 Thank you. I'll try to be brief. One the first question is to the question about India. The ethnic fragmentation of India was mentioned and I can quite understand that it's a convenient way to handle this kind of research but it seems to me on the other hand that it also perpetuates the fragmentation of Indian society because of this approach connected with that however ethnic fragmentation in India might be a blessing in disguise because if the Hindus were all united then the other minorities in India would suffer a great deal such as the Christians and the Muslims and so on so perhaps it's a God's blessing that you know fragmentation in India actually does exist which by the way will perpetuate it also by the British colonial power. Then I come to Guatemala and that is that what will be the consequence of this research on the future prospects for the indigenous people of Guatemala and also for the indigenous people in other parts of Latin America because the genocide has already more or less finished but the discrimination still goes on against the Amerindian population in practically all the Latin American countries. Secondly what is the reason for the discrimination against indigenous people in Guatemala and Latin America. Thank you. So I'm gonna collect a few more questions but please try to rapid-fire questions here. Miguel. Okay just very briefly I just want to raise two points in relation to what Francis was saying and in relation to the last presentation especially in the in the sense of the selection of your sample so I was quite surprised and concerned about the selection simply because the the characteristics of your students seems to me that don't really don't represent the distribution of preferences and views. So you most likely have lower bounds as a means of actually the views and perceptions of women in Pakistan but your response about how to deal that was really surprising to me the language. So I really encourage you to to do something about it. Thanks. This is Joe Barthi on the question on India and also related to what Francis was saying because your unit of analysis is at the district but we know that the publichood provision is tied to the constituency levels because and that's where the reservation and all those things happens like either the MLA or the MP members of parliament or members of legislative assembly constituencies. So if you move from a district level to a constituency level measure maybe things would change because reservation policies are all implemented at that level and which might be which might also give us some results on political influence in the publichood provision. Thanks. One more question here in the front. So also trying to be very very quickly and again on the same study on India. So one of the things that I know this is picking up on something that Francis said is how much of the saliency of the boundaries is actually what's causing this relation to fractionalization and lack of access of unequal access to the public services. And just to also reinforce something that you suggested and I think it would be a very good idea intersecting economic classes let's go back that quantiles in income with those identities and see whether that affects access to those services. So I'll ask a final question just picking up on Ricardo's last question. So I was also wondering about intersecting identities and if any of you have looked at that I know several of you have but if there's anything you want to comment on intersections between different types of identities in the work. So we are at time but I want to give each of you one minute to answer your favorite question. And then please continue the conversation over the break. Carla why don't you start. Thanks for the questions very quickly. So the reasons for our discriminations to indigenous people in Latin America they started at the colonial period right so they have been quite persistent mostly economic reasons. So during colonial period they were basically slaves in Latin America until the 50s or 70s indigenous peoples were sold with the land right that they were working on. We have the commodities of coffee plantations cacao plantations banana plantations indigenous people are chip hand labor. So it has been like that since colonial time until today they mostly work in agricultural sector and the consequence of the research for the indigenous population in Latin America in general is that you see is the point we wanted to make is we treat them as one homogeneous group and that's not the case. So if we are having a when pollution education for instance the fact that in the average we see that there are no difference anymore between the indigenous populations and the non-indigenous when it doesn't mean it's true for all the others. It's the same for the labor market right. Somehow we need to know why certain groups are falling behind compared with the others. So I'll answer the first question which asked about fragmentation can does not necessarily have to be a bad thing especially for the case of India. So there is some work which suggests that this negative impact of fragmentation is much less if there are cross-cutting identities. If the identity of caste cuts across the other identity of maybe language or maybe religion. So in the case of India for example Prayena Singh has found out that in Kerala where the identity of caste cuts with the identity of language this negative impact does not exist then. However in India in other states caste as an identity still remains important and does not cut across other identities in fact it is the other way around. The caste has very significant overlap with economic attainment of economic outcomes. So because of these overlapping identities between class and caste, caste again has a very important influence on provision of public goods and other economic outcomes. Second question regarding public goods provision we're not trying to explain variation in public goods provisions so that is not our concern we're just saying if public goods are there their impact they will help in reducing inequality that was our concern. Thirdly regarding the salience of boundaries we do find overlapping identities we do find an overlap between caste and class using a measure of cross-cuttingness and then in other paper we find that this because of this a disadvantaged groups particular scheduled tribes have participated because of their disadvantage have participated in Maoist insurgency in armed rebellious movements to address their grievances. So yes this cross-cutting identities have impact as extreme as participation in armed rebellious movements. So on the intersecting identities part so our context was very homogenous in terms of ethnicity but what we are trying to do is look across the 23 countries and that in that meningitis belt to see if the effect is generalizable and one of the things that comes up is that a lot of them you know you look at the anthropological data on on bride price practices much of that's what of the region does practice bride price. So there could be differences in the intensity there's definitely differences in the kind of amounts that are paid based on the education of the woman and the region of the country so maybe less in urban educated areas bride prices less of an issue in urban education areas and in poorer rural areas. So one of the things that that becomes a concern for us is trying to understand kind of differences by wealth and and basically trying to understand how poorer households basically view daughters as these tradable assets right which is a which is a big problem so if you view your daughters I say selling daughters which is the I'm an economist but anyway it's it's that's what the most direct way to describe really what's happening but if you view your daughter as a tradable asset what does that do as a what does that mean as a coping mechanism especially you know I mentioned that we've done some work on trying to understand how climate change will affect disease going forward basically the consensus is that it will worsen these epidemics in the future so so what does that mean in terms of already poor or inequality by wealth and already poor households that already view their daughters as this kind of because of cultural practices as these tradable assets for bright price so these are the kind of things that we are trying to think about in terms of intersecting issues and and identities. Yes on sample selection so languages it's it's much more difficult than you'd imagine we have a very sort of segregated system of education I went to a convent school because of my family background English is the only language I speak properly effectively I you know I can't communicate in I can communicate but you know not very in a very sophisticated manner and that is the problem with with so if I if you go to lumps most people will speak perfect English they'll have immense trouble with with Urdu which is the national language if you go to Punjab University most people will speak beautiful Urdu but they would have a lot of trouble with with English the problem then becomes that the measure of sexism is not really coming from economics it's coming from social psychology and we are not psychologists who we are worried about how the language of the statement is going to to you know so how do we actually translate so we need a translator that that is fluent both understands the context in which the statement was made in the sexism measure and then can translate it in Urdu and we really haven't been able to find some but it's certainly at the back of our mind we do realize the problem professors to which is already raised the issue of internal consist external consistency which is here is you know a little sample from an elite university in a little bubble can you really extrapolate to the population and and we really can't and it's just that we've actually tried several times to run experiments in other universities and it's been remarkably difficult and we can talk about that over the break intersectionality yeah that's the next