Last Friday morning, reading about China’s rare earths bombshell, my eyes caught the first light hitting the calm and beautiful mountain facing me across the bay. Yet it felt like Mount Vesuvius was erupting Chinese lava everywhere.
Xi is not a hip-shooter. He knew the effects that would have, particularly on Trump, the day of the Gaza deal.
But also on the corporate world, global economies and financial markets.
My sense is that this is a major warning after months of back-and-forths and tit-for-tats: deal or no deal, Mr. Trump?
Importantly, there was a lot more than rare earths erupting from what many thought was a rather dormant Chinese volcano:
- The semiconductor supply chain would be stopped impacting phones, computers, data centers, cars, solar panels and a large bunch of other production equipment. And military equipment. And AI development.
- On Thursday, Beijing also said it would require licenses for exports of certain lithium batteries and some equipment and materials used to make them.
- Also Thursday, Beijing added roughly a dozen organizations to its “unreliable entity list,” including TechInsights, a Canada-based semiconductor technology research firm that had released reports on chip-development efforts by China’s Huawei Technologies.
- China’s top market regulator said Friday that it had launched an investigation into Qualcomm for suspected violation of the country’s antimonopoly law. The probe is tied to Qualcomm’s acquisition of Autotalks, an Israeli startup, the regulator said. Qualcomm is one of the most exposed American company to China with nearly half its revenue from China in fiscal 2024.
- Then on Friday, Beijing said it plans to impose a special port fee on U.S. vessels docking at Chinese ports, retaliating against the Trump administration’s move to levy fees on Chinese ships.
China’s Commerce Ministry said the new measures were intended to “protect its national security and interests” and prevent the “misuse of rare earth materials in military and other sensitive sectors”.
Dai Menghao, a trade compliance partner at law firm King & Wood Mallesons, said China had put in place rules that closely mirror the US export-control regime. “It’s a systematic, comprehensive response to the US,” Dai said.
Vice Premier He Lifeng believed an informal “freeze” on new export controls had been agreed upon following recent talks in Madrid, according to people familiar with the discussions. But that understanding was shattered when the U.S. introduced new controls on foreign-owned companies.
Yesterday, Oct. 12, the Ministry of Commerce wrote: “Particularly since the China-U.S. economic and trade talks in Madrid in September, the U.S., in just 20 days, has introduced a string of new restrictive measures targeting China. (…) The U.S. actions have severely harmed China’s interests and undermined the atmosphere of bilateral economic and trade talks, and China is resolutely opposed to them.”
The Trump administration has tightened export controls on Chinese companies to make it harder to circumvent existing rules designed to slow China’s ability to develop advanced semiconductors. The US commerce department said subsidiaries of groups named on the government’s “entity list”, a compilation of foreign entities and individuals subject to export controls, would be automatically added to the blacklist.
US companies require a licence to export to groups on the list. In many cases, licences are denied for exports to Chinese groups that are listed. The new rule does not single out China. But among the targets are subsidiaries of Chinese chipmakers including Huawei and Semiconductor Manufacturing International Corp.
A preliminary analysis by WireScreen, a group that analyses Chinese companies, found the new rule would affect thousands of subsidiaries of Chinese state-owned groups and other businesses.
China’s Commerce Ministry said the new regulation was “extremely malicious” and “yet another typical example of the US generalising the concept of national security and abusing export controls”.
A decision was made by Xi himself to hit back—and hit back harder, some people told the WSJ.
China’s new measures all seem to mimic the US, perhaps to demonstrate the futility of the US measures.
While the rare earths restrictions seem global, the Commerce Ministry said that licenses would be granted on a case by case basis. A safe bet could be that American applications might take longer to process and carry smaller odds of success.
Trump’s reply was the usual: higher tariffs.
Yesterday, again from the Ministry of Commerce (my emphasis): “Willful threats of high tariffs are not the right way to get along with China. China’s position on the trade war is consistent: we do not want it, but we are not afraid of it. China urges the U.S. to promptly correct its wrong practices, adhere to the important consensuses of the phone calls between the two heads of state, protect the hard-won outcomes of consultations, continue to use the China-U.S. economic and trade consultation mechanism, and address respective concerns and properly manage differences through dialogues and on the basis of mutual respect and equal-footed consultation, so as to ensure the stable, sound and sustainable development of the China-U.S. economic and trade relationship.”
My reading of that is that Beijing says that agreements reached in the last call between Trump and Xi were not followed through by American technocrats, perhaps unbeknown to Trump (see “strange things happening in China” below).
The hope now is that the expected Xi-Trump meeting takes place in 2 weeks and that something, somehow happens to stop this nonsense.
Financial markets opened strong on Friday. The decline began after Trump “started the fun by announcing on social media midday Friday that “some very strange things are happening in China!” He said Beijing has turned “very hostile” and is sending letters to the world announcing tighter controls on the export of “every element of production having to do with” rare-earth minerals. “For every Element that they have been able to monopolize, we have two,” Mr. Trump wrote. “I never thought it would come to this but perhaps, as with all things, the time has come.”
Xi seems to have come to the same conclusion.
We’ll soon know who has the better hand. The hope is that Trump and Xi meet and clear the deck, once more.
Note that, so far this year, China’s exports have grown 6.1% YoY, faster than 2024’s 5.9% growth. Exports were up 8.3% in September, thanks to +14.2% in the EU, +15.6% in ASEAN, +15.2% in Latin America and +56.6% in Africa.
BTW, exports to the US were down 27% in September. YtD, the US is only 11% of total China exports.
Trump was right yesterday saying “Don’t worry about China, it will all be fine!”.
And in a sign of improving domestic demand, China’s imports accelerate to 7.4% YoY in September, up from 1.3% in August.
Imports from the EU (+9.4%), Japan (+20.9%), South Korea (+13.1%), Africa (+22.4%), and Latin America (+17.9%) all saw strong growth in September.
USA: -16.1%.
Goods inflation has already picked up in the US keeping headline inflation above 3.0%. It might get worse given the declines in Chinese shipments of apparel and accessories (-8.0%), footwear (-13.3%), furniture (0.4%), and toys (-27.9%).
All this entering a mid-terms year in the US.
Vance on Fox News’s Sunday Morning Futures:
It’s going to be a delicate dance, and a lot of it is going to depend on how the Chinese respond. If they respond in a highly aggressive manner, I guarantee you, the president of the United States has far more cards than the People’s Republic of China. If, however, they’re willing to be reasonable, then the US would, too.
John Authers today sets the cards game somewhat differently:
Beijing’s impudence stems from its capacity to hit Washington where it hurts most. To quote Gave, “China’s ability to disrupt US supply chains, especially in politically sensitive industries such as weapons manufacturing and autos, is now very high.”
Conversely, US efforts to derail China’s artificial intelligence efforts appear to be faltering and may prove counterproductive, as they have forced Chinese engineers to find ways to operate with less energy and computing capacity. Necessity is the mother of invention, as argued by Nvidia CEO Jensen Huang.
As China’s policymakers gather later this month to set the next five-year plan, few priorities loom larger than staking a credible claim in the global AI race. Expanded electricity generation gives it a head start, as Points of Return recently noted. Wang points out that China now produces roughly twice as much electricity as the US:
China this year will deploy about 500 gigawatts of solar alone. The US will deploy 50. So, just an order of magnitude difference. There are 33 nuclear power plants under construction in China right now. There’s zero under construction in the US.
Beyond electricity, research and development spending has surged by 48% over the past four years, above the 40% target in the current Five-Year Plan. Larry Hu of Macquarie notes that R&D has now topped the European Union.
Not to mention soybeans (China’s Soy Imports Hit September Record, Despite Avoiding US).
TACO coming again?
“The most likely scenario seems to be that both sides pull back on the most aggressive policies and that talks lead to a further — and possibly indefinite — extension of the tariff escalation pause reached in May.” (Goldman Sachs)
BUBBLING BUBBLES
This is happening amid a bubble in bubble calling. Is this a bubble?
Goldman Sacks’ Peter Oppenheimer last week:
There are elements of investor behaviour and market pricing currently that rhyme with previous bubbles, including the rise in absolute valuations, high market concentration, increased capital intensity of leading companies and the emergence of vendor financing.
However, we see key differences:
First, the appreciation of the technology sector has, so far, been driven by fundamental growth rather irrational speculation about future growth.
Second, the leading companies that have seen the strongest returns have unusually strong balance sheets.
Third, the AI space has, so far, been dominated by a few incumbents; most bubbles form in a period of huge competition as both investors and new entrants flock into the space.
Importantly, bubbles tend to develop when there is a combined surge in stock prices and valuations to an extent that the aggregate value of companies associated with the innovation exceed the future potential cash flows that it is likely to generate. Valuations of the technology sector are becoming stretched (here we look at P/E ratios versus other bubbles, PEG ratios, P/B vs. ROE and a DDM) but not yet at levels consistent with historical bubbles.
(…) the rise in EPS of the technology sector has been extraordinary relative to the EPS of all other sectors; the gap has particularly widened since the financial crisis. This pattern is unusual in the context of bubbles when the companies at the epicentre of the fervour are typically driven by expectations of future growth and market dominance rather than those that have already achieved it.
Of course, this does not guarantee that the dominant companies are not overvalued or vulnerable to competition, but rather their performance up until now has been a reflection of sustained fundamental success rather than excessive speculation about the future.
(…) it is underlying profitability and return on equity that has largely explained the rise in valuations. The ROE of the US market – largely driven by the technology sector – is well above its previous 20-year high and well above that achieved in other regions.
(…) the median 24m forward P/E ratio across the ‘Magnificent 7’ is 27x, or 26x if we exclude Tesla. This is roughly half the equivalent valuation of the biggest 7 companies in the late 1990s, while the dominant companies in Japan (mostly banks) traded at higher valuations still; the current EV to sales ratios are also much lower than those of the dominant companies in the late 1990s. So it is true that valuations are high but, in our view, generally not at levels that are as high as are typically seen at the height of a financial bubble.
(…) The US technology sector enjoys historically high ROEs, which to some extent does justify a high price to book. That said, the sector is starting to look more stretched on this metric (Exhibit 10). But it is not as stretched as in 1999/2000 – at that point price to book was similar to today but the ROE was much lower.
(…) Perhaps the most compelling argument for why valuations are not yet at bubble levels is to compare returns in this past year – disaggregated between earnings, valuation and dividends – with returns back in the 12-months before the peak in the dot-com bubble. We see from Exhibit 12 that US tech earnings growth was good back in 1999/2000 and certainly was a contributor to returns; but on top of earnings there was also a much greater rise in valuation that dwarfs what we have seen today. And for European telecoms the valuation ascent was even steeper, with returns driven almost entirely by valuation rather than earnings.
(…) Surging investment and capex has certainly been a typical feature of technology bubbles in the past. Companies at the epicentre of a major technological innovation often fail to achieve the returns that their high valuations imply as the marginal cost of the technology falls and capacity increases over time. Equally, a typical overlooked opportunity is that investors understate the returns available to new entrants in an industry that emerge after the initial investments are made, and can piggyback on the capex of others by providing new products and services able to leverage the new embedded technology.
There is a risk of over-investment by the current set of incumbents. They have generated extraordinary profit growth and ROE in recent years while adopting a relatively light capital intensity model, but have been rapidly increasing their capex spend since the emergence of Chat GPT, and there is a growing risk that the future returns that they achieve on this capital is less than the market is currently implying.
The problem of increased competition and overspend by incumbents was evident in the early days of the internet and in the bubble that followed. (…)
Ultimately, as with many other technology innovations, too much capacity was built – a problem that mimicked the railway bubbles and canal bubbles of previous periods.
According to the Federal Communications Commission, the amount of fibre-optic cable laid in the US went from one million miles in 1996 to 10 million by 2000. Increasingly, competition forced prices down and by 2004 the cost of bandwidth had fallen by more than 90%, despite internet usage doubling every few years. As late as 2005, as much as 85% of broadband capacity in the US was still going unused.
(…) While the capex to sales ratio for US tech is rising rapidly, the ratio of capex to free cash flow remains stable and significantly below levels experienced in the late 1990s.
Oppenheimer’s third point, that the AI space has so far not attracted much competition to the US incumbents omits China. The Chinese are coming with competitive, efficient, low cost and open models and applications. Whether the US biz models currently take this into account is unknown…
CONSUMER WATCH
The latest report from the Bank of America Institute is titled “The upturn in consumer spending continued into September”.
The details are important.
Total credit and debit card spending per household increased 2.0% year-over-year (YoY) in September, compared to 1.7% YoY in August, and the highest YoY growth rate since December 2024, according to Bank of America aggregated card data. Seasonally adjusted (SA) spending growth per household rose 0.2% month-over-month (MoM), continuing a solid run of monthly momentum since June 2025.
Retail spending (ex-gasoline and restaurants), on the other hand, declined 0.5% MoM in September.
This is a big slowdown in total spending (red + blue above) after 3 solid months, ending Q3 on a weak note. Is the wealth effect fading?
One way to illustrate the impact of these wealth effects in Bank of America data is to look at the difference in discretionary spending growth between the top 5% of households and middle-income peers. (…)
Indeed, Exhibit 11 shows that this difference has been fairly closely related to the growth in the S&P 500 since 2020. This suggests that spending for the high-income cohort should remain supported by wealth, absent a correction in equity markets.
America’s Manufacturing Resurgence Will Be Powered by These Robots China has more industrial robots than the rest of the world combined, but newer, more flexible robots are keeping smaller U.S. manufacturers in the fight
In small factories across America, agile automatons are making everything from parts for AI supercomputers to the hulls of America’s future autonomous naval weapons.
Once a luxury reserved for big manufacturers, smaller, smarter, more flexible and less expensive “cobots”—collaborative robots—are bringing automation to every fabricator, no matter the size. These robots aren’t just nice to have. The slow, fragile recovery of American goods production wouldn’t be possible without them.
The number of U.S. companies that make physical things reached a low point in 2014 and has grown since then. Yet they are trapped in a never-ending labor shortage as skilled workers age out, and young people fail to take their place.
China has become the de facto manufacturer of the world’s goods, owing not only to its enormous population of engineers, technicians and machinists but also its 2-million-plus army of industrial robots. Now the U.S. is attempting to claw back some of those contracts—a process called “reshoring”—and robots can in some cases quadruple worker output. (…)
“Automation is key to reshoring, plain and simple,” says Greg LeFevre, CEO and president of Raymath, a metal fabrication company based in Troy, Ohio.(…)
Per-worker productivity at his facility has soared, with individual workers producing as much as four times as many pieces per day as they could before. This makes Raymath more competitive, and has helped the company triple its revenue. LeFevre has also been able to slightly expand the company’s workforce, from about 130 to 145 workers. Humans have gone from doing all the grinding, welding, machine-operating and parts-moving to overseeing and collaborating with robots who take on the bulk of the tasks.
These robots have become radically easier to program over the past decade, and now people can use a simple tablet interface to instruct them to perform specific sequences of actions. (…)
Dozens of companies in the U.S. now offer manufacturers the kinds of specialized welding cobots that LeFevre uses, says Josh Pawley, co-founder of Loveland, Colo.-based Vectis Automation. (…)
These metalworking cobots are part of a broader trend in robotics: Specialized robots that use sensors to safely navigate human environments. They can cope with more variability than previous industrial robots, which had no sensing abilities. This has been essential to the rise of Amazon and its superfast fulfillment, and now it’s coming to manufacturing. (…)
China is indisputably the leader in high-volume manufacturing, and companies that want the biggest volumes of manufactured parts for the lowest possible price continue to send work there. And though many U.S. manufacturers can’t match their Chinese peers in volume, they are competing by using automation to tackle smaller batches of goods under tight deadlines. (…)
Bay Area-based Dyna Robotics is making a “foundation model” for manufacturing robots. This means they learn to perform tasks, instead of being programmed to do them. Today, the startup’s robot can fold dinner towels, but soon, the firm will partner with real-world factories to expand that skill set, says Lindon Gao, the company’s CEO. His company announced last month it had raised $120 million from investors including Nvidia and Amazon.
It’s a long-term play. Making these learning robots reliable enough for an industrial environment is next to impossible at present, says Christopher Müller, director of the statistical department at the International Federation of Robotics.
That’s not stopping Gao, who hopes to deploy them within a few years. Once robots can be set to work with a quick demo and some verbal commands, the country that can muster the most robots—rather than the most workers—wins.
The Washington Post carries a somewhat different story:
- The Chinese robots are coming China is miles ahead in the race to dominate a technology that could define the 21st century.
(…) China is on the cusp of a robotics revolution.
If embodied intelligence — think AI-powered robots that can navigate the real world — is the next frontier of AI, then China appears poised to dominate. (…)
Last year, China installed nearly 300,000 robots in its factories, more than the rest of the world combined, according to a September report by the International Federation of Robotics. More than half were made domestically. The United States installed only 34,000 robots, with most of these imported from Japan and Europe. (…)
Speaking to founders and engineers from Hangzhou’s “Six Little Dragons” — an online moniker for the city’s best-known tech start-ups — we could feel the ground beneath our feet shifting.
Part of the reason is sheer manufacturing prowess. From actuators to sensors to batteries, China has built a comprehensive supply chain that allows start-ups to rapidly fine-tune robot prototypes until a viable, affordable and scalable product emerges. Most of the companies we visited build their robots close to their research labs, creating fast feedback loops.
Companies are also prioritizing vertical integration, with many producing key components in-house and procuring base materials domestically. Entry prices for humanoid robots are dropping drastically. In July, Unitree released its R1 bot for under $6,000, about a third of the price of its G1 robot from a year earlier. (…)
Founders we spoke to talked about going from double-digit unit counts to tens of thousands within months. China is full of hubs — from Shenzhen to Hangzhou — that are “communities of engineering practice” where entrepreneurs, investors and engineers can mingle with the world’s most experienced manufacturing workforce to innovate quickly. (…)
Expect these innovations to charge abroad. Almost every founder we met wanted to expand overseas, including into the United States. Part of the reason is brutal price wars at home, with razor-thin profit margins amid competition against a glut of other companies (what some dub neijuan, or “involution”).
Despite these headwinds, the founders’ zeal was notable — they sincerely believed that “Made in China” was the sign of world-class quality, especially in high-end manufacturing. Already, Chinese companies such as BYD, CATL, DJI and Huawei are global leaders in electric vehicles, batteries, consumer drones and 5G networks, respectively. The emerging crop of start-ups in robotics and AI are determined to follow in their footsteps. (…)
While China’s older tech pioneers tend to idolize Silicon Valley, younger founders find their role models in people such as DeepSeek’s Liang Wenfeng and Unitree’s Wang Xingxing — both of whom studied exclusively at Chinese universities and are known to predominantly hire domestic talent. (…)
Much ink has been spilled about how the U.S. can reindustrialize. When it comes to robotics, the U.S. needs to invest in infrastructure to power the industries of the future — from building a better electrical grid to joint ventures with allies that can onshore manufacturing expertise for critical robotics components.
America also needs to double down on its strengths in ways that could prove difficult in this turbulent political moment: university-driven fundamental research, immigration policies that welcome the best talent from abroad, and a vibrant entrepreneurship ecosystem that’s not mired in regulatory hurdles.
The robot race isn’t over. But unless the United States ups its game, it seems clear that this is a race China will dominate.
Chinese cars coming to Canada?
Canada Will Fight to Save Auto Jobs Despite US Threat, Minister Says
The Canadian economy is too dependent on the US, the country’s industry minister said, as she vowed to protect jobs in its export-focused automotive sector.
The government is working on a new industrial strategy that seeks to open new markets for exporters and prioritize domestic procurement in the face of US tariffs, which have hurt steel, aluminum, forestry and automotive companies, Melanie Joly said in an interview aired by Canadian Broadcasting Corp. on Sunday. (…)
On Oct. 8, Commerce Secretary Howard Lutnick told a Toronto conference this week the US administration doesn’t want to see auto manufacturing in Canada.
The government “will fight for every single job in the auto sector,” Joly said. “These are our jobs, these are our people and we will make sure we stand with them.” (…)
China’s ambassador suggested that removal of the vehicle tariffs would lead to China scrapping its counter-tariffs on Canadian food products, according to Canada’s CTV News.
Manitoba Premier Wab Kinew asked Carney in a letter Saturday to take that offer, and eliminate tariffs on Chinese EVs “in exchange for restoring full access for Canadian agricultural products to the Chinese market.”
“While protecting our automotive sector is important, the current approach has created a two-front trade war that disproportionately impacts Western Canada,” Kinew wrote in his letter.
Anita Anand, Canada’s foreign minister, is about to embark on a swing through Asia that includes China.
Lutnick’s vision (not necessarily wisdom) is that American cars sold in Canada eventually are all built in the USA. He may get his wish if Chinese manufacturers set up plants in Canada as seems more and more likely.
BYW, FYI:
The biggest 5 technology companies in the US now have a collective value that exceeds the combined size of the Eurostoxx 50, the UK, India, Japan, and Canada.
Put another way, these 5 companies are worth around 16% of the entire global public equity market. The biggest 10 US stocks (8 of which are technology-related) account for nearly 25% of the global equity market, and are worth almost $25 trillion. (Goldman Sachs)
