 In answering the question, what's next? So that there were two ways we could look at this question. One would be, what I would call the academic way, would be we have done so much work over the last 50 years on measuring inequality, measuring poverty, using new data, new measures, new concepts that probably we could ask ourselves after this incredible achievement that we were able to do in the last 50 years, what should we do next? And I will say a few words about this. But the second way of looking at the question is to say, poverty and inequality are changing in the world. And we know more or less what is going on at the global level. We know more or less what is going on at the country level. We have written, many of us, on this over the last 10, 15 years. Now, the problem is to know what will happen next. For example, at the global level, we know that inequality in the world is going down, poverty is going down according to some specific concept. And we know also that inequality is going up in several countries. Now, what is next? Should we expect that the reversal in global inequality will continue? Should we expect that inequality will keep increasing in several countries? I think this is another way we could look at the question about what's next. On the first question for academics, the second one being more for global policymakers, my way of answering the question would be with another question, which is what inequality are we talking about? By the way, I will be talking more about inequality than poverty. Most of the issues that we'll be looking at are also valid in the case of poverty, but also Sabina has already done a very good job in talking about this, and I will not repeat what she has said. So in terms of inequality, what inequality should we be looking at? When you look at the literature, you have differences in terms of the measure that is being used. This is something that we have very much dominated over time. A second difference comes from the data that are being used. Household surveys, tax data, DHS, panel data, server of consumer finance, all these data sets will allow us to see or to look at a different aspect of inequality. And then we have differences in concepts. We can look at the inequality of outcomes, whether it is income, consumption, wealth, possibly. We can look at non-economic outcomes. Some people have been working on the inequality of happiness, and of course, inequality in terms of health, or possibly education. We could look at the inequality of circumstances, and this is all the literature which deals with the inequality of opportunity and the social mobility, for example, being one aspect of this. We can talk about horizontal inequality between groups of people, and within that we can make a distinction between procedural inequality in terms of discrimination against gender discrimination among ethnic groups, age groups, all these are different concepts which we have been working on so that today, when we say inequality has increased, what inequality are we talking about? And some people will say, oh yes, inequality in income has increased. We can see it with a genie, et cetera, but inequality of opportunities has not increased, and maybe that's going down. What do we say, and how do we handle all this? And this is an important question because we know that those various concepts do not move in the same direction. We know that the distribution of earnings may not be changing as the distribution of income. We know many cases where the earning distribution has become more and equal, but nothing of this type happened when we were looking at income. Basically because in between, we had participation behavior changing which meant that the increase in inequality in earnings was neutralized by this kind of behavior. When we look at the top 10% or top 1% or top 5% share of gross income, which is now one very important way of looking at inequality, it may be the case that this is not changing of a time and we know it because there's been a lot of debate about this as a genie coefficient based on equalized income and based on household surveys. So this is the issue. In front of different aspects and measurement, what should we do? And conceptual differences matter too. It is quite clear that the present focus, for example, on top incomes by my colleagues in the World Inequality Lab in the Paris School of Economics who are really working very much on this concept of top X% share in gross income is saying something about inequality which is not of the same type as equalized income and the genie. When we look at gross incomes, we're talking more about the command, the inequality in the command that people have on the economy when especially we look at top income, this is exactly the point. And it is really, of course, worrying to see that this command is increasing of a time in the extent of the type that we can see in the country like the United States. But this is not really what necessarily a reflection of what is going on with the inequality of welfare or as represented by equalized incomes. So this is quite important to keep in mind that there are differences. And when people say inequality has increased because the share of the top 1% increased, they're right, but they are describing one phenomenon and another phenomenon which is a parallel which may take place at the same time is the fact that inequality in standard of livings is remaining constant or maybe is declining in the same country. It's very, very important to keep that in mind. Now because of that, what can we do? There are two ways. One, we can try to integrate all those concepts. And in the same way as, and I plead guilty for that, we have tried to integrate various dimensions of inequality and various dimensions of inequality and the way in which the work by Sabina and others with a multidimensional poverty is doing. We may try to integrate in the same framework inequality of earnings, income, wealth, consumption because we know that they are linked in a very, very specific way. So we can think about these, but at the same time we know that this is very complicated. This will be relying on probably very strong assumptions and the data sets that we need to do that in particular that would integrate those various dimensions are simply not available. So I'm not sure that we can go very far in that direction and the result, even when we give an equal way to different concepts, I'm not sure that this is very important and very useful for policy makers. The other way, and this is the way I'm trying to define is really to use a dashboard approach. We have different concepts, fine. We have different data sets, fine. We have different measures, fine. What, let's try to think about one minimal set of indicators that would allow us to represent those various dimensions and let's maybe not impose to the media and to the policy makers the knowledge and the mastering of all those concepts but let's make sure that in one place in the knowledge that we have of inequality there is such a dashboard. And this is the only slide I wanted to show which is what I would propose as inequality dashboard which lists several aspects which I believe are important and aspects of inequality we're talking about every day. So I will not go in detail over the various items but you can see that the top, you have the top 1% or top X% gross income share as used by several people now in the profession. Of course you have the living standard genie based on equalized income or consumption data in household surveys, individual learnings which is very important and somehow is much less analyzed than other concept. Well, inequality indicator which is unfortunately not always and not frequently available in particular in developing countries. One thing which I believe is important is related to inequality of opportunities is it's not that difficult to compute what is a share of earnings inequality or living standard inequalities that is due to the family background of the people and this is an indication of the importance of what some people call circumstances in the present inequality and to check whether this is changing over time or from a cohort to another is not that difficult and this is not a statistic that you will be looking and you'll be producing every year but this is something that should appear in the dashboard. Now, we want also to take into account all the dimensions of welfare health but what matters is not the inequality of health's self-reported health status. What matters is the correlation with income or maybe the correlation with some other economic variables and we also can look at education but I believe that what matters is really more the inequality and the correlation with family income of the scores like the PISA scores, et cetera. We can talk, we must introduce something about informality especially when dealing with developing countries and then of course something about the gender gap and something about the ethnicity gap. Now, poverty should also be part of the story and to some extent this dashboard we already have it in the sense that as Martin said at the beginning we tend to look at what is going on in the country both in terms of inequality and in terms of poverty of course this dashboard approach is implicit in all that has been done lately to take into account the multi-dimensionality. I have two minutes left so it will be done for this so let me go back to that if you want to keep looking at it but let me say simply a few words about the second way of looking at the question. What will happen next with the global income distribution? In some time ago I called the drop in global inequality an historical reversal. Today I must say that I'm starting to doubt about whether it is really one direction or one way reversal. The reversal was very much due to China. China today is more or less at the middle of the distribution which means that statistically we know that the impact of faster growth in China than the rest of the world will not have very much impact on inequality and in a few years it will have a negative impact or it will increase in inequality. What I think is important is the fact that in the 2000s one of the big push towards less inequality of the world was due to very fast growth, acceleration of growth in South and Sub-Saharan Africa and also in Latin America. But the problem with this episode to some extent is the fact that this was very much linked to a very nice or high cycle, high part of the cycle in terms of international prices of commodities. Now if we look at the average growth or if we look at other long cycles of commodity prices we see that the growth of those countries Latin America and Sub-Saharan Africa which rely on commodities, the growth of which rely on commodities is not that great. And if we look prospectively to what may happen my calculation is very simple. The global economy is growing more or less at 4% a year. The demand for commodities will be growing more or less at 4% a year. This may include changes in prices at the same time it changes in quantity. But the point is that the demands, the total value of the demand will grow at 4% a year. The demographic growth rate in Sub-Saharan Africa is 3% more or less until the middle of the century which means that on average if we assume that there will be a full cycle or maybe several full cycles of commodity prices on average the region will be growing at 1% per capita. Now this is less than emerging countries. This is less than the long run potential growth rate in developed countries which means that in the long run what we observe today is more or less some forces toward some divergence. And I believe that this is a very, very important point. We cannot allow for ethical reason for people working on development. Development is before all convergence across the world. So this is going against this basic objective and something has to be done about this. And the second thing is that we also have to realize that there are some danger associated to that and with the migration pressure that we see today and the problem that it raises we have to realize that some more inequality in the world will probably go in the same direction. So I think that this is something that as a development community we have to take to keep in mind. I know that many people in this room work about this and we're sorry to miss John Page's presentation yesterday about industrialization in Africa but I would say that today this is really the most important issue on which we should work when we look at global inequality what next. Thank you very much.