 OK, so good afternoon, everybody. My name is Witness, as the chair just pointed out. So I am going to give my reflections on the presentations that have been given. Very interesting presentations. So for me, I think the topic of learning to compete, I think that is a very, very good and very timely research intervention. Because obviously, for many African countries, when you think of even issues around decent employment, decent growth, or growth with equity, one of the areas where people are looking to is actually sort of developing and strengthening manufacturing would actually help Africa to grow in a more balanced way. So one of the things that I think John pointed out or typed on earlier on was that their productivity of the differentials between agriculture and manufacturing is quite huge, which I think many people or many studies have sort of alluded to that. But I've seen some recent work which sort of looking at trying to measure productivity per hour spent rather than productivity where you divide output by the number of total hours. So if you take that route, so for example, Chris Barrett from Cornell, they have done quite a bit of work around that. And actually, the differential is not that huge. If you look at productivity per hour, so essentially, because most of the time in agriculture, you have productive season, especially in Africa. You have productive season and nonproductive season. So when you take productivity, the usual way we measure it, we'd look at probably for the year, which means you are including the period where there's no real activity. So as a result, you find a huge disparity. But again, it's something that is debatable. So on the issue of exporting behavioral firms and why some firms export and others don't, well, so I think one of the issues that was pointed out was that if you push a firm to export, it's likely to survive or it's likely to remain in the export market. And that's very good and that's very encouraging. But of course, what I would have wanted to hear more of is, so as a policymaker, for example, how do you actually push the firm into the export market? I think that's an area that perhaps needs to further investigation, but I think that's probably from a policy point of view, it's very, very important to understand if you were to go to a policymaker, what do we tell them? So yes, it's good to know that the firms, if they enter, they are likely to survive. And I think there's obviously issues around these stresses might also come into play there. But I think the challenge is how do we actually in the first place get them in? So I think that's one of the big issues that we might need to think of in terms of the safe development going forward. Okay, so that's one thing. So they know the issue of FDI, that FDI is important, there's evidence of spillover effects, which is great. And obviously, so one of the things is of course, one thing that I sort of thought about was, in looking at, we had Vietnam and we had Kenya, and of course, one of the things that was pointed out was that more engagement or direct interaction between export MNEs and local firms in Vietnam than in, for example, Kenya, where you tend to have more interaction among MNEs, sort of is probably understandable. It's probably something that you can intuitively sort of anticipate. But if I think of Africa also, the point that John was making that Africa is heterogeneous. So I would have actually sort of wanted to see the results for Tunisia, where you have a big component of, or a big share of manufacturing GDP. Probably it might correlate very closely with Vietnam. I would do the same for South Africa, where there's already a strong industrial base. So that the engagement between the MNEs, so essentially they sort of a natural, because you already have some big local firms who are also producing and probably themselves exporting. So if you bring in MNEs, you have a greater likelihood of getting more interaction between the two than when you have a very sort of different types of firms where maybe the local firms are very micro and then the MNEs, so probably the interaction would be much limited. So I think looking at giving examples, maybe I would have benefited from a prison. If you had included the examples of Tunisia, for example, South Africa. But I think it's also encouraging that where you, to know that where you have a better interaction, then you tend to get better outcomes in terms of productivity. So I think it's something, again, that from a policy point of view, would be interesting. So in those I was just thinking as I was sitting there, the issue of getting African firms to participate in global value chains, but I think that is also one way where you can increase the engagement of African firms with international firms and also probably encourage learning from these firms. I'm doing okay. Okay, that's good. Okay. So on a global issues, so it's actually sort of interesting again the results. So of course, most of them sort of confirm to what a priority I would sort of expect. So what was sort of probably interesting for me and probably from a policy point of view also, is the fact that clustering does not necessarily, is not necessarily the best option for all types of, for all firms. So I thought that was very, very interesting because probably sometimes we just promote clustering, but again, maybe one might want to take it a bit further and sort of try to figure out, so what does that mean from a policy point of view? So probably it's because from the presentation it came out as small firms might not benefit as much. True, but the question is whether, we are talking of clustering of all types of firms or you can also have clusters of different size firms. So you could, from a policy point of view, probably the recommendation might come across as, well, so it might not be very helpful to cluster big small firms. So you might actually want to think of a strategy which encourages clustering of firms of more or less the same size, which would allow probably then, of course, you want competition to be tough because if you want to be to be exporting or to end export markets, then you need to be very competitive. So that aspect is good, but obviously you also want to have a situation where the smaller firms are not necessarily put, to compete head on with the very big firms where they don't stand a chance. So maybe it might be that actually having different types of clusters might sort of resolve the tension between the small and the large firms. Then I also just want to sort of emphasize the points that John sort of on his second shot was raising, which is some of the sort of police failures sort of in terms of why we have sort of seen de-industrialization in Africa, which is of course the fact that infrastructure, people have paid or the policymakers have paid very little attention to infrastructure. And obviously what it does, it raises the cost of moving goods, especially if you think of manufacturing because you have to move things physically. Then of course the issue of skills, I think that's a challenge, which again in parts can be attributed to some of the consequences or the consequences of the structural adjustment programs. Then of course the issue of firm capabilities and of course the issue of investment climates. So one of the things that I just want to speak to is of course in terms of, yes. So in terms of building firm capabilities, I think this is an area where again it's not clear how government or how policymakers can actually come in to assist firms in building capabilities. But again the issue of ensuring that firms adopt global best practice in production, which in most cases is not just about the physical or technology, but actually the softer skills. Which in my sort of, at least in when I used to work in government, so we had some tours of where we were sort of going to check to see what's happening in some of the small firms. And obviously, so I think in that aspect the government is sort of proactive trying to bring in people from outside the kind of way, let's say Asia, where they have done these things of sort of improving the production line, if you want. So essentially you will find that firms actually, even the way people sit in the factory, you can actually, it can make a huge difference. Just the way people, the layout of the factory can make a huge difference in terms of your productivity. So obviously I think some of these things are very, very important. And of course one of the issues is, well they have to be funded. So who's going to fund them? That's one issue. And also I think the issue of political will and also the issue of I think in the morning session, I think it came across very strongly that ideology matters. So sometimes people don't believe that government or policymakers can actually assist. So in a way it does affect how much one can achieve. Thank you. Sorry. I would like to thank the presenters for very insightful set of presentations on their papers. I will just perhaps start by summarizing what we have learned today. We have learned that there are no doubts that export and FDI benefit domestic firms, especially firms in developing countries in terms of productivity and learning. But however, they have raised a very important point that the paths through which firms reach higher levels of productivity and learning are different and they differ by context. What they have also emphasized is that the paths in which domestic firms interact with foreign firms within specific contexts are also very different and it is important to acknowledge these different paths of interaction. For example in the measure that they have constructed for the first time in a study, I think I believe in Vietnam, they have actually established that it matters whether we observe direct or indirect interaction within domestic and foreign firms. And for example, in a study done in the manufacturing sector in Vietnam, they have found that if we use a broader indirect linkage measure, the effect of on productivity from interacting with upstream foreign firms is negative. But actually when we look at the same sample, same set of firms using their direct measure of interaction, they have found a positive effect from interacting with upstream foreign firms. Therefore, I really appreciated their presentation in terms of emphasizing how important the context is, which is actually seen in the way that foreign and domestic firms, they all interact with each other and it is also bearing serious policy consequences. And I think Mons emphasized that we cannot make any generalizations and that policymakers should actually look at specific cases made in each country. Yes, and why is it important that we look at the context is because at least two of the presenters mentioned that actually the largest gains are made by firms who enter the exporting sector and who are at the same time linked with foreign direct investments. Yeah, and so what I would like to perhaps ask or suggest in terms of maybe further types of analysis is actually to just ask, is there a possibility of investigating the level of linkages between domestic farms and their buyers in the final markets in a sense, whether their primary buyers are somebody they have regular contracts with or they even have some ownership share, basically distinguishing between performance of domestic firm depending on the level of vertical integration with respect to their main buyers. And also as Mons mentioned that there are different paths of how a firm becomes an exporter, I was wondering if there could be any scope for analyzing how the regulation in destination countries is perhaps a hindrance for less productive firms or actually a mechanism of making sure that the most competitive firms are the ones who are entering the export market. Okay, this was brief and thank you for your attention. Ready to discuss, comments, questions are welcome. If it can be brief, starting from there, one, two, three, four, five, yes please, six. Is there a mic going around, yes please. Okay, thank you very much for the very interesting presentation, so my name is Abdulisec from Sheik Antojob University in Dakar, Senegal. So I think one important issue when looking at the comparison between firms that export and those that do not, I think the evidence is, yes those that export are more productive than those that do not. And then the question would be why is that the case? And I think we have two competing arguments and you tend to focus on one, which is learning by exporting. Meaning let's say you have the same level of productivity and then one start exporting and then because of the competition it face in the international market, it becomes more productive through this process of learning, but there is another argument which is referred to as the self-selection effect. Meaning those firms that export are those that are more productive before starting to export. So my question is which effect tends to dominate the other and what would be the policy implication for each one of these two effects? And a second question which is also very quick is I think one thing that may prevent technology from being transferred from foreign firm to domestic firms would be a low labor mobility. And I see this as one of the mechanisms through which technology may go from one firm to another. So the fact is that we may not have a large labor mobility or turnover may be preventing this technology transfer within a given country. So exactly what can you say about this labor term as the potential mechanism through which technology may be transferred from one firm to another. Thank you. Thank you very much. Justin, proximity, safe time. First, congratulations for a very success for an informative studies. And my question is related to get more information from the empirical studies. And one of the findings from the empirical studies is that in East Asia the MMCs have a larger interaction with local economies and more spillover and a technological transfer. And in Africa country less. And I think one reason for that might be in East Asia the MMCs are mostly insectis consistent with local competitive advantages. And I use the East Asian country as an exporting basis. And under the kind of situation certainly they will have consistent local advantages and a local firm will be easier to enter and they also rely on more of the local supplies to build up their supply chains. And in Africa, many of the MMCs are in the natural resource sectors. They are very capital intensive and since they are so capital intensive, local firm would be harder to enter. And so that will certainly have less interactions, less technological spillover. Or sometimes local MMCs are very small scale and entered into the local markets and they don't have so much technology ideas. And that may also, two types of that. So I would like to see if you bring in these ideas maybe you can find the type of MMC would also be very important whether they are coming here to really use as an exporting basis or for resources or for entering into the local market. That's one thing. And the second is also related to in East Asian export procession zone are more successful. And in Africa export procession zone or industrial park were not successful. That may also related to what I just commented because in East Asian, spatial economic zone or industrial parks in general built to help the country to do export. And but in Africa, we did not see much of using spatial economic zone or industrial park for export except now in Ethiopia, they started to do that. And once they started to do that, I think that they are almost successful just like in Borodami and also the Eastern industrial parks. In that case, thank you. Thank you very much. Is the middle there? Thank you very much for presenters for a very successful lesson for us today. So I am going to hang out from Vietnam central in this deal for my economic management. I have two small questions. First question to professors on place when you see that is spotting firms have a higher productivity. But I wondering, can you consider the size in terms of labor? Does matter or not? Because this is a very important question for policymaker to subsidy or to promote HME or to promote small firm and medium firm firms to graduate to a larger firm. Because if larger firm more productivity, larger firm is tend to spot and have a higher profitability. So my question is does the size matter? So in terms of labor. And the second question is I return to the command by professor Justin M&E's demand and M&E's connection with the local firms. I have a different view of point here. I think that in Asian countries M&E's have more connection with domestic firms because they have demand that. If they don't have demand for local supply, they should use their current supply from their supply networks, supply networks. Why they don't have demand in Africa countries and why they do have demand in Asian countries, especially in China? I think that that is economy of scale. If those M&E's they sell on their products in the market near the country or they have in the proximity around their location. So they have demand to use the local supplies. And if their market share in the local market or around a neighboring countries is small. So they don't need to use the local supply from the, because they just use the current supply in the supply in their current supply networks. Thank you, brief, brief. Okay. Now, 10 seconds. Yeah, so the question is, yeah. So when we do empirical studies, can you take this argument into the consideration? So is the, there are sale volume, volume in the other producers. In, for example, in Kenya and there are sale volume in Vietnam. That's a matter for the connection with the local suppliers. Thank you. That side now. Yes, yeah. Okay, thank you. I also enjoyed the presentations a lot. Just some quick questions for Mr. John Rand. You spoke about the importance of linkages and complexities in industry, the Herschman linkages in terms of capturing the benefits of spillovers and knowledge transfer. I was just wondering though, if you could say a bit about what the storyline is for the Asian, Southeast Asian multinationals today in the electronic industry. Did they go along the same development path where they engaged with multinationals as local indigenous firms and didn't grow from there? Carol Newman also spoke about clustering a lot. I'm also interested in the response with regards to the implications of clustering around a value chain. This is because in small economies you won't have lots of players. You would want competition, but you don't have that many players. So what are the implications of building clusters around a value chain to the extent that the value chain is extends within, you know, downstream, upstream within the country before you access the export market. And I was also wondering, because there's a lot of skills mismatch that we see a lot of investment in education, but in the labor market you still find schools mismatches. So what would the optimal path be to the development of, you know, industry level, relevant capabilities, capacities in industries, while in the labor force to match industry needs? Thank you. Extreme land there. Thank you. Antonio Cruz, Mozambique. In Mozambique, the weight, the share of the manufacturing on GDP went down from 17% in 2004 to about 13% in 2011. Washington Consensus MDGs were not very helpful for industrialization in Mozambique. I'm also happy to observe that as a result of the studies, there is a weak FDI spillover in Africa. And that was also a perception. And another perception is that maybe it's important that local companies invest in manufacturing. So the challenge is how to achieve this. Thank you. Thanks. My name is Mona Zalatsuma from the South African Treasury. I'm still left wondering after the session. I think my curiosity would be what would be the best approach to industrialize? I mean, do you go for an export push wherein the market is very unknown, very uncertain, or do you approach it from an import substitution program, which we have not touched on this session, in that case where at least the market is known, is easier to capture? Or do you, of course, I mean, once you figure that out, what would be the most efficient way of support? Do you expose the fiscals to supporting whatever strategy you're going to follow? And most importantly, is this not determined perhaps by the stage of development? Like in a very underdeveloped country, perhaps it's best that you look in what you go for strategy that utilizes on the market that exists as dictated by your imports and less so about focusing on a perceived market which is a much more export push strategy. Thank you. Thank you. Now the last two. Hello. I'm from University of Gothenburg. I would like to make a point that has been emphasized quite often by Danny Rodbrick research that if I'm right, Vietnam came to industrialization when the WTO still have this domestic content preference and it's not the case with African country when they came to industrialization. And that may explain why we have a lot of linkages in Vietnam, which is not found in Africa. So my question is that, is there any wrong that global player like WTO, which Danny Rodbrick complained that it's not for development anymore, they only for trade now, play in explaining this picture or if we just blame the African countries themselves. Thank you. Now a finalist so we can... Thank you. I'm Sunil Bandu from University of Mauritius. Now, must say this is not really my area. So, but I would like just to ask one question. Maybe I don't know from the presenters or the research or we look at it or not. Now, I think John Page widely said it that China has changed the scene. Quite a lot and nowadays when we go and buy things, I mean, I would say we look at it three-quarter of it or more made from China. Now, why say that? Because one of our colleagues said that export processing zone has not been working in Africa, but in Mauritius we had one of the most successful EPZ sector, but today it's in a declining phase and it has contracted by more than 75%. And these firms are moving out or moving to the rest of Africa and you need to go up the value chain. If you don't go up the value chain, you don't survive, but going up the value chain is not easy. You need new technology, you need better scale labor and so on. So, an issue which I've seen some people in the area talking about is equitable trade. Now, what you got to say about this, but how can you compute with China? These guys are telling you that the labor cost is so low and you just can't compete. So, you can put it on the policy makers, on the government, but when they're starting, they didn't know about these new challenges and so on and this issue is coming up in many firms about equitable trade. So, I'd like to know what are your views on that? Thank you. Please. Thank you very much. My name is Montek Alwaly from India. Actually, so many questions have been asked and I was strongly tempted to duck, but since you've handed me the mic, I had a somewhat different question from what a lot of the others are. I mean, it's a very interesting presentation's made and maybe I'm directing this question at John Page. I mean, I agree with the general approach that if you interpret industrial policy to mean that, look, it's not just a case of liberalizing and then waiting for good things to happen. You have to do a lot of things and you emphasized infrastructure, special economic zones and skill development, all of which I agree. But I think industrial policy also used to be interpreted and I think certainly Danny Roderick, whom you invoked a minute ago, did suggest that at one time, that there are some industries that you need to pick up, kind of picking the winner's type of approach. I didn't find in any of the presentations a statement about whether the research suggests that that's a good thing or not. And of course, in large countries that always come, I come from India, so this is always raised and Danny's usually invoked, so I really wanna know what's your view? What does the research of the last 10 years say about countries saying somehow this is the kind of industry that we need to develop and therefore, let's have some industry-specific incentives. Of course, most countries give very high protection for automobiles and I'm not counting that. Other than that, when you get into the new technologies, robotics, all of fabs, this, that and the other, what's a good recommendation that comes out of research? That was my question. Thank you. Thank you very much. Now let's have a responses. Can we have two minutes each starting from whom? Who like start months? You can respond from there if the mic is working. Okay, maybe very quick. I thought they were excellent comments. Let me focus on one or two points that were made about exports. And actually, let me zoom in on this issue of how, what is this relationship between productivity and exporting actually look like? The gentleman from Senegal and also witness commentant on this. So I think that the effect, the relationship is very different depending on the context, right? So when Bernard and Jensen started to look at this thing in the US back in the 1990s, it was all selection. So you observed a positive relationship between productivity and exporting in the data because more productive firms participated in the export market. I mean, that's what drove them into the export market. If you look at this issue in Africa, in low income countries, it's the other way around. There is very little or weak evidence that high productivity gets you into the export market, but there is fairly strong evidence that exporting leads to productivity gains. And it's interesting also to look at our case studies on Tunisia and Vietnam because here you have a bit of both. It's like a virtuous circle. So productivity spurs exporting, but exporting also spurs productivity. So coming back to what I said earlier, again, the bottom line is that this is pretty context specific. I also would like to comment on this issue. So how do we get firms into the export market, right? Well, one of the things that we found is that entry costs are very high, okay? So that's a fact that we think we have established. So what do you make of that fact? Well, one kind of naive response to that is to say, well, let's try to reduce entry costs. That would be wonderful. But an alternative policy would be to, well, let's focus reforms on areas that are such that firms can grow and therefore accommodate high entry costs. Because there are falling average costs because these costs are primarily fixed. And without further analysis about sort of how did the cost side look here? There's no way you can choose between these two sort of competing policies. So more work needs to be done. Thank you. John Rand, followed by Carol and finalizing with the John page. Okay, thank you for the comments. I will also be a little bit selective in that. But I will take up this and before John steals it, the labor mobility and technology transfers comment. We actually started this in the papers. And this is I think one of the results that John likes the most is that we actually see a lot of labor mobility in these transfers. So we see that labor is previously employed individuals in multinational firms are actually establishing firms and they get actually often directly connected as either suppliers or customers of that M&E where they were employed. So we actually see these linkages being very strong in Africa. So this is one of our key results that we also have in the forthcoming book that some of these linkages can actually be strongly mobilized through labor, increased labor mobility and facilitating entrepreneurship of new firms, of new African firms and so for credit support and so forth. But you could maybe target previous employees of M&Es that could potentially be a successful policy. It turns out. So that was a very good question I think and there is a much more in the book about this. Then about the differential effects of due to sector differences that was Justin's comment. There's no doubt about that the type of M&E or M&C is critically important. But our study actually showed we focused in on the manufacturing sector leaving out these mineral resource rich countries. We have them in, but we do not focus so much on this and that may reduce the linkages that we find a little bit of a bias in the results. But other studies have shown that linkages from the mining sector, they are very weak. This is from the manufacturing sector and we just confirmed that the linkages are very weak. So we're actually confirming the results from the mining sector at least, but it's a topic that we need to explore more. There's a question about this, that Asian countries go through the same process of lack of linkages and you refer to electronic sector in the beginning. I can only cite some of our colleagues from the experience from Malaysia. I think it actually is combined with Antonio's comment a little bit that it also depends on how you structure ownership of these M&Es, how many linkages will be transferred to the domestic economy. We have been a little bit too hands off, I think, in actually demanding that M&Es have a joint ownership structure when they're entering developing countries. I know it's not, I will take that on my part, but I think we could actually demand more M&Es when they enter a lot of developing country settings and demand something about joint ownership, especially for lower income countries, to generate these billovers. Did we see that in Malaysia, in some sense, we did, yes. So we could actually demand more of the M&Es in the African setting, in my view. The final one, picking winners. I actually like this because, and I will just cite Hausmann and Hidalgo, their work, and Danny Rodrick as well, in the initial work of their economic complexity and linkages literature. They are actually stating, yes, we can pick winners, but we need to pick the industries that has the most linkages to the remaining economy. So it is not the same industry in every country. It's the industry that is best, most linked to other domestic firms in the economy. Do we have methods for doing this? Yes, we do. Has it been done in a lot of countries? No, it hasn't. So we have a lot of work to do in identifying linkages to pick the winners, in my view, to generate these billover effects. Thank you very much. Carol? Thank you. Thank you for all of the questions and comments. I'll be very quick so we can give the last word to John. I'll just answer the specific question about clustering that was asked by Nissen and by another speaker. So whether or not it makes sense to cluster around a value chain, or whether or not it makes sense to have clusters of similar firms within the same zones. And I think that, again, it comes back down to context. So in our work now, we didn't look at SEZs specifically. We looked at the whole distribution of firms in a country and looked at the pattern of clustering, developed clusters around them, and then looked at what was happening inside. So in Ethiopia, for example, the knowledge transfers are happening between similar firms, whereas in other contexts like Vietnam, they're happening between firms in different sectors. So it really matters, and it really depends on the context that you're dealing with. So in order for the value chain, clustering around a value chain for that to really work, I think it comes back down to one of our fundamental conclusions in all of this, is that it depends on the firm capabilities. Are the capabilities there for firms to actually benefit from those linkages within those clusters? So I think I'll leave it at that. Thank you, Lord. And now, John, you can fill any gaps and wind up, please. In two minutes, right? I can give you half a minute more. Let me just speak to three points and then fulfill my obligation, which my old friend, Paul Collier tells me, is that you never go to a public event without a shameless book plug. So if I forget, Carol, kick me and I'll do the shameless book plug at the end. Three things. One thing that didn't emerge so much from the discussion today, it comes very clearly from the body of the research, is, and since we're in the mood of nostalgia and looking back 30 years, what I might call chenery patterns rediscovered, it turns out for many of these issues, learning by exporting, your level of per capita income actually influences the way in which this takes place. For example, the reason why people dismiss learning by exporting is mainly based on advanced country, as Mon said, results. We find the middle income countries mixed results, what I think is interesting about our work here is that for low income countries, we find quite strong evidence of learning by exporting, which again, moves me back in the direction of saying, well, one of the things that's been a stylized fact of the international trade literature has been that relationships, again, vertical value chain relationships between demanding buyers and firms in developing countries quite often convey firm capabilities. And so I think we haven't yet understood fully the mechanisms of learning by exporting, but we do know they vary across countries, which means again, context is specific, not just country by country, but also according to level of income. Agglomerations, a very similar thing. It turns out that urbanization economies apparently, being in a city are more important in higher technology industries and in richer countries and localization economies and kind of near everybody who does something similar to what you do are more important than low income countries. That has to be set off against the competition issue because in lower income countries, tends to be a more localized market and more competition, driving me in the direction of saying, well, maybe you need an export oriented spatial industrial policy at low levels of income, but again, context is specific. So the chenery patterns matter and we've teased out a few, I think, which is useful. The question on the double to raise, to my mind, a broader issue, which is, is the international governance structure well set up to help support this industrialization effort in Africa? And my answer would be no. First of all, the two most important players in the game, the EU and the United States have quite dissimilar preference programs for imports of African products. They should be harmonized, they should be regularized and they should be made as generous as possible. They should be time bound, of course, but there's no reason why AGOA and the economic partnerships agreements couldn't be brought together and made more rational. Then there's the third big player, which is China. I think China has an important role to play in terms of leading in Asia, making its bilateral preference systems much more transparent and open in terms of what people are getting access to Chinese markets. And then leading Asia a bit if we're going to stay in a world of regionalization and finally the WTO itself. Getting a new definition of what constitutes a country eligible for practices, at least developed manufacturing economy definition would be quite useful, but probably is at least at the moment a bridge too far. And then finally Montek's question about picking winners. I mean, here I take slight difference with John in the sense that my reading of Danny's recent work is to suggest that in his mind, this process of what he calls self discovery, which is in a sense allowing individual entrepreneurs to identify where your comparative advantage is, has come to reduce his view on the selectivity issue and the idea that you can pick one or two industries. I think again, it may be well related to the level of income. If you're worried about the middle income trap in a country like South Africa or in India, you've got to deal with a very different set of priorities for industrial policy. And if you're worried about just getting some industry in a country like Tanzania or Kenya, so some more subtle thinking there. But I think what we came to the view of is that at a very minimum, you need to know where you're going and you need to have a strategy and it needs to take a broader view of what the role of government is than is currently at least the flavor of the month with the World Bank, the IMF and the donor community in Africa. And finally the shameless book plug. For those of you who are interested in kind of the bigger story, including things like what do you do if you have a lot of natural resources? There is a book coming out from Brookings Institution Press. It's called Made in Africa. It will be on the market late this year or early next year, but we will have copies of the book in December in case somebody's organizing an event and would like someone to come along and talk about it. We have a second book coming out from Oxford University Press, which will be in the spring of next year called Manufacturing and Transformation. And it contains the 11 country case studies along with an introduction and the synthesis chapter. And then finally a shameless plug for another wider project, which is that we hope to have, very shortly, a book which is called The Practice of Industrial Policy, which actually looks at business, government, coordination and communication in Africa and in East Asia as part of a compliment to the rest of the work on learning to compete. So, and a shameless plug for a special issue for Mons, which is the Journal of African Economy's special issue on learning by exporting. So, stay tuned for further developments and I have no doubt that the wider website will be flooding you with advanced information on where, when and how you can get copies of these books. So, thanks very much. Thank you very much. You agree with me that the panel has done a great job and the participants have done a great job. Congratulations. Thank you.