 Thank you very much to all three speakers. My question is for Hinden. I'm very grateful and very excited to learn more about what's going on inside China at the subnational level. I wondered a little more. I want to know a little bit more, I suppose, about the process that you mentioned of competition, the fierce competition that goes on between local provinces and municipalities for central government support. Could you just cast a little light on that process of how it actually takes place? And secondly, related to this, you presented a very positive picture of the role of local government in local industrial policy. But I want a little bit about the downside in this area. You hinted at it in the conclusion you talked about the environmental cost, but I wonder what other downsides there might be there. Thank you. Okay. Well, say I'm in then Justin and then we'll come back to, or maybe I'll take Carol as well, although I'm showing left-hand bias here. Thank you very much. And that was very fascinating. Maybe it's a blessing this guy started, John did not stick to cutting them short. So we benefited in the process. Well, now let me just ask about the incentive structure which makes provinces or municipalities perform. What is behind the performance dynamism there? But secondly, I think I like the idea of learning from good practices within the country. Like the case of Penang, it's very outstanding. What makes Penang stand out? And what slows down the learning from others? Because it clearly stands out and I'm not sure what the dynamic process are there. Three, support to training. What forms of support to training were seen in practice? Is it subsidization, free training, or what is it? Next is R&D, which is shown there. Is it based on broad LGA, local government support or is it firm based or a combination of that? Lastly, I think I like the idea of Eric coming through the back door to industrial policy as Paula said. I think the message is quite clear. What I was wondering is whether there is a balance between the cost of stagnating and the cost of making mistakes in trying to be strategic in intervening. Because normally the latter is emphasized very much, that no, no, no, government will make mistakes. But as clearly shown, liberalization has made mistakes. So I wonder whether either the next research or whether you have come across, can we weigh the traders and weigh the two sides and see whether we shouldn't really go back to tolerating mistakes, which later may lead to dynamism in the industry? My question now related to his presentation. And I'm happy to see that you had a very good study of how China support some traditional sectors. Like government, footwear, furniture and so on. But if you look into the success of China in the past 30 years, China from a very inward looking country to now the global, the world factories. And export increased from less than 10% of China's GDP, now to more than 50% of China's GDP. And China, the export, you will look into the competition actually the majority of the export are in new sectors. For example, ICD products, electronics. And those kind of sectors did not exist in China in the 1980s at all. And the success of China, to enter into those kind of new industries actually is also a result of very active the government and investment promotion to attract foreign direct investment. Or even the government started to play the entrepreneurship function, for example, to invest in those kind of sector and they were owned by the government locally. And so in that regards, I think that your conclusion to say that China's government only back the winners seem to be incomplete because a lot of sectors are new to China. And the government active support is also very important just like in the malicious case. To cultivate the multinational companies to come to China or to support to develop those kind of sectors by the government itself direct involvement. So I'd like to have your reaction about this kind of function. If China's government actually try to support the new sectors, they must pick the new sectors. And so how to balance the story that you have with the reality in China? I think we've got a long collection of questions there. So let me go back to the panel and I'll come back to everybody again for another round. But him I think. Thank you very much for the very, very perceptive questions. Let me start off with Justin's point because I think that's, in fact, it probably covers a lot of the other areas. As I mentioned at the beginning, the local government support in China is not one size fit all. It's actually varies from locality to locality. And there's no question that the local government have helped promote the production of new industries as well as the existing products coming from the private sector. I think what Justin was talking about was applying more recently in terms of the high tech. High tech industries, for example, in the electronic and the computer industry. But when China started out in the 1980s, or the 1990s, I think it was more the private sector initiative first. And I think I've seen, actually, I've talked to a lot of local government where they have an explicit, an explicit policies to promoting new products. That is true. But also behind each and every successful story in China, there's also other failing that we don't know. And I don't know to what extent the ideas of having new products will be entirely successful. And I think time could only tell. When I look at the existing better production in China today, there's a lot of progress in terms of technological, in terms of, you could see that by looking at the composition of Chinese exports over the years. It's very hard to separate our wish part of it that belongs to the local government's support for new products and which part of it is actually coming from private sectors. So I take your point, Justin. I think we probably should mention that that is also play a role in helping the variety of exports in Chinese to become better and better over time. That I think that's a well-taken point. There was a question on the dial size of the competition of the local governments. And I think the dial size of it I mentioned briefly is that when local governments are really compete to grow, then there was a disregard of the environmental impact that was back in the 1980, 1990s. I think that problem has been more or less fixed today with the very clear guideline from the central government and the environmental impact. But in the beginning, in the mid 1990s, when local governments was let free to just compete and all the promotion was based entirely on how fast your economy can grow, how fast your local community can grow. Not just your locality, but in relation to others, then they compete like a crazy to pollute. And that's why there was a saying somehow that there was a period in which local government compete to pollute. The pollution was related to the growth incentives given. So I guess that's it. Thank you, Sam, for your questions. The first thing Penang did in relation to inviting multinationals was to go and learn from China International Airport. The first export was heading down and subsequently Kaohsiung, which was the first one in Asia and Singapore, and to look at how they attracted multinationals. And one important point that remains useful for countries seeking to attract for the first time, multinationals, especially when you have a crisis. In the case of Malaysia, we had just come out of a bloody ethnic crisis. It was bloodshed in 1969. It was to provide incentives in order to write off uncertainties and risks before you can first prove to the multinationals that you can indeed perform. So the first two export processing zones were actually started in Penang, and they were actually the bargaining points the chief minister used with the federal government. The rest of the other export processing zones came as a consequence of him convincing them that these are important to create jobs. Now, that's the first part. And again, to me, it's not so much in relation to the state government incentives given to them. It's the pressure the Penang government faced in delivering to the constituencies because they had just defeated the federal government political parties then. Now, it's the pressure that required that they perform, which means to say that the state also hired officials to run all these massive organizations, Penang Development Corporation, subsequently Penang Skills Development Center, which very merit-based, capable experts. And this is a question John has been raising in a different Kwaika project about political capture. In here, you don't have the room for political capture that the federal government would face in the sense that they tended to hire people who were of the Boon Putra time, largely the Malays in the rest of the other organizations. So you don't have that detachment, including, we both admire Tung Mahade, but let me mention that he, from 1981, when he became prime minister, till the first major crisis, he was such a strong nationalist to the point where he just stopped giving incentives to multinationals. But while being a strong nationalist, he didn't take the steps Park Chung he did. He didn't really hire people with proven tacit experience in industry. He put all these guys there to run the show. We have protons there now, which for me, things shouldn't be there. National card that isn't really bringing profits, but it is still there for national pride. So that's a point I'd like to mention, but also you have to know that the Penang government, to some extent, was able to, this is like, if firms are not performing, you expect them to be liquidated. You don't want what Kaushik mentioned the other day to liquidate in firms that are, employment size are more than 100. It will be simply impossible because of the unions. But in this case here, they were willing to remove officials who did not perform. You didn't have that elsewhere. In other words, the chief minister himself was willing to go and remove these people simply because his own position depended on the performance of the state. Now likewise, the collaboration that came with the multinationals, that's part of the reason why I like this evolutionary argument, the need to understand the dynamics of industry, industry itself shaping industrial policy rather than one where you just create an environment and firms just respond. What happened was, an example of state failure will be when people were talking about a lack of diversification in the electronics industry. The state took it like, we need to move out from semiconductors. We need to have consumer electronics, industrial electronics, little did they realize consumer electronics and industrial electronics were more labor-intensive, less knowledge-intensive than say semiconductors. So they attracted these disc drive firms, firms that assemble television sets and so on. That only lasted for five years. They realized not only that these firms required large workforces, but they were bringing down skills. And now if you see in Penang, you don't see these firms anymore. In other words, Penang was ready for quick renewals. They made a policy mistake, they were able to correct the mistake. Governments fail all the time. We have to accept that element of it. And in the case of Penang, they were flexible enough realizing that unless they perform, the opposition parties would win. In any case, the opposition parties are ruling now in Penang. So I'd like to just in the interest of time discuss this point. But they didn't have, let me put it to you, they didn't have the capacity to access R&D grants, for example. Simply because there's all these national regulations that require that you hire the Malays. And you have, even to start a company and you need to have something like 30% equity from them. These were constraints they faced. Despite those constraints, they have performed to the level they have achieved. Thank you. Yeah, so the question was how do you balance mistakes in mistakes of liberalization with mistakes of intervention, essentially. And I don't have one size fits all answer to that question, but I think it's the right question to be asking, right? So thinking about these balance between say government failure and market failure is exactly what we should, what the discussion should be about, not just, and you're right, that I guess now it's changing quickly, but when I was coming up and I was in grad school, everything was about all government failures and so reduce the amount of intervention. The two slides I didn't show were about as different strategies in industrial policy. So one would be something like preferential tax treatment for targeted sectors. One thing that I'm working on with co-authors to try and develop is a sort of a matching grant program where firms have to come with innovative ideas, which then a board would decide, would select among them the most promising, which has the advantage that the board or the officials don't have to come up with the ideas themselves or it requires less knowledge about what's on the horizon, what are promising ideas. It's arguably more vulnerable to capture and to corruption than a simple preferential tax treatment for a sector. So that's, again, something else that has to be weighed, but that seems to be a promising direction to go in thinking about industrial policy. Okay, with your kind permission, I'm gonna take another round of questions. And yeah, I'll carol a call, but then I also gotta get over on this side. So John, and then back here, let's see, one, two, three, four, let's try that. The shorter the questions, the longer the answers can be. I'll be really quick. So my question is mainly for Raja, but also for Hinden, and it's about the fact that you both mentioned, especially in Raja's case, when the multinational companies led to a lot of spin-off local suppliers in the upstream sector supplying inputs. I'm just wondering whether that happened organically or whether there was policies in place to make that happen. And also do you have a sense of how well these suppliers survived and how did they do? And this is kind of a similar question for Hinden in terms of the development of these fully integrated supply chains. Do you have any examples of exactly how that happened? Was it through contracts? Was it through setting up new enterprises, state-owned enterprises, just a sense of how that happened would be interesting. Thank you. This is for Eric. I thought that was really intriguing story about the disappointing post-NAFTA growth in Mexico, that the liberalized trade skewed the industrial distribution and the composition in a certain direction. But I have questions both about the evidence and about policy implications. So you use plant-level R&D and look at the share of plants in the industries, but R&D is not usually a plant-level activity. It's a company level. And you have very high rates of R&D at the plant level in your survey. So lots of industries with more than 50% of plants carrying out R&D strikes me as kind of odd. But more fundamentally, productivity growth comes about, I mean, it's not necessarily that closely connected with R&D or with innovation. It takes place through adoption as well and various kinds of catch-up mechanisms. So I wonder why you didn't consider using some sort of international productivity growth by industry as a, it would be at least another thing to look at. About the policy conclusions, I'm a little bit confused because at one point you said that the government doesn't have any special informational advantage and therefore more horizontal kinds of approaches like R&D subsidies would make sense because of the externalities. But then at some point you seem to be supporting more targeted sectoral interventions and on your list, there were a couple of specific sectors like high-grade steel or something like that. And you mentioned that also in your response to your previous question. So I wonder where you want to come down on that more specific question about the type of industrial policy. I will do it very quickly. John ran from University of Copenhagen. First question to Eric. I'm not surprised, that's surprised about your result if you compare your results to the result of where you learned from exporting. Because if you compare the two exporters that are maquillators to the other exporters, they have lower innovation rates and that is what you show. But we expect lower in raising rates in lower value-added sectors because of if they export and learn less in lower value-added sectors as that literature is showing. So we should expect to see lower in raising rates if we believe in the learning by exporting story. If that translates into knowledge, we can talk about that later. Then I have a general question to all of you. It's about acclimation. A lot of the academic literature has difficulties in finding acclimation effects. And we are still promoting industrial parks. We have an example in Vietnam where we see a lot of unfilled space in industrial parks and all. So some of the firms are actually picking up that it might not be as beneficial as we are all saying all the time. So why is that? And why can we not find the positive effects in the acclimation literature? So thank you. Thanks, John. Dona Limmar from Repo. My question regards the last bullet by Eric that we need more research to establish the positive or negative evidence of industrial policies. But in my opinion, they do exist. China, for example, seems to be a positive evidence of working industrial policy, South Korea. There are some selective sectors also in Africa, for example, in Ethiopia. But there are also, of course, areas where industrial policies failed. So what I think we need more research is why industrial policies works in some countries or in some sectors and not in others. That is my question. Okay, with the full knowledge that they're standing between you and lunch, I'm gonna start with Eric and just work my way down. Yeah, thanks for the questions. Yeah, John, so this R&D measure, part of the reason I put the whole language on there is because I think it's different from other R&D measures that people typically use. And in general, it's really capturing any sort of upgrading, essentially asking the plant whether they're doing any sort of upgrading. It's true, we might like to have that at the firm level, but the data is at the plant level and most of these guys are single plant establishment. So I think that explains part of it. And I think in particular it includes whether they're adopting new technologies, not just whether they're doing R&D in the sort of the way we typically think of it. So I think that is included in the measure, which is part of the reason why I like it. But your point about shouldn't we do TFP measures is also well taken, right? And so there is evidence that I cited the Lopez Cordova paper about that NAFTA was having positive effects on TFP. So that would be a much more positive sort of view of effective NAFTA on productivity growth. But I agree, there's more work to be done there. You asked about where do I come down on the sectoral promotion policy? So I cut off a little bit and I mentioned the matching grants. So I think my basic answer to that is that this sort of matching grant models seems to me more promising than the sectoral promotion model. I listed what Vietnam was doing because it seems sort of in line with what I was talking about. But I like the matching grant idea. Now with the caveat that you need that subject to that's vulnerable to capture. And so you only wanna try that I think in situations where you can be pretty sure that your board or whoever it is that's making decisions about these things is not gonna be captured, which is not true in many places. And so I don't wanna push it too hard, but that's the sort of direction I'm going about the learning by exporting. So I'm not sure that's exactly what's going on. Because these guys are exporting a lot, right? Even the Maquilas are, you know, there's 95, everything's exported, right? Which isn't true about all the, you know, everybody, all apparel firms in Mexico, but so I don't think the story is that they're, so if it were just learning by exporting story, then this would have been the right strategy because those are the goods, Mexico is specializing in the goods that's gonna export to the US, which are the apparel and electronics and that sort of thing. So I think it's something else is going on. Thank you for the question. If you're looking at suppliers that actually evolve in Penang, because I mentioned that Penang itself doesn't have the capacity to introduce its own industrial policy, it's at the national level. They have the first industrial mass supply and the second and the third. Penang felt discriminated in relation to the targeting of say incentives for those who are willing to participate in the supply process. What I'm meaning here is really the workforce that evolve in the multinationals. It's somewhat like those guys who enjoyed all these experience and went back to Korea and Taiwan from the US. We also have that kind of individuals not brought back, but here people who worked in the multinationals themselves, a lot of them are in Singapore now because the incentives are there for them to perform there. In fact, in some firms, about 50% of the engineers are from these kinds of firms. Now, these guys recognized in the mid-80s that there was this capacity to supply. What Michael Best calls constrain in one firm becomes an opportunity in another firm. So if you have a problem of, say, suddenly having to fix something to do with adapting a machinery and so on, then it's an opportunity for a supplier firm to do simply because the supplier firm can amortize investments by supplying many other firms. Now, in that sort of sense, from the mid-80s, he saw the growth of that through largely the relocation of engineers, production foremen, and so on because they saw the opportunity. And the Penang government was very happy to support by way of arranging licensing and so on very quickly. So in other words, you would classify Penang governments as facilitative rather than really aggressively intervention is no. Now, that's simply because Penang state largely run by Chinese, Malaysian Chinese. Now, what really happens subsequently is that you don't see the transformation of that to suppliers who became, like, say, what Mauritian pioneer in Taiwan, that you have contract manufacturers in a wave of fabrication only because you need all this upfront grant. It's an extremely expensive start wave of fabrication. And in the case of Taiwan too, the E3 Labs also, electronics research and service organization, this organization actually acquired RCA in 1970. That became UMC in 1980. He was the Sintu Science Industrial Park director then. Then subsequently, when they had this joint venture with Phillips, 87, that's TSMC, he moved on to do that. Now, Penang didn't have that sort of opportunity to upgrade supplier capacity to that sort of level. The highest level they would have got is to become global service providers. An example would be transcapital, which was also having a subsidy that I heard from Rafael that Intel is leaving Costa Rica. I was in Costa Rica in San Jose, only one semiconductor firm, largely because the president negotiated, the Intel CEO negotiated directly with the president. They went there. But in this case, the suppliers, including Malaysian suppliers, are there. That's the highest level they went. And the role of Penang state is really facilitative here. Thank you. There was a question on specifically how China actually built up the upstream and downstream industries. I think they've done it through a number of channels. One of them I mentioned is, for example, in the case of ill-meat products, not only they provide subsidies on land, subsidies on credit for the meat product factory, but they also give subsidies to the farmers producing feeds for hogs using in the meat product. They use hogs. They use hogs. So that's one way. And then the very same local government, Igor, also built a very large market, they call it market, for the metal subsectors to bring so that all the producers of the metal products can bring their products into one place. I was visiting that one, and I saw a couple of ladies from Kenya who was asking what Kenya was doing there. And the lady was actually going all the way from Kenya to buy the logs. And that's one place that they can instead have gone to various producers and not knowing how much and not knowing they just come to a huge marketplace where you could see hundreds and even thousands of little small logs coming from Igor. So these are the kind of channels that they produce. They also end to build a lot of the facilitating infrastructure that can help the downstream activities. So for example, in the case of these shoes, I think it was the Chandu shoes, they started building the so-called the auxiliary material for the downstream manufacturing by providing a lot of subsidies to the industries. And as a result, within the period of eight, nine years from a very obscure producer of shoes, they actually produced shoes that Obama's, the wife of President Obama actually was wearing in the first inauguration. All of that is described in the book, and I'll leave it there.