Arthur Kroeber on Dwarkesh Patel: Why BYD Beat Tesla at Home

1. 0:00 – 22:07

   We should reconcile with China

   1. AKArthur Kroeber0:00

      The communist leaders in China have a lot in common with the techno-optimists of Silicon Valley, in their, what I would call, technological fetishism. The number one job of the government is to figure out how to mobilize the resources of Chinese society to maximize technology acquisition. Think of China as like a giant VC fund that is just willing to lose huge amounts of money for a really long time on the assumption that a few of the bets will pan out. The Chinese have seen this coming for a long time. They have long had the view that the US was gonna try and constrain them, and they say, "Well, how do we get around this?" And their answer is not, "You build a block, I'll build a block against that." Their answer is, "I will operate so that it is impossible for you to build the block that you want."

   2. DPDwarkesh Patel0:46

      Today, I'm interviewing Arthur Kroeber, who is the founder of Gavekal Dragonomics, which is a research consultancy focused on China, and author of China's Economy: What Everybody Needs to Know. A friend, while I was in China, recommended it to me, and it's been the most valuable and useful resource that you can get today on how China works. So Arthur, thanks for coming on the podcast and taking the time to chat with me.

   3. AKArthur Kroeber1:11

      It's great to be here. Thanks.

   4. DPDwarkesh Patel1:12

      First question: What really is the problem if China becomes as wealthy, or if its economy grows as big as America's, or grows even bigger? I know that maybe it's not your perspective to be a China hawk, but I've never really understood why this is a problem in the first place.

   5. AKArthur Kroeber1:26

      Uh, yeah, it's a very good question. It's a very good question. Uh, there's a lot of criticism of China from the standpoint of, okay, you're trying to get rich. That's fine, okay. But you're basically trying to get rich on the backs of everyone else in the world by running this gigantic manufacturing export machine, where it seems like the Chinese ambition is to produce all the manufactured goods in the world, uh, for everyone, uh, run an enormous trade surplus, which means that they are depending on other people's buying power to support them. And that this is basically not fair, not sustainable, not a stable way to participate in the global economy. Uh, so that's more of a question, not of if China does get rich, but how does it get rich? Does it get rich by essentially operating on the same rules as everyone else and having a, uh, a market that other people can participate in? Or does it grow rich by essentially, um, making it impossible for anyone else in the world to have the kind of production structure that they wanna have and relying entirely on these ever-growing trade surpluses? In general, in principle, from a welfare standpoint, it would be great if China got to be, you know, as rich as everywhere else in the world, even as rich as America. But that is likely to have some pretty difficult and destabilizing political consequences because of the vast differences between the Chinese political system and the US political system, and the political systems of all the other major industrialized countries.

   6. DPDwarkesh Patel3:01

      Mm-hmm. I mean, the trade surplus point, to the extent that it is made possible by, um, the government involvement in industry, which is actually not even clear to me that that's the case. I mean, just like, if you have high savings and not enough investment domestically, just like the accounting identities are adjusted, you will have a trade surplus.

   7. AKArthur Kroeber3:20

      Right.

   8. DPDwarkesh Patel3:20

      Um, but suppose that's even the case. On paper, it just seems like what is happening, the, the Chinese taxpayer, or the Chinese saver is, like, subsidizing foreign importers.

   9. AKArthur Kroeber3:30

      Right.

   10. DPDwarkesh Patel3:30

       So on paper, it just seems like we're getting a good deal. I'm sure some people are upset about this, uh, specifically people who manufacture outside of China. Um, but it's certainly not, like, something obviously insidious.

   11. AKArthur Kroeber3:43

       Right.

   12. DPDwarkesh Patel3:43

       And so it seems like if it wasn't for this, there'd be some other reason that, you know, China can't grow as wealthy as us. Um, and I'm, I am playing a little bit of a devil's advocate here, but I'd just like... I don't really understand why this is, like, such a big issue that there needs to be a great power competition about it.

   13. AKArthur Kroeber3:58

       I think there are a couple of things beneath that. One is, the systemic difference politically is really important, right? So the US self-identity is, uh, of the leader of democracies around the world, and we are very invested, at least for now, in our democratic system. We'll see how that evolves over the next two or three years. There's some question marks around that. Um, but historically, that's been a really huge part of the US identity. And basically, the morality play of World War II and the Cold War, which has really been the crucible in which US foreign policy has been formed, is that you had an alliance of democracies that fought back, first against fascism, then against communism. They won, and the result was that you created a world where, you know, most major countries were democratic. They operated on market economic systems, and we were heading towards sort of a, a world of convergence of systems, where everyone's essentially playing by the same rules. And China really gets in the way of that narrative because it is an incredibly successful authoritarian system, where the ruling party still calls itself communist. US elites have never, to this day, really gotten comfortable with the idea that the Chinese political system is legitimate, right? And as long as you have that kind of underlying problem, I, I think it's gonna be very difficult to, to erase these difficulties. And then on the economic side, I think it is a problem, the way that China has chosen to organize its economy. Yeah, in a, in a strictly financial sense, if you will, um, the rest of the world comes out ahead in the sense that China's willing to subsidize a lot of low-cost production, so the rest of the world gets a lot of cheap goods, and the overall welfare, uh, of the world improves a lot. I think that's clearly true. But I think if you look at what has happened to the US politically over the last 20 years, there are a lot of flies in that ointment. And the fly is that it is important for large countries to be able to maintain a diversified production structure, uh...... and maintain kind of social cohesion. And if you lose the capacity to run a manufacturing economy that employs large numbers of people, there's a lot of disruptions that come as a result of that. And the purely financialized economy that we had grow up in the United States in the early 2000s as a result of this bargain with China wound up being pretty bad for the social compact. So I think there is a legitimate question about, how do you integrate China's growing power, wealth, industrial might into a world in a way that societies around the world can, can tolerate? Right? And it's, ah, it's a legitimately difficult problem, and I, I think a lot of the problems have to do not with China, but with US domestic policy. So I think a lot of what's going on right now is scapegoating of China as a way of diverting attention from decisions that m- need to be made domestically about income re- redistribution and, and macroeconomic policy and so forth. But there is also, I think, a legitimate question to be made about China now accounts for 20% of the global economy, probably will be larger in future, a third of the global manufacturing economy. There has to be an agreed set of rules about how it inter- interacts with the rest of the world so that everyone feels that they are benefiting, not just in financial terms. And we don't really have that agreement now.

   14. DPDwarkesh Patel7:37

       Mm-hmm. Okay, so two points. On the political system, I wonder if we've learned a bad example from, uh, World War II and the Cold War, which is that the way in which great power conflict, um, culminates is that the other person totally collapses.

   15. AKArthur Kroeber7:56

       Right.

   16. DPDwarkesh Patel7:56

       I think that was actually necessary, obviously, for Hitler and even for, um, you know, the Soviet Union. I think they were evil regimes. I think, uh, China today is an evil regime in a way, but it's just not, not in the same order as Stalin or Hitler. Um, and so the, the end state for any great power competition for America, you know, history is long, right? So there will be-

   17. AKArthur Kroeber8:17

       Yeah, right.

   18. DPDwarkesh Patel8:17

       ... more than just these in the next few decades.

   19. AKArthur Kroeber8:19

       Right.

   20. DPDwarkesh Patel8:19

       Cannot be that if there's a different political system, that it has to collapse the way-

   21. AKArthur Kroeber8:25

       Right.

   22. DPDwarkesh Patel8:25

       ... that the Soviet Union collapsed or that-

   23. AKArthur Kroeber8:26

       Yeah.

   24. DPDwarkesh Patel8:26

       ... Hitler collapsed. And then, then there's a question about regardless of their political system, there's this economic dislocation.

   25. AKArthur Kroeber8:32

       Right.

   26. DPDwarkesh Patel8:32

       I mean, the first thing to note there is, this could be true of any country, so I think people can play these two arguments, whereas it's worth noticing that if Australia was, like, producing everything the world consumed and had an economy the size of America's, these arguments should apply to them as well, and I don't think people have this sense of, "Well, if Australia is producing a bunch of stuff for us, we need to, we need to form a coalition against them and have this adversarial attitude." But suppose, suppose we did. I think, um, there's a question of, okay, how could you prevent this dislocation, and is it Australia's fault? Um, the analogy is breaking down, so let me just go back to China. (laughs)

   27. AKArthur Kroeber9:07

       (laughs) Right.

   28. DPDwarkesh Patel9:07

       Um, so there's low value-add manufacturing, where labor cost is a big fraction of, um, cost, and that's, that kind of stuff was shipped off to China, but if it wasn't for China, there's many other countries in the world that have much lower labor costs than-

   29. AKArthur Kroeber9:21

       Right.

   30. DPDwarkesh Patel9:21

       ... the US. So if it wasn't for China, it'd be in, like, Vietnam-
2. 22:07 – 38:08

   BYD, Tesla, & Chinese EV industry

   1. AKArthur Kroeber22:07

   2. DPDwarkesh Patel22:07

      Do you want to tell the story of BYD and Tesla in 2018 and 2019? Because I think this really illus- illustrates the point or what could be done in reverse-

   3. AKArthur Kroeber22:16

      Right.

   4. DPDwarkesh Patel22:16

      ... very well.

   5. AKArthur Kroeber22:17

      So if you go back to the early 1990s, China recognized that pretty much every other country that had gotten rich had done so in large part by building up an automotive industry, uh, that then served as the, uh, a mechanism for creating innovations in other- other sectors. So you look at the US, you look at Germany, you look at Japan, you look even at Korea, which is very successful. So they said, "Okay, we have to have a big auto industry. This is one of the key industries that we have to support." So starting in the early 1990s, they had this strategy of bringing in foreign companies, making them do joint ventures on a 50/50 basis with Chinese auto companies. Uh, and the theory was that eventually, the Chinese companies would learn and that you would be able to, you know, develop your own national champion auto companies and, you know, kick the foreigners out, right? And basically, this failed. Uh, it failed massively. Uh, over the next 25 years, foreign companies came in, General Motors, VW, Toyota, um, Honda, and so forth. Um, and they did very, very well. And the joint venture partners in China basically just sat around and clipped coupons from the dividends that these companies, that these joint ventures were producing. All of the technological inputs continuously came from the foreign partners, all of the design ideas that the local partners just were never able to, uh, succeed. And in volume terms, you had a lot of smaller-scale, locally, uh, sponsored vehicle companies, uh, in China building kind of low-end vehicles. So a lot of the cars that wound up getting made in China were made by local firms. But if you look at the share of profits-

   6. DPDwarkesh Patel24:10

      Mm-hmm.

   7. AKArthur Kroeber24:10

      ... and total value in the system, it was dominated by these joint ventures, which were in turn dominated by the foreign partners. So China, you know, circa-... 2010, 2015, had been doing this for 25 years and they were basically no closer to having globally competitive conventional car makers than they had been 25 years before. So, um, in the late 2000s, so 2005, 2010, they started thinking, "Well, this and other related kind of industrial policies are not working very well. We are not generating the national champions that we thought that we were gonna do, so what do we do about this?" And their answer was, "Leapfrog. Let's, let's try and figure out, what is the next stage of technological development that people really aren't working on yet? Let's work on that and then maybe we can get in on the ground floor." And so what they came up with, in large measure, was a renewable energy, specifically electric vehicles. So they started having a pretty comprehensive set of subsidies and other kinds of industrial support for companies like BYD to produce electric vehicles. BYD is basically a private company.

   8. DPDwarkesh Patel25:19

      Right.

   9. AKArthur Kroeber25:20

      This worked okay for 10 or 12 years, uh, and BYD got pretty good and, and particularly what they started to figure out was the supply chain, so they got very good at making the batteries. But still, when you got to 2018, 2019, BYD was still not that exciting a company. Most people in China really didn't want to buy electric vehicles. They seemed really inconvenient. Uh, the sexy cars were, like, the big SUVs that the joint ventures were, were putting out and so forth. And then in 2018, uh, the Chinese government made the decision that they would allow Tesla to come in and build a wholly-owned gigafactory in Shanghai. They had never approved a wholly-owned automotive company before. They started cranking out cars in 2019, which became immensely popular, and it seems like one of the things that happened was that the Chinese companies, by 2019, had gotten pretty good at the underlying technology of electric cars, specifically the batteries and then the software systems that governed the cars. But they were terrible at consumer design. They had not figured out how to make these appealing products that people would want to buy, and Tesla did have appealing pro- products that people wanted to buy. They were big status symbols. So BYD and its competitors said, "Okay, this is the part we have to figure out. How do we, how do we achieve this?" And basically, they recognized that they had to, uh, up their design game. Um, they, among other things, went to Germany and loaded up on a lot of German car designers, um, that were able to transform their, kind of, technology base into much more appealing packages. Uh, and then by 2022 or so, they were, you know, able to compete with T- Tesla, um, both on price and on quality. So, you know, I think what that shows you is that, number one, the, the Chinese government made a pretty good bet on this kind of leapfrogging idea that there are technologies of the future, we need to get in early, subsidize the heck out of them. You know, some of the estimates that have been made about the level of subsidies given to the EV industry and, and related supply chains, it's $200 or $300 billion. Huge, huge amounts of money that did not generate that much of a return, financial return for a really long time, but they stuck at it. And then, the thing that finally flipped the switch was this catalytic foreign investment that showed the Chinese companies what they needed to do to actually compete in the consumer market, and then they got very good at that very quickly.

   10. DPDwarkesh Patel28:19

       How much was the $200, $300 billion actually relevant, given the fact that BYD, as you mentioned, is a private company?

   11. AKArthur Kroeber28:25

       Yeah.

   12. DPDwarkesh Patel28:26

       Um, w- was that all ... Like, how much of that was actually necessary or counterfactually important-

   13. AKArthur Kroeber28:32

       Yeah.

   14. DPDwarkesh Patel28:32

       ... to creating this outcome?

   15. AKArthur Kroeber28:33

       You know, I think th- the thing is, it, it's like the old slogan about advertising. Half of it works, but you don't know which half.

   16. DPDwarkesh Patel28:38

       Mm-hmm.

   17. AKArthur Kroeber28:39

       Yeah, did China need to spend that much in subsidies? Probably not, right? But had they spent less, would they have had the same effect? Maybe, maybe not. They, you know, they tried various different types of subsidies. So initially, they were producer subsidies to just, like, get companies to produce more. Then they had a lot of buyer subsidies, so at the individual level, but also the city government level, to promote, you know, things like electric buses. Uh, and so I think all of that played a role, and I think what was important was that the government set a very clear direction and they said, "We want this to happen, and we really want it, and we don't know how long it's gonna take, and we're just gonna keep trying different things until it works, because the goal of creating this electric vehicle sector and the whole supply chain for it, that's what we want at the end."

   18. DPDwarkesh Patel29:34

       And why are they able to identify these kinds of sectors in advance? Because, I don't know, central planning isn't supposed to work, right?

   19. AKArthur Kroeber29:40

       Right. Yeah.

   20. DPDwarkesh Patel29:41

       So ... And there's, there's been many cases of countries which have tried to do this. Germany, um, mis- Germany, Japan, many other countries missed out on the internet because there was a centrally directed effort towards these heavy industries or manufacturing, which actually turned out not to be relevant-

   21. AKArthur Kroeber29:57

       Yes.

   22. DPDwarkesh Patel29:57

       ... in the 21st century.

   23. AKArthur Kroeber29:58

       Yeah. No, it- it's- it's definitely perilous. So if you look at ... China has a list of key industries.

   24. DPDwarkesh Patel30:04

       Right.

   25. AKArthur Kroeber30:04

       It's called the Strategic Emerging Industries. They came up with it, basically, in 2010. It built on previous iterations of industrial policy. Most of it is not really an effort to predict in- with high degree of specificity what comes next. They're mostly pretty obvious things, like-... semiconductors, industrial automation, new materials, all this kind of stuff that I think would be on anyone's list if you were just sort of running a VC fund and saying, "What are the sectors that we think are-"

   26. DPDwarkesh Patel30:37

       But most VC funds are, like losing money. And so like actually, it's like maybe like a couple of firms in the world which are able to make this prediction correctly.

   27. AKArthur Kroeber30:44

       Yeah.

   28. DPDwarkesh Patel30:44

       So it's like a couple firms in the world plus the-

   29. AKArthur Kroeber30:48

       Right.

   30. DPDwarkesh Patel30:48

       ... Ministry of Information Technology in China are the only people who can (laughs) predict the future?
3. 38:08 – 45:26

   Will China have a Japan-style financial crisis?

   1. DPDwarkesh Patel38:08

      I just interviewed Ken Rogoff, so Japan is top-

   2. AKArthur Kroeber38:09

      Yep.

   3. DPDwarkesh Patel38:10

      ... of mind for me. Um, that's really interesting that even though Japan had this export discipline, maybe the reason that the growth didn't continue after the '80s is because you had this, um, was it the convoy system where the-

   4. AKArthur Kroeber38:23

      Yes, right. Yeah.

   5. DPDwarkesh Patel38:24

      ... uh, banks were incentivized to lend to people they knew in these conglomerates, and these conglomerates have been around for decades, even before World War II.

   6. AKArthur Kroeber38:31

      Right.

   7. DPDwarkesh Patel38:32

      Now, a question I have about China is now that you do have these companies which have become national champions-

   8. AKArthur Kroeber38:39

      Right.

   9. DPDwarkesh Patel38:40

      ... and are conglomerates, right, the same company is producing my- a phone and a car and everything in between. W- a- and g- you know, given this sort of like, um, intrinsic nature of maybe authoritarian systems or systems with financial repression and, um, will we again, w- will we return to, or I don't know if return is the right word, will we go to a system that is closer to what Japan had in-

   10. AKArthur Kroeber39:05

       Right.

   11. DPDwarkesh Patel39:05

       ... the '80s where, okay, now you're going to give money to Huawei because Huawei is the pe- H- Huawei are the people, or BYD are the people, um, or will it remain dynamic, especially in the sectors which are coming up over the next few decades?

   12. AKArthur Kroeber39:18

       No, it's, it's a good question. I, I guess I would start by emphasizing the ways in which China is very, very different than Japan.

   13. DPDwarkesh Patel39:26

       Mm-hmm.

   14. AKArthur Kroeber39:26

       Um, the sort of central thing that is really different that lies at the root of a lot of this is that China is an independent geopolitical actor, and Japan was not.

   15. DPDwarkesh Patel39:37

       Mm-hmm.

   16. AKArthur Kroeber39:38

       Japan, at the end of the day, they could rely on the US for security. They were demilitarized. China basically is on its own. They're in a very dangerous neighborhood. They have 14 land neighbors that they share borders with. They have names like North Korea, Russia, Pakistan-

   17. DPDwarkesh Patel39:58

       Yeah.

   18. AKArthur Kroeber39:58

       ... Afghanistan, a lot of dangerous actors, um, several of them have nuclear weapons, North Korea, Russia, Pakistan, and India. So China, just in a very narrow sense, lives in a very dangerous neighborhood, and so they quite legitimately have, uh, I think some pretty significant national security needs, and they also have aspirations to be a great power, and they want to do it on their own. Um, so the incentive to get things right is much more existential in China than it was in Japan. Japan essentially could take the choice to, to sort of say, "Oh, well, we're going to have a stagnant economy, and that's bad and, and whatever, but actually, we'll be fine," right? And for China, it's not fine, right? So there is underlying it, I think there's a spur, um, that increases the chances that the Chinese leaders will do the things that they need to do to keep the system more dynamic.

   19. DPDwarkesh Patel40:55

       Mm-hmm.

   20. AKArthur Kroeber40:55

       It's, it's not a guarantee, but it's a pretty strong incentive.

   21. DPDwarkesh Patel40:58

       Mm-hmm.

   22. AKArthur Kroeber40:59

       But then if you look at some of the specifics, so the, the big difference between the way that China and Japan are economically organized is essentially the relationship between the financial and the corporate sector. So Japan had this system where, uh, banks owned equity in, uh, these giant industrial companies and in the trading companies and vice versa. There were these cross-shareholdings, which meant that it was essentially the entire economy of Japan, the entire corporate and financial sector had one big balance sheet.

   23. DPDwarkesh Patel41:32

       Mm-hmm.

   24. AKArthur Kroeber41:33

       And all of that balance sheet rested by the late 1980s on land values that had just gotten completely detached from any form of reality. There was an investment thesis among investors who were, you know, playing the Japanese stock market in the late '80s that you did not look at the earnings of the Japanese companies. Those were irrelevant. What you looked at was the value of the land that they controlled, so essentially a capital ratio, where you assumed that this, um, uh, the value of their underlying land capital-

   25. DPDwarkesh Patel42:06

       Right.

   26. AKArthur Kroeber42:06

       ... was a permanent contribution, uh, to their wealth and prosperity. And so the valuations of these stocks, which got, again, crazy, which but you couldn't justify them by earnings, and you say, "Well, it doesn't matter because the land that they're sitting on is so valuable." Then the land values collapsed by 80%. The stock values had to collapse by a similar amount, 80%, and everyone was holding shares in one another, so the banks, their capital got eroded.... because a lot of their capital was tied up in the land and in the stocks which were collateralized by land, and so they were unable to lend. They just had to deleverage as fast as possible. The companies had to deleverage as possible, fast as possible. And so the entire economy got engulfed by this phenomenon of d- debt deflation, which basically means that in the act of paying down the debt, 'cause you're doing fire sales of assets, you're reducing the prices of those assets, introducing deflation. So the real value of your debt continues to grow even as you are supposedly reducing it-

   27. DPDwarkesh Patel43:12

       Right.

   28. AKArthur Kroeber43:12

       ... because of the deflation effect. It's very, very difficult to, to get out. China does not have this problem at all because they looked at the problems that Japan got into. They looked at similar problems that Korea had in the late 1990s, and they said, "We will never allow this kind of cross shareholding between financial companies and corporations. This is way too dangerous." So they have kept the two quarantined. Uh, it is illegal for industrial companies to own a bank. Um, they, they're allowed... The bigger ones are allowed to have internal financing subsidiaries, but they can only sort of manage financing flows within the group. Um, and it is also illegal for, uh, uh, for banks to load up on shares of industrial companies, right? So you have a distinct financial system and industrial system, which means that the particular type of macro problem that Japan got into is very, very unlikely to occur, uh, in China today.

   29. DPDwarkesh Patel44:13

       Mm-hmm.

   30. AKArthur Kroeber44:14

       So you have, you have huge debt problems, right? So the property developer's very overleveraged. Huge problem, you've had a massive property crash in China over the last five years. That is a, that is a big, big problem. You have local governments that, you know, borrowed a lot of money, invested in infrastructure which is now delivering very low returns. They have a big debt problem. So you have s- very substantial debt problems in China that have significant negative macro consequences, but they're all isolated, and they can be dealt with in, in, in, uh, sort of one by one. If you look at the industrial sector, it's not very highly leveraged. Debt levels, in fact, are not high. Most private companies in China have learned for many years they couldn't get access to bank credit, um, because the banks only wanted to lend to state-owned enterprises-
4. 45:26 – 58:23

   Local debt situation is manageable

   1. AKArthur Kroeber45:26

   2. DPDwarkesh Patel45:26

      Interesting. I mean, this is just stuff from your book that I'm citing back to you, but you talk about these local government financing vehicles which are backed, so gov- the local government takes out a lot of loans in order to build this infrastructure, and that is backed by the presumed appreciation of the land which should go up as its infrastructure goes up.

   3. AKArthur Kroeber45:47

      Right.

   4. DPDwarkesh Patel45:48

      I mean, that sounds actually very similar to the problem you're describing-

   5. AKArthur Kroeber45:50

      Right.

   6. DPDwarkesh Patel45:51

      ... with Japan, in that, well, you say, look, this is debt that's on the local government's balance sheets.

   7. AKArthur Kroeber45:56

      Right.

   8. DPDwarkesh Patel45:57

      But I mean, fundamentally, I don't know, it's like one country, and especially if you have this system of, you know, the government can just, like, hand the debt to somebody. So how immune is the private economy really from... Like, will, will it just take down the local government, and then nobody will make a fuss, and nob- these other companies will just never have to hear about it again?

   9. AKArthur Kroeber46:18

      I guess my view on this is that the... this local government land-based financing model was actually a pretty smart thing to do initially, and for a decade or so, it worked quite well. Um, local governments were sitting a lot of, uh, on a lot of these land assets. Uh, it was perfectly... The, the land assets originally were undervalued-

   10. DPDwarkesh Patel46:37

       Mm-hmm.

   11. AKArthur Kroeber46:38

       ... quite substantially, and so it made sense for the local governments essentially to try and capitalize some of the future value of-

   12. DPDwarkesh Patel46:48

       Mm-hmm.

   13. AKArthur Kroeber46:48

       ... of these things and use them to finance, uh, infrastructure development. It actually... It was... This was not a, sort of a rogue policy, um, that local governments came up with on their own. It was actually sponsored by the central government and the China Development Bank back in the early 2000s which was grappling with the problem that China, at that point, had a severe shortage of housing and infrastructure of all kinds, including in- urban infrastructure, and the question was like, "We need a lot more of this stuff. How do we build it? How do we finance it?" And they realized, "Oh, actually, as we build this stuff out, the, the values of this land are gonna skyrocket. Uh, so let's, you know, essentially take some financing against that future land price appreciation-

   14. DPDwarkesh Patel47:37

       Yeah.

   15. AKArthur Kroeber47:37

       ... and use that to build the infrastructure that makes everything possible." And if you look at the first few cases of where this was done in the early 2000s, they had these estimates of what the land would ultimately be worth that seemed ridiculous and astronomical at the time. They were, in fact, way too low. They were way, way conservative on what the actual land price appreciation would be. So you had a period of about eight or nine years where essentially the development banks, the state-run development banks, were doing this in a, in a fairly controlled way, um, very, very successfully. And then what happened was that after the 2008 financial crisis, the government said, "Oh, we just, like, have to spend a lot of money to, um, gin up growth because there's been this global economic catastrophe." And so they did a massive cash infusion into the banks and to the state-owned enterprises and said, "Spend, spend, spend, and basically, all infrastructure is good infrastructure. Just, like, build whatever you want."And so what happened at that point was that all of the commercial banks got into this business of lending money to local governments for these infrastructure projects. And their underwriting standards were much less robust than the China development banks had been, and so they... And the local governments just, you know... It was free money to them, so they just, like, took all the money and, and they ran with it. And then that created this kind of Frankenstein's monster. Even so, I think you had quite a lot of that money in the first two years went into, you know... It probably generated pretty good return because China was still underbuilt-

   16. DPDwarkesh Patel49:18

       Mm-hmm.

   17. AKArthur Kroeber49:18

       Relative to what it needed to be. But after that had run for three or four years, it became clear that a lot of this was going into wasteful projects and it didn't deliver good return, and now they've been struggling for the last decade to, for, y- how to unwind that. But I think it's important to un- to understand that the underlying model was actually an okay model. It just wound up being done in a, in a way that was completely un- uncoordinated and excessive and, and led to a lot of... And you didn't have incentive structures that, that enabled local governments to, you know, rein it in when they needed to.

   18. DPDwarkesh Patel49:55

       Yeah.

   19. AKArthur Kroeber49:56

       Right? So I still think that that is fundamentally... That, that is a local government fiscal problem, which ultimately is a central government fiscal problem. If you look at the consolidated balance sheet of the central and local governments, put them all together, uh, China's government debt is still probably less as a share of GDP than US federal government debt, right? The US is over 100%. Uh, China is probably under 100%.

   20. DPDwarkesh Patel50:23

       Oh, so my, my pr- previous guest, Victor Shi, s- I think estimated that the local government debt alone is somewhere between 100 and 150% of GDP.

   21. AKArthur Kroeber50:33

       Yeah.

   22. DPDwarkesh Patel50:33

       And you add it to the, um, the central government debt, and I think he estimated that government debt in China is 200% of GDP.

   23. AKArthur Kroeber50:40

       Yeah. I, I would, yeah, I would respectfully disagree with, with Victor on that. I mean, it's, it's impossible to know for sure, and the, and the problem is that you have the, the problem of formal debt-

   24. DPDwarkesh Patel50:51

       Yeah.

   25. AKArthur Kroeber50:51

       ... um, which is debt issued by, um, you know, government entities, uh, and then you have contingent debt, which is debt issued by corporations that may or may not essentially be fronts or windows for local governments, right?

   26. DPDwarkesh Patel51:07

       Right.

   27. AKArthur Kroeber51:08

       Um, there is also a very significant problem of double counting, um, which is, uh, which is hard to resolve, where, you know, debt gets counted at different levels in the system.

   28. DPDwarkesh Patel51:22

       Mm-hmm.

   29. AKArthur Kroeber51:22

       It's, it's one piece of debt, but-

   30. DPDwarkesh Patel51:24

       Sure.
5. 58:23 – 1:05:53

   If CCP is so competent, why isn’t China richer?

   1. DPDwarkesh Patel58:23

      I think the valence of things you've said so far has been, look, um, they've been remarkably competent even at the things which we, which, um, economists criticize them most for. For the majority of the period of these schemes, they've, they've actually worked out quite well. For example, this, uh, local government financing through land sales on estimated future income and so forth. And then, obviously, we were talking about Chinese industry and how that's been successful in many s- many key sectors. So, I guess the big picture question I have is, if that's true... Look, I mean, they're still at like a fifth of American national income per c- per capita, a third of similar countries in East Asia like Japan, Taiwan-

   2. AKArthur Kroeber59:03

      Right.

   3. DPDwarkesh Patel59:04

      ... uh, and South Korea. So how do we explain the relative poverty on a per capita basis in China? How do we explain obviously bad decisions like zero COVID?

   4. AKArthur Kroeber59:15

      Right.

   5. DPDwarkesh Patel59:15

      Um, yeah, I guess I'm having trouble squaring-

   6. AKArthur Kroeber59:18

      Right, yeah.

   7. DPDwarkesh Patel59:19

      ... the, the circle of like, uh, you know, if, if they're making all these great calls-

   8. AKArthur Kroeber59:22

      Yes.

   9. DPDwarkesh Patel59:23

      ... why, why, why, why isn't China more successful?

   10. AKArthur Kroeber59:25

       They're not successful enough. I thought pro- one of the problems in the world that we're facing today is that they're too successful at too many things, and that-

   11. DPDwarkesh Patel59:31

       Yeah.

   12. AKArthur Kroeber59:31

       ... this is creating difficulties for other people. So, yeah, that, that, that has come up, you know, from time to time, and, and during the period in the n- in '90s and 2000s where they're growing at 10% a year, um, which is the fastest growth rate recorded by any eco- economy-

   13. DPDwarkesh Patel59:45

       (laughs)

   14. AKArthur Kroeber59:45

       ... in the history of mankind, um, and they were doing it just not, not for one or two years, but for 20 years in a row-

   15. DPDwarkesh Patel59:53

       Right.

   16. AKArthur Kroeber59:53

       ... people often say, "Well, there's all this waste, and there's too much state-owned enterprises, and they could grow much faster-"

   17. DPDwarkesh Patel59:58

       (laughs)

   18. AKArthur Kroeber59:58

       "... if they were just more efficient." And it's like, "Really? Could they?"

   19. DPDwarkesh Patel1:00:01

       (laughs)

   20. AKArthur Kroeber1:00:02

       Uh, they're already growing faster than anyone has grown before.

   21. DPDwarkesh Patel1:00:05

       Right.

   22. AKArthur Kroeber1:00:06

       Is it really the case that they could have grown a lot faster than they have? And maybe, but I doubt it. I doubt it. And, and the reason is I think there's what I would call kind of an efficiency fallacy.

   23. DPDwarkesh Patel1:00:20

       Mm-hmm.

   24. AKArthur Kroeber1:00:20

       Um, which is that economists, who remember, have all been basically tr- all global economists, macroeconomists have been trained in the United States, you know, since the end of World War II, and in particularly since, you know, the 1990s, right? And so they view sort of the, the US in its developed state as the norm.... um, rather than as the result of other processes. So it's like the way we operate now is the way we got here.

   25. DPDwarkesh Patel1:00:54

       Mm-hmm.

   26. AKArthur Kroeber1:00:55

       And actually, that's not quite the case, right? The- the reason China was able to grow so fast for so long was that they were able to mobilize these huge swaths of resources, domestic savings, and throw them in a kind of uncoordinated way and a lot of problems, and there was a lot of waste along the way. That was part of the equation. So if they had tried to be maximally efficient and said, "No, no, no. We're not gonna allow banks to lend unless they can prove that this project is definitely gonna have an ROI of X," they would have lended a lot less, um, and they would probably have had a lot less growth-

   27. DPDwarkesh Patel1:01:30

       Mm-hmm.

   28. AKArthur Kroeber1:01:31

       ... um, because they were not willing simply to, um, create that- that wall of money that would- that would create broad-based growth. And they might, in fact, have had substantially more inequality than they did because this WASTEFUL method allowed them to spread out investment in a lot of areas w- which were geographically disadvantaged, which might not have done so well in a purely efficiency-driven system. So I think we need to be careful about assuming that inefficiency is bad. I think inefficiency sometimes, uh, can be, uh, as long as it is a, uh- uh, basically the- the result of a desire to be effective-

   29. DPDwarkesh Patel1:02:12

       Yeah.

   30. AKArthur Kroeber1:02:13

       ... um, that can be sort of a byproduct of the successful growth story.
6. 1:05:53 – 1:34:32

   How China keeps tech under control

   1. DPDwarkesh Patel1:05:53

      I- I was gonna ask, you said, uh, the government has decided to prioritize this high-tech development rather than the growth rate. And you said it a- as if there's a trade-off. But naively, it seems like isn't high tech supposed to lead to growth, right? Like what- what- why- w- what is the-

   2. AKArthur Kroeber1:06:09

      Right.

   3. DPDwarkesh Patel1:06:10

      ... wh- why are we on this Pareto frontier with, uh, uh, uh, t- technological development and growth?

   4. AKArthur Kroeber1:06:15

      Yeah. Well, and here you get to, I would say, a- a conceptual sort of fork in the road. I would say that in many respects, the communist leaders in China have a lot, you know, in common, uh, conceptually with the techno-optimists of Silicon Valley in their what I would call sort of technological fetishism.

   5. DPDwarkesh Patel1:06:33

      Mm-hmm.

   6. AKArthur Kroeber1:06:34

      So let me see if I can, uh, I can break that down. The Chinese leadership throughout the last 40 years, uh, but even going back further, has had this notion that, you know, technological upgrading...... is the key to wealth and power, right? Mm-hmm. And this, th- the fundamental idea here goes back to the 1840s when you had this great Chinese empire that had been sitting around for a thousand years, doing great. They thought they were the best in the world. And then these upstarts from England come and crush them repeatedly, militarily, and impose this, you know, very humiliating political settlement on them through the Opium Wars. And so the basic analysis among the Chinese elites was, "Oh, okay, we fell behind technologically, and so therefore, we must catch up technologically. This is the, the central thing that we need to do." And this was kind of recognized, you know, among some elites, but it ran into a lot of incrusted political pressure in the late imperial Chinese regime. They didn't get very far. Um, after the empire fell in 1911, you had a republican government that basically, if you look back at the things that they were saying and doing, they're pretty similar to what the Chinese Communist Party has been doing, saying and doing. Uh, they just did them somewhat less effectively. They were dealing with civil wars and various other problems. But the, the basic insights remai- remained the same, that we need to catch up with the West technologically. This is the, the fundamental thing. It's just a question of how do we do this? And so what the current regime in China has, has been able to do is to take that basic idea and actualize it. They have been successful at this task of technological upgrading and catch-up and so forth. So this is really central to the conception of Chinese elite, uh, thinking, um, that the, the number one job of the government is to figure out how to mobilize the resources of Chinese society so as to maximize, uh, technology acquisition, creation, and, and upgrading, right? Starting in the 1980s, you know, that was a big part of the philosophy, but the other part of the philosophy is we just have to grow as fast as possible. Uh, and we are fairly agnostic as to where the growth comes from, and we realize that there are many different sources of growth, you know, foreign investment, exports, construction, infrastructure. We're gonna do all of these things, and basically, the incentive structure that we will set up is just to tell local officials, "Your job is to maximize GDP growth," right? And that will make all of these other things happen. If we, if we tell local government officials, "Your job is to maximize technology development," they'll do a lot of dumb stuff- Yeah. ... and waste a lot of money. That's not the right way to do it. Let's just GDP as the KPI, the end, right? Yeah. And that turned out to be incredibly successful. Um, then Xi Jinping comes along in 2012, 2013, and says, "Well, okay, the problem with this is that it's now gone too far. It's led to a lot of corruption, it's led to a lot of inequality, it's le- led to a lot of local governments doing wasteful and duplicative things, and now we need to, like, hunker down and focus a little bit more on the really central task, which is this technology task." So less of growth at all costs. Basically, the implicit idea there is we've got this figured out. Growth will take care of itself. We have a growth machine in China. The resources of the government now need to be focused on specifically the technology question. So everything that he has done, particularly since about 2015, 2016, when you really started to see this come into focus, has been, "Let's worry less about the growth target and more about the technology targets." And particularly after Trump launched his trade war and the Biden administration followed that up with a lot of controls on exports of technology to China, I think the, the, the focus got intense because they were saying, "Oh, the US really wants to prohibit, constrain our technological rise. They kind of have the ability to do so, and so therefore, we have to double down on our efforts to replicate all of these core technologies so that we're not reliant on, on the US or its allies anymore, and we just have to, like, go full tilt on technology at all costs," right? Um, so that was, uh, that was kind of an accelerant of, um, of this tendency. But then there is also, there is kind of a macroeconomic theory. So if you go back and, and read some of the, the statements that they put out in the late 2010s about their vision of the future, basically they said all of these technological developments, by which they meant investments in... They didn't mean Alibaba and Tencent and all of these great internet companies. They meant physical technology, semiconductors, new material, green energy, industrial robots, all this kind of stuff. Essentially, their view is this is gonna be the productivity engine of the future, um, and so growth and, you know, uh, uh, growth in the overall economy and incomes, it's all gonna be driven by these technological investments. Uh, and I think that is essentially what they sincerely believe, right? And I think there are some problems with that, which is that, you know, the vast majority of people in e- in economy do not work in those sectors, right? So you can get very, very good at making semiconductors and industrial robots and, and whatnot, uh, and the people who work in those industries do extremely well. Um, but that is a tiny, tiny fraction of the totality of the population, and how do you have spillovers, um, that, um, enable everyone in the economy, regardless of how remote they are from this, you know, engine room of the high-tech sector, um, to enjoy...... gainsed income, right? And the answer in the United States his- historically has been, we're just going to have this great consumer economy, right? We are essentially going to be demand-driven, and so consumers will tell us what they want, and companies will arise to, you know, make what they want, which is increasingly not stuff, but services, experiences, and so forth. Uh, and we will create, uh, a financial system that produces very good financial rewards for the companies, for the people that produce these kinds of services. And then you generate huge amounts of actually very high-paid, um, you know, employment in the services economy. And I think the- the- the Achilles heel of any sort of industrial policy-driven growth model like China, but you can see variants of this basically throughout East Asia, is they're very materialist. They're convinced that it's only really the physical stuff that matters. So you get really good at making the physical stuff, um, that will somehow magically, uh, spill over into the rest of the economy, and I think the evidence that we have is that that actually doesn't work that much, that most of the demand in a modern industrial economy, the really high-wage, high-income economies comes from intangible services. And so there's a more complicated interaction between this high-tech core and wider economic growth, and I don't see in China how they set up the linkage between all these great high-tech industries that they have and the 90 or so percent of the economy which is doing something else.

   7. DPDwarkesh Patel1:14:27

      I guess I just don't understand. So point taken that maybe in the short run, these other sectors don't generate as much growth as just building or housing or something, but by that same token, they don't take that much capital, uh, in, like, the scheme of a national economy-

   8. AKArthur Kroeber1:14:44

      Right.

   9. DPDwarkesh Patel1:14:44

      ... to sustain. So if China wants to do a $100 billion fund for semiconductors, that shouldn't detract from its abil- I- I just don't understand why the macro growth number has to go down for it to be able to do that 100 billion semiconductor fund or something.

   10. AKArthur Kroeber1:14:58

       Well, it doesn't have to, but the way that they have, in fact, executed this tends to lead in that direction, and so here's why. So at the same time that they have been doubling down on industrial policy, they have also been much more tightly regulating their service sectors. So a few examples. The famous one is that they woke up in November of 2020, realized that Alibaba's financial, uh, subsidiary was about to do this gigantic IPO, the premise of which was that China should build a financial system that resembled that of the United States in 2005, and that this was a really smart idea. And they said, "Oh, actually, that's a really dumb idea, uh, 'cause look what that led to the US. Um, we don't want securitization. We do not want this, like, financial risk, um, generation that- that Ant Financial is trying to talk about. We really need to crack down on these guys." So they did, but then they were also simultaneously concerned with a lot of other things that the internet platforms were up to, and so there was this broad-based, uh, regulation of the internet sector, which weirdly, for about 20 years, had operated in a complete free-for-all land of almost no regulation whatsoever, except for political censorship. But pretty much anything else you wanted to do, you could do it with very little government interference. It was a strange exception to the sort of regulatory norm in China. They finally caught up and said, "No, no, no, we have to regulate all these guys." And so they went around to all of the internet companies and basically told them, "There are things that you can do that are fine, and there are things that you cannot do, and there are a lot of business models that you might be interested in doing, such as fintech, such as telehealth, um, such as kind of wider social media applications, um, that you may not do because we think it creates too much financial risk, we think it creates too much social stability risk, we think it creates too much political risk for us, the Communist Party." And so a lot of the avenues of growth, um, for the, uh, internet companies were shut down. So that- that was... Uh, they still did okay. They're still very big companies. I think their share prices, you know, fell way more than was justified by the actual loss in revenue. That's starting to correct. Uh, Alibaba, Tencent, they're still great companies, they're figuring it out, but they were severely constrained. They were much, much more tightly regulated. Uh, you look at the financial sector generally, it was significantly deregulated from about 2008 to about 2017, with some bad consequences, increased financial risk, shadow banking, lots of bad stuff, but also lots of good stuff. Uh, private companies which were essentially shut out of loan finance got the ability to finance themselves through debt in a way that they had not been able to previously. Um, you had a lot of, uh, new financial services going to households. Um, and basically, the government said, "No, this is a little bit too dangerous, so we're gonna re-regulate it." And- and now most financial activity in China has been constricted back into a small number of basically state-owned banks, so a severe re- re-regulation of the financial sector. Uh, uh, there were a lot, there was a lot of talk about healthcare sector, opening that up to, uh, much more to private entrepreneurship. That, again, has been reined in. So you can go down the list, and there has been a massive re-regulation of the service sector, which limits the opportunities for profitable expansion by companies, limits the opportunities for entrepreneurship, and limits the, uh, opportunities for- for employment.

   11. DPDwarkesh Patel1:18:45

       But- but the examples you mentioned all seem like, um...... that the potential growth was constrained not because it was cutting off against high-tech development in other more physical fields, but because it was a threat to political power-

   12. AKArthur Kroeber1:19:00

       Right.

   13. DPDwarkesh Patel1:19:00

       ... or just perceived social stability. So th- to the extent that growth has gone down because of all these actions, not to the extent, it has gone down because of these actions, and it doesn't seem like it's happened because they needed that to happen in order to, you know, get SMIC up to snuff.

   14. AKArthur Kroeber1:19:17

       Right, correct.

   15. DPDwarkesh Patel1:19:17

       It just seems like it's because-

   16. AKArthur Kroeber1:19:18

       No, it's a political choice.

   17. DPDwarkesh Patel1:19:19

       That's right.

   18. AKArthur Kroeber1:19:19

       It was a political choice.

   19. DPDwarkesh Patel1:19:20

       Yeah.

   20. AKArthur Kroeber1:19:20

       That's right. And- and I think, you know, just, you know, so this is what happens if you say technological development, particularly in hardware, will solve all of our problems. It will generate all this growth and so therefore, we don't need this service sector, right? So we have a real-life experiment in what happens if you do that. And what happens if you do that is you get a lot of progress in the technological sectors and you get basically a- a persistent shortage of aggregate demand, which means that even the people who are making all this great technology stuff don't have any pricing powers. They have to keep lowering their prices, so you get deflation. When you get deflation and less employment growth, people say, at the household level, "Oh, people aren't hiring, so I guess I need to save more and spend less-

   21. DPDwarkesh Patel1:20:10

       Right.

   22. AKArthur Kroeber1:20:10

       ... because my income is not gonna go up." And you create the risk of this deflationary spiral, which they're sort of on the- on the edge of right now. And so I think this is- this is something, a- and again, this gets back to the sort of like the techno-optimist view of things, is that if you, um, if you have this very narrow view of the economy as- as purely the expression of, um, you know, the emanation from this technological core, you're missing a lot of stuff.

   23. DPDwarkesh Patel1:20:42

       Right. Yeah.

   24. AKArthur Kroeber1:20:42

       Right? And you will basically take yourself into a blind alley-

   25. DPDwarkesh Patel1:20:47

       Yeah.

   26. AKArthur Kroeber1:20:48

       ... um, if you think that that is gonna solve all your problems. And the Chinese government in the last two months, I think, has begun to recognize this-

   27. DPDwarkesh Patel1:20:55

       Mm-hmm.

   28. AKArthur Kroeber1:20:55

       ... and say, "Oh, actually, we need to have some kind of a demand strategy." This is very evident in their top-level statements. And the problem is they've been working for 45 years to build this magnificent machine of investment and industry and manufacturing, and they're very, very good at it, and they have basically devoted no time in the last 45 years to figuring out how to create a demand machine.

   29. DPDwarkesh Patel1:21:20

       Mm-hmm.

   30. AKArthur Kroeber1:21:21

       So they're just at the very early stages of figuring out that part of the problem.

Episode duration: 2:27:29