China Rare Earth Dominance
and Western Response

Rare earths as a topic got blown onto the table in 2025 by China's export controls. Before that, it occupied a space in policy circles best described as "everyone knows it's a problem but nobody's actually in a hurry." It briefly entered public view during the 2010 Sino-Japanese dispute, then the heat faded fast, because China loosened its grip quickly, prices came back down, and the panic dissipated. This time is different.

First, clear out a deeply entrenched misconception. Rare earths are not rare. Cerium is about as abundant in the Earth's crust as copper. The US has deposits, Australia has deposits, Brazil has deposits, Vietnam has deposits. In terms of reserves, China holds a little over a third of the global total. That ratio doesn't constitute a monopoly. But China processes about 90% of the world's rare earths and manufactures about 94% of rare earth permanent magnets. Twenty years ago the latter figure was around 50%. All the nervousness should be directed at this second set of numbers, not at reserves. Being able to dig up rocks and being able to turn those rocks into magnets used in an F-35 are separated by a deep trench.

How deep is that trench.

The process used for rare earth separation is called solvent extraction. The 17 rare earth elements have extremely similar chemical properties, requiring hundreds or even thousands of sequential extraction stages, each exploiting minuscule differences in partition coefficients between elements. This is not the kind of thing where you buy the equipment, hire consultants, and start running. Every ore source has a different impurity profile, which means the process has to be adjusted accordingly. Even within the same mine, ore composition varies between different veins, requiring continuous parameter tuning. What concentration of extractant at what temperature, what ratio of organic to aqueous phase, what modifier to add at what stage to prevent emulsification, all of this is knowledge built up over decades of running production lines.

Mountain Pass is the only rare earth mine in the United States, producing about 47,000 metric tons per year, roughly 10% of global output. Before 2022 these ores had to be shipped to China for processing. America has the mine. It didn't have the processing capability. In this business, having the mine is useless by itself.

France's Solvay tried independent separation. Japan's Shin-Etsu Chemical tried too. Progress in both cases was slower than expected, precisely because the accumulation of process know-how cannot be skipped. You can buy equipment blueprints. You cannot buy the twenty years of hands-on feel that an operating team has developed. This point gets glossed over in too many policy discussions, as if building a factory and allocating funds will naturally produce output. It won't. Before output materializes there is an extremely painful ramp-up period during which your product yield and cost structure simply cannot compete with China's mature production lines.

The Molycorp story deserves to be treated as a full case study, not compressed into a passing mention of "and then it went bankrupt."

This American company decided in 2010, when rare earth prices were surging, to restart full-chain processing capability at the Mountain Pass mine. The logic was sound: America needs independent rare earth processing, prices are high, timing looks right. They invested over $1.5 billion in modernization. The money arrived, equipment was installed, the technical team was hired, and then China started flooding the market. Not a modest increase. After 2012, Beijing deliberately pushed up output to suppress prices. Global rare earth prices crashed more than 70% from their 2011 peak. Molycorp's cost structure could not survive at those prices. A factory built in California, subject to American environmental regulations, paying American wages, competing against a processing system that had been running in Inner Mongolia for thirty years, where environmental costs were heavily externalized and labor costs were a fraction of America's, the cost curves were simply not on the same plane. In 2015 Molycorp filed for bankruptcy protection. The assets were later acquired and restructured into what is today MP Materials.

This is not a story of mismanagement. This is a story about market structure. Under existing conditions, without some form of policy protection (tariff barriers, long-term procurement contracts, cost subsidies, price floor mechanisms), any Western rare earth processing project is exposed to exactly the same risk. China can push prices below your cost line at any time by increasing output. This is not theoretical reasoning. It is what happened between 2012 and 2015. Everyone talking about investing in Western rare earth capacity today knows this, which is why their first question to the government is always: how will you protect me?

In 2022 China executed a similar move, pulling up processing volume by 25% to suppress prices. Same effect. Overseas competitors got squeezed again. This kind of behavior is called predatory pricing in economics. In theory it can be challenged under the WTO framework. In practice, pursuing legal remedies against a sovereign state's industrial policy is extremely difficult to make work.

In April, China imposed comprehensive export controls on medium and heavy rare earths including terbium, dysprosium, samarium, and yttrium. In October the controls were escalated, requiring any foreign company using Chinese-sourced rare earth materials or Chinese rare earth technologies to produce components to also apply for export licenses. The second provision is the heavy weapon. Because global rare earth processing technology is extensively dependent on Chinese technology licensing and equipment exports, the condition of "using Chinese technology" has a far broader reach than "using Chinese materials." It extends the controls from the material layer to the intellectual property layer.

Export volumes collapsed in April and May. Automakers in Europe and the United States couldn't get permanent magnets. Some factories shut down temporarily or ran at reduced capacity. Trade volumes later recovered partially. Prices did not come back. European rare earth prices at one point reached six times the Chinese domestic price. Six times means that assembling rare-earth-containing products inside China is inherently far cheaper than doing the same thing anywhere else. The effect of the controls goes beyond restricting supply. Through the price differential, they are pulling manufacturing toward China.

In March 2026 the controls were extended to Japan, with 20 Japanese companies restricted from importing permanent magnets.

The 2010 episode with Japan, when analyzed after the fact, showed that actual export reductions were far smaller than the level of panic they generated at the time. That round of controls was brief, narrow in scope, and closer to a diplomatic gesture. 2025 is entirely different. A formal licensing system was established. A technology tracing framework was built. The scope of application was continuously expanded over subsequent months. It is a persistently operating control regime, not a one-time signal.

Every time the controls are used, the voices in Western politics calling for investment in alternative supply chains get a burst of fuel. The 2010 round gave birth to Japan's fifteen-year rare earth diversification effort. The 2025 round got several previously stalled appropriations bills through the US Congress. China should be able to see this dynamic.

The West is spending money. The question is whether money solves this problem.

The US Department of Defense allocated $400 million to MP Materials in July 2025 to build a domestic magnet production line. Australia's Iluka Resources received A$1.25 billion in government loans for its Eneabba refinery in Western Australia, expected to come online in 2026. Arafura's Nolans project received $840 million in federal funding. Lynas commercially produced dysprosium oxide in May 2025, the first time anyone outside China had done so. The volume is negligible in terms of global supply and demand. The significance lies in proving the technical pathway works.

Mineral diplomacy is also being laid out. Trump put mineral cooperation at the center of his Saudi state visit agenda. MP Materials and Saudi Arabia's Ma'aden signed a full-chain cooperation agreement from mining to magnet manufacturing. Australia passed a 10% critical mineral processing tax credit.

Bloomberg Intelligence's report this year says new capacity could potentially bring China's share of the NdPr market from 90% in 2024 down to 69% by 2030. The same report says that global NdPr will still face a supply gap of approximately 36% in 2030.

Money is arriving. Capacity is being planned. So where's the problem? The problem is time, cost structure, and the fact that China will not sit still while you build.

A new mine takes ten to fifteen years from exploration to full production. Processing facilities need several more years of commissioning and ramp-up. The window of fastest demand growth (driven by net-zero targets requiring mass deployment of EVs and wind power) is precisely when alternative supply is at its weakest. The gap between the two curves is not shrinking. In the short term, it is widening.

The cost structure issue has already been covered. Projects without policy protection can be destroyed at any time by Chinese pricing behavior. And Chinese rare earth companies are seeing their profitability rise, not fall. China Northern Rare Earth Group forecasts net profit growth of up to 135% in 2025. Beijing Zhong Ke San Huan, based on its earnings guidance, expects net profit growth that may approach 900%. The opponent's ammunition is increasing.

Discussion of alternative materials and motor design is usually compressed into minimal space. Within the entire rare earth security conversation, upstream mining absorbs most of the attention and funding. Attention to the demand side is severely insufficient, yet the demand side is exactly where results come fastest.

Grain boundary diffusion technology is already in industrial use. It is not a laboratory concept. The principle is to precisely diffuse dysprosium or terbium onto the grain boundary layer of NdFeB magnets, rather than uniformly blending it throughout the entire magnet as in the traditional approach. The benefit of precision diffusion is that very small amounts of heavy rare earths can achieve the same high-temperature coercivity, reducing heavy rare earth consumption by 50% to 70%. Dysprosium and terbium are the two elements hit hardest by the 2025 export controls and are also the most volatile in price. Directly reducing demand for these two elements is, in effect, partially neutralizing China's export controls.

Chinese companies lead in grain boundary diffusion. Japan's TDK and Shin-Etsu Chemical also have deep expertise. This is one of the few technical segments where the West has not been entirely left behind. In terms of time efficiency, scaling grain boundary diffusion is far faster than opening a new mine to production. A mine takes fifteen years. Upgrading an existing magnet production line to grain boundary diffusion processes takes a much shorter cycle.

Why does policy discussion place its center of gravity on upstream mining rather than here? Part of the reason may be that mining investment is easier to communicate politically. "Build a mine" is easier for voters to understand than "improve magnet manufacturing processes." It translates more readily into television footage. Appropriations bills for mining are easier to pass in Congress than appropriations for process R&D. Resource allocation is sometimes driven by narrative convenience rather than technical rationality.

Tesla used an AC induction motor on the Model 3 front axle, containing no rare earths. The PM-SynRM design in subsequent models combines reluctance torque with permanent magnet torque, substantially reducing rare earth consumption. Switched reluctance motors require no permanent magnets at all. Noise and torque ripple issues are still being improved, though they are already usable in low-speed industrial applications. Fe₁₆N₂ magnets have theoretical performance approaching NdFeB using only iron and nitrogen. Wang's team at the University of Minnesota has worked on it for over a decade. It remains far from industrialization.

Consumer electronics and some automotive applications can be engineered around rare earths. The defense sector cannot. Fighter jet guidance systems, nuclear submarine silent propulsion motors, high-power direct-drive offshore wind turbines: the demands these applications place on power density and temperature resistance mean NdFeB magnets have no substitute in the near term. So even if civilian demand is partially absorbed by technical innovation, defense procurement's rigid dependence on Chinese rare earths remains. The Pentagon knows this. The $400 million allocated to MP Materials was aimed squarely at this.

Global rare earth recycling rate: approximately 1%. Recycling requires having something to recycle. Large-scale EV deployment only began gaining volume around 2018. Motor design life is ten to fifteen years. Offshore wind turbines, twenty to twenty-five years. A recycling feedstock pool of meaningful scale will not emerge until the mid-to-late 2030s. The explosive growth in rare earth demand is happening now. Discussing recycling's contribution to supply security at the current stage has no basis in reality.

HPMS and the end-of-life motor recycling work being done by Cyclic Materials both show technical progress and both remain in the validation phase. Recycling may occupy a meaningful share of the supply structure from the 2040s onward. In this current decade, its contribution is approximately zero.

The severity of this problem exceeds the funding gap. Specialized talent in rare earth metallurgy and separation chemistry is globally concentrated in China. For the past thirty years the industry has been in China, the jobs have been in China, so the universities there have the students, the advisors, the publications, the internship opportunities in these fields. In Western universities these disciplines have withered to near non-existence because there was no domestic industry to absorb graduates.

Japan after 2010 built rare earth reserves equivalent to several months of domestic consumption while simultaneously advancing its recycling legal framework. Among non-Chinese players, Japan has done the most systematic work on rare earth security. It got burned once, then never stopped. The United States and Europe before 2025 were largely watching.

A UN Industrial Development Organization projection puts China at 45% of global manufacturing value added by 2030. Rare earths are one facet of a larger picture. Solar-grade polysilicon, cellular IoT modules, legacy semiconductors, rare earths: they all point to the same underlying structure, the systematic hollowing-out of Western manufacturing and systematic dependence on the Chinese industrial system. Rare earths became the focal point first because China chose to play its hand on this commodity first. Next could be tungsten, antimony, or gallium and germanium. China already imposed export restrictions on gallium and germanium in 2023.

China's rare earth advantage was built over forty years. Separation capacity, magnet manufacturing capability, industry consolidation, the talent pipeline, pricing instruments, all of it required time to accumulate. The Western response will necessarily be measured in decades.

On the supply side, separation and magnet manufacturing are the chokepoints, not mining. Money spent on opening mines without matching investments in processing yields a half-finished product. On the demand side, scaling grain boundary diffusion technology delivers results far faster than any new mine. On the institutional side, stockpile systems, procurement coordination among allies, and price protection mechanisms are all necessary, because market mechanisms in this domain have already been demonstrated to fail.

China's use of its rare earth leverage to date appears precisely calibrated. Make a move, create an impact, pull part of it back, maintain the other side's sense of insecurity. Not a full cutoff, but no reassurance either. Maintaining a dependency relationship that is just sufficient. If this calibration persists, the West's decoupling motivation may permanently remain in a half-committed state of "willing in principle but lacking urgency," with policy commitments repeatedly diluted across electoral cycles.