 So let me start by thanking the presenters for very insightful presentations and I'll focus my comments on just three key messages that I picked up, the three key takeaways. One is that agriculture matters and remains an opportunity. Both David and Michael found evidence that it's driven growth. At the same time, there are indications of reducing poverty. So this with Ethiopia and Malawi, but in Burkina we also saw evidence of increasing poverty with rising food prices, increasing quote-unquote, but basically as a following through from reduced purchasing power by the poor. Unfortunately productivity remains low. That's a consistent message. There are also questions around sustainability, whether it's of subsidies in Malawi. You have land expansion, driving agricultural expansion in Burkina Faso. You have massive aid floors in Rwanda and it's arguable that that has helped the country be able to achieve gains in agriculture. At the same time, Murray also noted that the cocoa price collapse can actually link agriculture to political unrest. So given this background so to speak, the question that I have is what is the opportunity for new institutional arrangements? So for example, at the World Economic Forum, we've been supporting initiatives that promote public-private cooperation. We have one in particular in agriculture that covers some of the countries that were presented and essentially seeking to see how these arrangements can foster or accelerate private sector investment in agriculture and this takes into account smallholder farmers. And we've seen through these arrangements over 10 billion being committed by the private sector reaching millions of farmers just in terms of impact and jobs being created. That said, over the last two days of the conference, we've had quite a bit about striking the right balance between government failure and market failure. So this already questions the merits of this particular approach as a means of addressing some of the challenges that the presenters put forward. So my question to them, I promise them that I would ask them some questions, is whether we can develop effective measures of the effectiveness of such public-private cooperation interventions that are taking place. My second comment is, or take away was with respect to structural transformation. The presenters, for example, Michael, noted that both the agricultural sector and non-agricultural sectors have low productivity. In Ghana, we saw that manufacturing growth has been low and it was noted that in Ethiopia, the productivity of manufacturing is low. Although I spoke to David earlier and said over the last three years, we've seen tremendous growth in manufacturing in Ethiopia, but then we get into issues around, is it around Addis, is it spread out throughout the country? So in terms of the impact on poverty, that's questionable. We also had an observation about the role of the labor market. For example, in Kenya, where you have informal sector employment growth rising, but also lack of clarity about what exactly is happening in that particular sector. So just to add to this, again, over the past two days, it's very clear that over the past two, at least two decades, probably a little bit more, Africa has been de-industrializing and at the same time, we've seen the services sector grow. So as we pursue reinforced or renewed enthusiasm around industrialization policies, Justin Lin was very eloquent about that. I have to wonder a little bit, so if I cite a few statistics from a recent report from ANCTAD, this is economic development in Africa report, they noted that between 2009 and 2012, the services sector in Africa grew at more than twice the average rate for the world. It accounted for one-third of formal employment, and 21 African countries had a source of services in output greater than 50%. Unfortunately, similar to agriculture and manufacturing, productivity in the services sector also remains low. So as we address some of the bottlenecks to a takeoff in industrialization, energy being a major one, I wonder whether we can explore transformation paths that can link this movement from agriculture to services and then forward to manufacturing, for example, through the use of regional and global value chains, there may be others, and that's a question to the panelists. The third key takeaway that I had was that it's not just income to borrow from marries onwards. We saw mention of the role of political violence, social unrest, and Cote d'Ivoire. Arguably, there's the role of post-conflict recovery in the case of Rwanda, in their case it was positive. The non-monetary measures of poverty, for example Madagascar, with its mixed performance still showing improvements in stunting, for example. In Burkina, you saw the impact of population growth rates, also having an impact on both growth and arguably also on the poverty front. And there was also a reference to the effectiveness of public service delivery in particular with respect to health and education. I'd like to extend that a little bit by looking at the impact of technology, which did not come through, and this probably was not the objective of the work that you were doing. And one question that I often get is what is the next EMPESA in Africa? And EMPESA not being the only innovation, but clearly one that has had a significant impact in terms of banking, the unbanked in countries like Kenya in particular, Tanzania as well, less impact in other countries, but the ability to transfer resources between the urban and rural areas is something that would be of interest to see what exactly has happened to poverty there, where the incomes not being generated in the rural areas, but is being facilitated by technology. At the same time, if we look at the impact of the MOOCs, for example, so technology in education, it's clear that technology alone is not sufficient. We need to take into account other factors, right? So the basics and the context remain important, but a key question, at least in my mind, is what amount of investment needs to be made, because there is a school of thought that Africa does not need to invest much in technology and should just adopt or adapt technology that's been developed in other places, particularly the developed world. And so my question, my last question really is, should Africa go tech or not? Thank you. Thank you. Could the speakers, you can sit up here, please. Can you please come up here, please? Great, time for questions. Since lots of countries and lots of issues will have to be brief, but Eric can start. Very quick questions. A couple of quick questions. First of all, the speakers said very little about which poverty lines they were using. Is it $125 a day? Is it national poverty lines based on food energy intake, basic needs? I'd like to know more about it. And incidentally, these different methodologies do not necessarily track. Together with Andy McKay, we compared the poverty estimates using both the $125 a day and national poverty line, and it was far from a perfect tracking. Secondly, some of the findings that were reported today are implausible. There was one, I forget, which countries where poverty went from 35% to 50% back to 35%. Clearly, this reveals surveys that are not comparable. So again, one has to be very careful in terms of being sure that surveys are comparable. And finally, I was really surprised that in at least three of these countries, the data that were used ended up 10, 15 years ago, Bokina Faso 2003, Cameroon 2007, Kenya 2005. Can you not get some more recent data? And again, I would have thought that progress over the last 10 years, at least in some of these countries, might have been significant. Thank you. Thank you, Rofos FEO. Very interesting presentations and very informative. I have three points I would like to make, maybe require a bit more careful assessment. First, a common point was low productivity in agriculture. They have to be more precise what kind of productivity you look at. Actually, in Africa, on average, land productivity growth has been much faster than any other region over the past 25 years. And even that's also among smallholder farmers. What has stayed flat is labor productivity in agriculture. A question, how does it compare in each of the countries, which I haven't looked at, country by country, or I wouldn't have the numbers. But I think that's something to look into what's happening there. It could also be that the lack of labor productivity growth has to do with further fragmentation of land, except maybe in Bokina Faso where it seems to be an expanding land frontier still. So that's one point I would like to get some more reaction on what you mean by that. Second is on food price inflation. In the case of what Michael was presenting on Bokina Faso, it seems if food price inflation is so important that all poor are net consumers of food. So it doesn't seem to have an impact on the farmers. So where does the price increases end up in whose pockets do they end up maybe in intermediaries? We may also, on price inflation, I didn't hear any part of the story talk about food subsidies, food price subsidies, which is important in most of the countries. And also if you look at the global market prices, you see a lot of volatility, but that's much more muted in average domestic food prices in African countries. So in between there's, among other things, the food price subsidies and how does it work out. But underneath that, what I didn't hear because probably only use annual data, is the volatility in prices in each harvest cycle. What you see a lot of problems in the countries is that all farmers produce in the same cycle. There's no storage space, lack of access to markets, they dump all the products, they produce at the same time in the market and then it plunges and then it goes back up again. And that causes a lot of problems for income security as well as I guess should have an impact on poverty. Final point in the comparative notes, maybe could highlight a bit more systematically. I think one point you raised for David raised. For one of the countries, but not for all of the countries, spatial differences across the countries, like Ghana, it's very different if you look at the trends in northern Ghana. And as compared to the average of the country and I think that applies to all of the countries. So maybe that could be looked into further. And last but not least, you mentioned the role for the social protection program is important to Ethiopia. It also plays a role in the other countries, but in different ways. So see how that impacts on the poverty numbers would be important to get to the common factors, also the difference across countries where some countries, these programs work better than the others. Sorry to have taken so long. We could do another one, maybe. Okay, go on behind you. Thank you. I'm Susan from Makeda University. Yeah, I have a concern about reconciling the results in the different countries. In particular, the rural and urban poverty. We see that in some countries, the rural poverty is declining in others. It's increasing. And when we compare Uganda with Burkina Faso, we see that in Burkina Faso, growth was not poverty reducing, but it's poverty reducing in Uganda. Is it an issue of a threshold of the growth rate? Or it is something to do with the price of agricultural products? Because we note that in the two countries, there were no productivity increases. It was just increase in agricultural production. We see a success story for Uganda, but we don't see a success story for Burkina Faso. So probably we need to know more about the prices as well. Is it because of the price of these agricultural products? Given that in both countries, there was an increase in productivity. Then there was an interesting finding in... Mari mentioned that in Madagascar, the poverty rates were lower for females than males. I think that's rare. Stunting. Stunting. Stunting. Okay, so sorry about that. I got it wrongly. And the other issue is about... David made a comment that monetary poverty in Ghana was... There was a slow response to monetary poverty to growth. I wonder why... I mean, what did you mean by that? That there was a slow response to monetary poverty to growth? Thank you. Okay, so thank you for the questions. First, for Eric, the poverty lines that we're using in all of these studies are national poverty lines, basic needs approaches. And yeah, so in quite a few of the case studies, we also used this utility consistent approach to estimating regional poverty lines. And so the emphasis here was on the particular case study and not trying to make it something that would be comparable across countries. In terms of the volatility of the poverty estimates that we observe in the Ethiopia case. This particular dataset was the Ethiopian Rural Household Survey. And so the comparability of that, I would argue, is actually quite good. It's a small sample of 1,500 households. But I think the point is really that our monetary poverty numbers can be so volatile. They're not capturing long-term trends in poverty, but you can see these enormous changes due to shocks such as food prices and droughts. Just a comment on food price inflation and seasonal variation in food prices over time in Ethiopia. What's interesting is we're finding that with the improved infrastructure that there's less variation in seasonal prices because of the integration of the markets. And so that's just a comment on that particular question. And then I'm going to leave it there for now to get my thoughts together. Thank you very much for these comments and questions. I start with one point raised by NC. You asked about the scope for private partnerships. And the problem specifically now in the case of Burkina is that there are many ideas around. And if you read the poverty reduction strategy papers, there's a lot on how one could increase productivity in agriculture. But then indeed there is a problem of, I think, one commitment and second also the capacity to implement these. And so they do indeed now, the donors, go more and more into the direction of public-private partnerships. And maybe that can make a difference. I think that's well possible. But now again, specifically in the case of Burkina, of course, the first thing they need now is stable government, right? Good. Then I go to Eric's point. So the same Burkina, it's a national poverty line, which is a bit lower than the $1 poverty line. But we tested all the robustness towards alternative lines. And the trend you saw is quite robust against this. We take into account regional differences and prices. So we have price deflators for the various regions in Burkina. So that's in the line. It's not a utility consistent line. So we do not apply the concept that was mentioned by David simply because we have not data that would allow to do that. And they personally also have a few reservations about this, but maybe not for now. And yeah, you mentioned the end line of the data. In fact, in our case, it's 2009. And for the DHS 2010, for the agriculture survey 2010, it's just, I have more detail in 2003. And so there was one or two tables where indeed it ended in 2003. But the poverty trend and the story about the agriculture productivity that's covering the period until 2010. So it's relatively recent. But otherwise I fully agree that for many African countries that should make quite a difference whether you stop in 2000 or in 2010 or so. Good. Then Rob's points also very valid. So regarding the productivity, in my case, that was indeed land productivity. So the yields per hectare, right? And or the production per hectare of land. And yeah, the story, I mean, of course, one could think that if prices rise so tremendously, so why do farmers not benefit from this? And I showed you a few numbers about the share of households that sell these products on the markets and onto the share of the consumption that is purchased on the market. And this is quite enormous. And the story is really that most households are not in a position somehow to store their harvest and then somehow to play on the price. Most of them really sell what they produce right after the harvest. And of course prices are very low and then most of the benefits go to traders. And there are some programs on paper that would allow farmers to store their products and somehow to work more through associations and so on. But for the moment, that's really a problem. Good. Regarding the food subsidies, that's also something that is on the agenda. It was not in the earlier 2000s, but I mean, there was first unrest in 2009 and then again in 2012. So in the course of these social unrest, the government has started to experiment with food subsidy programs. And some of them have shown a little bit of success. Others were not so successful, so that's still something that is somehow in the experimental stage. Good. And then the last point that was raised regarding the structural transformation. The first, I want to make clear again, it's not that in Burkina there was no poverty reduction. There was poverty reduction. It was just very low compared to the growth they experienced. It's not a country where you have zero poverty reduction. It's just very slow. And yeah, it's again, to make this point, it's in my view, basically a story of say a lack of structural transformation in connection with this high demographic growth. And of course, these two are interrelated. I mean, the demographic growth is endogenous to this. And I think if you want to trigger now faster development and more poverty reduction, it's really that we have to work on this structural transformation. And then many other things will follow, including lower demographic growth. And this will then also take a bit away the pressure from the the agriculture prizes again in conjunction with the increase in productivity. Thank you. Just talking a little bit about the government failures and market failures and linking it to your point about technology. Because one of the things that strikes one about this, almost the way we write about it, but probably the way the political economy pans out, is that even liberalization is a government driven thing. And that's not innocuous. It's driven from a particular set of political circumstances and a particular set of political actors and plays out from there. So, you know, that's one of the reasons I think why one sees these these exercises where you say this is a break and then you somehow drift back to to the old way of doing things. The infrastructure investment in Cameroon is a great example that got completely distorted in a sense by the political configuration. Whereas the impact one of the great things about in PESA, of course, is is it almost zoom past government to to people and capacitated people again and opened up some dynamic. You know, to some extent, it's not so much about government failure in the market. It's about capacity, whether your focus is on capacitating people or not. I also think you're making a really good point, though, about the need for some more theory. It's a bit of a Stiglitzian type of a point, but I think it's true. We need to do some thinking about these processes of structural transformation and to make sense of all of this. You know, the facts don't tell their own story. And it's a very hard and very agated story. Even within even this within group approach of ours has thrown up a lot of variation. All I can all I can say to Eric was that the project tried to be very self conscious about data. In fact, one of the chapters, the way they organize the way the book is organized, the final grouping of countries is called low information countries, which is sort of an acknowledgement that in this case it was the DRC that we can't believe anything that's out there. But still there are these anomalies. Certainly in the Kenyan case and in the Cameroonian case there was a lot of attempt to tell a story that wasn't just about the head count ratio. The head counts or the poverty estimates made sense in the context of a story that was being told. I think that's a strength of the methodology. Let me stop there. Okay, just add something. We talked about structural change. And I think one of the things we try to do in the Kenyan chapter is look at structural transformation and how it's driven by changing factor proportions. And if you look back in Kenya over time, look for example the accumulation of capital relative to the growth of the labor force. It was going up until 1980. K over L was going up. It sort of flooded out and even fell to some extent. Which means that there is a structural transformation going on. People again pushed out of the agricultural sector into the urban sector. But the formal sector, there's not enough investment in the countries that the formal employment is growing very slowly. And people end up in the intermediate informal sector. So if you look at employment outside the agricultural now, 80% is in non-formal or not in the formal sector. Which means that you shift people from the rural areas into mostly not so productive activities in urban areas. And we try to sort of, problem with Erica's point, we have data, survey data from 1944 to 2005, 2006. It's a very short span. The next service coming next year. So we'll write a new paper then. But still, if you look at this period and try to explain what's happened. I mean, there is a very modest growth in per capita incomes over this period. Poverty is going up over this period. And this is driven by inequality going up quite a lot. And what is explaining inequality going up then in the case of Kenya? Well, first is that the composition effect, the urban sector is way, the weight of the urban sector is higher. The inequality particularly in the urban sector is increasing. And the urban rural income gap is increasing quite a lot. And if you look back in the history of Kenya as what has been driven inequality, if you have the urban rural gap, you have quite a bit of the story in there. And that seems to play out even now. We don't know for sure after 2005, 2006. We have some DHS numbers to look at, but we don't have any income. So we don't say much about that. But the structural transformation, I said the factor for calculation, driving structural transformation, explaining what's happening. I think that's my take of things that are going on, going back to Louis and all the rest. Okay, I think we are all hungry. If you wondered how we did the estimate of poverty in 1914, you have to read the old paper in World Development 1986. Never mind. Thanks a lot, everybody.