How Far Is Rare Earths' Break From China? The Progress and the Gaps in Alternative Supply Chains

Many people are puzzled: rare earths aren't actually that rare in the earth's crust, so why can China wield them as a weapon? The key isn't "who has the ore" — it's "refining."

Mining rare earths is only the first step; separating the dozen-plus elements mixed together and purifying them to an industrially usable grade is an extremely polluting, technically demanding, and thin-margin dirty job. Over the past thirty years, the West, for reasons of environmental cost and economic incentive, was happy to outsource this step to China; China quietly absorbed it and built up roughly nine-tenths of the world's refining capacity. So the truth of "rare-earth hegemony" is this: it's not that China monopolizes the resource — it's that China monopolizes the ability to turn the resource into a usable material. That also explains why substitution is so hard — you can dig ore out of the ground in Australia or the United States, but you can't build, in three to five years, an entire refining-and-magnet industry able to rival thirty years of Chinese accumulation.

For the past decade, "de-China rare earths" was almost a joke — every time the West tried to mine its own, China would dump ore at low prices and drive Western mines out of business. In 2026, this deadlock showed its first real crack, thanks to two breakthroughs.

First, the hardest segment — heavy rare earths — has been taken. Dysprosium and terbium, the "heavy rare earths" that keep magnets performing at high temperatures, are also the segment China has held tightest. In May 2025, Australia's Lynas, at its Malaysia plant, became the first company outside China to commercially produce dysprosium oxide ^[3]. The most tightly held link has loosened.

Second, the United States simply made the state the buyer. The U.S. Department of Defense bought a $400 million equity stake in MP Materials, helped it build a heavy rare-earth separation plant, and did two more critical things: it set a price floor and committed to buy its future magnet output; right after that, Apple also signed a roughly $500 million long-term supply contract with MP ^[3][5]. This directly broke the old pattern of "China dumps, and the Western mine dies" — when the buyer is a defense department and Apple, neither of which will go bankrupt, a price war can't kill it.

By early 2026, the consensus among analysts was: capital is flowing in, governments are intervening directly, and Western projects are moving from "intent" to "execution." This is the first time in a decade that the sentence "we can make our own rare-earth magnets" has had some real backing behind it.

Don't rush to declare decoupling a success, though — even as the West has been loosening its grip, Beijing has been upgrading its own tools.

The control list is expanding. In April 2025, China imposed national-security export licensing on 7 categories of medium and heavy rare earths and related permanent-magnet materials — samarium, gadolinium, terbium, dysprosium, lutetium, scandium, and yttrium; by October, it added holmium, erbium, thulium, europium, and ytterbium — 5 more categories — bringing the total scope to 12 categories ^[6]. After the April controls took effect, export volumes of yttrium, dysprosium, and terbium fell by roughly 50% from the previous year's average, and the impact of this new round of restrictions has been progressively showing up ^[3][4].

The more toxic part is the "0.1% rule." The new rule states plainly: NdFeB magnets containing terbium or dysprosium, samarium-cobalt magnets, and any part, component, or finished product containing 0.1% or more Chinese-origin content, are subject to control when re-exported to a third country ^[6]. This is a kind of "extraterritorial effect" — meaning that even if you buy a non-Chinese magnet, as long as it has even a trace of Chinese material mixed in, selling a product containing it to another country can still get blocked by Beijing. This indirectly ties the whole world's downstream manufacturing — Taiwan included — into the net.

Between late 2025 and 2026, after a summit in Beijing, the U.S. and China reached a kind of détente: China announced it would pause enforcement of the October round of controls, for one year, until November 2026 ^[3]. Markets breathed a sigh of relief.

But that relief shouldn't run too deep. There are two reasons: first, the April licensing regime was never paused — meaning the most core controls are still in effect; second, during this very "truce," China's Ministry of Commerce, in June 2026, added 10 U.S. companies including MP Materials and USA Rare Earth to its export-restriction list, banning the export of rare earths and magnetic materials to them ^[3].

This "truce" looks much more like "a pause that can be called off at any time." For Taiwan, mistaking a brief loosening for the new normal is dangerous — no one can guarantee what happens after November 2026.

Pulling the camera back, the gap is large enough to be sobering.

The real bottleneck was never "digging" — it's "refining" and "turning it into a magnet," and China still holds roughly 90% of both ^[1]. The timeline is even harsher: consulting firm Benchmark Mineral Intelligence estimates that by 2030, the West's demand for heavy rare earths will still be about 91% reliant on Chinese supply — a figure that has fallen only 8 percentage points from 99% in 2024 (this refers to the West's demand-side dependence, a different measure from the 99% refining-and-processing capacity share cited in "The Rare-Earth Chokehold") ^[7][10]. McKinsey and CRU also judge that by 2035, non-Chinese supply of heavy rare earths will still meet less than a fifth of global demand.

The West can dig up the ore, but it still doesn't have enough capacity to turn that ore into magnets; and for the most critical heavy rare earths, truly breaking free of China is probably a matter for the 2030s or later. The two red teams meet on this point. Red Team East says: "Ninety percent of refining is still in China, and by 2030 it'll still be 91% reliant on China — don't get ahead of yourselves." Red Team West says: "Heavy rare earths have had their first production outside China, and the state itself is now the buyer — that's a qualitative shift, stop writing it off." Both are right — the progress and the gap are simultaneously true.

It's worth noting that being "choked" hasn't just driven a search for "other sources of rare earths" — it has also driven innovation aimed at "not needing rare earths at all."

Facing China's controls, companies in the U.S., Germany, and Japan have already invested in developing alternatives such as "iron-nitride magnets" and "magnet-free motors," trying to route around dependence on rare earths at the material level ^[9]. These technologies are still early-stage and will need time to commercialize, but they point to a more fundamental possibility: when a resource is weaponized to its fullest extent, the world's strongest countermeasure is often not "finding an alternative source" but "making that resource no longer needed." For Taiwan's motor and EV industries, this is also a direction worth a long-term bet.

Taiwan is not a rare-earth producing country, so at first glance this issue looks unrelated to Taiwan. But that's an illusion.

Taiwan makes heavy use of NdFeB permanent magnets — in motors, offshore wind power, electric vehicles, and all kinds of electronics. Rare earths' hand is ultimately played against Taiwan through "magnets." The good news is that, per an assessment by the Ministry of Economic Affairs, the elements covered by this latest round of expanded controls differ from what Taiwan's semiconductor processes require, so there is for now no significant impact on chip production ^[8]; the Ministry has also stated it will continue monitoring and strengthen supply-chain resilience ^[2]. But the bad news is that Taiwan's traditional motor, wind-power, and electronics supply chains remain genuinely exposed to the risk of rising magnet prices, delivery delays, and that "0.1% extraterritorial rule."

Don't underestimate that unremarkable little NdFeB magnet. It's hidden in many places in Taiwan you can't see: EV drive motors, offshore wind-turbine generators, servo motors in machine tools, hard drives, speakers, and all kinds of precision industrial equipment. Taiwan's prized precision-machinery, EV supply chain, and domestic wind-power localization all need a stable, affordably priced, high-performance magnet supply. So when rare earths' hand is played against Taiwan through magnets, the ones who get hurt won't just be the big companies in the headlines — it will be the large cluster of precision-machinery SMEs in central Taiwan that make motors and components. They have weak bargaining power and limited inventory; the moment magnets get more expensive or run short, their margins get eaten away and their delivery schedules blow open. That is exactly why "breaking rare earths' hold on China" is not a distant geopolitical issue for Taiwan, but a matter of survival for an entire domestic industrial chain.

So the alternative supply chain taking shape has a very concrete meaning for Taiwan: in the medium-to-long term, Taiwan's bargaining power on magnets and its exposure to supply cutoffs should improve; but in the short term, Taiwan remains constrained by China. The key is — whether Taiwan can move early enough to get its procurement into "non-Chinese magnet" supply chains like Lynas and MP, while simultaneously investing in recycling and alternative technologies.

Pulling all these threads together, there's one date worth circling in red: November 2026. That's when China's "pause" on the October round of controls expires. No one can guarantee what will happen then — the truce could be extended, full controls could be reinstated, or things could escalate further. For Taiwan's magnet supply chain, this means: whatever inventory needs building, whatever non-Chinese sourcing needs negotiating, whatever substitute materials need validating, all of it has to be done ahead of time — not scrambled together after the rope has already tightened. The thing to fear most in geopolitical risk isn't that "it will happen" — it's that "it happens before you're ready."

But history also reminds us not to bet everything on "substitution." China understands perfectly well that if it pushes too hard and forces Western alternative capacity and "rare-earth-free" new technologies into existence, the value of the card in its own hand permanently depreciates. Some analysts therefore believe Beijing actually prefers "intermittent, adjustable" controls — applying pressure without forcing the other side into full decoupling. This is a subtle reminder for Taiwan: China's rare-earth card won't disappear anytime soon, but it also won't be played all at once, to the very end. What Taiwan needs to do is build up its own resilience, bit by bit, over this long game of "tightening and loosening."

For the state: bring "non-Chinese rare-earth and magnet supply chains" into critical-materials strategy, connect with friendly-shore supply chains in the U.S., Australia, and Japan, and compete for non-Chinese quotas on magnets and heavy rare earths; simultaneously build up critical-mineral inventories and "urban mining" recycling capacity. The date to watch most closely is November 2026, when that "truce" expires — if China reinstates full controls at that point, Taiwan's magnet supply chain needs to already have a backup in place, not be scrambling at the last minute.

For industry (motors / wind power / EVs / electronics): bring magnet-source diversification (non-Chinese magnets from Lynas/MP) into procurement strategy, and invest in rare-earth recycling and use-reduction design — using less and recycling more means being controlled less. At the same time, keep an eye on alternative technologies like iron-nitride magnets and magnet-free motors, and treat "not needing rare earths at all" as long-term insurance.

For SMEs: rising magnet prices and delivery volatility will pass straight through to downstream component makers. Track the availability and certification of non-Chinese magnets, and move early to secure a position in "de-China magnet" supply chains — pay special attention to that "0.1% rule": if your product has even a trace of Chinese rare-earth material mixed in, it could get blocked at export. This window only exists while the alternative supply chain is taking shape, and it is a risk that must be inventoried early — falling behind by one step now could mean falling behind by several years later.

A year ago, China's rare-earth chokehold looked unbreakable. A year later, the West has proven for the first time that this chokehold can be loosened, inch by inch — only it isn't loose enough to breathe freely yet, and Beijing's other hand is pulling the rope tighter and braiding it more intricately. For Taiwan, the homework isn't to worry about when China will tighten its grip — it's to use this open window to connect itself into the magnet supply chain now growing outside Beijing's reach — and to remember that this window may well be shut again, at the latest, in November 2026. In the game of geopolitics, there is never a "decoupling that solves everything once and for all" — only the continuous, ongoing effort of "making yourself harder to choke." That is work Taiwan has to do, bit by bit, every single day this window stays open.