 This is ongoing work, which I'm doing with Colundo, Moabu and Vambugo, who is not here, Vambugo. We have not quite managed to sort out the toolkit and all the years for which the focus is on. But we have a large long background and we have got a lot of bits and pieces, so I'll tell you the story as we see it right now. First, going back to some older work starting from 1914. I was interested in this idea of structural change and how the structural change of the economy drives the genie. And until 1976, you had data for Kenya by race. Then all became Kenyans, but you had whites, Asians and blacks. So you could compute and you had data for various years on employment levels and wages by race. And you had some other estimates you could use. So I computed GDP and allocated it to 12 or 11 income groups by race, by type of employment, agriculture, etc. And also I allocated capital income, which means that these genies are going to be pretty high. That was the most random. But anyway, what happened in Kenya was that the genie coefficient basically, if we believe this estimate, was increasing while we modernized the economy. So up to about 1950, genie was continuously going up because the modern sector of the economy was employing more and more people and more and more people getting higher incomes. And you also had this inflow of Asians and whites. So it was a very racial society at the time. So for example, in 1950, about half GDP went to non-Africans. That was like one, two percent of the population. So I think in 1950, if you have a long-term perspective here, we can level what was one to ten to a hundred between Africans, Asians and whites. So that's within the whole rollback since then to some extent. But the whites and Asians mean much less than before. If you look at what drives the change in inequality, the national genie of the first period, it is the compositional change to some extent, but also, of course, relative income changes in different sectors. For example, you can see that when the urban, rural and agricultural, non-agriculture gap shrinks, the genie comes down. So the two things that are important here are the relative size of the sectors on the income groups and the gap between them. From independence in 1963, the income distributions among the groups were overlapping much more. But one thing we pick up from this part, which we want to bring into the sort of second part is this important of the structural change and how that drives the compositional change, how that drives inequality. Another paper in the Waldin is we tried to measure factor of income distribution change in Kenya between 1964 and 2000. What is wrong with that basically? Where we have estimated wages, we have backed out capital returns from production function estimates and we have land prices as the proxy for land rents. And we also have the proportions of those. And we try to see what drives these relative prices between wages and rentals, for example, over time. And we were interested in the globalization. Opening up does that change relative prices, which drive these prices? Basically, we don't find anything like that. We find that the quantity of factors drives prices up. So the fact that land is not really expanding drives the price of land up dramatically. And capital returns are more or less constant and wages are pretty much constant until the mid-90s. So one thing that happens which we have a hard time to explain which is probably important for our later income distribution analysis is that from about mid-90s wages start to shoot up. So from about 95 when the labor market was liberalized until 2007 real wages in the formal sector went up by 75 percent but from 2008 it's going down again by 25 percent. And this is, there was crisis in 2007-2008 of course but still this is, we had a hard time to explain that in our time series econometrics that we put in a dummy to think about it later. But the most interesting thing from this part of the analysis which we extend into the latest period is that if you look at the capital labor ratio that peaked in 1980. So K over L has been going down since the 1980s for 30 years. And what's happening in Canada is basically that we have a structural transformation where people are leaving agriculture going to the rest of the economy and since K over L is going down they have to go to something which is less capital intensive that is the informal sector. So if you look at what's been happening into the pattern of labor over the last 20 years so formal employment has barely risen at the same time rate as labor force or less while informal employment has increased very rapidly. So half labor is out of agriculture but most of them are in the informal sector and that is important then for the outcomes. We have some data here for 94 in 2005-06 there is some data for 97 we'll squeeze in in between but basically what we find here is that inequality well first this is not a great growth period all the growth after 2002 starts to increase. So growth goes up quite good after 2002 but from 94-2005 per capita income so maybe 5-6% higher so it's not much. What happens is that over the years period this low growth has led to higher inequality somebody increased poverty all the most of that happened in the first period 94-97 so we will then try to explain and analyze this to a larger extent why we have this pattern but for example this shift in the structure of work that people are leaving agriculture not being absorbed by the formal sector by going into a wide informal sector explains I think what we see here that urban inequality is going up a lot rural inequality is actually going down in most of the regions and if you aggregate that together the national genie goes up so in a way it's the same kind of similar story that we saw in the good old days that this compositional changes drive the genie up poverty has gone up a little we also have some estimates from DHS about social indicators they look a bit better actually most of them so we need to sort this out and next time we meet we might have Thank you