Why China's manufacturing economy is dominating — Arthur Kroeber
What this covers
Arthur Kroeber and Dwarkesh Patel discuss why China has achieved manufacturing and technological dominance, and what the US strategy toward China should be. Kroeber frames China's approach as functioning like a giant VC fund—a state willing to absorb losses on many bets to secure wins in critical sectors—enabled by an export-driven economy that forces genuine technological competition rather than rent-seeking. He traces this orientation back to post-Opium Wars thinking: the conviction across Chinese elites that technological lag caused national humiliation and that catching up is the state's primary task. The conversation moves through concrete cases: how informatization and electrification built general-purpose infrastructure that catalyzed progress across sectors; why Tesla's Shanghai Gigafactory unlocked Chinese EV makers when twenty-five years of joint venture technology transfer had failed; and why export discipline, not careful sector selection, explains China's industrial policy success. A central claim: China prioritizes effectiveness over efficiency—the wasteful mobilization of resources achieves outcomes and reduces inequality in ways a lean, efficiency-obsessed system would not.
The conversation addresses China's weaknesses and geopolitical implications with equal force. Rather than inheriting Japan's debt deflation spiral, China learned that lesson and separated banking from industry. Its real vulnerability is chronic underdeveloped domestic demand—a supply-side machine without a demand machine to match it. On strategy, Kroeber argues that Cold War-style containment is empirically obsolete: China accounts for $600 billion in US corporate investment and is the top trade partner for roughly 140 countries, an order of magnitude of integration the Soviet Union never approached. He contends the US could revitalize manufacturing by inviting Chinese competitors to invest and compete domestically, mirroring how China industrialized. The discussion also turns on contingency: whether China's rise followed design or a "random walk" of adaptation, and whether technology progress like AI emerges from directives or from unplanned, recursive steps driven by other demands.
Kroeber argues that China's manufacturing and technological dominance results from a 'giant VC fund' state model that prioritizes effectiveness over efficiency atop an export-disciplined competitive ecosystem, and that the US should pursue managed coexistence rather than a Cold War-style containment that cannot work given China's deep global integration.
- China's state functions like a VC fund willing to lose money for a long time on a few winning bets, enabled by an export-driven, internationally competitive ecosystem
- China is too big and too integrated (140 countries' top trade partner, $600B+ US investment) for containment or a bloc strategy to succeed
- China's weakness is an over-reliance on supply-side industrial policy and chronic deficient domestic demand, not a Japan-style cross-shareholding financial fragility
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Chinese elite thinking since the Opium Wars of the 1840s has held that falling behind technologically caused national humiliation and that catching up technologically is the central task for wealth and power; this idea—shared by the late imperial reformers, the 1911 republican government, and today's CCP—makes maximizing technology acquisition the number one job of the Chinese government.
“The basic analysis among the Chinese elites was, 'Okay, we fell behind technologically. Therefore we must catch up technologically. This is the central thing that we need to do.'”
Contrary to 1990s Western belief that the internet would be the death knell of authoritarianism, China bet early that the internet could be an instrument both to accelerate technological development and to enhance social control—creating a panopticon in which an online population lets the leadership see what everyone is thinking, solving a problem of governing a vast empire that had bedeviled Chinese administrations for thousands of years—and China was proved correct on both counts.
“everyone was saying the Internet was the death knell of authoritarian regimes... The Chinese government very, very early said, 'No, that is wrong.'... We can use it to... enhance our control of society. They have been proved absolutely correct in that bet on both counts.”
China's industrial policy succeeds where import-substitution regimes fail because of export discipline: in a closed domestic market firms can rig outcomes and buy government favor without upgrading, but in an export-driven economy you cannot rig the global market, so you must compete on raising your technological game rather than merely on price.
“If you have a closed domestic market, you can rig the domestic market and buy off your connections in the government... That is what dooms all import substitution regimes. If you're an export-driven economy, you can't rig the global economy. Ultimately you have to compete.”
China succeeded in industrial policy less because it picked the right industries and more because it was very adept at building the right enabling infrastructure—exemplified by 'informatization' (networking the whole society early, which both sped technology and enhanced social control) and electrification—general-purpose decisions that, once scaled, made many downstream good things possible regardless of sector-by-sector winner-picking.
“China has been successful in its industrial policy less so because they picked the right industries and more because they were very adept at building the right enabling infrastructure.”
For a country of China's scale, what matters is not the efficiency of using resources but the effectiveness of achieving outcomes; the wasteful mobilization of huge domestic savings thrown in an uncoordinated way at many problems is what enabled 10% growth for 20 years and actually reduced inequality by spreading investment to geographically disadvantaged areas a purely efficiency-driven system would have neglected.
“For a country of China’s scale, what matters not is efficiency of using the resources, but effectiveness of achieving outcomes”
China functions like a giant VC fund: as a state it is uniquely able to throw money at an identified universe of potential winners for a very long time without caring about returns, on the assumption that a few bets will pay off and the successes will pay for the failures.
“Think of China as 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.”
If the CCP vanished and were replaced by something that successfully governed China, it would almost certainly share the CCP's core traits—determination to be an independent geopolitical actor with a strong military and a commitment to maximizing technological progress and domestic production of core technologies—because these interests are rooted in China's institutional, cultural, geographic, and geopolitical position, not in this particular regime.
“If that something else were to be successful at governing China, it would almost certainly share many of the characteristics of the Chinese Communist Party. It would have a strong determination for China to be an independent geopolitical actor”
After Chinese EV makers had mastered batteries and software by 2019 but remained poor at appealing consumer design, the 2018 decision to let Tesla build a wholly-owned Gigafactory in Shanghai catalyzed the industry: BYD and competitors learned what consumer appeal required, hired German car designers, and by 2022 could compete with Tesla on both price and quality.
“in 2018, the Chinese government made the decision to allow Tesla to come in and build a wholly-owned Gigafactory in Shanghai. They had never approved a wholly-owned automotive company before.”
China's strategy from the early 1990s of forcing foreign automakers into 50/50 joint ventures to absorb technology and build national champions failed massively: over 25 years foreign partners (GM, VW, Toyota, Honda) supplied all the technology and design while Chinese partners merely clipped dividend coupons, leaving China no closer to globally competitive conventional carmakers.
“The joint venture partners in China basically just sat around and clipped coupons from the dividends that these joint ventures were producing. All of the technological inputs continuously came from the foreign partners”
The root difference between China and Japan is that China is an independent geopolitical actor in a dangerous neighborhood of 14 land neighbors (several nuclear-armed) that must provide its own security, whereas Japan could rely on the US and was demilitarized, making the incentive to keep the system dynamic far more existential for China.
“China is an independent geopolitical actor. Japan was not. Japan, at the end of the day, could rely on the US for security. They were demilitarized.”
If the US is serious about revitalizing its industrial base, it will not happen unless it invites in the world's leading players—including Chinese companies in EVs, green energy, and industrial automation—to invest and compete, mirroring exactly how China itself industrialized by getting leading global companies to invest and learning from them.
“if the US is serious about revitalizing its industrial base—and I think there's a good case for trying to do that selectively—it is not going to happen unless you invite in the world's leading players and have them compete. That is how China industrialized.”
The Cold War 'red telephone' worked because it addressed a two- or three-dimensional problem of nuclear holocaust between two self-contained, highly centralized systems, but AI presents a more-than-three-dimensional problem that is diffuse, decentralized, and unpredictable, so a stable game-theory communication solution like Schelling's for nuclear balance is far harder to construct.
“the specific problem that we were dealing with there were these very self-contained systems that were clearly under highly centralized control by a couple of people. So having that one-to-one line of communication was an effective response... With AI, it's much more diffused, much more decentralized, much more unpredictable”
Beneath China's turbulent surface of construction and high-speed rail lies a deep, slow-changing ocean of ideas and bureaucratic tendencies, such that China's fundamental social attitudes (gender relations, patriarchy, social cohesion) changed less from 1985 to today than US attitudes on race, sexuality, and lifestyle did over the same period—reflecting both the US's intrinsic churning dynamism and China's lower surplus for non-economically-productive activity.
“The changes that have occurred in the United States over the last 40 years are substantially greater than the changes that have occurred in China over that time period in terms of people's fundamental understanding of how we interact together in a society.”
Cross-border investment is an even better communication channel than trade because it forces you to go to the other party's place, put down roots, and learn how things work, so a permission structure for Chinese investment in key US industries would simultaneously help rebuild US manufacturing and enhance mutual understanding.
“investment is even more than trade a very good communication channel because it forces you to go to the other guy's place and put down roots there and figure out how things work and figure out ways to communicate with one another.”
The Chinese Communist Party faces a central, enduring dilemma between maximizing growth and maintaining a high level of control, and in recent years the pendulum has swung much more toward control, which is associated with slowing growth.
“There's always been a trade off between growth and control in China... 'We want as much growth as possible, but we also want to maintain a pretty high level of control.' That pendulum has swung much more towards the control side of the equation the last few years.”
A Cold War framing is wrong empirically because the Soviet Union never accounted for more than about 1% of US trade with near-zero investment flows, whereas China peaked around 17% of US trade and there is $600 billion or more of US corporate investment in China generating sales far larger than US exports, an order of magnitude of integration without precedent in economic history.
“the Soviet Union never accounted for more than about 1% of US trade and investment flows were basically non-existent. China at its peak about 10 years ago accounted for 17% of US trade, about the same amount as Japan did in the late 80s and early 90s.”
China's economic ascent was not the result of a carefully crafted master plan but a random walk of much groping, doing one thing for one reason and another for another, that coalesced through interaction; the amount of intelligent design was actually quite low, with policy credit due mainly for getting a few big directional bets right and adapting frequently when things failed.
“this was not the result of some carefully crafted master plan that they knew all along. It was a lot of groping and a lot of random stuff... The amount of intelligent design behind that was I think actually quite low.”
A Cold War-style containment via an anti-China bloc cannot work because China deliberately operates so that no such bloc can form: more countries (about 140) have China as their number one trade partner than any other country, which was not true of the Soviet Union, so Southeast Asian and Latin American nations dependent on Chinese inputs and goods will refuse to join an anti-China alliance.
“Their answer is, 'I will operate so that it is impossible for you to build a bloc that you want.'”
The US gave a single loan to one solar company, Solyndra, and treated its failure as proof that industrial policy doesn't work, whereas China had the right answer: you must be willing to accept failures and losses of government money along the way to reach the end goal.
“the US government gives one loan to one company, Solyndra. It goes bad and everyone says this is proof that industrial policy does not work. China actually has the right answer to that question. We had the wrong answer to that question.”
China deliberately quarantined its financial and corporate sectors—making it illegal for industrial companies to own banks and for banks to load up on industrial shares—after studying Japan's and Korea's crises, so the cross-shareholding/land-collateral debt-deflation spiral that engulfed Japan in the late 1980s is very unlikely to occur in China.
“They have kept the two quarantined. It is illegal for industrial companies to own a bank... It is also illegal for banks to load up on shares of industrial companies. So you have a distinct financial system and industrial system. That means that the particular type of macro problem that Japan got into is very, very unlikely to occur in China.”
China's fundamental problem is too much supply-side strategy and not enough demand-side; stronger domestic demand would raise corporate profits and pricing power, generate a little inflation to erode the real value of debt and produce cash flows to pay it down—exactly the growth-and-inflation approach that resolved their late-1990s bad-debt problem.
“they have too much of a supply-side strategy, not enough of a demand-side strategy... If they had stronger domestic demand, that would generate more profits for companies... A little bit of inflation would erode the real value of the debt.”
China's pre-Xi incentive structure of telling local officials simply to maximize GDP growth was incredibly successful because it remained agnostic about growth sources and avoided the waste that would result from directing officials to maximize technology development; Xi shifted toward technology targets, accepting slower growth in exchange.
“If they tell local government officials, 'Your job is to maximize technology development, they'll do a lot of dumb stuff and waste a lot of money. That's not the right way to do it. Let's just have GDP as the KPI. The end.' That turned out to be incredibly successful.”
China has spent 45 years building a magnificent machine of investment, industry, and manufacturing at which it is very good, but has devoted essentially no time to building a demand machine, so even as the leadership now recognizes it needs a demand strategy, it is only at the very early stages of figuring that out.
“they've been working for 45 years to build this magnificent machine of investment and industry and manufacturing... But they have basically devoted no time in the last 45 years to figuring out how to create a demand machine. They're just at the very early stages of figuring out that part of the problem.”
Technological fields like AI develop through highly contingent, unplanned steps—gaming demand drove decades of GPU progress that later enabled AI, and social-media posting created the data 'fossil fuel' for AI models—so a 2000-era directive to 'make progress toward AGI' could not have produced what actually emerged.
“People wanted to play video games and so we had decades of progress in making these graphics processing units which contributed to AI development. People were just posting on Reddit, on Twitter and so forth. That fossil fuel of data has been powering these AI models.”
Washington's consensus to block Chinese investment rests on the fact that any modern manufacturing process is also a data-creation machine and that the old division between civilian, military, and a small slice of dual-use technologies has collapsed—now essentially everything is dual-use—making investment patterns riskier when you don't know where data flows into another country's defense base.
“back 30 or 40 years ago, you used to be able to divide the world into technologies which were essentially for civilian use, a few technologies that were military, and then a very small proportion of dual-use technologies... Now basically everything is dual use.”
China's trade surplus may not chiefly result from government industrial involvement; given high savings and insufficient domestic investment, the accounting identities mean a trade surplus follows, and on paper this amounts to Chinese savers subsidizing foreign importers, so importing countries are getting a good deal.
“If you have high savings and not enough investment domestically, the accounting identities are such that you will have a trade surplus.”
There is an information asymmetry disadvantaging the US: China understands America far better than America understands China, evidenced by ~300,000 Chinese students in the US versus a struggle to reach a thousand US students in China, plus mutual consulate closures (Houston/Chengdu) and journalist expulsions—a reversal of Kroeber's longstanding earlier skepticism of the Chinese claim to understand America better.
“You have 300,000 Chinese students in the US until their visas get revoked. We're struggling to get to a thousand US students in China.”
China's emerging strategic bet is electrification: starting from prosaic factory power shortages, it built renewable-heavy generating capacity now more than double that of the US (its renewable capacity alone equals total US capacity), pushing electricity to about 30% of energy consumption, on the logic that abundant cheap domestic renewable power makes it self-sufficient forever and gives it a competitive advantage in manufacturing, AI, and anything electricity feeds into.
“China has generating capacity that is more than double that of the United States. Their renewable generation capacity alone is as big as the entire generation capacity of the United States. Electricity as a whole is now about 30% of China's total energy consumption.”
The US-China relationship has no endpoint, no winning or losing, only managing; the US must accept this is a relationship that goes on forever and do a far better job keeping competition and friction in as contained a zone as possible, because the Chinese already recognize the US is not going anywhere.
“This is a relationship which will go on forever. There's no winning, there's no losing, there's just managing. We need to do a much better job of figuring out how to manage it and keep the conflicts and friction in as contained a zone as we can.”
Large countries need to maintain a diversified production structure to preserve social cohesion; losing the capacity to run a manufacturing economy that employs many people creates major disruptions, and the purely financialized US economy of the early 2000s that resulted from the bargain with China was bad for the social compact.
“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. The purely financialized economy that we had growing 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.”
China adds on average roughly the equivalent of France or the UK's entire power generation capacity every year, an electricity buildout pace that, extrapolated, underpins its potential advantage in power-hungry sectors like AI.
“they add basically a France or UK on average every year.”
Reducing AI catastrophe risk requires tight contact between competing human groups—analogous to how vaccination spreads immunity and to how the Spanish conquered the Americas because conquered peoples couldn't share knowledge of prior conquests—so that if one side's AI attempts something dangerous, the other can learn and do the opposite, making a bifurcated US-China AI world valuable for safety only if accompanied by high-trust communication.
“different human groups should be in tight contact with each other and have the same kind of mechanism that vaccination creates with diseases.”
China's lower per-capita income is largely an arithmetic problem of scale: moving 1.4 billion people to US income levels takes far longer than the generation it took Taiwan (20 million) or South Korea (40-45 million), even though China urbanized at the unprecedented rate of 20 million people a year and built the equivalent of New York plus Philadelphia plus San Francisco every year for 20 years.
“in terms of housing and urban infrastructure, they were building the equivalent of New York plus Philadelphia plus San Francisco every single year for 20 years. No one has ever done anything like this before.”
A fully representative Chinese government could be more reactionary, nationalist, and militaristic than the current regime, because the government spends most of its time tamping down the unfettered nationalist sentiment visible on the Chinese internet rather than amplifying it.
“There is a lot of unfettered nationalist, highly militaristic thinking there that the government spends most of its time keeping in check.”
China's local-government land-based financing model was initially a smart, centrally sponsored policy (via the China Development Bank in the early 2000s) that capitalized future land appreciation to fund needed infrastructure and worked well for about eight or nine years; it only became a Frankenstein's monster after 2008 when commercial banks with weaker underwriting standards flooded local governments with free money for wasteful projects.
“the local government land based financing model was actually a pretty smart thing to do initially. For a decade or so it worked quite well... It was actually sponsored by the central government and the China Development Bank back in the early 2000s.”
While doubling down on industrial policy, China simultaneously re-regulated its service sectors—cracking down on Ant Financial's IPO and internet platforms, re-regulating finance back into state-owned banks, and reining in private healthcare—for reasons of financial, social-stability, and political risk, which shut down avenues of profitable growth, entrepreneurship, and employment and contributed to deficient aggregate demand and deflation.
“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 opportunities for employment.”
China will likely make the same bet with AI that it made with the internet in the 1990s—treating it as a basic enabling technology that is dangerous but ultimately beneficial and controllable—and favor high-speed AI development, even though grassroots successes like DeepSeek (a quant hedge fund effort outside the official plan) cause some disquiet in Beijing.
“My hunch would be that they would make the same bet with AI that they made with the Internet back in the 90s... we'll probably be able to figure out how to control it because our control systems are so good.”
The US dismantled its roughly 100 working-level government dialogues with China (Trump cut all of them, leaving only trade talks), and the loss of channels like CDC staff embedded in China's CDC—a relationship grown after SARS—was catastrophic in early COVID because the US had no way to understand what was happening, plausibly contributing to many deaths and social chaos.
“he inherited a structure where there were approximately 100 working-level dialogues... He got rid of all of them, 100%. He then replaced them with one, which was the trade talks”
No strategy premised on China accepting arbitrary limits on what it can do—such as ceding AI and semiconductors while taking solar, EVs, and batteries—is viable, because China refuses to be pigeonholed and would view such limits as an absurd infringement on sovereignty that the US would never accept for itself.
“I don't think that any strategy premised on the idea that China should accept arbitrary limits on what it can do is viable. Would we accept that? Absolutely not.”
Past results are no guarantee of future returns for political-economic systems: the US's enduring political stability, long one of its core strengths, now seems less assured (Xi has declared himself president for life, untested for China in a long time), so confidence that either country will keep making the right calls into the AI era should be tempered.
“A few years ago, I would have said I had very high confidence that the US will be able to retain the basic political system and structure that it's had for a really long time... I'm much less confident about that.”
The core problem is not whether China becomes as rich as America but how it does so: whether it gets rich by operating under the same rules and offering a market others can participate in, or by making it impossible for others to maintain their own production structures while relying on ever-growing trade surpluses.
“The real question isn’t just if China gets rich, but how it gets rich. Does it get rich by operating under the same rules as everyone else and having a market other people can participate in? Or does it get rich by making it impossible for anyone else in the world to have the kind of production structure that they want to have, and relying entirely on these ever-growing trade surpluses?”
The attraction to extreme narratives about China—total domination or total collapse—is a product of ignorance: the less one knows about a place, the easier it is to take some truth and absolutize it, whereas deeper knowledge reveals a messier reality that is neither extreme, which is itself an argument for deeper engagement so people engage with realities rather than fantasies.
“The attraction that people have towards extreme narratives about China—total domination or total collapse or so forth—is very strong in my view… Again, it's a product of ignorance. The less you know about the realities of place, the more you can construct these fictional stories which take some truth and then absolutize it.”
China's relatively open, cutthroat domestic market—unlike Japan, which kept foreign companies out—created a crucible in which the government's industrial ideas could be tested against international players, so successful bets received validation and could be built upon regardless of whether the government's initial ideas were right.
“you have a crucible in which their ideas can be tested. You have a very competitive domestic market where you have a lot of the big international players doing their own thing as well. So the things that work get some validation and you can keep building on those successes.”
Much of the current adversarial posture is scapegoating of China to divert attention from domestic decisions the US needs to make about income redistribution and macroeconomic policy; the decline in US manufacturing employment share is a straight line driven by technological progress, not by China, Japan, NAFTA, or the Cold War.
“A lot of what's going on right now is scapegoating of China as a way of diverting attention from decisions that need to be made domestically about income redistribution and macroeconomic policy and so forth.”
Although AI's economies of scale would make it advantageous to concentrate China's limited advanced chips in one trainer, the Chinese government cannot force Huawei, ByteDance, Xiaomi, and Alibaba into a single national-monopoly solution, because each has large revenue streams, capital access, and central and local political connections to move independently—national monopolies almost never occur in China, which may lower the probability of China winning AI given its compute scarcity.
“As powerful as the Chinese government is, they do not have the ability to force everyone into one room for this. Huawei's got its own thing going. Bytedance is going to have its own thing going... You will not have a national monopoly solution to this. That almost never occurs in China.”
The Achilles heel of any industrial-policy growth model like China's is its materialism: it assumes getting very good at making physical things will magically spill over to the rest of the economy, but evidence shows it doesn't, because most demand in high-income economies comes from intangible services—and China has not built the linkage between its high-tech core and the 90% of the economy doing something else.
“They're convinced that it's only really the physical stuff that matters. They think if you get really good at making the physical stuff, that will somehow magically spill over into the rest of the economy. The evidence that we have is that that actually doesn't work that much.”
Direct experience of China inoculates against caricatured views: in Kroeber's executive MBA classes, when a few students had spent time in China, discussions were far more nuanced, but in post-COVID years with zero students having visited, the baseline view of China was far more negative, stylized, and cartoonish—showing that without people going back and forth, perceptions are formed entirely by media and preconceptions with no reality check.
“when I had a class that had a few of these people sprinkled in that had that experience... the class discussions were always much more nuanced. People did not retreat into these sort of cartoon stereotypes”
Most metrics of return on investment in China are declining and the number of zombie companies—firms that survive but don't generate enough revenue to cover the interest on their debt—is rising, creating mounting financial pressure from the obsessive supply-side focus that gives China self-interested reasons to adopt a more balanced, demand-oriented economic model.
“If you look at most of the metrics that I and my team follow of return on investment in China, they're going down. The number of zombie companies that still exist but don't even generate enough revenue to pay the interest cost and their debt, that's going up.”
Because Chinese banks historically lent mainly to state-owned enterprises with collateral and state guarantees, private companies learned for years to finance investment out of retained earnings, so the industrial sector is not highly leveraged—debt levels are low and only creeping up recently due to the industrial-policy push—leaving Chinese firms in a very different position from over-leveraged Japanese companies.
“Most private companies in China have learned for many years that they couldn't get access to bank credit. The banks only wanted to lend to state-owned enterprises... Private companies for years have learned to make do by basically financing their investment out of retained earnings. Their leverage ratios are not that high.”
Betting that hardware technology will solve all problems produces persistent shortage of aggregate demand, so even successful tech producers lack pricing power and must cut prices, generating deflation; with weak employment growth, households save more and spend less, risking a deflationary spiral that China is on the edge of now.
“You get basically 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.”
China's total debt across all sectors is on the order of 300% of GDP—in the ballpark of highly developed economies but extremely high for a middle-income country with per-capita income like Brazil's—so while the leverage is a serious constraint on growth and policy, it is unlikely to cause financial collapse because the system is contained and debt is denominated in local currency.
“we do have a pretty good sense from the macro data of how much total debt there is in the economy. It's on the order of 300% of GDP.”
Despite the rise of strong Chinese firms like BYD, Huawei, and CATL, US companies remain incredibly powerful and adept across many technological sectors and often hold strong positions even within China; the notion that the US faces existential commercial threat (e.g., no auto industry in five years) is overstated, and any such outcome would result from bad US policy decisions rather than the competitive threat itself.
“US companies are incredibly powerful and adept across many technological sectors... In many cases, they still have very strong positions in China because they're so much better at what they do than the Chinese competitors.”
If China hawks genuinely believe China cheated the US by absorbing technology when US companies built factories there, then the reverse—Chinese companies building factories in the US and transferring their technology—should be an incredibly compelling opportunity to capture the same arbitrage.
“if you actually believe that, it should just make the reverse of that incredibly compelling. Because you get a chance to do that again.”
If the most economically important part of AI is the specific applications built on top of large language models rather than the models themselves (the substrate), then the US has a fairly strong edge because Western AI operators can deploy across the entire world while China's increasing self-enclosure confines its applications to a Chinese walled garden, making interesting application innovation more likely outside China.
“What really matters are the specific applications that get built on top of that. That's where most of the economic benefit is going to accrue.”
China's growth from 2000 to 2020 was not clearly below potential, but it is now below potential because the government has gone too far into supply-side industrial policy, sacrificing the average Chinese person's catch-up in living standards in order to achieve technology goals.
“They are sacrificing the ability of the average Chinese person to do more rapid catch-up with the living standards of the West, in order to achieve these industrial policy technology goals.”
The current US-China negotiation is unlikely to produce much because the US side has a collection of grievances that have not coalesced into a coherent agenda of what it wants from itself and from China, and you cannot negotiate constructively when you don't know what you yourself want.
“we have on the US side a collection of grievances and complaints that have not coalesced into a very coherent agenda of what we want from ourselves and what we want from China. If you don't know what you yourself want... it's very hard to engage in a constructive negotiation.”
Within the next 5-10 years SMIC will likely be able to produce H100-equivalent chips comparable to TSMC's 5-nanometer process; export controls may still have been net good, but China matching leading-edge chips is something that will happen, after which it becomes a matter of scaling chip production, energy, people, and data—all things China excels at.
“eventually SMIC will be able to produce the kinds of H100 equivalents that the 5 nanometer process a TSMC can produce. I don't know how many years away it is... But this is not something that will not happen within the next 5-10 years.”
The current cycle of US-China escalation could, in retrospect after 50 years, appear as mystifying and senseless as the outbreak of World War I—where motives like Germany's fear of encirclement and contingent accidents are hard to reconstruct—raising the question of why an adversarial relationship was necessary at all.
“I worry that this sort of cycle of escalation will seem in retrospect to people... as mystifying as World War I seems to us now.”
One can simultaneously hold that authoritarianism is morally wrong and has had bad impacts on China's growth and daily life and culture, while also pursuing a productive relationship with the country—avoiding both the hawkish caricature of China as a new Stalin/Hitler and the opposite attractor of concluding that central planning and authoritarianism 'work'.
“you can have a productive relationship with a country and still think that not only is authoritarianism morally wrong, but it actually has had bad impacts on growth or on the day-to-day life of people and their culture”
Estimates of the level of subsidies given to China's EV industry and its supply chains range from $200-300 billion, a huge amount that did not generate much financial return for a long time, yet the government stuck with it because the strategic goal was the sector and supply chain, not near-term returns.
“Some of the estimates that have been made about the level of subsidies given to the EV industry and its supply chains range from $200-300 billion. It was a huge, huge amount of money that did not generate that much of a financial return for a really long time but they stuck at it.”
The speaker offers to tell the story of BYD and Tesla in 2018-2019 to illustrate a point.
“Do you want to tell the story of BYD and Tesla in 2018 and 2019?”