 Good evening, good morning, or good afternoon, depending on where you are. My name is Steve Seng. I am the director of the Sewers China Institute. Let me welcome you to another of our webinars on China. Before I introduce the speaker and the subject, let me just remind you that if you would like to ask a question at the end of the presentation, please use the Q&A function at the right hand bottom of your screen. When you do so, it would be very helpful if you would provide information about yourself for my benefit. If you would like to stay anonymous while you raise a question or a comment, and if you indicate that you would like anonymity, it will be respected. But the information will still enable me to pick questions more effectively than if I don't have that information. For the presentation, I'm delighted to have Professor David Bartman to speak to us on the subject of aspects of defence industrialisation in China from 1949 to 1989, a very important period of China's development. Our speaker hosts the Henry Jackson Professorship of International Studies at the Jackson School of International Studies University of Washington in Seattle. He was educated at Stanford University where he received his PhD. He then taught at Stanford and at Princeton before he moved to his current position at the University of Washington. He is the author of many learned papers. I will only mentioned the two books that I think should be mentioned in particular. One is Chinese political system, bureaucracy, economy and leadership in China, the institutional origins of the Greek leap forward, which is a book that he published entirely himself. He's also the co-editor with Professor Dali Yang of Yan Jiaqi and China's struggle for democracy. He is currently working on a book on China's defence industrialisation, which is very much the subject of what he is going to be talking to us in the next 40 to 45 minutes. Over to you, David. Thank you, Steve, and thank you for the China Institute for inviting me, obviously for multiple reasons. I'd love to be doing this in person, but this is the next best thing. So let me move on. So this is a book project that's been in the works for way too long. It's part history of defence industrialisation in China. It examines the effect of defence industrialisation on China's political economy. And it also looks at how defence industrialisation fit in with China's foreign and defence policy in terms of strategy and so on. Sort of a supplement to the kind of work that Taylor Fravel did with his book on active defence. He's providing the sort of doctrine and strategy side. This is a look at weapons procurement and building of factories to produce the weapons. The book is based on a wide range of documentaries and statistical and industrial census sources. From these sources, I've built a database that argues or that that I think I can show that there were 1161 large and medium scale defence enterprises in China in 1985 out of a total of 8285 large and medium industrial enterprises. And I got this basically from the 1985 PRC industrialisation industrial census. The presentation today is based on the history and political economy of defence industrialisation of the first and second parts of the book. So why write on defence industrialisation? So now, of course, is well known for the saying political power grows out of the barrel of a gun and the party must command the gun. In the English language literature on this saying, this statement has been interpreted in two ways. One, it reflected a statement of necessity that the CCP was almost wiped out in 1927 because it lacked an army. And therefore the conclusion was, of course, that the CCP Chinese Communist Party had to have its own army. Others have taken this statement and the follow on about political power growing out of the barrel of a gun and the party commanding the gun as a sort of metaphor for tensions between party and army in the PLA in the CCP and PRC period. But I'm focusing on the third and much more direct kind of issue here. And that is if political power grows out of the barrel of the gun and the CCP must have an army, where and how does the party and its army get its guns? Building on that, how does obtaining guns impact the overall economy? How do the weapons procured fit into strategy? What does the amount of investment in the rate of production tell us about perceptions of threat and these kinds of issues? And certainly Mao was not the only Chinese leader to believe this. Chinese modernizers since the 1860s had made developing an arms industry a core goal of the Chinese state. I'm not going to talk about that today, but that many factories that were important during the early 40 years of the PRC trace their origins back to the self-strengtheners and the modernizers of the 1860s and beyond. So the short answer to where the guns came from was that after 1949, they overwhelmingly came from the Soviet Union. And even after the break in 1960, Soviet weapon systems would be the basis for Chinese military power into the early 1990s. And this is a major theme of the first part of the book. You know, almost no sector of the Chinese political and economic system after 1949 had as deep a Soviet imprint on it as did the defense industrial sector. Some of this is deliberate and some of it is by default, as I'll get to. This may not be an Earth shaking kind of argument, but there are elements to the Soviet defense industrial system transferred to China that are not so widely appreciated. I don't have slides on this, but I'm going to give you a kind of path dependence argument that suggests that the Soviet style of industrialization is quite different than American, English, German, Japanese during World War Two and so on. So let me speak for a few minutes about the Soviet defense industrial model and China. So based on World War Two, the Soviet economy and defense industry was less advanced than the U.S., U.K. and Germany in World War Two, especially if we use GDP per capita as a proxy for technological ability. Yet the Soviets were able to defeat Germany through a profligate use of men and resources and the intense mobilization of Soviet society. There were big gaps between the Soviet Union and China, but certainly less of a gap in terms of technical levels between the Soviet Union and China than China and the United States and U.K. With Soviet and Stalinist ideology, there was a firm belief in the Soviet Union that Mao shared that the socialist states were embedded in a capitalist world economy, war with capitalism was inevitable. The Soviet Union must be ready for war. War and defense per operation was a core element of the Soviet economy after 1927, if not before. World War One and World War Two were both decisive in different ways for the Soviet Union. There was these were massive industrialized, prolonged conflicts that posed existential threats to the regime in power and this was an assumption and reality built into the Soviet system, and I would argue transferred to China. The Soviets fought World War Two and the Russia fought World War One based on massive land-based combat, prioritizing infantry, artillery, armor, and increasingly tactical air. War was fought through attrition and maneuver. The system had to generate lots of weaponry before conflict began, anticipating a war of attrition. And after war, after conflict broke out, the system had to generate hugely greater quantities of weapons and other requisite materials. Soviet experience emphasized land-based combat, artillery and armor was unique among major combatants in World War Two. They spent much more of their defense budget on these things than did other major combatants during World War Two. The others emphasized air, air defense, and navy for where most of their money went to in fighting World War Two for Germany, Japan, UK, US. And so the Soviet experience of land-based war with most of the resources going to infantry and their equipment largely matched China's war with Japan and the Civil War. Soviet military personnel were generally poorly educated. Soviet military equipment was generally rugged, reliable and relatively easy to maintain. And in some cases, close to state-of-the-art weapon systems characteristics, in other words, roughly corresponded to the skill levels of military personnel who would use them. For the Soviet Union, there were relatively few basic few modifications in basic weapon systems designed. Once weapon systems entered production, they were produced in huge numbers at scale. And because of scale and not changing them all the time, you had quite cost-effective or cheap weapons. Moreover, the Soviets produced many fewer weapons types than most other combatants. When it came to land warfare, the Soviets produced about 34 basic types of weapons. The Germans were producing more than one hundred and thirty, which is why the Germans couldn't produce nearly as much as the Soviets did. The Soviets imposed a strict quality control system and ability to surge production in as part of the basic nature of the system. The Soviets built large factories to produce military equipment, yet it was recognized that normal production would not be enough for wartime needs. Therefore, Soviet defense factories had excess capacity, which in times of relative peace could be used for civilian production. Moreover, in the event of full-scale war, non-defense factories would be tasked with defense production as well. The Soviet system also separated defense industrial production from defense industrial research and design, a system that carried over directly to China. And sort of to sum this up, the Soviet victory in World War II reinforced many of these elements of the Soviet model. Stalin stopped being able to purge military heroes in World War II. I could put them aside, but he couldn't purge them. Leaders who successfully mobilized to fight the Germans became became the leaders of the Soviet Union for the next 30 or so years. World War II was the most formative experience for these leaders when it came to political, military and economic affairs for the Soviet Union. And I would argue also for China, the lessons of World War II would be reified and perhaps frozen in place. By extension, this was the system that was transplanted to China from the Soviet Union. And remember when the PRC was established in 1949, World War II was just four years before the Soviets had tested nuclear weapons weeks before October 1st, 1949. No one else could supply or would be willing to supply China with weapons. And so China was was stuck, if you will, with the Soviet system. But again, they weren't unhappy with that in 1949. And they would, in many respects, buy into that system without necessarily reflecting on or knowing some of these background factors that affected how the Soviet system worked and that the biases that were inbuilt into it. So if I were to briefly sort of characterize the history of defense industrialization in China from 49 to 64, you had the Soviet phase of defense industrialization. China was building Soviet factories in China was using Soviet designs and specifications in all weapon systems. There were Soviet experts in major Chinese factories. Chinese managers and students went to the Soviet Union. These were all fairly well known factors. Soviet quality control techniques were imported. Even after the Soviets left, Soviet hardware and software largely remained in place. And indeed, you know, even after the Soviets left, the Soviets sold China the plans to make twenty ones in 1961, 62. So in this way, the Soviet defense industrial path was embedded in China, I would argue. From 65 to 71, in contrast to the the Soviet phase in defense industrialization, you had the defense sectors greatly forward. This was, of course, the third front period. It was characterized by excessive haste, waste, poor planning, massive expansion of the defense industrial sector affected, particularly in 67 and 68 by cultural revolution violence. We could ask whether, in fact, this this period of heavy, very heavy defense industrialization actually made China more secure. I'll come back to that in the conclusion to the talk. And from 72 to 89, basically was was roughly equivalent to the period after the great leap forward for the rest of the economy, a period of drift and readjustment. Some factories were closed. Others shifted entirely to civilian production. All factories increased civilian production. There was very limited military procurement in the 1980s. Arguably, the ordinance industry survived only because of exports and civilian production. And with that, I'm going to show you a number of tables about defense industry, development and so on. They don't come across as well as they might on PowerPoint, but I'll talk through them as best I can. So let me look at show you some of the data from my project and show you some of what I've found. Again, apologize if it's hard to see. This is a work in progress and I fear I'm caught up with the details. Big picture may be lacking. And hopefully in the questions and answers, you'll help me figure out some of the big details. So let's start with what the defense industry looked like at the start. So in 1985, from my database of 1161 factories, I've got details about which sector of the defense industry they were attached to for 1152. So for ordinance or land-based combat, 372 electronics, 281, as you can see. The last column is nuclear with 49. So obviously ordinance was the number one subfield of the defense industry followed by electronics and the PLA. PLAs I've already mentioned, experience was largely ground-based and for bureaucratic, organizational and experiential reasons, not surprising that the Chinese high command would favor land-based weapons. Mao following Fravel's interpretation and active defense also strongly boosted ordinance industries in the 1960s so that they could be the basis for resisting revisionism as part of the emerging cultural revolution or the basis of local guerrilla warfare against the US or Soviet invasion. Electronics is somewhat more surprising. Perhaps there was a particularly obscure period during 69, 70, 71 when someone, perhaps Chen Boda, was strongly advocating building electronics factories for radar communications and other military uses. PLA enterprises were a diverse set of factories. For example, there was a PLA printing press in every major major regional PLA headquarters. PLA tailoring facilities were 60% of all large and medium tailoring enterprises in 1985. But other PLA enterprises were naval shipyards and repair yards, aircraft maintenance facilities, armories for repair and maintenance of tanks and so on. Prior to 49, there had been very limited Chinese production of modern warships and aircraft. So most of these factories and shipyards are a post-1949 vintage, and that is absolutely true of missiles and nuclear weapons. So the second table gives you a sense of data about what I have in terms of number of people employed, its output in 1985, the fixed assets embedded in production in the facilities of the factories. So about two and a half million people working in the defense sector, which was a little less than 4% of all industrial employment. It was more than 11% of large and medium enterprises. Output almost 3 billion yuan, which was about a little less than 4%. Of national industrial totals and close to 8% of large and medium enterprises and fixed assets, 5.3% of all national industry and 8% plus of large and medium enterprises. Probably if I had all the data, we'd be closer to 12% of all labor in large and medium, 8% or 9% of output and 10% or so of fixed assets. Okay. This third table gives a sense of the provincial distribution of defense enterprises by provincial level units. This is 1985, so there were 29 provincial level units. Haidan becomes a province in 86 and Chongqing becomes essentially administered city in the late 90s. So we're talking about Chongqing being part of Sichuan for the entire purpose of the talk. So not unsurprisingly, we know that there were the most defense industrial enterprises in the Northwest and the Southwest, what we know about the third front. I'll talk more about Guizhou, Sichuan, Shanxi and Gansu later on, but here we see that Sichuan was the overwhelmingly largest center of defense industry followed by Guizhou and Shanxi and then moving on down and we'll have more to say about these details later on. There are quite different profiles across different provinces and the percentage these made up of all large and medium enterprises also shifts quite dramatically. In general, we'll get to provinces that had more than 20% of their large and medium enterprises in defense enterprises were in the bottom half of per capita GDP for an extended period of time. Provinces that had less than 10% of their economies in large and medium defense enterprises tended to do extremely well in the reform period and have to this day very high per capita income. So on average, there were 40 defense enterprises per province. They averaged about 14% of all large and medium enterprises on an average basis. Okay, so moving on from this again, I'm sorry if this is hard to read, if we look at the temporal development of the defense enterprises by different sectors, we had prior to 1949, 70 enterprises, most ordinance, some electronics, some of the BLA, and so on. And I want to clarify with the 1985 database, this is when enterprises got started, even if they weren't producing defense products. So for example, capital machinery in Beijing, which produces missiles, was established in 1939. And that's what it says in the senses materials, industrial senses materials. It certainly wasn't producing missiles in 1939. It was obviously Beijing was under Japanese occupation in 1939. So it wasn't even producing weapons systems for the Chinese military at that time. But in the 1960s, it became a missile producer. And that's what we know it as. So some of these things were not actually producing weapons at the time, but just go through this. We can see that from different time periods, the production of weapons, weapons factories. So 70 existed in 49, you added 78, 49 to 52. You added 61 during the first five year plan, 104 during the Great Leap Forward, 65 in 61 to 64, 543 during the third front of 65, 71 and 206 from 1972 to 1985, when census materials ends. And I don't know about 34. So this averaged the production of 28 and a half or 29 factory defense factories a year from 49 to 85. And it breaks it down by various sectors here. So. Yeah, let me just move on since I'm using time. So this compares the development of defense industries versus non defense industries. And what we can see here is that prior to 1960, there was a steady development of defense industrialization in the PRC, but defense industries were never 10% of all the factories built during the period. That began to change in 1961. And particularly 62 when the rate of defense industries coming online began to grow quite rapidly, where it became 20% of 20 more than 20% of non defense enterprises. And 17.5% of all large and medium enterprises. The third front period, obviously a huge increase both in the relative share and absolute share of defense, large and medium enterprises. And that continued from 72 to 1985. And so from 61 on their above average in terms of when defense enterprises or how they compare to both overall industrialization, large and medium industries and compared to civilian industries. Some more detail about this in the next chart again, this is hard to follow. But if we look at the three main sectors and the 1985 industrial census used 39 categories of defense industry, sorry, of industry, the three major categories were machinery for defense industries were machinery category 34 sector 34 transportation sector 35 and 37 electronics. So 34 would include ordinance and space enterprises. Transportation would include air naval and many PLA factories, 37 electronics military communications and so on. As we can see here, defense electronics outnumbered non defense electronics in every history in every period in the PRC. Defense machinery enterprises only greatly increased significantly compared to non defense machinery factories. After nine 1960 should be noted that machinery as a category was the largest single one of large and medium enterprises in the 1985 industrial census with 1855 enterprises. Defense transportation was always a higher relative priority than defense machinery. Defense transportation was never less than 30% of non defense transportation. And for about 25 years was 100% of non defense transportation. The table suggests that the three largest sectors of heavy industry, non defense construction greatly diminished after 1971. More research would have to be done here. But on this basis, I'd be willing to argue that, that the period of heavy extensive event industrial expansion had started to add by 1972 in PRC economic history. And while many factors can contribute to an understanding of why the slow down of non defense construction was taking place in these three sectors. It does argue that the Soviet style of industrialization in China was beginning to come to an end after 1971. Okay, so this, this table shows you the ranking of based on official figures that were published in the late 2000s of the ranking of defense industry investment and defense science and technology investment compared to other sectors in the Chinese economy. From 52 to 65. And based on this, we can see that the defense sector was always a very high priority except for one year and the great leap forward. It averaged between ranking three and four out of 12 sectors surpassed only by metallurgy, electric power and rival by coal, all of which were very capital intensive sectors requiring lots of investment. So this gives us a sense of the budgetary priorities for investing in heavy industry. And I think that's a good point for heavy industry and the defense sector. And let me move on. I tried to construct estimates of defense, defense industrial production. How, how worthy these are of any sort of use remains to be seeing. I'm getting a little short of time. But I'm, I'm arguing that we could estimate that defense industrial production was about 10% or so, 9% 10% maybe a little higher in, in the great in the third front period. And this is of all state owned enterprise production at that time. I have a lower estimate based on another methodology. I'm happy to explain this during questions and answers, but I think I should just not go into this in particular detail at this, this time. As mentioned with part of the Soviet system that a lot of factory output from different enterprises went into civilian production. And I think that's one of the things that I've found based on materials from five provinces, we can see that civilian production and flowed out of their defense enterprises. As noted there, this generally excludes PLA and electronics firms. But during the early 60s, we can see civilian production. Civilian production goes down from 42% of total output from these factories to 17% to 12%. And with the third front period down to 8% and stays below 20% until 1975 stays below 25% until 1979. And then by 1989, it's 73% of all output from the factories in these provinces for which their data. Okay. And then we have data on exports by the defense sector for the 1980s. So pay attention here to the left column that exports from the total ordinance industry as a percentage of all production. These are based on two different series of data, so they're somewhat different, but nonetheless, in 1985, 25% of all output from the ordinance industry was exported in 1985, 38% of all output from the ordinance industry was exported in 1985, 35%, 87%, 85%, 88%, 31%, and 89%, 17%, or so. Most of this was going to Iran, Iraq. But this gives you a sense of, you know, what the defense industrial sector, at least the ordinance industry was doing in the 1980s. And this is the data of the outing province distribution across different defense industrial sectors, aviation ordinance, naval missiles, and nuclear. The outing province distribution across different defense industrial sectors, aviation ordinance, naval missiles, and nuclear. The outing is a major center for China's aviation. It was a Shanyang aircraft corporation is the premier jet fighter production facility or it was from 49 to 85. It's also a major ordinance producer. It had significant naval shipyards in Daliand, and the nuclear submarine facility also on the Bohai Gulf. And so from this, we could look at relative priority and perception of threat. And we could see that the, that aviation for most of the period was the priority planes being more expensive than tanks, other things equal, but it was aviation, conventional ordinance, and then naval and missile production. These were not significant or terribly significant missile production facilities in Liao. So let me move on. So let me let's include the sort of data presentation with looking at four provinces, Guizhou, Sichuan, Shanxi, and Gansu. So with Guizhou, Guizhou was the province that was most affected by defense industrialization. You can see that there was one defense factory which actually wasn't producing defense goods in created in Guizhou prior to 1964. Guizhou obviously a poor province, not much industrialization. But if we look then at 65 to 71, you build in Guizhou, 40 large defense factories, 27 medium size, and by 1985 you end up with 83 defense factories, only one of which existed prior to 1965, out of a total of 175 factories. So 46% of large and medium enterprises in Guizhou were defense factories, all of them built after 1964. So Guizhou was, you know, it generally was in the bottom five for capital construction investment in 53 to 64, its average capital construction rank was 23, 64 to 71, it was top 10. Again after 73, bottom 10. But for the entire period from 53 to 95, it's GDP per capita ranked in the bottom five. So there was dramatic structural economic change caused by defense industrialization, but there was little change in the position in the national economy, so that you had much more industrial employment, but you really didn't have a fundamental change in how the economy worked for the people of Guizhou in some ways. And indeed, basically from 1965 to 75, there was no change in per capita income at an absolute level going out. So here we have a case where the third front period was absolutely decisive to the industrial structure of Guizhou, turning to Sichuan. So with Sichuan, the situation is somewhat more complex. Let me take a drink. It had the most defense enterprises in the country by more than 50 over Guizhou and Shanxi. Sichuan, of course, had been the KMT's redoubt during the war against Japan. The sagas of factories being removed from Wuhan and other cities and transported up the Yangtze during the early days of the Sino-Japanese war of 37 to 45 is well known. These would include some enterprises that would be core and ordinance plants well into the history of the PRC. As well of the 287 turnkey factories the Soviet Union was to sell to China, 88 were defense factories for defense purposes. One of these might have included four projects related to nuclear weapons. In any event, 10 of the 88 Soviet factories were in Sichuan. Sichuan had a fairly large and important industrial base in the 1950s and a significant place in the national defense economy as well in the 50s and into the 60s. Nonetheless, almost 100 new defense plants would be built in Sichuan 65 to 85 with more large defense factories built over this period than large non-defense factories. Yet, as with Guizhou, there was not much of a change in per capita GDP from 53 to 85 or 95 stars. Shanxi was the province that received the most Soviet plants, 53 to 60, 21. It was already a high priority defense industrial base by 1960 and would see substantial increase in its defense establishment from 65 to 85. But the increase during the third front period and after was not as large as in Guizhou or Sichuan in relative terms. And defense industrial construction outpaced, sorry, non-defense industrial construction, outpaced defense production, construction 65 to 85, substantially so during the core third front period. But again, there was no major change in per capita GDP on a rank order basis. Finally, Gonsu. So, Gonsu received four Soviet defense plants and the two large nuclear plants in Gonsu may also have been on the Soviet contracts list or another specialist of Soviet projects. Gonsu had a limited industrial base prior to 1985. It had the seventh fewest large and medium enterprises among provincial level units in 1985. The third front and after did see an upsurge in defense industrial construction in Gonsu, but it was much smaller than the non-defense increase in Gonsu, especially 65 to 71 and less so 72 to 85. Again, this had little perceptible impact on GDP per capita. So with that, let me thankfully relieve you of data charts and talk about what are the key takeaways from this presentation and defense industrialization in China 49 to 89. So first and most basically defense industrialization was a substantial and fundamental aspect of Chinese economic history from 49 to the mid 1980s. Defense enterprises constituted 14% of all large and medium enterprises in China. If the defense industrial sector was a separate category in the 40 category system used to characterize Chinese industrial enterprises in the 1985 industrial census, it would be the second largest sector after machinery subtracting defense machinery plants. If we include factories that provided substantial inputs to defense production, or that during the third front period produced weapons that were organized to defense mobilization production lines, another 1,000 or so large and medium enterprises were deeply connected to defense industrialization at one time constituting well over a quarter of all large and medium enterprises. Second, the pattern of defense industrialization is generally in keeping with the degree of threat in China's external security environment. There was a steady buildup in defense enterprises in the 1950s as part of China's overall industrialization drive. But the Sino-Soviet alliance meant that China was with not on its own and it did not need to concentrate on China's defense industrialization. While there was a significant increase in defense enterprises coming into production during the Great Leap Forward, the non-defense industrial sector grew substantially more. But with the effective ending of the alliance and increasingly worrisome and increasingly worrisome external and internal developments, defense industrialization became a much higher priority. Third, although this has been asserted more than presented today, the defense system was profoundly influenced by the Soviet pattern of defense industrialization, as well as Soviet designs, weapon systems, defense factory management systems, specifications for machinery, etc. When politics was in command, Soviet-type managerial systems were ignored or attacked. And when it was time for rectification of the excesses of politics in command, at least for the defense sector, China had no other real model to turn to except Soviet managerial practices. Fourth, defense industrialization was a lasting legacy to the political economy of every province. In 1985, there was tremendous variation in the proportion of output from large and medium enterprises to total provincial gross value of industrial output. With little more than 22% of Zhejiang's gross value of industrial output coming from large and medium enterprises, while 72% of Gansu's came from large and medium enterprises. This speaks to Kelly Tsai's argument on capitalism without democracy, that there are a number of different patterns of provincial political economies and that in the places where state-owned enterprises dominated, as in the Southwest, Northwest, and Northeast, reform policies emphasizing marketization, light industry, and smaller enterprises were generally stymied. These are the places in general with the lowest per capita incomes in China. This is not all due to defense industries, but it is not unrelated either, particularly for Sichuan, Guizhou, Shanxi, and Gansu. Conversely, those provinces with the lowest proportion of GVIO coming from large and medium enterprises have the greatest income gains leaving aside Tibet. As well, the third front with its emphasis on dispersed production, building factories or workshops and caves and other remote locations, and building tremendous redundancy added considerable inefficiencies to the already inefficient planned economy. China would spend additional billions in the 80s and 90s trying to remove factories to move factories in remote locations to major urban centers, closing down other factories and otherwise absorbing lots of other costs of adjustment to try and salvage something from these factories. Finally, we might ask, did China's history of defense industrialization make China more secure? China, of course, was not invaded, but its military fell increasingly behind the technological state of the art as the combination of the break with the Soviets and Maoist attacks on professionals and higher education stymied the creation of an effective managerial system and the production of large numbers of high quality scientists and engineers. The nuclear and missile programs got first call on remaining top scientific and technical talent. There was little left over for other defense industries, much less non-defense industries. There was very little innovation in the other elements of the defense sector and what innovation there was was fairly marginal changes on the basis of pre-existing Soviet designs. Arguably, the nuclear and missile programs and the Sino-American rapprochement of the early 1970s did more for Chinese security than did the construction of the rest of the defense industrialization system during the Third Front period. Moreover, whether the US and the Soviet Union ever seriously considered invading China, which is what China's military strategy was geared towards, is an open question. Certainly in the 1950s and into the 60s, US war plans for China were generally nuclear, not conventional. In that regard, China's defense industrialization was essentially moot. The Soviet Union was thought to have seriously considered an attack on China at 69 to 70, but chose not to. But it is far from clear that their contingency planning, what their contingency planning might have been, and it is doubtful that the Soviets would have stuck around to occupy China and thus be subject to guerrilla war. So with the benefit of hindsight, it is far from clear that much of China's defense industrialization did much to enhance Chinese security. Only in 1999 did defense budgets begin to increase again. Newly trained scientists and engineers educated it close to the state of the art, and Soviet Russian equipment began to really modernize China's defense industries. And with that, I've gone on more than too long. Let me stop and take your questions. Well, thank you very much indeed, David. I think it's extremely interesting presentation that you have with enormous amount of data. There are specific issues that one could raise with you. And before I go to the questions that we already have, let me start off by asking you something much more general. You decided to end the project in the year 1989. And I'm trying to push you here as to why 1989 is significant for the military industrialization of China. Was it political, dividing lines that you see as significant? Or was there something with that military industrialization or industrial development of the Chinese military that marked 1989 as a particularly important year for the project to conclude? I mean, you seem to be suggesting that, in fact, it really gets bit more into the 1990s before the big changes would happen. So there are a couple of reasons why I stopped in 89. One is why the couple are convenient. One is that most of the data that I have sort of disappears in 85 or 1989 so that I have much less material to go on. Another is that I do see 1989 as a real fundamental inflection point where after Tiananmen, you have a rethinking of strategy where defense spending begins to go up again where you begin to purchase Soviet and then Russian weapons systems to modernize, particularly after 91 and the Gulf War, a recognition on the part of the Chinese about how truly backward their military industrial production is and their equipment is. All of these push China in very different directions. And so, and I also thought 40 years was a nice convenient ending point. So that's why I stopped in 89. And, you know, at least in the US, there are huge arguments among people who have the kind of specialized knowledge about the capability of weapon systems about how big a threat the Chinese military is to American interests. I don't have that kind of specialized knowledge. I didn't really want to engage in those kinds of fights about, yes, it's a very modern military threatening the US or no, it's still backward and not a major threat. I mean, that kind of argument, I didn't have a comparative advantage in speaking to and didn't actually find them all that interesting in terms of how we, you know, we judge a J-20 fighter versus a US F-35, you know, I'm in no position to speak to that. And I don't, you know, it all depends on who shoots first and a whole bunch of other contingencies that can't be predicted. So essentially that's why I stopped in 89. Okay. Thank you very much. That's very helpful. Let me just remind everybody that if you would like to raise a question or pose a comment, please, if you have any questions or if you have any questions in the Q&A box, you can do that anonymously, but it would be helpful if you could say who you are for my benefit. I will not reveal your name if you would like anonymity. The first question I am really knowledge to you comes from Fabio Moline in Rome, Italy. Were there moments of confrontation between the political leadership and the Chinese high command and far as industrialization and military policy guidelines, as far as industrialization and military policy guidelines were concerned? You know, this is one of the sort of major questions I'm thinking about. At least in the 40s and 50s, the Defense Industrial Plan actually was drawn up by the Military Affairs Commission. That we have lots of evidence that people like Peng Dehuai, Xu Xiangqian, Nierong Zhang, and others were meeting with Soviet officials talking about Soviet aid in the defense industrialization process. These were incorporated into broader first five-year plan, second five-year plan data, so that it appears that there was not direct confrontation for the first five, six, seven years of industrialization in the PRC. Mao did, of course, intervene at various points in 56, 57 with the 10 great relationships, talks about building fewer conventional defense factories to build nuclear weapons. There's a debate within the military in 6061 about whether to emphasize conventional versus the high-tech or nuclear and missile programs. Mao comes down on the side of Nierong Zhang and some of the others supporting the nuclear and missile program. We do know that by 64 it looks like the planning apparatus has taken over defense industrial planning and Mao intervenes to insist on much more defense industrialization beginning in 6465 to 71. And as a consequence of the purge of Lin Biao in 1971, the defense industrial investment, again, seems to go down. So it plays into it, but it's not clear that this is a party versus army type of issue as a whole, much more sort of an attempt to institutionalize within first the military planning system and then the core planning system and then Maoist interventions at various points that speaks to politicians sort of as they do in China intervening in what are supposed to be more institutionalized processes. And this continues with Deng Xiaoping in both in his, in 1975 calling for rectification in defense industries and then when he's the premier or the number one leader in China holding defense budgets flat, putting the four modernizations with defense industrialization as the lowest priority and using those broad strategic parameters to set things up. So it's not a straight on party versus army in terms of planning here. Can I, can I push you a bit there, David? It may not necessarily be a kind of party versus army competition, but there was a reality there that if you were in command, you would have seen the positions commanding the PLA in the 1950s, early 1950s. You would have noticed that the nascent PLA Air Force was flying state-of-the-art jets, MiG-15s, as good as any of the Soviets were flying and they started having MiG-17s not that long after the Soviet Air Force were flying them. By the 1980s, by your say 1989, you were commanding an army equipped with museum pieces. It's a pretty stark contrast to commanding an army with state-of-the-art weaponry, to commanding a military with museum pieces. And it didn't have to be a matter of army versus military versus party. It was quite simply that the generals and air marshals and the admirals have every reason to be unhappy with whatever government it was that reduced them from the state of the 1950s level of equipment to the 1980s level of equipment. Why was there so little articulation of unhappiness? You know, I think there was unhappiness. It was subterranean in some ways. That it was clear that with Pung De Pi and Lin Biao, that if you were perceived as challenging the chairman, it didn't go well for you. And many of the top military figures, particularly those most associated with modernization and conventional weapons had modernization professionalization, had fallen by the by over the course of the 50s and 60s. So that it was sort of politically fraught to raise this as an issue directly. You know, this has always puzzled me. I mean, these were people who presumably faced death on the battlefront from 27 to 49. And yet we're unwilling to sort of with rare exceptions to sort of stand up to Mao and later on to Deng and say this is an unsustainable sort of position. But at the same time, it's not clear what the answer was. We know, of course, that China tried to piecemeal modernize over the course of the 70s and 80s, buying the Rolls Royce spade jet engines that were supposed to be the basis for a new fighter, but they never could build an airframe around the Rolls Royce spade jet engines. And so whether broader lessons were learned from that, I think there was a profound recognition that you didn't have the scientific and technical capabilities to catch up very rapidly. And so, and as people like Evan Fagenbaum have argued in various writings, that defense planning and indeed much of the planning for the 1980s came from military figures who were protected during the Cultural Revolution who tended to be associated with nuclear weapons and missiles. And they emphasized nuclear weapons and missiles, but they also emphasized building up science and technology and other kinds of things to create the basis for a more sustained catch up later on. And Deng's own experience was that the defense sector was a mess. It needed profound internal rectification. You needed new doctrine. You needed a whole bunch of things before you could begin to upgrade the equipment for the PLA. The prescription was in some ways, the software had to come first before you built the hardware. And so I think there was a lot of politics. It wasn't all that obvious, but for a variety of reasons, you ended up with, again, as I say, software first before you went on to try to modernize hardware. Okay. Next question comes from Philip Mead. During the period, where was the locus of decision making on three bits? A, defense spending. Two, military asset priorities. And third, production centers. I mean, by locus, he's really referring to whether you're talking about civilian versus military, or povins versus central or that kind of dichotomy. Okay. So these are good and difficult questions to answer. Let me, in some ways, start with the last one first, production centers. So we know in 1964 that the general staff, the operations department of the general staff command, did a study of where China's defense enterprises were located. And something like 60% of them were located in the 14 or so PRC cities in 64 that had a population of a million. This was seen by the PLA and picked up presumably by Mao as a profound source of vulnerability. And it was one of the stimuli that led Mao to argue for the third front in the interior. Although it should be noted that Chengdu, Chongqing and Xi'an were among the cities that were, had a population of more than a million in 1964. So there already was material there. And clearly Mao argued for location of defense industries in the interior. He also argued as Taylor Fravel argues, and which I've argued in an earlier piece, much earlier piece than China quarterly, for a little third front within every province. And those it seems like there was a sort of what in the US we call port barrel politics that that provincial leaders who were also chairs of military district commands saw a chance to enhance their resources to their province and would were quickly jumped on the bandwagon to build little third front defense enterprises, which could be used for guerrilla war type of strategy. And began to greatly assemble those. So once you Mao basically sort of interior, and then provincial chiefs then said, well, in our interiors, we too should build little third fronts to enhance their resources. So you did get decentralization down, but the big third front, at least in 64, 6465 was was a centrally planned endeavor. Efforts made to employ geologists and others good at figuring out good locations for things. 6465 into 66 it was a fairly controlled policy, although now as usual, insisting that things move much more quickly. But so there was some control. After 69 things ran amok. We have one source saying that that for the ordinance industry, Lin Biao and his people in the PLA high command tried to increase investment in ordinance to be three times the total invested from 1949 to 1964. So control was lost. It seemed that 69 to 71 with the state council sort of destroyed by the cultural revolution, the party deeply damaged by the cultural revolution that Lin and his followers controlled decisions about spending and about where production would go. And so in terms of military asset priorities, I don't have a clear sense about who was making decisions about how much relative share was going on. Clearly there was mobilization across the defense industries. The, you know, the some of this is based on, of course, output data. These are artificial prices for weapons systems. So we have no sense about whether the prices Chinese used for fighters versus tanks reflected any kind of scarcity values or not. So there's an artificiality to using these output data. In terms of physical quantities, I don't have a lot of data about physical quantities of weapons produced. And so it's difficult to answer. I do have a few sort of things that in 1972, 250 jet fighters were produced, of which 50 crashed in 1972. This could be due to pilot error. It could be due to quality issues. We don't know, but it speaks to one of the fundamental problems of the high tide of defense industrialization. So I really can't give you a good answer about military asset priorities. We do know that for things like the nuclear and missile program, they had first call on scientific and technical manpower and on things they needed. After that, it appears that from 65 to 72 or so, aviation and ordinance came next. Naval was a last priority during that time. In terms of spending, again, much of this was based on the physical planning system that spending followed what physical plans there were and targets there were. So that again, that's arguably a lagging indicator. It doesn't, particularly during upsurges in investment and production. It seems clear to me that spending decisions were influenced by general policy lines, set by both the political leadership and the military leadership, but sort of something that could be caught up with sort of general patterns of economic construction and development. So on spending, I'd say there was a general plan put forward at least in the fifties and into the sixties by the PLA about what should be, should be agreed to spend in terms of new facilities. And then the political leaders could intervene increased decrease in more or less ad hoc ways. And I'm sorry, I can't give you a more sort of precise kind of answer to those questions, but that's what the data seems to indicate to me. Okay. Philip got a second question. I'll come back to that later. In the meantime, I'll raise a question from a student in one of the London institutions. Have you come across evidence of China exporting its own military capabilities in the period you are studying? Yes, there were Chinese military exports from in some ways from the early days to into the 1980s. That certainly the Chinese were providing military training and equipment to the Vietnam in Vietnam from 50 on that, that various there were small exports. There are books on this by Bates Gill and others about Chinese military exports in the 50s and into the 60s. In the cultural revolution in some of the unofficial Mao speeches, Mao meets with defense industry leaders. He gives a speech where he said we should be the armament, the armament factory for the world revolution. He says there are even some weapons where a made in China would not be put on them. Which I read if, you know, particularly in light after 2009-11-2001 as suggestive of Mao thinking that maybe China would proliferate nuclear weapons or missiles. Needless to say, this has not been incorporated into more official collections of Mao's writings either on military works after 49 or other works. Certainly I showed you exports of conventional weapons in the 80s, also the Saudi, the export of the missiles to the Saudis in the 1980s. But there was a sort of friendship cooperation, military sales to or gifts to North Korea and other countries that China had good relations with throughout the period. So there was, you know, in value terms prior to 1980, I would say not all that significant. But it was a part of Chinese defense industrialization. Thank you. Let me return to the second question from Philip. He's asking you about something a bit rather more contemporary, a bit at least after your period. So I hope that you don't mind. How have the PLH learn the military lessons from the U.S. strategic or technical operations in the 1990 Gulf War? Did it change China's industrial defense model of 1949 to 89? Absolutely. You know, Taylor Frevels, active defense goes through this quite extensively. And I would definitely recommend you look at the chapter on the doctrinal changes to take place after the Gulf War. And given the kinds of weapons systems that were deployed, the precision guided materials, this had a, you know, a shocking effect. It's widely reported. It had a shocking effect on the PLA. There's also, of course, the Li Deng Hui trip to the United States and the confrontation with Taiwan 95 to 96, where it became apparent to PLA leaders that, you know, you could threaten Taiwan with missiles, but that was all you had. You didn't have a credible kind of conventional pressuring of Taiwan at that time. All of this called for a fundamental rethink of both doctrine and capabilities. And so for that, Frevel gives you a sense of what the doctrinal changes were. And there were sort of closing down of many sources of information. Closing down of many smaller, smaller defense factories closing down, particularly ordinance factories, buying Russian equipment, both for naval equipment and aviation. And then efforts to call state of the art knowledge through open source acquisition of, not intellectual property, but of what were scientific and technical papers, begin to incorporate them, entice scientists and engineers to come back. There had, of course, been military cooperation with the United States in the 1980s that led to some arms transfers and some cooperation between US military and Chinese PLA that were a forerunner of this. You had a great reduction in the size of the military industrial complex, but more attention paid to what's now the byword, civil military fusion, a fundamental revamping of the kind of machinery that were used in defense factories, moving to, you know, computer-aided design, computer-aided manufacturing. So while, you know, Shenyang and Chengdu remain the major aircraft producers in China, the floor plans have been fundamentally revamped from all accounts so that you did begin to fundamentally change the quality of what was being produced. You know, we don't know, you know, or at least people without clearances don't know other than, you know, what things like the Institute, International Institute of Strategic Studies tells us about more advanced equipment that China is now deploying. We have a good sense about how many ships they have and numbers of aircraft, how effectively they'll be used, we don't know. This is something I'm sure CIA, U.S. intelligence services pay a lot of attention to. They're saying there's been a fundamental upgrade. Rand and others have done studies about the upgrading of aviation in particular. So we have a sense that things have made a fundamental change for the better, but how much, you know, some of these things are only told are answered by actual use in combat and I hope we never know the answer to those questions. Indeed. Could I push you here to clarify it on something which is that you implicitly highlighted, well, you explicitly highlighted how important the Gulf War was in terms of changing the Chinese military thinking, which is absolutely spot on. I agree with you totally. But the implication of that was that before the Gulf War, the PLA didn't really realize how much they were behind the American military. Given that Saddam's army were using a lot of equipment that were in use in China, some of them in fact were Chinese made, and it was utterly devastating for them. Did they really not know how badly behind they were by that point? You know, I think some of them did. But I would argue that you, you know, with the retirement of older countries in the 1980s, you have the phasing out of the people who knew that they had done well, who could, who had the kind of clout, you know, people like Xu Shuiou and others who could go to Dong and say, look at how bad our stuff is. You know, they knew that they, the army did not perform well in the war against Vietnam. And so they, you know, that they phased out the sort of generation of three-star, four-star and marshals who were still alive in the 1950s in the 1980s. And Dong played, I would argue, the patronage game quite well with the younger commanders coming up, may have played them off against each other. They were in some ways beholden to them as you got new leaders of the PLA Air Force, and maybe the ground forces, chiefs of staff, defense ministers and so on. And that defense minister, they kept old timers in for the most part, but they weren't sort of politically confident enough. They didn't have the kinds of connections that would lead them to say to Dong, this can't be maintained. I would still argue that people like Song Jian and Chen Xue-sun tended to still sort of dominate the thinking of civilian leaders about the defense industry. So you built up your missile force, you continue to expand your nuclear weapons production, you try to diversify the kinds of nuclear weapons you were producing. And that remained the priority. And it, you know, I think there is a sort of political explanation for that. But again, Dong could come back and say, so what do you want me to do? We don't have state of the art regular industry. We don't have a large, but a growing scientific and engineering cohort. Once we do, then priority will shift back to national defense. And I think for many, that was a compelling kind of argument that at least one that could kick the can down the road. 1989, the need, the PLA intervening in 1989 sort of made it more possible for the PLA to say, we've got to start doing something about this. And Dong and the more conservative leaders facing a more hostile international environment could make the case. Yes, indeed, we do have to start doing things. Yes, we did. Well, thank you very much, David, for these really interesting and thoughtful discussions with us. We have reached the times we have allocated for the webinar. So I have to draw it to a close. And let me also thank all those who have raised questions to you and engage in this conversation. I look forward to seeing some of you at the webinar next week. So thank you and goodbye. Thank you.