Rush Doshi retweeted
1/2
Caixin: "“It took me a long time to make up my mind, because I’m a deep believer in open trade,” Pascal Lamy told Caixin. But he believes that China’s combination of industrial subsidies, weak domestic consumption and price competitiveness across supply chains has distorted the traditional logic of comparative advantage."
caixinglobal.com/2026-11-07/wee…
Caixin: "“It took me a long time to make up my mind, because I’m a deep believer in open trade,” Pascal Lamy told Caixin. But he believes that China’s combination of industrial subsidies, weak domestic consumption and price competitiveness across supply chains has distorted the traditional logic of comparative advantage."
caixinglobal.com/2026-11-07/wee…
14
76
306
30,310
Paul, appreciate the reply and enjoyed the back-and-forth. You have not contested the central point of this exchange: Beijing is non-compliant with Busan while Washington has complied with it. Indeed, you even admit that Beijing is now linking Busan to issues outside of it, even though this too is a violation of Busan. Notably, to me at least, you finally acknowledge that tariffs may have triggered the 2025 escalation, which is welcome.
A few replies:
First, you've once again appealed to your own authority without clarifying it. As I raised earlier, if that authority is based on your work helping clients obtain MOFCOM licenses, I think it is fair for readers to ask whether your need for MOFCOM access influences your ability to be objective in public discussions about China's policy, Busan, and MOFCOM. It doesn't mean you are wrong, but it does warrant disclosure and clarification when you are citing your own experience as a source of authority, as you do multiple times here.
Second, the entire debate was about whether Beijing was compliant with Busan. You never respond to the arguments I offered on how Beijing was non-compliant with Busan, including the general-license commitment and MOFCOM's own December general licenses. You seem to concede the point Beijing is non-compliant when you write, "And yes, issues not officially part of the Busan process are at clearly at play, this is obvious to those who know where to look." Indeed, it is obvious! If Beijing is now linking Busan to items outside of Busan, it shows they are not compliant with Busan. Thank you.
Third, earlier you claimed that the U.S. licensing backlog might be as existentially threatening for China's economy as the rare earth mineral and magnet throttling was for the U.S. economy. This was always dubious, and you now admit "it turns out that rare earths and magnets are more of a 'chokepoint' technology than semiconductors and SME." Exactly, so this shows you are wrong to justify Beijing's non-compliance with Busan with reference to the licensing issue, which was neither part of Busan nor proportionate.
Fourth, I am glad you have also shifted your position on what caused Beijing's 2025 rare earth escalation. Now you admit that "of course the tariffs may have been a trigger" in Beijing's 2025 rare earth escalation. In your last post, and for over a year, you have suggested Biden's semiconductor export controls in 2022 were what caused Beijing's response to Trump in 2025. This was always a stretch, or perhaps a kind of motivated reasoning since you (and your clients) oppose the export controls. But you have not been able to explain why China's rare earth actions were narrow and targeted under Biden and massively disruptive and sweeping under Trump. The reason, obviously, is indefensible tariff escalation by Washington, which was at the heart of Busan.
Fifth, despite conceding that tariffs were the trigger, you argue this is still all really about 2022, writing, "there is no evidence Beijing intended to play the rare earth card in serious way before October of 2022. None." This is a bold statement, dripping with confidence, but entirely bereft of evidence. Indeed, there is plenty of evidence against it. Let me list just some of it.
(A) You totally ignore that in Trump's first term, China made these threats. Xi even visited a rare earth magnet maker and the People's Daily reported it and noted China's ability to strike back, saying, "don't say we didn't warn you." Again, that was in response to tariffs.
(B) As we all know, China did play the card against Japan in 2010 and then consolidated the industry to strengthen its ability to play the card again.
(C) There are plenty of authoritative and semi-authoritative writings from PRC presses speculating about using rare earths, or even pharmaceuticals, against the United States for political reasons dating back to the Obama and Trump administrations.
(D) Of course, China did not play the card at any point in 2022, 2023, and 2024 the way it did in 2025, which suggests perhaps semiconductor export controls were not a sufficient trigger and you are overindexing on 2022.
You further claim they built the regime under Biden, which is only partly right: (1) they actually began assembling it before Biden, and (2) they didn't need a legal regime to throttle rare earths, so the construction of "the weapon" is less critical than the decision to use it. You claim this was inevitable without acknowledging, again, that Beijing could have used it sooner, did not use it sooner, and was reticent to do so even in 2025. Again, this was about Trump's existentially threatening tariffs.
Sixth, you argue Biden officials ignored the risks of Beijing's retaliation when they pursued export controls and assumed none would be forthcoming. I'm not sure who you talked to, but the Biden administration expected Beijing would retaliate to export controls and chose to manage those responses with diplomacy, unlike the Trump administration, which poured fuel on the growing fire. This is why Beijing's response under Biden was targeted controls on gallium, germanium, graphite and antimony and why, under Trump, it went in a far more sweeping and dramatic direction, threatening the supply of all rare earths to the United States.
A few replies:
First, you've once again appealed to your own authority without clarifying it. As I raised earlier, if that authority is based on your work helping clients obtain MOFCOM licenses, I think it is fair for readers to ask whether your need for MOFCOM access influences your ability to be objective in public discussions about China's policy, Busan, and MOFCOM. It doesn't mean you are wrong, but it does warrant disclosure and clarification when you are citing your own experience as a source of authority, as you do multiple times here.
Second, the entire debate was about whether Beijing was compliant with Busan. You never respond to the arguments I offered on how Beijing was non-compliant with Busan, including the general-license commitment and MOFCOM's own December general licenses. You seem to concede the point Beijing is non-compliant when you write, "And yes, issues not officially part of the Busan process are at clearly at play, this is obvious to those who know where to look." Indeed, it is obvious! If Beijing is now linking Busan to items outside of Busan, it shows they are not compliant with Busan. Thank you.
Third, earlier you claimed that the U.S. licensing backlog might be as existentially threatening for China's economy as the rare earth mineral and magnet throttling was for the U.S. economy. This was always dubious, and you now admit "it turns out that rare earths and magnets are more of a 'chokepoint' technology than semiconductors and SME." Exactly, so this shows you are wrong to justify Beijing's non-compliance with Busan with reference to the licensing issue, which was neither part of Busan nor proportionate.
Fourth, I am glad you have also shifted your position on what caused Beijing's 2025 rare earth escalation. Now you admit that "of course the tariffs may have been a trigger" in Beijing's 2025 rare earth escalation. In your last post, and for over a year, you have suggested Biden's semiconductor export controls in 2022 were what caused Beijing's response to Trump in 2025. This was always a stretch, or perhaps a kind of motivated reasoning since you (and your clients) oppose the export controls. But you have not been able to explain why China's rare earth actions were narrow and targeted under Biden and massively disruptive and sweeping under Trump. The reason, obviously, is indefensible tariff escalation by Washington, which was at the heart of Busan.
Fifth, despite conceding that tariffs were the trigger, you argue this is still all really about 2022, writing, "there is no evidence Beijing intended to play the rare earth card in serious way before October of 2022. None." This is a bold statement, dripping with confidence, but entirely bereft of evidence. Indeed, there is plenty of evidence against it. Let me list just some of it.
(A) You totally ignore that in Trump's first term, China made these threats. Xi even visited a rare earth magnet maker and the People's Daily reported it and noted China's ability to strike back, saying, "don't say we didn't warn you." Again, that was in response to tariffs.
(B) As we all know, China did play the card against Japan in 2010 and then consolidated the industry to strengthen its ability to play the card again.
(C) There are plenty of authoritative and semi-authoritative writings from PRC presses speculating about using rare earths, or even pharmaceuticals, against the United States for political reasons dating back to the Obama and Trump administrations.
(D) Of course, China did not play the card at any point in 2022, 2023, and 2024 the way it did in 2025, which suggests perhaps semiconductor export controls were not a sufficient trigger and you are overindexing on 2022.
You further claim they built the regime under Biden, which is only partly right: (1) they actually began assembling it before Biden, and (2) they didn't need a legal regime to throttle rare earths, so the construction of "the weapon" is less critical than the decision to use it. You claim this was inevitable without acknowledging, again, that Beijing could have used it sooner, did not use it sooner, and was reticent to do so even in 2025. Again, this was about Trump's existentially threatening tariffs.
Sixth, you argue Biden officials ignored the risks of Beijing's retaliation when they pursued export controls and assumed none would be forthcoming. I'm not sure who you talked to, but the Biden administration expected Beijing would retaliate to export controls and chose to manage those responses with diplomacy, unlike the Trump administration, which poured fuel on the growing fire. This is why Beijing's response under Biden was targeted controls on gallium, germanium, graphite and antimony and why, under Trump, it went in a far more sweeping and dramatic direction, threatening the supply of all rare earths to the United States.
As always, appreciate the repartee. Everything I noted is accurate, and reflects the reality on the ground, rather than an assessment of what should be happening or is not happening. The situation is much more complex than described here, as Beijing views all of the issues as connected, and this explains why things are currently playing out the way they are. I left out a lot of detail here that is required to understand the full picture.
Also, the Biden administration was warned early on about exactly how Beijing would retaliate to an escalation of the type that occurred in October 2022, and chose to ignore it. At the time the line I heard frequently was “anything China tries to do to retaliate for the export controls will hurt China more than the US. Wow was that wrong as you admit with the gun analogy, which is still inaccurate of course. It turns out that rare earths and magnets are more of a “chokepoint” technology than semiconductors and SME. But there is no evidence Beijing intended to play the rare earth card in serious way before October of 2022. None.
And of course the tariffs may have been a trigger but they were clearly not the reason Beijing had the licensing regime in the works. If a Harris administration had rolled out more export controls, as it almost certainly would have, Beijing would have responded in just this way. There is no conceivable world where this would not have happened. All the tariffs did was move up the timeline. Any serious student of Beijing, its redlines, how the Chinese systems works, and the underlying motivations understands this.
The challenge now is avoiding any further “escalate to de-escalate” actions, and having a realistic and deeper understanding of why Beijing seeks to maintain leverage, while allowing most of the flow of materials to happen for clearly civilian end uses. And yes, issues not officially part of the Busan process are at clearly at play, this is obvious to those who know where to look.
0
1
10
1,413
Rush Doshi retweeted
@RushDoshi Re: "intolerable levels of corruption and compromise in China policy." No one should be surprised if Trump's approach to China is like his approach to most other countries: focused more on his financial opportunities than on US national security.
1
3
19
4,319
Rush Doshi retweeted
This is insane. Huge national security threat. WTAF?
The story keeps getting crazier. To recap everything that has been reported so far:
- The US intelligence community discovered that Jensen Huang, the CEO of the largest company in the world, was sleeping with Alice Huang, a Chinese spy and an executive at Megaspeed. (FT)
- Multiple independent investigations have concluded that Megaspeed is likely a Chinese cutout, which helps Chinese companies access AI chips or circumvent export controls. (NYT and Bloomberg)
- The US intelligence community also concluded that Alice Huang’s goal was to convince Jensen to lobby the U.S. government to loosen export controls on advanced AI chips. (NYT)
- The intelligence community relayed this information to “a small group of top US government officials” earlier this year. (FT)
- Over the last year, Jensen Huang has repeatedly lobbied the U.S. government to loosen export controls on advanced chips. @BISgov has done so, not only by permitting the export of H200 chips to China, but also not enforcing export controls that would have required a license for exports of the most advanced Blackwell chips to Megaspeed and Southeast Asia.
- Although @CommerceGov leadership was presumably briefed on this information by the U.S. intelligence community, @BISgov has taken no actions to tighten controls on Megaspeed or Southeast Asia.
- Megaspeed remains a preferred cloud partner of Nvidia, and continues to receive Nvidia Blackwell chips despite employing a Chinese spy who reportedly is romantically engaged with the CEO of Nvidia.
- Tens of billions of dollars worth of AI chips have been exported to Southeast Asia this year alone, and Malaysia and Singapore are now two of the four largest importers of AI chips in the world.
What is going on here??? We need answers. The question of who gets the most advanced AI chips in the world is literally the defining strategic question of our time. And not only does it seem like the MSS may be directly influencing the debate on export controls in the United States, but also @CommerceGov is refusing to listen to the warnings of the IC on this issue. What is a POSSIBLE good explanation for why Megaspeed is still receiving chips? Why is there not a license requirement for AI chip shipments to all of Southeast Asia? This is beyond strange at this point.
The fallout from this story will be massive. Obviously, the motivations underpinning Nvidia’s policy views opposing export controls must now all be questioned. But also, how did @BISgov sit on this information and has still taken no actions? We need answers.
258
1,157
4,542
801,879
Good time to keep in mind this important research from @SamBresnick and the CSET team.
1/3 @nvidia spokespeople are (again!) claiming (falsely!) that its products are not aiding China's military modernization. We @CSETGeorgetown have a bunch of evidence that, in fact, the PLA wants to buy a range of Nvidia chips, including those that are export-controlled.
0
7
27
2,646
Paul, thanks for the reply here. I often enjoy our exchanges, online and in-person, so please see this in that spirit.
I do think much of what you have written is misleading, incomplete, or in some cases, inaccurate.
To begin, a general point. You dismiss others and appeal to your own authority by stating you are "in the trenches" on this issue. By that, do you mean you're actively helping clients secure licenses from MOFCOM, as you've mentioned elsewhere? If that's the case, I do wonder whether your analysis essentially defending MOFCOM here is at all shaped (even implicitly) by your need for continued MOFCOM access. Doesn't mean you are wrong, but since you're the one who raised the point, it warrants clarification.
On the substance, a few points:
(1) You claim, as always, that the Chinese decision to play the rare earths card wasn't about Trump but was about Biden. But a few replies.
(A) It's true China built the licensing regime over time beginning in 2020 (beginning under Trump), but it's also true they did not pull the trigger until Trump's liberation day tariffs in 2025. In fact, their April actions came two days after those tariffs, and they were in the same package as the 34% retaliatory tariff. This was about tariffs, and later, the affiliates rule.
(B) Under Biden, China never threatened to cut off rare earths to the entire US or global economy. They did so *only* under Trump because he took tariffs to 145% and they were backed into corner. Even now, Beijing officials indicate it was a hard decision they made in the moment. By the way, many admit it was modeled off our sanctions and not export controls because it targeted something so fundamental.
(C) It is true China took some rare earth actions in response to semiconductor controls under Biden. But you know well there is absolutely no comparison between the limited controls they took under Biden and the massive, globally disruptive ones they announced under Trump. The scale of licensing slowdown is wholly different, too.
(2) I suspect the reason you keep saying China's rare earths actions in 2025 were about Biden and not Trump is because you want to undercut the case for US export controls on semiconductors. It seems like motivated reasoning since you and your clients oppose those controls. Separately, your claim that the controls don't work, which was empirically questionable earlier, is particularly thin today, when we've learned that the Chinese government ran an espionage operation against Nvidia's CEO to undercut that same policy.
(3) Your framing treats the PRC's rare earth leverage as a reaction to others. Let's retire this idea. Jessica's piece, in my view and as I've argued elsewhere, is incomplete here. It is true that Beijing didn't plan weaponization in the 1980s and 1990s, but they then did so in the late 2000s ahead of its deployment against Japan. They then used it in 2010 against Japan. They even say so in primary sources. The idea that the formal regime was built beginning in 2020 doesn't detract from the idea they were already applying this tool earlier.
(4) Your oddly spirited defense of China’s non-compliance at Busan rests on two claims: that “we still do not know what China agreed to do,” and that Beijing only had to issue licenses under its own regulations, meaning Busan just required it to hold off on the October controls but allowed it to keep going on the April regime. Both are questionable responses. The basic bargain was clear: China suspended its October controls in exchange for lower US tariffs and suspension of the Affiliates Rule. Interestingly, the US fact sheet also said China agreed to issue general licenses for rare earths and other critical minerals to US end users and their suppliers worldwide, and Beijing didn't deny that. You say the current case-by-case throttling is just fine, but it's contrary to a general license which would mean blanket authorization. Notably, Beijing's actions disconfirm your interpretation. Weeks after Busan, as you well know, MOFCOM began issuing a new category of general licenses to magnet makers, replacing case-by-case approvals. Why create a new licensing channel if the deal only meant returning to the post-April status quo? It's a bizarre reading to say China's April actions were totally off the table: it makes sense only if you now are trying to justify Beijing's non-compliance.
(5) Nor has Beijing delivered on October. MOFCOM’s suspension of its October controls expires November 10. The US says the truce runs to January 10, but Beijing has published no extension to January 10, and we're coming up on the deadline in two days. If China considers itself bound, it should say so in writing, right? The fact it hasn't is interesting.
(6) You bring up military end use, but you know (from your own clients) as well as I do that Beijing isn't just throttling military end users. You then say it's a pipe dream to go back to pre-April 2025. Perhaps, but that isn't really the point. The point is that China isn’t delivering what it agreed to at Busan.
(7) I did say “US companies are not getting licenses.” But if you read the context, you'd know I didn't mean an absolute ban was in place but rather that some companies are struggling to get them, which is why export volumes are down (you and I are in fact cite the same figures here). I think you cherrypicked for a cheap shot reply and didn't look at the whole context.
(8) It is odd that you claim that the US backlog of licenses justifies China violating the Busan agreement. That backlog was not part of Busan. It's quite a reach to justify Beijing's non-fulfillment of Busan by pointing to a third issue that was longstanding.
(9) You disputed my claim that China's Busan non-compliance poses serious risks to the US economy while the US backlog on licenses (which is not relevant for Busan) is not existential for China's economy. You dispute it with two points, both which are dubious.
(A) The first thing you do is say that the licensing backlog may in fact be existential for Chinese companies. You do not prove this, and even if it were true for some companies, the backlog does not directly target a massive segment of the entire economy like manufacturing. Put differently, the United States has no licensing block on something as elemental as rare earths. And as for your points on chips, the United States has in fact licensed billions in chip exports to China. None of this should matter though, because the licensing backlog at Commerce was not part of Busan.
(B) The second claim you make, after saying China faces big threats from our licensing backlog, is to say the US *does not* face a major economic disruption from China's rare earth controls. You argue rare earths access is not a gun to the head of the US economy, and instead of showing why that is true, you simply say it is too complicated to explain and we should take your word for it. This is a bizarre claim: no one else, not Beijing, not Washington, and not industry would agree with you. Moreover, Beijing understands this is serious leverage and that’s why it used it. Just consider that several high-employment manufacturing sectors, autos above all, faced shutdowns last year. You say the US economy is mostly services, so who cares? That's quite an argument! Even a services economy can’t absorb mass plant closures from a shortage of rare earths, and manufacturing isn't a trivial sector (probably at least 10% of GDP). Indeed, the fact this mattered is the very reason there was a Busan deal in the first place.
In sum, I don't find much of what you wrote accurate or compelling. And I do think the proximity to the issue you enjoy, far from being validating, raises questions.
I'll end there.
I do think much of what you have written is misleading, incomplete, or in some cases, inaccurate.
To begin, a general point. You dismiss others and appeal to your own authority by stating you are "in the trenches" on this issue. By that, do you mean you're actively helping clients secure licenses from MOFCOM, as you've mentioned elsewhere? If that's the case, I do wonder whether your analysis essentially defending MOFCOM here is at all shaped (even implicitly) by your need for continued MOFCOM access. Doesn't mean you are wrong, but since you're the one who raised the point, it warrants clarification.
On the substance, a few points:
(1) You claim, as always, that the Chinese decision to play the rare earths card wasn't about Trump but was about Biden. But a few replies.
(A) It's true China built the licensing regime over time beginning in 2020 (beginning under Trump), but it's also true they did not pull the trigger until Trump's liberation day tariffs in 2025. In fact, their April actions came two days after those tariffs, and they were in the same package as the 34% retaliatory tariff. This was about tariffs, and later, the affiliates rule.
(B) Under Biden, China never threatened to cut off rare earths to the entire US or global economy. They did so *only* under Trump because he took tariffs to 145% and they were backed into corner. Even now, Beijing officials indicate it was a hard decision they made in the moment. By the way, many admit it was modeled off our sanctions and not export controls because it targeted something so fundamental.
(C) It is true China took some rare earth actions in response to semiconductor controls under Biden. But you know well there is absolutely no comparison between the limited controls they took under Biden and the massive, globally disruptive ones they announced under Trump. The scale of licensing slowdown is wholly different, too.
(2) I suspect the reason you keep saying China's rare earths actions in 2025 were about Biden and not Trump is because you want to undercut the case for US export controls on semiconductors. It seems like motivated reasoning since you and your clients oppose those controls. Separately, your claim that the controls don't work, which was empirically questionable earlier, is particularly thin today, when we've learned that the Chinese government ran an espionage operation against Nvidia's CEO to undercut that same policy.
(3) Your framing treats the PRC's rare earth leverage as a reaction to others. Let's retire this idea. Jessica's piece, in my view and as I've argued elsewhere, is incomplete here. It is true that Beijing didn't plan weaponization in the 1980s and 1990s, but they then did so in the late 2000s ahead of its deployment against Japan. They then used it in 2010 against Japan. They even say so in primary sources. The idea that the formal regime was built beginning in 2020 doesn't detract from the idea they were already applying this tool earlier.
(4) Your oddly spirited defense of China’s non-compliance at Busan rests on two claims: that “we still do not know what China agreed to do,” and that Beijing only had to issue licenses under its own regulations, meaning Busan just required it to hold off on the October controls but allowed it to keep going on the April regime. Both are questionable responses. The basic bargain was clear: China suspended its October controls in exchange for lower US tariffs and suspension of the Affiliates Rule. Interestingly, the US fact sheet also said China agreed to issue general licenses for rare earths and other critical minerals to US end users and their suppliers worldwide, and Beijing didn't deny that. You say the current case-by-case throttling is just fine, but it's contrary to a general license which would mean blanket authorization. Notably, Beijing's actions disconfirm your interpretation. Weeks after Busan, as you well know, MOFCOM began issuing a new category of general licenses to magnet makers, replacing case-by-case approvals. Why create a new licensing channel if the deal only meant returning to the post-April status quo? It's a bizarre reading to say China's April actions were totally off the table: it makes sense only if you now are trying to justify Beijing's non-compliance.
(5) Nor has Beijing delivered on October. MOFCOM’s suspension of its October controls expires November 10. The US says the truce runs to January 10, but Beijing has published no extension to January 10, and we're coming up on the deadline in two days. If China considers itself bound, it should say so in writing, right? The fact it hasn't is interesting.
(6) You bring up military end use, but you know (from your own clients) as well as I do that Beijing isn't just throttling military end users. You then say it's a pipe dream to go back to pre-April 2025. Perhaps, but that isn't really the point. The point is that China isn’t delivering what it agreed to at Busan.
(7) I did say “US companies are not getting licenses.” But if you read the context, you'd know I didn't mean an absolute ban was in place but rather that some companies are struggling to get them, which is why export volumes are down (you and I are in fact cite the same figures here). I think you cherrypicked for a cheap shot reply and didn't look at the whole context.
(8) It is odd that you claim that the US backlog of licenses justifies China violating the Busan agreement. That backlog was not part of Busan. It's quite a reach to justify Beijing's non-fulfillment of Busan by pointing to a third issue that was longstanding.
(9) You disputed my claim that China's Busan non-compliance poses serious risks to the US economy while the US backlog on licenses (which is not relevant for Busan) is not existential for China's economy. You dispute it with two points, both which are dubious.
(A) The first thing you do is say that the licensing backlog may in fact be existential for Chinese companies. You do not prove this, and even if it were true for some companies, the backlog does not directly target a massive segment of the entire economy like manufacturing. Put differently, the United States has no licensing block on something as elemental as rare earths. And as for your points on chips, the United States has in fact licensed billions in chip exports to China. None of this should matter though, because the licensing backlog at Commerce was not part of Busan.
(B) The second claim you make, after saying China faces big threats from our licensing backlog, is to say the US *does not* face a major economic disruption from China's rare earth controls. You argue rare earths access is not a gun to the head of the US economy, and instead of showing why that is true, you simply say it is too complicated to explain and we should take your word for it. This is a bizarre claim: no one else, not Beijing, not Washington, and not industry would agree with you. Moreover, Beijing understands this is serious leverage and that’s why it used it. Just consider that several high-employment manufacturing sectors, autos above all, faced shutdowns last year. You say the US economy is mostly services, so who cares? That's quite an argument! Even a services economy can’t absorb mass plant closures from a shortage of rare earths, and manufacturing isn't a trivial sector (probably at least 10% of GDP). Indeed, the fact this mattered is the very reason there was a Busan deal in the first place.
In sum, I don't find much of what you wrote accurate or compelling. And I do think the proximity to the issue you enjoy, far from being validating, raises questions.
I'll end there.
Yes, lets indeed worth correcting the scene here, as this is too important issue in US China relations to be mischaracterized. The perspective here is from someone in the trenches on the issues on a day to day basis, rather than from people who are assessing at a healthy distance, and may not appreciate the nuance of the situation.
First. The Chinese export control licensing regime announced in April 2025 was not a response to the tariffs. As Jessice Chen Weiss and Gloria Xiong lay out brilliantly in the FA piece, the critical minerals controls started in the Biden administration, and really only in earnest in response to the October 2022 US export controls. foreignaffairs.com/china/how-chin…
For people who talk to Chinese officials regularly, the answer was even more clear that the rare earths were a direct, and of course delayed, response to the complexity and sweeping nature of the US controls in 2022, 2023, and 2024. It took MOFCOM and other Chinese ministries time to gain control of a fractious and complex industry of suppliers and traders.
One Chinese official commenting in 2026 noted that China’s addition licensing requirements rolled out in October 2025, were the key piece in “China’s full response to the October 2022 US controls.” So anyone trying to still argue that the Trump tariffs were the cause of Chinese rare earths regime is simply engaging in wishful thinking. The facts on the ground tell another story. In any case, no country could have even an incomplete licensing regime in place after just a few days, so the planning and preparation, including controls on gallium, germanium, graphite, indium, and other critical minerals well predates April 2025.
Second, “China was supposed to let rare earths flow, and they have not” is a fairly wildly inaccurate characterization of the reality on the ground. We still do not know exactly what China agreed to do, there was not statement from the Chinese side specific to rare earth licensing. What we know, again, from first had experience and not unsubstantiated conjecture, is that Beijing agreed to issue licenses, as the US does on controlled items, according to China’s export controls regulations. These regulations state that exports of rare earths and magnets are prohibited for military and defense industrial end use. So, China was not “supposed to” do anything, other than abide by its own regulations. The idea that the situation would go back to what held prior to April 2025 is a pipedream, and the fact that people continue to assert this shows how out of tough with reality they remain.
“US companies are not getting licenses.” This may be the most patently inaccurate of assertions here. Even USTR Greer and Treasury Secretary Bessent acknowledge that roughly 2/3 of licenses are being issued. I know personally of dozens of licenses for critical components and materials issued by MOFCOM just over the past month. MOFCOM has not issued guidance formally extending the suspension of its October 2025 rare earth controls. Until last Thursday, the Commerce Department had not issued a formal Federal Register notice of the extension of its suspension of the Affiliates Rule. Important context.
Third, the reason many people, not just Kyle, in industry rightly raise the issue of the massive backlog of BIS licenses that have not been issued to US companies, is that it seems rather strange to be criticizing MOFCOM for its licensing process, which is just as much, or more complex than that of BIS, and ignoring the licensing process in Washington. Of course these are not part of the formal process, but the are part of what every company impacted by licensing delays experiences, and some US companies are on both ends of these delays. Hence from an industry point of view, the US government should be raising these licensing issues in the context of asking about China’s licensing process. More important context.
Fourth, how does a former US government official know that “the licensing queue…is not existential for Chinese companies? Talk to any of them recently? The licencing queue, coupled with other existing controls, which many believe have produced more damage to US companies than any national security gain, are in fact existential to Chinese companies, and further controls, such as the MATCH Act, are clearly so. But understanding the linkages here is critical, and it is this broader context that matters here.
“But rare earths access is a gun to the head of the entire US economy.” Wow, quite the hyperbole. Understanding how wrong this is would be complex and take some time to explain, and also requires understanding of multiple complex supply chains.
First of all, a licensing regime is just that, a licensing regime. Were the US controls in October 2022 “a gun to the head of the entire Chinese economy?” One could make an argument that they were, even more so than with rare earths and the US economy. China responded with the rare earth controls, suggesting that the US controls were pretty serious.
“The entire economy”” Well, first of all, as noted, the vast majority of licenses are being issued, things are not as easy as they were before the licensing regime, but licenses are being issued, so no, not the “entire economy.” Hint, much of the US economy is services, not advanced manufacturing. What would be more accurate is to say that there are no licenses being given for rare earths and magnets for military and defense industrial end use. Just as the US is controlling semiconductors with potential military end uses. This is what great powers do.
So what does this all mean?
Yes, without real understanding in Washington that Beijing can and will respond to any escalation, and wants to see the Busan truce extended until the end of the Trump administration, we are in a world where the relationship will not be very stable, strategic, let alone constructive. But lets be clear about the causes of the situation, once you start weaponizing supply chains without a clear understanding of the potential consequences, you have to live with the consequences. Here we are.
Note: implementation of the Affiliates Rule and China's October 2025 controls and rare earths and magnets would be catastrophic for global supply chains. Here is the real gun held to the head of the global economy. So why don't both sides rescind the rules completely? That would show real strategic thinking. End Thread.
4
17
71
13,624
Rush Doshi retweeted
Commerce approved a company called New Matrix this year to assemble Nvidia's H200 chips into AI servers. But when the Times visited the site of its listed office, it was mysteriously vacant. People there had no records of this company.
New: Trump administration cleared Nvidia H200 chips to Chinese firms with empty offices and military ties nytimes.com/2026/11/07/bus…
11
157
392
32,340
Investigations seem likely, per the next Chair of SFRC.
The more I learn about the insane relationship between the Trump Administration, Chinese spies, and AI billionaires, the more I’m convinced there is nothing this administration does that isn’t corrupt. Congress will investigate and get answers.
3
15
93
9,376
Rush Doshi retweeted
The more I learn about the insane relationship between the Trump Administration, Chinese spies, and AI billionaires, the more I’m convinced there is nothing this administration does that isn’t corrupt. Congress will investigate and get answers.
2,067
2,254
10,305
330,191
More calls for an investigation into possible corruption at the heart of US China and AI policy, this time from @JerusalemDemsas
There is no better avenue for incoming House & Senate Dems to take than hearings on wtf is going on with NVIDIA + Trump corruption + AI policy in the U.S.
It has the virtue of riding the wave of anti-tech populism and directing that energy towards the productive need to stop selling these chips to China.
1
2
28
2,694
If the reporting is right, this is the most significant story in US China policy in years.
If the USG knew Jensen Huang had a romantic relationship with a Chinese spy and looked the other way, all while that spy ran a company backdooring US compute to China, it warrants a Congressional investigation.
There may be an innocent explanation to all this.
But if not, then it reveals intolerable levels of corruption and compromise in China policy - to the detriment of AI safety.
If the USG knew Jensen Huang had a romantic relationship with a Chinese spy and looked the other way, all while that spy ran a company backdooring US compute to China, it warrants a Congressional investigation.
There may be an innocent explanation to all this.
But if not, then it reveals intolerable levels of corruption and compromise in China policy - to the detriment of AI safety.
22
151
554
50,696
To quote Xi Jinping, "the world is experiencing great copes unseen in a century."
Everyone saying “Oh Jensen got got. Jensen fell for the honeypot”
Ever consider he knew the whole time? That actually HE was honeydicking China and feeding them false intel?
It’s a new Cold War. There’s levels to this. Don’t jump to conclusions. Don’t rule out a honeydick
11
29
327
27,002
Rush Doshi retweeted
“Barring an extension at one of the upcoming summits, we are looking at a period of possible economic tension and escalation in January without a summit on the horizon to put the brakes on a spiral.”
This raises an interesting question. The Trump administration feels Beijing has not fulfilled the Busan agreement. Others disagree.
So who is right? And what does this mean for the next few months and for next year?
It's worth first setting the scene. The Trump administration erred in starting a trade war, taking tariffs on China to 140%, and then folding abruptly when Beijing retaliated with rare earths. That resulted in the Busan deal with Beijing last year. But it implementation is mixed.
First, the United States was supposed to suspend the affiliates rule and lower tariffs, and it did. The affiliates rule remains suspended, and on tariffs, as @Brad_Setser notes, "the U.S. has actually over delivered on its Busan commitments (the term 2 tariffs fell to 12.5%, well below the 20% cap)." SCOTUS played a role, of course.
Second, China was supposed to let rare earths flow, and they have not. The administration claims, and data indicates, that those flows are below what they should be by 30-50%. U.S. companies are not getting licenses. US consultancies and trade associations are trying to help them navigate MOFCOM. What's more, the "truce" was extended to January 10, and as far as I can tell MOFCOM hasn't issued a guidance formally extending the suspension date of its rare earth controls.
Third, the items Kyle (and others ) raise aren't part of the deal. Clearing the BIS license queue was never part of Busan, and even so, the US did license H200 exports. Implementation of the 30-for-30 tariff cuts are not part of Busan either, and they were designed to take effect after going through domestic processes, by MOFCOM's own admission.
Fourth, there is a lack of proportionality here, too. The licensing queue and thirty-for-thirty deal are not existential for Chinese companies. But rare earths access is a gun to the head of the entire US economy.
So what does all this mean?
In November, Trump and Xi will have an awkward meeting in Shenzhen. In December, they'll have an awkward meeting in Miami. And then in January, they won't have any meetings. The truce will expire on January 10th.
Barring an extension at one of the upcoming summits, we are looking at a period of possible economic tension and escalation in January without a summit on the horizon to put the brakes on a spiral.
So if 2026 was the year of so-called detente, 2027 may bring something different. I don't think this is priced in by analysts or the markets.
0
1
4
1,118
I'm not sure about this.
First, the Trump administration doesn't seem to be to be acting as if it has the upper hand. They fear the Busan deal is breaking and they are trying to negotiate a way out.
Second, Beijing knows well what tools the US has left, and has not yet used, and they are concerned about them. So there's an incentive on both sides to fix this.
If in 2025 the US miscalculated and pushed Beijing into rare earths coercion, in 2027 I think the risk is Beijing miscalculates and pushes Washington into further escalation.
First, the Trump administration doesn't seem to be to be acting as if it has the upper hand. They fear the Busan deal is breaking and they are trying to negotiate a way out.
Second, Beijing knows well what tools the US has left, and has not yet used, and they are concerned about them. So there's an incentive on both sides to fix this.
If in 2025 the US miscalculated and pushed Beijing into rare earths coercion, in 2027 I think the risk is Beijing miscalculates and pushes Washington into further escalation.
Busan truce or otherwise, the bigger picture is the US still thinks it’s negotiating from a position of strength where it’s clear China is content with managing rare earth flows at its own pace. Low cost to them and outsize impact on US industry. Beijing has little incentive to heed to US demands now. In the meantime, US inaction on export licenses to China is surrendering China’s market to our competitors.
1
9
31
7,628
Rush Doshi retweeted
a restart in tensions (or deterioration in the detente) is definitely underdiscussed on the street.
trade policy too narrowly discussed thru 1) goods inflation and central bank reaction functions, and 2) selling dollars.
broaden your horizons. think about the impact to growth and uncertainty.
cc: Tiff Macklem
trade policy too narrowly discussed thru 1) goods inflation and central bank reaction functions, and 2) selling dollars.
broaden your horizons. think about the impact to growth and uncertainty.
cc: Tiff Macklem
This raises an interesting question. The Trump administration feels Beijing has not fulfilled the Busan agreement. Others disagree.
So who is right? And what does this mean for the next few months and for next year?
It's worth first setting the scene. The Trump administration erred in starting a trade war, taking tariffs on China to 140%, and then folding abruptly when Beijing retaliated with rare earths. That resulted in the Busan deal with Beijing last year. But it implementation is mixed.
First, the United States was supposed to suspend the affiliates rule and lower tariffs, and it did. The affiliates rule remains suspended, and on tariffs, as @Brad_Setser notes, "the U.S. has actually over delivered on its Busan commitments (the term 2 tariffs fell to 12.5%, well below the 20% cap)." SCOTUS played a role, of course.
Second, China was supposed to let rare earths flow, and they have not. The administration claims, and data indicates, that those flows are below what they should be by 30-50%. U.S. companies are not getting licenses. US consultancies and trade associations are trying to help them navigate MOFCOM. What's more, the "truce" was extended to January 10, and as far as I can tell MOFCOM hasn't issued a guidance formally extending the suspension date of its rare earth controls.
Third, the items Kyle (and others ) raise aren't part of the deal. Clearing the BIS license queue was never part of Busan, and even so, the US did license H200 exports. Implementation of the 30-for-30 tariff cuts are not part of Busan either, and they were designed to take effect after going through domestic processes, by MOFCOM's own admission.
Fourth, there is a lack of proportionality here, too. The licensing queue and thirty-for-thirty deal are not existential for Chinese companies. But rare earths access is a gun to the head of the entire US economy.
So what does all this mean?
In November, Trump and Xi will have an awkward meeting in Shenzhen. In December, they'll have an awkward meeting in Miami. And then in January, they won't have any meetings. The truce will expire on January 10th.
Barring an extension at one of the upcoming summits, we are looking at a period of possible economic tension and escalation in January without a summit on the horizon to put the brakes on a spiral.
So if 2026 was the year of so-called detente, 2027 may bring something different. I don't think this is priced in by analysts or the markets.
0
3
9
2,980
This raises an interesting question. The Trump administration feels Beijing has not fulfilled the Busan agreement. Others disagree.
So who is right? And what does this mean for the next few months and for next year?
It's worth first setting the scene. The Trump administration erred in starting a trade war, taking tariffs on China to 140%, and then folding abruptly when Beijing retaliated with rare earths. That resulted in the Busan deal with Beijing last year. But it implementation is mixed.
First, the United States was supposed to suspend the affiliates rule and lower tariffs, and it did. The affiliates rule remains suspended, and on tariffs, as @Brad_Setser notes, "the U.S. has actually over delivered on its Busan commitments (the term 2 tariffs fell to 12.5%, well below the 20% cap)." SCOTUS played a role, of course.
Second, China was supposed to let rare earths flow, and they have not. The administration claims, and data indicates, that those flows are below what they should be by 30-50%. U.S. companies are not getting licenses. US consultancies and trade associations are trying to help them navigate MOFCOM. What's more, the "truce" was extended to January 10, and as far as I can tell MOFCOM hasn't issued a guidance formally extending the suspension date of its rare earth controls.
Third, the items Kyle (and others ) raise aren't part of the deal. Clearing the BIS license queue was never part of Busan, and even so, the US did license H200 exports. Implementation of the 30-for-30 tariff cuts are not part of Busan either, and they were designed to take effect after going through domestic processes, by MOFCOM's own admission.
Fourth, there is a lack of proportionality here, too. The licensing queue and thirty-for-thirty deal are not existential for Chinese companies. But rare earths access is a gun to the head of the entire US economy.
So what does all this mean?
In November, Trump and Xi will have an awkward meeting in Shenzhen. In December, they'll have an awkward meeting in Miami. And then in January, they won't have any meetings. The truce will expire on January 10th.
Barring an extension at one of the upcoming summits, we are looking at a period of possible economic tension and escalation in January without a summit on the horizon to put the brakes on a spiral.
So if 2026 was the year of so-called detente, 2027 may bring something different. I don't think this is priced in by analysts or the markets.
So who is right? And what does this mean for the next few months and for next year?
It's worth first setting the scene. The Trump administration erred in starting a trade war, taking tariffs on China to 140%, and then folding abruptly when Beijing retaliated with rare earths. That resulted in the Busan deal with Beijing last year. But it implementation is mixed.
First, the United States was supposed to suspend the affiliates rule and lower tariffs, and it did. The affiliates rule remains suspended, and on tariffs, as @Brad_Setser notes, "the U.S. has actually over delivered on its Busan commitments (the term 2 tariffs fell to 12.5%, well below the 20% cap)." SCOTUS played a role, of course.
Second, China was supposed to let rare earths flow, and they have not. The administration claims, and data indicates, that those flows are below what they should be by 30-50%. U.S. companies are not getting licenses. US consultancies and trade associations are trying to help them navigate MOFCOM. What's more, the "truce" was extended to January 10, and as far as I can tell MOFCOM hasn't issued a guidance formally extending the suspension date of its rare earth controls.
Third, the items Kyle (and others ) raise aren't part of the deal. Clearing the BIS license queue was never part of Busan, and even so, the US did license H200 exports. Implementation of the 30-for-30 tariff cuts are not part of Busan either, and they were designed to take effect after going through domestic processes, by MOFCOM's own admission.
Fourth, there is a lack of proportionality here, too. The licensing queue and thirty-for-thirty deal are not existential for Chinese companies. But rare earths access is a gun to the head of the entire US economy.
So what does all this mean?
In November, Trump and Xi will have an awkward meeting in Shenzhen. In December, they'll have an awkward meeting in Miami. And then in January, they won't have any meetings. The truce will expire on January 10th.
Barring an extension at one of the upcoming summits, we are looking at a period of possible economic tension and escalation in January without a summit on the horizon to put the brakes on a spiral.
So if 2026 was the year of so-called detente, 2027 may bring something different. I don't think this is priced in by analysts or the markets.
Slow pace of China’s REE and CM exports, but no mention of U.S. foot dragging on implementing tariff reductions, or wait for it, the billions of dollars in export licenses sitting on Kessler’s desk. Got it.
8
16
64
67,521
Rush Doshi retweeted
It's been pretty clear for a while that Beijing has been slow-rolling the Busan "consensus"/"agreement", while taking a lot of actions not covered explicitly, such as other inputs, equipment, and talent.
I think this area is a good example of where PRC leadership thinks more in terms in systems than the US tends to do.
I think this area is a good example of where PRC leadership thinks more in terms in systems than the US tends to do.
SCOOP -- US warns China it must ‘immediately’ meet terms of Busan trade deal (#USChinaScoop). Comes as Xi Jinping prepares to host Trump at Apec in Shenzhen.
ft.com/content/391bb7…
1
8
23
5,574