 Well, thank you for coming to the session. Well, I just want to say that this is a joint work with two of my colleagues, Fintar and Larry Rue from the Oxford University. The second announcement is that the work we are presenting today is just work in progress. So we are using the latest data that we got a few weeks ago. So this is very much the first results and by any means are conclusive. But we want to present what we are finding, which is surprising, but nevertheless it's something interesting. Well, of course, one of the questions that we were asking when we were thinking of writing this paper was within the discussions on inequality, the importance of inequality, and a number of things. Obviously the connection between inequality and growth is well established. But there are other factors as well that inequality can negatively affect in terms of consumer demand, national savings, human capital formation and so forth. At the same time inequality has other evasive effects on social cohesion and crime, conflict, corruption, governance and other aspects that have been covered by other scholars. But within the UN there have been a lot of discussions and one of the most recent reports for the preparation of discussion on the post-2015 development agenda, they actually stressed the importance of inequality. They said one of the key challenges for the future is actually inequality. So obviously it's very important to understand what is happening within countries but also globally. And also to identify the trends, the directions of this has become a very contested issue. Perhaps one of the last most authoritative reviews on these issues of global inequality says that it is not possible to reach a definitive conclusion regarding the direction of change in global inequality. So this is a very clear statement about how we are right now and this was one of the motivations that we have to write this paper. So when we start to look at other work, of course there are a number of studies looking at trends within country inequalities. The work done by Andrea has been influential and of course other studies have been looking at between country inequalities, so how different countries diverge or converge at the time. Perhaps a few studies have been measuring global interpersonal inequality while decomposing the within and between components. And well documented or known studies by Salem Martin, by Francois Bouignot and others, they have been suggesting different approaches to measure global inequality. When we were deciding the ways we want to measure, of course we came across a few measures and we decided to use the TEL because as you know it is the measure that can be the composable and this is what we are using. We are using as well genies but then of course the mean law deviations is the measurement that we are using for measuring interpersonal global inequality. So just to give you the main findings that we have which is closely related to Rahul's findings as well is that although we find that changing the quality over time and in particular in China and India have been pushing down the global inequality estimates in particular because from 75 to 2005 we saw a reduction in between country inequalities simply because China and India were growing faster than developed countries so of course the gap between these countries were narrowing. But when we started to work on the latest version of the grid we saw something interesting coming out from the data. It's something that we want to show you. And we are not quite sure what is behind that. So after the crisis of 2008 we observed a fall in inequality globally but particularly in countries like China and other countries that have even further reduced trends in global inequality. So just briefly to tell you what we did is of course we used the conventional genie that as you know measures the cumulative share of income relative to the cumulative population share so the problem as you know with the genie is that it's not composable and that's why we are using the mean law deviations to get the between and within country components. So how we are defining global interpersonal inequalities it's actually very simple. For example if you assume that ycqt is the average per capita income in quintile cube of country c in your t then we can actually get a domestic inequality in a given country year which is estimated on the assumption that all the individuals in the same quintile have the average per capita income for that particular country you're quintile. So it's a very strong assumption and we will discuss I would tell you a bit what we are trying to do but of course once we do that the word distribution of income in year t can be constructed by compiling all the available consequential data in year t. So once we waited this by population we get an estimate of the global interpersonal inequality. So of course as you may be thinking I suppose we are making a very simplifying assumption because we are putting together all the individuals in every quintile and we are giving them the same income. We are not the only ones who have adopted that approach there are other studies that have adopted the same approach with a few exceptions they are trying to construct some smooth distributions within the country components and we assume that our estimates are perhaps biased downwards this is perhaps a conservative estimate of the global interpersonal distribution but nevertheless as a robustness check we are as well computing the Sherrocks and one algorithm that actually help us to smooth the distribution and this is what Salah and Martik have done previously so we are doing it so the smoothing exercise just as a comparability exercise. At the same time we constructed a few counterfactual scenarios the first one was we consider the situation in which India and China's income per capita and distributions of income have remained unchanged let's assume between this period and keeping the population growing at the same rate as they did the second counterfactual scenario was assuming that China and India had been able to grow the incomes at the same rate as they actually did but they managed to keep the inequality as they observed back in the 75 so again these are counterfactual scenarios that we wanted to see what was the effect of those big countries across the global interpersonal inequality Right, so as I said we are using the last version and then of course we did some judgments because of course it is unclear and it still can be a source of debate about whether to use income or consumption and I think what we did was we followed a dener and squire approach of in a way identifying the deviation of the genus based on expenditure in relation to incomes and we just awaited and adjusted those gaps to bring all the genus to income inequality So the correlations are as you may expect relatively closer but they are not perfect and that really depends on the point at which you are in the distribution and for the number of individuals per country quintile we use different sources for the population the United Nations Population Division the Eurostat the US Census Bureau and so forth and for income levels per capita per country quintile we use GDP per capita just for purchasing power parity from the World Bank the reason is because in the way we don't have income for all the countries across time so that was one of the limitation that we have in the data so we decided as imperfect as it can be use GDP per capita Right, so just to give you the results Well, the people that is available we have done this until 2005 so you can go and take a look and we were explaining what we think is behind the default in global interpersonal inequality primarily driven by a very strong force from between countries inequalities but then when we estimated the last data Fin and Larry were pondering what was happening not only the between country inequality was falling but also the within country component was falling so we were not expecting that so as you can see the change is significant so we are trying to figure out what is driving this fall in within country inequalities we will give you some potential ideas about some possibilities or clues about what is behind this decline in between country inequalities So we look at regional inequality we observe a considerable variation across countries whereas Latin America is Asian South Asia observed a decline in inequality recently we observe an increase in inequality in North America and South Africa and as you can see the trends are not homogeneous at all so then when we look at the between country component then it seems like to be more the trajectories are more homogeneous and all the regions consistently have observed in different degrees of decline in between country inequalities within these regions so we also observe some obviously strong negative correlations between GDP per capita and genies as you may expect and although we find a modest positive correlation between the increase in genies and growth in GDP per capita although this weak correlation is primarily driven by China so the what is in China is driving these results is something that we wanted to look at more carefully and so we look at Indian China so what was happening more recently so I didn't present the graph in India but I can describe you what is in there so from you look at India's genie they have been increasing until 2004 and they remain constant throughout 2009 so it seems to us that what the data or micro data shows is that not much is happening in terms of inequality in India however when we look at China there is a steady decline in within country inequality or domestic inequality which is reflected precisely after the financial crisis so we were discussing these results with some Chinese colleagues and they gave us one story which we are not sure if we buy it but this one story so the story is that it's related to two reasons or explanations why inequality is falling in China and therefore it is pushing down the global interpersonal equality estimates as well so first is that there were some domestic policies that were introduced in the 2000s before the crisis with the aim of reducing the inequalities within the country in particular between the gap between the rural and urban areas so these policies were for example aiming at increasing the minimum wages extending the social protection and anti poverty policies in particular the devout which as you know was introduced in urban areas but then expanded to rural areas in 2005 then they started to introduce agricultural support policies because they wanted to prevent migration from the rural areas to the urban areas and then they started to introduce tax reductions to provide incentive for people to stay and invest in those more deprived areas but on top of that which is not necessarily related to the crisis the second component is that this major stimulus package that the Chinese government introduced that was aimed at increasing investment tax costs but in particular increasing in social expenditure so this is one of the potential clues that may explain what is happening in China and therefore is pushing down the global interpersonal inequality that we observe so I will jump the counterfactual scenarios because we are not these are just between 75 and 2005 but what we just want to conclude here is that what we observe nevertheless whatever the trajectories of inequality we have seen is that global inequality is incredibly high so it's higher than every single country that you observe in Latin America, South Africa so the global inequality is higher than the inequality that we observe within countries so it's incredibly high so it has been falling primarily because between country inequalities have been pushing down these estimates but then as I just explained we are observing more recently not only in China also in Latin America and other regions but in China it was interesting to see the effect because obviously our estimates are population weight so it has a lot of effects on our estimates so it seems to me that domestic policies are playing a role here in explaining why global inequality is falling and this is something that we want to explore and we are looking forward to your views and comments on this potential way of explaining what we observe