 I should start by saying, I'm very grateful for this opportunity to get a chance to present some of the work that we've been doing on the country study on India for the past I guess year or so. The work is still underway so I'm very grateful and looking forward to getting comments and suggestions from you all. I should also say I'm quite miserable about making this presentation because I normally go way over time when I present a single paper and now I'm supposed to summarize five different papers and I'm supposed to do all of that within 20 minutes so it's really painful but it's a really nice opportunity and let me start by saying, so the kind of context within which this presentation is going to be is sort of starting from the point of view or from the perspective that well inequality in India is a topic that's actually currently under quite a lot of discussion and debate, it gets a lot of attention not just in India but also around the world, not least because of a number of very kind of big splash publications that have come out recently. There's a paper by Chancel and Piketty that sort of documents India's inequality over a very long period of time since the early 1920s, it's entitled Inequality in India from British Raj to billionaire Raj and documents this very rapid and extraordinary increase in inequality over this time period. Another book that was recently published by James Crabtree who's a columnist with the Financial Times called The Billionaire Raj talks a lot about what's going on in India and focuses like the Chancel and Piketty book very much on what's going on at the very top end of the income distribution and documents this really rapid accumulation of enormous wealth amongst the very top of the income segments in India. And so that's the main contention of these papers and also the subject of quite a lot of this debate which is around the economic growth in India has been noticed to have been rising and been at very high levels in recent years and these publications are coming out and suggesting that this has been associated with a very very significant concentration of wealth at the very top end of the income distribution. There are skeptics these views are not entirely endorsed by everyone an example here's a recent article by Surjit Bala quite an influential columnist in India who claims that there's no evidence that India has experienced an above-average increase in inequality. So this is still a debate but it's certainly something that's put Indian inequality on the sort of on the map and sort of in the public domain. What we're trying to do with this project is to complement this activity, this research by asking a number of additional questions that have perhaps been posed in or answered in these publications I referred to. The first one is this consensus of inequality being high. Is there additional evidence that corroborates some of these views or is this something that is just coming out of these individual studies? There's certainly no question that India has achieved significant poverty reduction in recent years and so the question is whether there's an analog of change occurring in inequality the kind of forces that have been underpinning the poverty decline. What has there been their impact on inequality? We're looking at the trends of inequality in other dimensions than just an income. What's happening in rural areas? What's happening within the country, not just at the national level? And what are some of the patterns of income mobility that we can discern from the recent evidence? So in some sense this project I guess and this was not done deliberately but in some sense it's kind of complementing the discussion of the previous session where François Bouguignon was proposing that there's a dashboard of indicators that we should be looking at. We're trying to at least look at some of those items that are on his dashboard as well as perhaps some that haven't been included in his dashboard but that perhaps in the context of India could also merit attention. So the project consists essentially of six studies at the moment of which five have more or less been done. They're all in draft form still and one is still very much underway. There's a paper by Professor Himanshu of JNU and Rinku Murgayos at the World Bank Office in Delhi which looks, it basically assembles the secondary data and looks closely at all kinds of evidence on what's been happening to inequality in India over times. A second paper is the way by myself and Chris Elbers at the Freyja University where we're looking at inequality and structural change in the context of a single village and trying to see what's going on at the very local level and I'll try to motivate and justify that a little bit further. A paper by Abirup Mukhopadje who's here and David Orzaynke at the Free University of Amsterdam as well looks at the spatial decomposition of inequality. We have a paper by myself and Haiyan Dang where we're looking at intergenerational mobility. So we're trying to get at some of these ideas of what's been happening to mobility over time. And we have a paper by Roy Funderveda from the World Bank and his co-author Melody V, I believe who's in Berlin looking at intergenerational mobility and human capital, looking specifically at education, intergenerational mobility and education. We have a final paper that's trying to look again at the kind of findings of the Chancel Piketty study which is based on methods that are quite complex and somewhat non-transparent and possibly not entirely, one may not necessarily be entirely at ease with all of the assumptions that underpins it. So we're looking at the same question of what's been happening at the top incomes in India but based on a different methodology, one that involves looking at house price data to try to fill in the gap of what's going on at the top end of the consumption distribution. And that's being carried out by Kertan Rangan who's also at the Free University in the Netherlands. So this is, I suspect that this may not be legible to you all but this is kind of the way I see these five papers filling together with the paper by Himanshu and Rinku, sort of giving the big picture of what's been going on in India with respect to inequality based on whatever available evidence we can put together, wealth data, earnings data, income data, consumption data, non-income dimensions and so on to try to fill in as much as we can from the available evidence and it's very important also to look at trends. So we're not just trying to get a snapshot of what's going on in India at the moment but also to document how things have been changing in this particular paper since the early 1980s. The second paper by myself and Chris Albers looks at this, it goes from the big picture to the very, very micro picture to try to isolate and identify some of the things that are going on at the local village level. We then turn on the lower left hand side to the paper by Abirup and David which looks at spatial decomposition of inequality. We try to basically unpack inequality asking the question of whether it's really that compelling to only look at inequality at the level of India as a whole or should we be trying to look at inequality also at the level below the country level and motivated by the analysis of the village in India. We basically try to look at the importance of village level inequality within India as a whole in this particular paper. The Dang and Lanyal paper looks at intergenerational mobility as I said but tries to employ given the fact that we don't have the kind of panel data that are necessary to do this kind of analysis properly. We try to get around that by invoking or employing some sort of approximate methods of constructing synthetic panels to try to get at some of these mobility questions. And then the final paper by Roy from the Veda and his co-author as looking at intergenerational mobility and documents that in India intergenerational educational mobility is currently in international terms very low but there is a positive sign of improving or increasing intergenerational mobility. So turning to the Himanshu, I'll try to give you just a feel for some of the findings that are coming out of these papers. The Himanshu and Murugai paper summarized this fairly large literature on inequality in India and document the evidence from a variety of different sources and they generally point to indeed inequality rising in India and they illustrate the sort of sectoral transformation that's occurring in India of a move away or out of agriculture that's sort of declining importance of agriculture in the Indian economy towards the sort of non agricultural sector and distinguishing also between the sort of formal and the informal sector as being important parts of understanding the puzzle of inequality in India. So here's a picture that extracted from their paper documenting an increase in inequality. It's not a dramatic increase in inequality and this is based on consumption data, which is sort of the bedrock of data that people have been looking at in terms of inequality and poverty in India is coming from the national sample surveys and it's the various rounds of national sample survey data starting in 1983 and extending up to 2012. So they document two important things. One is that indeed in recent decades, so the last say decade or 10, 15 years, inequality has been rising with some leveling off perhaps in the most recent period, but there was also an important period of declining inequality in the very early years of this series of cross sectional data. So inequality actually declined somewhat during the 1980s and the early 1990s, which is important because it means that it's not necessarily inevitable that with economic growth, we should be seeing economic inequality rising as something that Martin has already pointed to earlier that it can rise with economic growth, but it needn't rise and in India the experience of the early of the 1980s was that inequality actually very slightly, but did actually decline over this time period, particularly in rural areas. The incidence of growth is sort of captured as captures this as well in the pictures on the right hand side here. When we look at indicators other than consumption, which has been sort of the mainstream indicators of inequality or data that have been used to understand inequality, we can look at income inequality and there are very few good data sets that capture overall total income in India, but there is one survey called the IHDS, the Human Development Survey in India, which was collected in two periods in 2004 and then again in 2011. And those data can be used to look at income inequality and they suggest that income inequality is in terms of levels, but at a much higher level than what we typically see with consumption inequality and that they also document some increase in inequality over time. It's not dramatic during this time period between 2004 and 2011. The consumption data also suggests it's not been dramatic during that time period with most of the increase in inequality actually occurring before 2004. Curiously in the income data, the suggestion is that inequality in urban areas is lower than in rural areas. This is not supported by the consumption data where you look at urban and rural inequality. So there's some issues there of comparability and so on, but the general trend of rising inequality is supported and income inequality levels seem to certainly be considerably higher than the consumption inequality measures that we have. Wealth inequality is something that can be looked at. It has been looked at in the right hand corner. We see the picture that's taken from the Champs-Eau Piketty studies, which documents an increase in quite a significant increase in inequality in recent decades. And to the point that the levels of top income inequality nowadays are similar to what was the case in the early 1920s. So it's something that's really quite striking. But it's also supported by other data that look at wealth more directly. These all India, the AIDS data have their own problems are not as complete as one would like, but they do seem to document as well this rising inequality in wealth. We have a very important feature of inequality in India. And this is something that's been long known and long understood is that there's very important group differences and a crude way of dividing the Indian population to different groups is either on the basis of caste status or on the basis of sort of religious affiliation. And so these two tables just document that inequality differences between these groups are very significant and have remained significant over time with the schedule tribes and the schedule castes typically receiving a much smaller fraction of income than their population share than their share of the population, whereas the others and the other backward castes and the remaining groups have done much better. And similarly, in the case of religious affiliation, we do see some evidence that the Muslims have a sort of less share of total income than their population would suggest with the Christian very small relatively small Christian population doing particularly well. And this is also something that has certainly not diminished sharply over time. Other indicators of well being that are of potential interest, of course, are health education. We have just here a couple pictures, one picture documenting the significant higher stunting rates among the schedule tribes in the schedule casts. And although there's been some decline, those differences haven't diminished or haven't changed sharply in terms of stunting. And similarly, when you look at dropout rates from school, we see that the schedule tribes are just much, much more likely to drop out and say secondary school in particular, but at all levels of schooling and the schedule casts also certainly in terms of secondary schools are dropping out more than the rest of the population. So that's a real whirlwind tour of a paper that could actually receive a lot more time in this presentation. But I want to move on to the Elberso and Laniow paper, which looks at one village and the village we're looking at here is a village called Palampur in this province or in the state of Uttar Pradesh. And it's been studied very intensively since the late 1950s. So we have a very detailed picture of what's been going on in this one little village since the late 1950s up until 2015. And the big picture there is that we've seen economic growth has occurred. We've seen in terms of big forces of change. We've seen green revolution technologies and rural non farm diversification of having exerted a big influence on the village economy. And that's reflected in a declining headcount rate and also an improvement in income mobility. People are moving out of their particular income fractiles over time as we go from one year to the next. But we also see and this is what I want to underscore. This is potentially quite important. We see a quite a considerable rise in inequality within the village. And one could ask the question is that important or not. And I think it's important for multiple reasons. And one of them is that it could potentially be one of the reasons that we see people claim that inequality is rising even though objective information at the country level doesn't necessarily suggest that to be the case. People's reference group might be less than the country as a whole. It might be a local community and if inequality in the village is rising that may well influence how people perceive inequality to be evolving. So their reference group might be the village rather than the country. Another important reason to think about and to take seriously inequality at this very local level is that work we've done work to document that inequality at the local level has been shaping and has been shaped also by institutional changes in the village level. So things that are going on at the village level are shaped by the inequality at the local level. If we want to understand what's going on with people and we want to understand their institutional context, their political environment and so on, we need to also understand what's going on with respect to inequality in those villages. A final point to make about inequality in Palampur, which is potentially a real interest, is that we document rising inequality, falling poverty, but we also document a decline in intergenerational mobility over time. So a father's income is better able to predict a son's income today than it was 25 years ago. And we're able to document these changes in intergenerational mobility because we have data that goes back all the way to the 1950s that covers at least two generations. So we have this insight. It's a glimmer of a possible issue of real concern, which is with rising inequality, we might be seeing declining mobility of a kind that is really closely associated with our ideas of inequality of opportunity and so on. I won't talk further about an ongoing work that we're trying to do to explore the possibility that this village is, in some sense, the kind of forces shaping the income distribution in the village are possibly in some sense more broadly valid on the basis of sort of simulation work. That's something that's underway. But we have another paper, the paper by Abiru Mukhupade and David Orzanke, which looks explicitly at this question of inequality within villages and essentially this paper decomposes national level inequality into a between village inequality and a within village inequality component. This is done not with just sort of dry use of existing data because the existing data don't allow us to do such an exercise. And so the project involves measuring inequality basically at the village level, measuring average income at the village level by estimating an imputation model that relates consumption at the village level against nightlights intensity data as well as district level variables. And so this study produces an estimate of average income for each village of India 600,000 villages. And because it has an estimate of average income for each village, it is able to sort of back out the between share of between village component of total inequality. And thereby the residual from that calculation is the within village component of total inequality. And the evidence suggests that this within village component of inequality is actually a very large proportion of total inequality. So if you decompose inequality in this way, you get an contribution to total inequality from village level inequality of something like 75, 80 percent. So much of what was going on in terms of inequality is actually occurring within the village. And again, a reason to think about what's so, you know, what's going on at the village. Why? How are these village inequalities influencing all kinds of behaviors and all kinds of outcomes? They carry this out at the national level. And then they also carry this out at the state level. And they document that in a very large number of states, the within village component is actually increasing quite sharply. There's an interesting example of the state of Jarkand here where overall inequality in Jarkand is not found, not observed to have changed at all. But within within the state of Jarkand, we see within village inequality component actually having risen quite sharply. So you can have quite a disconnect between your assessment of inequality. If you look at the village level inequality, or if you look at state level or national level inequality. Turning to the Dangan Lanyao paper, what we do here is basically construct synthetic panels using multiple rounds of NSS data, which are cross section data to construct synthetic panels, which then allow us to explore patterns of mobility. And what we basically observe is that there has been an increase in mobility over time. So if we compare 1993 to 2004, we observe that something like 60% of households are basically still alongside along the diagonal of this particular type of transition matrix where we distinguish between the poor population, a vulnerable population, and the sort of middle class or secure population. And we document that between 93 to 2004, something like 60% of the population was was was was sort of on the diagonal of this transition matrix. Moving to 2004 to 2011, which is actually a somewhat shorter period of time, the actual transitions have been have been have increased. There's a smaller fraction of the population is sort of on the diagonal here. So we have greater intergenerational intergenerational mobility that's occurring. And the correlates of this inter intergenerational mobility can be explored. And we tend to find, not terribly surprisingly, that factors such as high education, residents in urban areas, employment in the regular wage sector and so on is associated with upward mobility, whereas schedule tribes, schedule castes, and certain states, for example, are more likely to be characteristics that are associated with with downward mobility over time. Finally, the paper by Roy van der Veide and his co-author Vee, we investigate they investigate intra intergenerational educational mobility, the income information that would be necessary to explore intergenerational income mobility just are not available. And it's not obvious what methods would exist to allow us to try to get at that with the existing data sources. But educational mobility can be explored. And this is on the basis of of information where we look at co-residents of fathers and sons co-residing and looking at their education, their respective education levels. And we this this study is based on also on the NSS data going back to the 1980s. And the exercises involves calculating sort of intergenerational regression or correlation coefficients that capture the extent to which education levels persist over time. And the general message from this study is that intergenerational mobility in the educational dimension is actually very low relative to say the experience in other countries and van der Veide and co-authors at the World Bank have just completed a very exhaustive study of this intergenerational educational mobility, where they're able to look at rates of intergenerational mobility in the education domain across the world. And India comes in as a country where intergenerational mobility really is quite low in this in the education dimension. But and this this graph shows there's some evidence of that declining. So that's about improving intergenerational mobility is increasing. The correlations are actually declining. So that's that's an encouraging sign. One interesting finding from the from the work that van der Veide and co-authors carry out is that they by employing a regional sort of panel that they've constructed with these data, they look at the relationship between intergenerational mobility and growth. And they distinguish between the growth rates of different quintiles in the consumption distribution. And they show that higher into lower intergenerational mobility is associated with lower or bigger declines in growth rates amongst the poor and less of a of an impact on the growth rates of the of the rich. And so this suggests that intergenerational mobility is associated with income inequality in the way that we might expect. And that's also reflected in something like the the Gatsby curve, for example, where where higher intergenerational mobility is associated with lower inequality and and and and the reverse. Finally, they collect, they carry out some work to try to look at what are the possible correlates of intergenerational mobility. And they find that public expenditures at the state level, state level public expenditures are associated with higher intergenerational mobility. More political competition, interestingly, is found to be associated with greater mobility. And they also found that, and this is almost somewhat slightly mechanical, that the percentage of parents with no education is also positively associated with mobility. But that's possibly something that's somewhat mechanical out of this exercise. Anyway, I'm so sorry for having raced through this. But thank you very much for your attention.