face of gold price shocks. So here I'm talking about a paper by Balotra, Czechováty and Gulesky. And first, a little bit of history. Dowry was, historically, it's very important. It was primarily a bequest, a parental property to the bride. It has disappeared in most of the words world. It persists in contemporary India despite being prohibited since 1961. It's become more common in Bangladesh, Pakistan, and Sri Lanka. Nowadays, instead of the dowry necessarily going to the bride, it actually is often appropriated by a groom and his family. So it's really acting as a groom price. Okay, this is the price for the men. In India, the dowries are very large, sometimes four to eight times annual income, household income. Gold is a very important part of these dowries. Some estimates range from 70 to 90% of households in India give dowry and 95% of those who give report giving gold. Given how expensive it is, families start to save for dowries upon the birth of a girl. So what the authors do is to study the negative effects of the interaction between having this dowry tradition and what happens when gold price increases. And the main idea they're exploring is that an increase in the price of gold is going to make daughters more expensive. It might make sons less expensive since now the son is getting and you as his parents are getting some of that dowry and it's potentially going to lead parents to want fewer girls and more boys. I want to note from the outset that this argument, this main idea, requires the amount of gold not to adjust completely when the price of gold changes. So a price of gold is going to make your girl more expensive but if it adjusts it completely, you'll be left paying the same dowry price as before. And the first thing they show is the dowry's value increases by 80% of the percentage price of increase of gold. Now the price of gold is not determined and it's determined on the world market. It's highly variable. I'll just show you the two pictures. It follows around a mock. I won't show you more than that. What the authors do is that they study what is the outcome of the second born child. Why the second born child? Well, there's evidence that families want one girl and if you were to study third or four for fifth born there's potentially greater selection in unobservable family characteristics when family have more than two kids. They differentiate between two periods, 1972 to 85, which is pre-altersound and then the period 85 to 2005 in which ultrasounds allows prenatal sex selection. They control for a large set of variables that you can see there, rainfall state year, birth month fixed effects, the sex of the first child, et cetera. And they show that increases in the price of gold for the pre-altersound period are associated with higher neonatal mortality for girls. So they're more likely to die in the first year after birth if they thought there was an important price increase in gold in the first month after they're born. For the second period after ultrasound exists they don't find any effect on neonatal mortality but instead what they find is a much lower probability that the second child at newborn is a girl. So the sex ratio becomes a much more skewed towards boys. So they show further evidence regarding the mechanism. They differentiate between those families with a first born boy relative to a first born girl and they show that the results are driven by those, the second born child would have had a sister as a first born rather than a brother as a first born. It's driven by those. They differentiate on religious grounds, Muslim, Christian, and Hindu since this prakari is much more prevalent in Hindu families than it is in Muslim and Christians and shows the results are driven by the Hindu households. And as an interesting further note, they show that the stature of individuals who are between the ages of 15 and 50 prior to being able to have the ultrasound technology, if you were born in a year in which you had higher gold prices this is associated with women but not men being lower height than average. So again, the disfavor, the discrimination against the girl child. Okay, so I've done all this on culture and showing how it interacts with some policies. Now I want to talk about how culture can change. So first of all, there's nothing static about culture. This is often something I have to say to people even study culture because often people define it as something that is unchanging over time. There's nothing static about culture. The rate of change of culture is an endogenous variable and that's a way it should be thought about. So one question that you might ask yourself is why does culture change? And since we're economists, we think it's going to change in response to changes in incentives and those changes in incentives are given by, we can call them loosely shocks, changes in technology, policy shocks, technology shocks, knowledge, environmental shocks. It also can change in response to things that make the ability of others to monitor or to punish transgressors change. So again, culture's not just an internal belief, culture's a social belief and as such is maintained through rewards and punishments. So the reason that men might here not be wearing skirts today aside from the fact that it's cold might be because if you came in in a skirt everyone would stare at you. So that's a way that culture is going to punish transgressors. We will monitor you and kind of look at you strangely. So, but these are just after all cultural beliefs. There's nothing to say though. Men can't wear skirts just like women sometime back could not wear pants of that change as well. So culture can also get stuck and that's important thing to understand is a cultural change like almost any other change will create winners and losers. So when people resist cultural change it's not just because of embedded beliefs but also because of the damage that changing those beliefs might do to maybe their economic prospects or simply their span of control. So my co-author Natalie Bow studied cultural change with respect to pension policies and this is a very interesting study. She studied the introduction of a pension plan in Indonesia and in Ghana. Indonesia has matrilocal and neolocal ethnic groups whereas Ghana has patrilocal and neolocal ethnic groups. Again matrilocal means that they go and live with the wife's parents, the neolocal they live elsewhere and patrilocal they live with the groom's parents once they're married. Now why do these practices exist? One of the reasons that people speculate these practices exist is because it provides parents with care in their old age. It gives them an additional incentive to invest in the human capital of their kids because they're going to reap some of the marriage market and labor market returns of their kids by having them live with them. Now in Indonesia which is matrilocal so they go and live with the parents of the bride, daughters relative to sons in the same household. So now we're at the household level so we can be sure that we're not missing some information in terms of household wealth or other attributes of the household. Daughters relative to the same sons in the same household are more likely to be enrolled in school in matrilocal ethnic groups than daughters versus sons that come from the neolocal ethnic groups. So we do see more investment in girls in those groups. In Ghana which is patrilocal, sons relative to daughters in the same household are more likely to be enrolled in school if they come from patrilocal ethnic groups than sons versus daughters from neolocal ethnic groups. The question that they study is what happens when a pension plan is introduced? And the main idea that they're exploring is a pension plan is going to reduce the dependency of parents on their kids which then might affect the reasons that you want to invest in your kids in the first place. Now this was completely stuck, cultural practice wouldn't be effective but this investment in kids is going to be affected here showing how much the incentives matter. So they explore both cohort variations in particular daughters who would have been too old to receive more education in Indonesia and different intensity of treatment, how the pension plan was rolled out and in particular how many pension plans offices they are in that district. So they're basically doing a triple difference by also comparing across ethnic groups and what they find is for Indonesia that women's education fell in matrilocal relative to neolocal ethnic group when these pension plans were introduced they fell more in those places which had larger pension plans or a more intense pension plan rollout and there was no differential effect on males education by ethnic group practices. So the pension plan affected culture here in a negative fashion and that they invested less in their daughters but you can think maybe education is not a cultural practice and I would agree in some sense but would it really affect it was a practice of matrilocality and what they show is matrilocality decreased more for those cohorts that were treated more intensely. They do a similar study for Ghana which also had a pension plan introduced and they find similar results but this time for men from patrilocal groups. Those men's education decreased as did the practice of practically locality. The last thing I'm going to tell you before I turn it over to my discussant is to talk to you very briefly about a different sort of modern family although I won't touch upon the family part which is that between people of the same sex. So this is a paper on why did attitudes towards same sex relationships change and what I show you in that graph this is a paper with Sahar Parsa and also Virango what I show you in that graph is that ever since this question was asked is it wrong for same sex adults to have sexual relations. You see that for basically a span of 20, 30 years nothing happened. Basically the share who thought it was either never wrong or only sometimes wrong for this for same sex adults to have sexual relations didn't change. What I'm graphing here just to be clear is a share who we call approved. Same sex relationship which are the people who answered it's, those are the people who answered never wrong or sometimes wrong. What I'm leaving out is the people who answered it's always wrong or almost always wrong. And as you can see there's a break in the data one can show this quite rigorously between 91 and 93 there's a big upwards jump and then it takes off. So the people are increasingly approving of same sex relationships and I have it graphed until 2017 here. So by 2017 this is in the United States representative sample, general social survey about 60% of people approve. So the first question we ask is why did approval jump in between 92 and 93? We don't have any information for whether it was whether it happened in 92 or 93. We have 91 and 93 as our data. So in 1981, so 10 years earlier, it was the beginning of the AIDS epidemic. Now the AIDS epidemic did not make people more sympathetic towards same sex relationships. In fact, if you look closely at the data you will see that approval went down during this period. What it did though was it organized same sex individuals behind a common cause. You might have been before a gay Republican, a gay Democrat, this was no longer the case. Now you were united. The most important thing was to find a cure for AIDS. Other important things was to find insurance for health insurance for your same sex partner, et cetera. So this rallied everyone behind a common cause and the Democratic Party was more sympathetic towards it on the ground. And so they really started to create packs, a very important pack that still exists today and gave money to the Democratic Party. So our argument is this shock, AIDS, over time created a very potent political group that had money because it spanned all the income distribution in the U.S. and that was really ready to be courted. And this happened at the national level in 1992 during the presidential election which in the U.S. was Bill Clinton versus George H.W. Bush, George Bush Sr. where the Republican and Democratic parties for the first time ever took explicitly opposing stands about gay people serving openly in the military which was prohibited at that point. So as you know, Clinton won the election and this gave rise to a public debate in Congress that went on for all of 1993. It ended up being the compromise don't ask, don't tell which is we're not gonna ask you if you're gay and you're not gonna tell us, don't tell us whether you're gay or not. So it wasn't really what was wanted but it was better than what existed before. And what these two graphs want to show you is that there was a very large jump in media coverage during this time period. If any of you are old enough to remember when we used to get the news via evening news at least in the U.S. then that's what people watched. So what I'm showing you on the left hand side in that blueish purple line is a number of news segment dedicated in each year to gay-related topics where gay-related is by searching for certain key words. And the one in yellow is doing the same thing but for AIDS-related stories. Obviously there's gonna be an overlap between gay-related stories and AIDS-related stories and that dashed line is showing you that overlap. As you can see, 1993, there's a huge increase in gay-related stories. That large increase is not due to AIDS-related stories, as you can see by the fact that that dashed line is going down. That same point, there's a lot more media attention, mainstream media attention, can be seen for newspaper articles which look at a balanced panel of I think 59 newspapers from 1987 to almost 2005. And since 2003 and the share of newspaper articles on average that are dedicated to gay-related topics. Again, searching for key words. And again, you see the spike in 1993. So there's a lot more attention being given to these topics. What we hypothesize is that the public debate and the far greater salience of gay-related issues let people to reconsider their positions towards same sex individuals and that this initiated a process of cultural change and a fusion of different values over time. Now this is a national event and so we can't get variations since it's national except over time. So what we do is we argue that individuals from places, that is from states and one part counties is another data set with greater exposure to gay individuals and thereby with greater exposure to mobilization, greater number of grave friends and creintices. People are coming out during this period. Local news, greater local news. Then these places, these people will be more affected. And you could take this to be either from salience or from contact theory. How do we measure exposure to gay individuals? It's not so easy. We measure exposure to gay individuals via cumulative HIV cases in 1992 or by the proportion of households in the 1990 census who for the first time are asked a question about whether there's another adult of the same sex living there as a partner. And what we show is that those places of greater exposure are the ones that change their minds the most. That's where you see the greatest impact. This is of course controlling for a lot of things that are different across counties, et cetera. And in particular, there's a state and county fix effect across these decades but also income, size of the city town, sex, race, education, et cetera. So we've considered this as an example of how incentives for people to come out, to mobilize and to organize ultimately led to a process of cultural change once the institutions of the national parties and the media changed their incentives in terms of how much attention they gave to these issues. Okay, let me conclude on a bit over time, I think. So social beliefs and family cultural traditions matter a lot, I hope you've understood to economic outcomes, whether it's women's labor force, gender segregation, time spent with children, especially things that relate to women. The importance of culture in the family's essential role is often overlooked in economics. I think it's very important. I encourage you to think about it when you're thinking about your research. And I hope that what I showed you in the last two slides is that culture is not destiny. Beliefs change and often radically in response to other changes. And but that nonetheless policy when you're formulating policy, you need to take culture into account without becoming a slave to it. Thank you very much. Thank you very much Raquel for a very, very nice, very, very informative paper and presentation. I'm going to set my timer for 10 minutes but I see there's a clock over there so I'm going to watch there as well. Because I still have 10 minutes. But I'm going to be discussing this paper. I highly recommend, Raquel mentioned going in and looking at the book chapter. Highly recommended it is very rich, very informative. And really kind of adds to our understanding which I'm very happy to see the literature has progressed since I started studying this in grad school. But really adds to our understanding of the organization of family and its role in the transmission of culture for economic outcomes. So Raquel had a very, very rich discussion already and I won't spend too much time summarizing what you already said. But I'm going to highlight three takeaways and I'm going to give you three main comments as a discussant for this chapter. So first, first takeaway, I think I'll mention this and reiterated this very well throughout. Culture is endogenous, very, very important to note this and its implications for how we think about policy. Number two, family is the main structure for transmitting culture. Number three, culture has very real effects on economic outcomes and inequality and vice versa. And this also then I mentioned has very important implications for how we frame effective policy for economic development. So I'm going to give you three main comments that I mentioned focused on thinking about how we define culture and why we should care. When is culture beneficial to economic development? So I think there are a number of papers that have shown when culture is detrimental to economic development and trying to see the flip side of this. And also suggesting future or further areas of study for people interested in doing more on this topic, which I think we should all be interested in doing more on this topic. It's again, very important topic. All right, so defining culture and why we should care. So again, Raquel had a very, very nice discussion on this. I just want to highlight some more things here. So oftentimes my favorite definition, or I should say my favorite definition, is to think about culture as a set of informal institutions, right? So you think about social norms, think about beliefs that influence group behavior and decision making. So that said, it is very, very difficult to define concretely what it is, right? So this is me trying to type in culture into our econ database. And this is a very, very simple figure that I came up with, right? This is from Alberto Biscin and Verdiere, not simple at all, right? It's very, very non-simple. And they were trying to kind of outline this link between formal institutions, as Raquel referred to them, informal institutions, culture and how these things interact with each other over time. But again, just to highlight how difficult it is to really pin down this definition of what culture is. Okay, on this point on culture being endogenous, then the question is, where does culture come from? So Raquel mentioned this, and this is something that we are learning more about culture is affect and is affected by environment. So thinking about the natural environment, it affects formal institutions and is affected by formal institutions, which then affect and turn economic outcomes. So again, going back to the Biscin and Verdiere graph from the last figure. So there's a nice paper that I want to highlight on the environment side, which I think, maybe we didn't get to cover too much in the chapter, I think this, so just new paper from Basi, Fisbian and Gribresilasi. I think it came out in Colometrica, 2020. And they basically try and link American individualism to this American frontier, right? So you see the picture, you see the settlers that go to the West, right? And highlight the links between the terrain, the isolation of the environment, how difficult it is to survive in the American frontier, to values around that prize American individualism today by looking at the surnames, I think, of the children. So of different generation of children, sorry, not surnames, first names. And it's very interesting, right? So people start giving their kids much more creative names, and this is what they use as a measure of individualism. And very nice paper, kind of trying to let us know again how environment is linked to these informal institutions and culture. So Raquel very nicely covered the role of formal institutions. So thinking about marriage markets, which is something that I also do some work on, and thinking really about the role of bride price versus dowry, right? So I like to, you know, I think we might know what they are, you know, this is the transfer from the groom's family to the bride's family. This is Ghana, this is in India, this is the transfer from the bride's family to the groom's family. And Raquel covered this in thinking about how economic shots, negative shots, can then have very disparate effects, depending on whether you're in a bride price culture or you're in a dowry culture, right? So formal institutions, environment, and these are the things that affect and are affected by culture. So why should we care? So again, Raquel did a very, very nice job going over this. I'm going to just highlight some things again and highlight this point, especially in particular, that not accounting for culture can make your policy interventions fail, right? So what does that mean? So, you know, Raquel presented this school constructions paper, a very nice paper from Ashraf Et al that looked at the role of bride price traditions on female education, right? So by the school constructions that were happening in Indonesia versus Zambia. Also very, very nice discussion. I did highly recommend everyone read this paper on looking at intimate partner violence, IPV, and how increases in or declines in kind of cyclical male employment is associated with IPV depending on where you are and the cultures in where you are. So they look at developing countries as the Balotra et al paper, and then they look at the US and the UK and you get very, very different signs, right? On the links between cyclical employment of men and women and IPV depending on where you are. And they highlight again this idea of male status in societies, this cultural factors has been important in understanding why your signs might flip when you're looking at these effects. Also another big one, I know there are a lot of development economists in the room, right? Development economics conference. So cash transfers, right? So there's a new kind of body of work trying to understand the effects of cash transfers on household welfare and IPV and how that might depend on culture, right? So how women status within the household, women's bargaining power in the household is viewed and also how kind of challenges to the male image as a breadwinner, especially in very patriarchal societies might affect then the effects of these cash transfers to households. So we also have very new work, right? So on online labor markets in Nigeria, I do a lot of my work in Nigeria. And one of the things I stood up towards in the descriptive data paper for coming was that if you look at just within hiring managers, so looking at hiring managers, looking at applicants that are applying on these online labor markets, and you said, okay, what is the effect of gender on hiring in these online labor markets, right? So we're talking about structural transformation. Online labor markets are quite big and they're growing, everyone has access to ICT these days. So if you just look at gender and you look at the outcomes for hiring, you don't see any effects when you add hiring manager fixed effects. However, when you look at the interaction between gender and it and co-ethnicity with the hiring manager, then you see quite significant effects, right? So essentially, hiring managers are much more likely to hire co-ethnic men, much less likely to hire co-ethnic women. And what's happening, these women are applying for senior positions within the firm. And so what we think is going on is that, again, cultural norms around, you know, can a woman be senior to me from my group in my workplace is playing a role here in terms of the hiring. But again, if you just looked at gender without considering this, you would say, oh, there's no gender bias. There's nothing, you know, no effect here. So just highlighting again why it's important to study culture. Okay, on the second comment, when is culture a beneficial to economic development? There is a nice body of work going back a few years, so Sam Bulls at Al have talked about, you know, cultural norms as being important. And I think Raquel touched on this in promoting pro-social or cooperative norms, right? So like my favorite story, so this is a, he has this book, my favorite story from the book was my study on preschool, parents picking their kids up from preschool, if you guys have heard this story. And essentially, you know, economists, the teachers were very upset that parents would come and pick up their children late from preschool all the time. So of course, you know, talk to the economists, what do economists say? We're all economists, oh yes, provide economic incentives, make it costly for them to pick up their children late. Of course, then what did the parents do? They stopped picking up their children even later. Why? Because now they have completely foregone the pro-social norms of not trying to, you know, leave your students or your kid's teacher, they're very late because they have families they want to go back to. But I say, no, you can just pay, it's now a fee. Pay the teacher and you can leave your kid there for as long as you like, right? So this concern of, you know, economic incentives crowding out pro-social norms and the role of culture in promoting pro-social norms, I think it's something that is also very, very important to study here. Okay, my final comment in my last minute. So one of the, they're kind of thinking about the further areas of study. I think this is less covered in the chapter, which is again, very rich. So I would probably understand, not much space to cover everything, but then about the role of religion, right? I think this literature and the economics of religion is very, very exciting, very, very important considering the huge role religion has in many of our lives, right? So we can think of religion as maybe both in the realm of formal and informal institutions. And one of my favorite recent papers comes from Oriole et al. I believe it was published in the QJE in 2020 and they essentially look at the relationship between religiosity and religion and economic outcomes using evidence from Ghana and highlight how, for example, religious donations are viewed by many people as a substitute for formal insurance, right? So when you don't have access to formal insurance, you pray a lot. And you say, we hope that God will protect us, right? And you go to the church and you give donations in the church because partly you are trying to insure yourself against negative shocks. And so what they find is that when they introduce formal insurance, people then take up formal insurance and reduce religious donations to their churches. And that's my clock telling me I'm out of time. Almost done. So very, very important aspect. I'm thinking again, going back to Raquel's point on the role of policy in shocking cultural norms and also then the importance, right? People are using maybe what might be somewhat suboptimal ways of insuring against shocks. Maybe we should have more formal insurance so that people don't necessarily have to depend on the church. Same thing again. There's new papers on thinking about how shocks increase religiosity. Very, very big area. This is one of the things I hope that any grad students or junior scholars or senior scholars, anyone looking for a new research agenda, I think this would be a fantastic thing to study because we just know so little about this aspect of culture. And then finally, on race at an ethnic discrimination. So there is a new, not new, it's old, but hopefully you're thankfully getting much more study with the new increased study on group based inequality. But this is this literature on stratification economics that tries to understand the origins of persistent group based inequality. So you have this new issue in the GEL that came out, solidarity, et cetera, and co-authors are trying to understand again, what is the role of these informal institutions, formal institutions, and also what we don't really understand is trying to understand the role of the family, right? Given that the family again is an important center of transmitting values. What is the role of the family in transmitting suboptimal values around race or ethnic discrimination that might then lead to persistent negative outcomes for group based inequality? So two areas before their studies. And again, thank you very much to Raquel for a very, very interesting, very informative paper. Thank you very much. Well, it was a great talk and a great set of provocative thoughts about how to move forward. Let's open now the floor to discussion. And I can't see anything from here. Who wants to start it? Too overwhelmed. Okay, that's all there, over there. So thank you Raquel for a very interesting presentation. It was really nice to see a very concise presentation of how culture and family interact and how it evolves. So my question is more about, you know, the culture being endogenous, right? So since the process is so slow, right? It takes a long time. And especially thinking about it being transmitted through family members and you know how it happens given that people who are going to be the losers from this cultural change are the ones who are in power at the moment, right? So it's going to be very difficult to do this because it's not incentive compatible for them to support such cultural changes. So do we have any evidence of how this cultural change, you know, which is going to be very slow, how it can be accelerated or you know how? Do we know anything about that? Could you also introduce yourself briefly and anyone else that asks questions as well? Sorry, yeah. I'm Nikita from the Indian Statistical Institute. Thank you. Can I? Okay, I'm Mariano Tomasi from Universes San Andres in Argentina. First I just remarked to Raquel, at my age I have difficulty sitting 40 minutes listening to a talk but this time it was very easy, it was very engaging, very, very motivating. Just two minor comments. In terms of motivation for why we should study family, perhaps we might add the fact that all modern knowledge suggests that human development is so temporal dependent and everything that happens very, very early matters for your whole life and obviously what happens very, very early happens in the context of the family. That's why the family is the main producer of human development in some sense. I'm connected to that and we were chit chatting here with Santiago about that. Another crucial component of family formation cultural and characteristics of different societies are these patterns of assortative mating. In some sense of the extreme version of some of the things you mentioned, of the more society, what version of that. And as we were discussing here, there are little simulations there that show that if you were able to change the norm of positive assortative mating you would reduce inequality in a dimension that no successful welfare state could ever do. Thank you. Thank you very much. And I think first just to welcome the presentations. I mean I found it quite insightful. My name is Aya Bongataui, I'm from the University of Badwata Shrine in South Africa. I have a question and I think it came through from Belinda's presentation around intimate partner violence and the link between that and some of the studies she looked at between that and cyclical male unemployment. And I'm quite interested in whether or not you've come across any studies that are able to look at the link between intimate partner violence and actually structural rather than cyclical male unemployment. Well I think one of the things we're dealing with in South Africa is not only the significant incidents and prevalence of intimate partner violence but also the interface between that and structural rather than cyclical male unemployment. And then maybe I think the second part of the question is around just the comment you were making around the role of social transfers in tilting the balance of power within the household. So I'd be interested to hear just a bit more of your reflections on that. Thanks. Thank you and one final one maybe? Like to start. Why don't we start with you? Okay. Okay. Yeah, it works. There's no way I'm remembering anyone's name and I barely could see where the fourth person was sitting. I never found him. So let me start with cultural change. Again the idea that it's going to be slow because it's going to take through generations and then it's resisted by the older generation and that then contributes to its slowness. I mean I think there's a bit of truth on that but there's also evidence in other directions. First of all, not all cultural change or maybe not even most cultural change leads to intergenerational conflict. It can be within the same generation of the conflict and so you would have one group that might want it and one group that may not. If you think about women's role in society then there was this nice study by Avonya Washington which showed that congressmen who had daughters were much more sympathetic about voting on female issues, positively on female issues than those that did not. So again there is an aspect of intergenerational altruism which might get more turned on purely for selfish reasons because you care about your daughters or because knowledge, having daughters makes you understand more what the consequences of certain views are or certain cultural practices are. Second in terms of examples the one of social change being fast when you look at attitudes towards same sex relation and not just attitudes but legal changes that have occurred in many countries in the world not just the United States, not just Europe and Argentina and I'm not sure if Uruguay also changed but in any case these are changes that were very rapid in that scheme of things and while it might be true that older generations and you can show that we're more conservative if you look at the same opinions the same question asked by generation older ones are more conservative. Servitive nonetheless everyone changed it's not just young generations who changed their minds. So even cultural beliefs that one holds at the age of 40 are subject to change. Mariano thank you very much for your questions. So I guess one question I would have is is assortative mating, assortative marriage a norm or is it an incentivized practice? And I would argue that there is room for both in the sense that it's not so much I would call it a norm but I would say one paper that I did with Nezzi Gunnar and John Knowles and I was studying assortative matching looked at the correlation, simply a correlation but nonetheless very interesting between the degree of assortative matching across countries and the degree of inequality across countries. The idea being I assume you're not Bill Gates for a second the idea being that if you are giving up more to marry down you're more to be less willing to do that and giving up more means giving up more money. So if you're going to have a husband or wife who is less educated than you so I'm talking about sorting only at the educational level here you're going to be less willing to do so and there's a very strong relationship in the data. On the other hand what else goes on you don't hopefully marry your spouse just for money but you also love your spouse in the best of cases and love is something that's fermented by spending time with people so that's why people marry in college or they marry with their graduate school classmate you only marry the people you know and you tend to know people of the same background think about your PhD program how many unfortunate PhD graduate women and I'm kidding by the way married PhD men because they were in the same class of them but it's true not the unfortunate part but women because there's so few of them end up being disproportionately married in economics to another economist and anyone who's ever tried to hire a female economist knows that a big question is can you get a job for her spouse so it's another burden that women face not by choice and not because there's discrimination when they are in the economics job market so please take that into account when you choose to marry so I think if you mix people more and I'm not talking about the elite levels now college and graduate school we don't care so much necessarily what happens there though it has consequences but if you think about secondary schools where you also meet people imagine that you meet people from many more backgrounds than from say an elite background or for a poor background and of course this relates to Chetty's recent paper that founds at schools there were more mixed people who were from poor backgrounds tended to do better not necessarily through marriage but maybe simply from peer effects and studying but also information knowing that you should want to go to a good college knowing that it's not that hard to get into a good college or to get into a good grade or having something to strive for so in any case I do think it can be very much affected by institutional structures in the country I'm going to let Belinda talk mostly about IPV and structural issues good luck I'll just add my own thing to that which is that that work is a little bit all over the place in terms of what it finds so when we look through that literature we find one conclusion it's very hard to do structural because you need to find some source of variation to be able to do a rigorous analysis and that's why a lot of the analysis looks at you know, cyclical fluctuations I really would think that there's a nice room for papers that look at I think that they look at how easy it is for men and women to divorce and also how easy it is for women to support themselves if they walk away but right now the evidence is mixed and lastly on transfers and balance of power so I think a lot of economists think that you give more money to women let's say with cash transfers and they tend to be made better off that makes them have a greater bargaining power and the family is questionable and we've seen more studies that say that's questionable because in the end it depends how appropriable that transfer is and so if you can be appropriate or if the man resents it as it happens to be the case with some of the studies on APV actually they find that when women's employment goes up women suffer more IPV incidents because there is some people seem to think that it's a psychological mechanism a backlash mechanism against those things and lastly but not least I really want to thank Belinda for her excellent discussion and for a lot of the papers that she brought to my attention which we should have incorporated probably but it's very hard to it's like a running race that you're always behind but we will certainly take them into account for the future and I certainly want to read them including very much yours so I will just add one more thing thank you very much again Raquel I will add one more thing to the cash transfer IPV I don't know that literature as well so I'm Nigerian and the women in my family always said a woman should have a separate bank account that the man doesn't know about completely I take no stance normative stance on that I'm just stating the statement and one of the things that this literature has started doing is to say we'll have the cash transfers to women and this isn't the experimental stuff I see coming out recently but we'll make it so that women can hide the amount that she's getting and so their starting to actually see that then women's savings spending etc behavior changes very drastically depending on whether she can hide the amount that she's getting and this is in sub-Saharan Africa that's the context but depending on the amount that she's getting in these experimental studies so this just goes back again to highlight this point that really understanding the cultural context is very very important for us especially as development economists that are interested in influencing policy because it can very very much change how the impacts of your policy and for good for ill and we hope for good but by studying these informal institutions you can then much much much better a frame effective policy for development so let's add that there Thank you very much We have time for a couple quick questions if anyone wants to come back into the debate Okay then I have there's someone there, okay sorry the lights are impossible there's someone right there Microphone please Thank you so much for the amazing talk I'm really excited about it and my name is Yasmin Abdul Fateh I'm from University of Prince Edward Island Cairo campus and I just want a little elaboration on the saving for the women I believe it's really crucial especially for developing countries so if you would like give us more on the literature that's been done that would be great Thank you Thank you very much for the question and yes I don't know if we can send if we post some of the resources definitely Raquel's paper and we can post some of the references for this new experimental work on that Please translate Thank you I'm Lima Jaime from the University of South Colombian in Eiba Well, they said there are cultures where when there are gifts women give them more education yes My question is if that education that the girls get improved in quality how to reduce those violences of gender so that they have more empowerment and that the families that family culture of gender violence where only the man is the one who sends and the woman is the submissive has been changing, has been improving in that sense I suspect Raquel can definitely do that and there was someone else if not here then there's someone in the middle just behind Thank you My question is to Raquel and it relates to the definition of culture where you include social norm but when I was reflecting on the lecture and thinking of certain social problems we face such as early marriage there is, and again female labor force participation, there is equal emphasis and often separation of culture and social norm for example issues of sanction and stigma therefore papers and interventions targeting changing community social norm leaving aside family specific cultural practices so therefore as if there is a two horse race and if you were to bet which one would you put your money on that well for example if you were to address this problem of mass early marriage which is a problem in South Asian sub-Saharan Africa in certain parts of the world whether you would advocate reforms that would bring about changes in family specific culture or would you promote interventions that would focus on shifting community specific social norms Thank you Thank you very much and we have about 30 seconds for replies So the question and IPV that was a question that was asked in Spanish I'm going to answer in English is that okay because I want everybody so the the question was when we see these women's education go up do we see not not IPV but necessarily violence towards women go down and there are some studies that have looked at this I know that in the cross-section more education is associated with less reported violence at least in the household but I don't know that for these sort of changes which are induced by policy I don't know of any study that and then for culture versus social norms you know I know Jan Elster relatively well and we would always get into discussions about social norms versus culture I do not make a distinction I know that some people do I do not make a distinction between the two and so whether you should focus on changing family culture versus the norms say of society culture of society I mean they kind of go together so I think it's what what you're trying to incentivize or to change and how you would target it and it may not be by trying to show oh you should have a different norm or that your social norms are bad but by maybe the economic changes that incentivize the social norms although there's evidence for example there's newspapers that explore soap operas and the penetration of cable TV one in Brazil for the soap operas cable TV in the context of India and they show for the cable for the soap operas they depict it families that were much smaller in size than the traditional families that these people actually had and what they show using variation the cable TV's ability to penetrate that area so actually receive the signal they show that seems to have changed people's fertility outcomes so people ended up having fewer kids in those areas where they watched women having maybe zero kids or one kid in the soap opera and then for the cable TV they also found changes in for cable TV in India they found changes in people's ideas of what women's roles are and also the amount of agency that women had so particularly how much they were allowed to go outside on their own unaccompanied and things like that let me turn it over to Belinda I don't know okay then I have to bring this to an end we're a little bit late I'll ask you for another round of applause yes