 I think it might be best to think of the issues under two headings, sort of one sort of more conventional set of issues, which we've always worked on, and the other less conventional set of issues, but we've worked on those before, but I think they appear in a new light as we go forward. So under the conventional set of issues, I would include the old questions of the growth elasticity of poverty reduction. How much does growth translate into poverty reduction? This is in one sense a very well worn area, but on the other hand I think it is worth continuing to work on this and to refine it and analyze it further in the context of Africa, especially given the new data that we're getting in the African context. What is it that maximizes the impact of growth on poverty? In particular, how does inequality mediate the effect of growth on poverty? And I think with new data sets that are becoming available in the African context, we can begin to, we can look at this question in far greater detail than we've been able to. So there's not really a new question, it's actually an old question, but I think it's an important one for us to keep on working on and to work out in the African empirical context to assess what the determinants are of a high or a low impact of growth on poverty, and in particular how inequality makes the impact high or low. So that's one type of conventional question. The second type of conventional question, but equally important for us to carry on working on it, is the causal impact of inequality on growth? And again, this has been discussed in the literature in the past, but I think it's very important in the African context and that using African specific data, we begin to analyze this question in greater detail. Is there a positive causal impact of inequality on growth? In particular, do high levels of inequality impede growth? This is, I think there's been an evolution of the literature, and I would say that globally, although the evidence is not overwhelming, there's a growing consensus that inequality impede growth. We would like to look at this question in greater detail in the African context. So those are two examples of conventional questions, which I think we should go on working, and they're very important ones in the African context. Let me take up a couple of questions which are less conventional. Again, I don't want to suggest that these have not been worked on before, but I think they appear as becoming important as we move forward. In particular, I think we have to think of different dimensions of inequality in the African context, and I don't just mean the old distinction between income and non-income dimensions of well-being. What I mean is the distinction between inequality as seen as being interpersonal inequality versus inequality as seen as being inequality between broadly defined groups in society, in particular salient socio-economic, socio-political groupings like ethnic groups, religious groups, regional groups, and so on. And I think perhaps one distinction between, let's say, a Nordic type setting and an African type setting or indeed an Asian type setting is the high degree of heterogeneity on these dimensions that you find in an African context than you do in some northern and particularly in a Nordic type context. So I think one can have a situation, for example, where the standard measures of inequality like the genie coefficient or whatever which are calculated on differences between persons may remain unchanged, and yet there are major structural shifts in how different groups are doing relative to each other, and that can lead to the political economy of a country turning in one direction or the other and will be completely innocent of it if we just focus on the standard measures of inequality like the genie coefficient or the variance of logarithms, etc. So I think that whole area of how the group, the distribution of income by salient socio-political groups has changed in Africa, I think is going to be very important, is an important topic and is going to be an increasingly important topic. As for example, we begin to talk about how the rents from oil and minerals are being distributed not just between rich and poor, which will be the conventional way of looking at it, and it's an important way of looking at it, but as between salient political economic groups and socio-political groups in society. So that's an example of a non-conventional type question that I believe we will need to work on in the context of inclusive growth in Africa. And while I believe we should, as analysts and technicians, work on fine indices of inclusive growth, and I think that's an interesting exercise and can lead to some insights, I myself don't believe we should get too hung up on this or that index of inclusive growth or this or that way of measuring inclusive growth. I think we should view this as an exercise touching on a broad range of issues. For example, the three issues that I just mentioned to you, two conventional and one non-conventional issue. Let me give you another example actually of an old new issue. And that, which again touches on the point that you made about the question that you raised about what is inclusive growth. The rural-urban divide in Africa, again, is an old question. But I think it hasn't gone away and it will appear in the new guise because of course of all the different developing areas, Africa is the least urbanized and has the most urbanization still to happen, so to speak, over the next 20, 30, 40 years. And this leads to a very important class of questions about if one had a fixed amount of resources, either financial resources, financial capital or political capital, for example, where would we advise policymakers to spend it? Would we say spend it in rural areas where the largest numbers of poor live and where poverty is at its deepest, perhaps? Or would we say spend it in urban areas, building up urban infrastructure because the argument might be this is the future, this is where the future growth is going to come from and that's the trade-off that one faces. I think we're quite good in the literature both in Africa but also more broadly in saying why we should spend money in rural areas. We're also very good at saying why we should spend money in urban areas. But I don't think we've confronted fully the question about where would be the highest priority to spend this money or to direct policy attention and I think we need frameworks to think those things through and we need frameworks to understand the trade-offs between addressing poverty in rural areas versus addressing poverty in urban areas and whether addressing the growth of urban areas is addressing the poverty of tomorrow versus addressing incomes in rural areas which is addressing the poverty of today. These are tough questions both from the positive analysis point of view, how do different types of expenditures feed through but also equally from the normative point of view which is how do you weight the reduction of poverty today versus perhaps a greater reduction of poverty tomorrow. Again, these are not new questions but I think these questions will appear in a new guise in the African context as we move forward in this high growth era of Africa and presumably with a high rate of urbanization in Africa.