The Magnet Trap
In June 2017, a hedge fund manager named James Litinsky stood on the steps of a federal courthouse in Delaware and, by his own later account, scraped together “a few million dollars” in the final hours of a bankruptcy auction to keep a mine alive.
The mine was in the California desert, three miles from the Nevada border, fifty miles from the nearest town. It had been operational, on and off, since 1952. It had once supplied most of the world’s rare earth elements.
By the spring of 2017 it had eight employees, no production, and was days — Litinsky says literally days — from being permanently shut down by its creditors and bulldozed back into the ground.
He paid $20.5 million for it.
Almost nobody in the rare earth industry thought he was making a sound investment. Most thought he was making a terrible one.
The conventional wisdom in 2017 was that the rare earth game was over. The United States had lost it three decades earlier, and there was no economic case for trying to win it back. China had cornered the market, locked in the processing technology, and was selling rare earth products at prices no Western producer could match.
Litinsky thought the conventional wisdom was wrong.
Eight years later, on July 10, 2025, the United States Department of Defense purchased $400 million of preferred stock in the company he had built around that mine. The Pentagon became its single largest shareholder. The Secretary of Defense personally announced a ten-year guaranteed price floor on the company’s primary product. Five days after that, Apple signed a $500 million long-term supply agreement.
By May 2026, the company is worth approximately $10 billion in public-market capitalization. The man who paid $20.5 million on courthouse steps had built — or, more accurately, rebuilt — what is, today, the single most strategically important industrial asset in the Western Hemisphere.
This letter is about that company. And about why, despite an apparent 50x return already in the rearview mirror, we believe the most asymmetric play in this complex is no longer that company itself — but a set of second-tier names that the market has not yet rerated.
It is also about a quote from 1992, a lake in northern China that you can see from space, a marriage in Beijing that nobody outside the rare earth world has ever connected to U.S. national security, and the most aggressive piece of American industrial policy since the Manhattan Project — which most readers of the financial press have, somehow, almost entirely missed.
By the time you finish reading, you will understand why we believe the window for positioning against this thesis is measured in months, not years.
We start, as these stories often do, with a quote.
“The Middle East has oil. China has rare earths.”
In January 1992, the most powerful man in China climbed onto a train.
Deng Xiaoping was eighty-seven years old. He had been the de facto leader of the People’s Republic for fourteen years. Three years earlier, the Tiananmen Square massacre had triggered global outrage and a partial freeze of foreign investment into China. Deng’s reform program — the one that would, over the following thirty years, lift more human beings out of poverty than any economic policy in human history — was in genuine trouble.
So he climbed onto a train.
The trip is now called the Southern Tour. Over the course of about a month, Deng traveled through Shenzhen, Zhuhai, Guangzhou, and several inland cities, giving impromptu speeches to local officials. He told them to keep building, keep opening, keep reforming. The Soviet Union had collapsed two weeks earlier. China, he said, would not collapse — if China kept moving.
One of his stops was Baotou, a city in Inner Mongolia about 400 miles west-northwest of Beijing.
Baotou is not a tourist destination. It is a steel town built on a desert plain, with prevailing northwest winds that bring dust storms 46 days a year. It exists because, sixty miles to the north, in the foothills of the Gobi Desert, sits the largest rare earth deposit ever discovered — the Bayan Obo mine, which holds nearly 60% of the world’s known rare earth reserves.
Standing in Baotou, in January 1992, Deng Xiaoping is supposed to have said one of the most consequential things any leader has said about industrial strategy in the past half century:
“The Middle East has oil. China has rare earths.”
It is worth pausing on what Deng was actually saying.
He was not saying “China is rich in rare earths.” That would have been a banal observation. He was saying, by direct analogy, that rare earths would be to the twenty-first century what oil was to the twentieth — a strategic instrument. Not a commodity. An instrument.
OPEC had, in living memory, brought the United States to a halt in 1973. The price of oil had quintupled. Gas lines stretched for blocks. The strategic vulnerability of the dependent country was so absolute that the U.S. had been forced to build the Strategic Petroleum Reserve, sign the Camp David Accords, and reorient its entire Middle East foreign policy around a single fact: somebody else had what it needed, and could turn the tap off.
Deng was telling the Communist Party that China was about to be in that position. On the other side.
The Party listened.
Over the next two decades, in a sustained, methodical, lavishly-funded campaign that almost no one in the West paid serious attention to, China built — element by element, stage by stage — the most concentrated industrial monopoly in modern economic history.
The Bayan Obo mine was expanded. The Baotou Iron and Steel Group, the local state-owned champion, was given subsidies and credit lines to expand into rare earth refining. The Baotou Rare Earth Research Institute was massively funded to develop separation technology. State-owned enterprises were instructed to acquire foreign rare earth companies and bring their technology home.
One acquisition in particular is worth knowing about.
In 1995, a Chinese state-owned consortium acquired Magnequench, an Indiana-based company that had been spun out of General Motors and that owned the patents on the most important magnet technology of the late twentieth century — sintered neodymium-iron-boron permanent magnets, the magnets that today make every electric vehicle, every wind turbine, and every guided missile work. The deal was approved by the Committee on Foreign Investment in the United States (CFIUS) on the condition that the Indianapolis manufacturing facility remain in operation for at least five years.
The five years passed.
In 2003, the entire Magnequench operation — every machine, every patent, every blueprint — was physically packed up, shipped to China, and reassembled in the city of Tianjin. The new operation was led by Archibald Cox Jr., son of the Watergate special prosecutor. Behind him, an executive named Wu Jianchang took the senior Chinese position.
Wu Jianchang was Deng Xiaoping’s son-in-law.
It is not an exaggeration to say that the most important industrial transfer of the 1990s — the one that put the West roughly thirty years behind on the magnet technology its modern military, automotive, and energy industries would depend on — was effectively closed by a member of Deng Xiaoping’s own family.
By 2010, the structural result was complete.
China accounted for roughly 90% of global magnet rare earth production. It controlled more than 90% of global rare earth refining and separation. It produced approximately 95% of the world’s neodymium-iron-boron permanent magnets. In the heavy rare earths — the small but absolutely critical elements like dysprosium and terbium that make magnets work at high temperatures — China and its supply tributary in Myanmar controlled above 98%.
The deal Deng had described in 1992, in a single sentence, in a Mongolian steel town, was complete.
Then, in September 2010, China decided to demonstrate what it had built.
The lake you can see from space
To understand what happened in September 2010, you have to first see Baotou as it exists today.
About six miles north of central Baotou, in the floodplain of the Yellow River, sits the Weikuang Dam. It is also called, by the residents of the small farming villages around it, “the tailings lake.”
It is not a lake in any natural sense. It is a man-made impoundment, roughly twelve square kilometers in area, three and a half kilometers long and three kilometers wide. It is filled with the slurry waste of seventy years of rare earth refining. Every tonne of rare earth oxide produced upstream generates an estimated seventy-five tonnes of acidic wastewater, contaminated with heavy metals, processing chemicals, and trace radioactive thorium. All of it ends up here.
The dam was built in the 1950s, before the West developed the modern industrial practice of putting a thick waterproof liner underneath such impoundments. Baotou’s lake has no liner. It has had no liner for seventy years. The slurry seeps directly into the regional groundwater table.
Chinese environmental researchers — Chinese researchers, publishing in Chinese journals — have measured thorium concentrations in the dam at levels approximately thirty-five times higher than in the surrounding soil. They have linked the dam to crop failures in the villages downwind. They have documented intellectual development disorders in children. The reservoir, according to one of the few BBC reporters ever allowed inside the industrial zone, “smells of sulphur and looks like hell on Earth.”
You can see the Weikuang Dam from satellites. You cannot photograph it from the ground without security stopping you.
Here is why it matters.
Every iPhone you have ever owned. Every Android phone, every Tesla, every Apple Watch, every wireless earbud, every hard drive larger than 500 gigabytes, every wind turbine built since 2005, every electric vehicle, every guided missile fired by the United States military in the past fifteen years, every MRI machine that diagnosed your father’s tumor or your mother’s stroke, every drone that has ever delivered an Amazon package, every speaker in every Bose headphone — almost every single one of these objects contains rare earth magnets that, on their way to becoming finished products, passed in some form through the watershed that ends in the Weikuang Dam.
This is the cost the West externalized. This is what not having a rare earth industry meant.
In September 2010, the Chinese government decided to demonstrate that the externality came with leverage.
A Chinese fishing trawler had collided with two Japanese Coast Guard patrol boats near the disputed Senkaku Islands. The captain was arrested. Diplomatic relations deteriorated. And, without an official announcement, China stopped shipping rare earths to Japan.
The embargo lasted approximately two months. Japanese auto manufacturers — Toyota, Honda, Nissan — held emergency board meetings. Prices of certain rare earth elements rose by factors of three, six, ten. Hitachi, which had built the world’s last large stockpile of these materials in the 1990s, became briefly, accidentally, the most strategically important Japanese company outside Toyota.
Then the embargo ended. The captain was released. Trade resumed. Prices, over the following two years, normalized.
And the West, faced with the most visible demonstration of strategic vulnerability since the 1973 oil shock, did almost nothing.
By 2015, when the U.S. Office of the Secretary of Defense was confronted with the fact that the F-35 stealth fighter program — the most expensive weapons program in human history, with a lifecycle cost north of $2 trillion — could not be completed on schedule because there were no available American suppliers of the neodymium magnets that move its rudders, the Office signed a waiver. The most advanced fighter aircraft in the world would, going forward, contain Chinese parts. The Pentagon was unhappy about it. The Pentagon also had no choice.
Then the world moved on.
The mine that came back from the dead
By 2017, a hedge fund manager in Chicago had been studying the rare earth industry for nearly seven years.
James Litinsky had not started his career thinking about minerals. He had a JD and an MBA from Northwestern. He had spent the 2000s at a series of hedge funds, eventually founding his own firm, JHL Capital Group, in 2006. He had survived 2008 — barely — and had spent the years afterward looking for the kinds of contrarian, mispriced situations that hedge fund managers of his generation had been trained to find.
In 2010, in the immediate aftermath of the Japanese rare earth embargo, JHL Capital took a small position in Molycorp, the California rare earth company that was attempting to revive Mountain Pass. The investment thesis was simple. The 2010 embargo, Litinsky believed, was not an isolated incident. It was a preview. The strategic vulnerability of the West would, sooner or later, become impossible to ignore.
Molycorp went bankrupt in 2015.
A lesser investor would have written off the position and moved on. Litinsky did something stranger. He spent the next two years buying up the bankruptcy claims against Molycorp, putting himself in a position where his consortium would have effective control over the mine’s mineral rights when the bankruptcy proceedings concluded.
In the summer of 2017, the proceedings did conclude. At the final auction, on the steps of a Delaware courthouse, a competing group of creditors moved to push Mountain Pass into formal reclamation — meaning it would be permanently shut down, the equipment scrapped, the site bulldozed and returned to its pre-mining environmental state. Litinsky says the consortium had perhaps “a few million dollars” in liquid funds, on the day, to keep the operation alive. They wrote the check. They won the auction. They paid $20.5 million for the entire property.
At the time of the purchase, Mountain Pass had eight employees.
Eight.
For the next four years, the work was largely invisible. The company restarted refining. It built new separation circuits. It developed in-house metallurgical capabilities. In 2020, MP Materials went public via a SPAC at an equity value of approximately $1.5 billion. Litinsky told investors a three-stage roadmap that was, at the time, considered ambitious to the point of implausibility.
Stage one: produce rare earth concentrate from Mountain Pass and sell it. Achieved 2022.
Stage two: build separation capacity to produce refined rare earth oxides — the actual high-value product. Achieved 2024.
Stage three: build downstream capacity to produce finished neodymium-iron-boron magnets — the final, highest-value link in the supply chain, which had not been manufactured at commercial scale in the United States in twenty years. Achieved at the company’s “Independence” facility in Fort Worth, Texas, by the end of 2025.
This is, to be clear, what the conventional wisdom in 2017 had said was impossible.
By the spring of 2025, MP Materials was the only producer in the Western Hemisphere doing all three stages, on one continuous chain, at industrial scale.
And then April 4, 2025 happened.
The April that broke the silence
On April 4, 2025, the Ministry of Commerce of the People’s Republic of China issued a notice that, to anyone who had been paying attention to the rare earth industry, marked the final and unambiguous answer to a question that had been hanging in the air since 1992.
The notice imposed export controls on seven heavy rare earth elements — dysprosium, terbium, samarium, gadolinium, lutetium, scandium, and yttrium — along with all related compounds, all related metals, and all finished permanent magnets. The mechanism was a case-by-case licensing regime. Every export now required individual approval from Chinese regulators. The approvals could be granted, denied, or simply not processed, with no public criteria and no appeal.
The effect was immediate, and it was visible in factories you have probably driven past.
Within seventy-two hours, the procurement managers at Volkswagen’s Wolfsburg plant were on emergency calls. Within a week, Ford had quietly informed its top dealers that delivery dates on certain EV models would slip. Within ten days, the European Association of Automotive Suppliers issued a statement warning of “severe near-term disruption.” Several manufacturers — including major German auto OEMs — temporarily reduced production utilization or shut down assembly lines outright.
But 2010 had been a demonstration. 2025 was a policy.
The April 4 notice was not framed as a one-off action. It established an infrastructure — a licensing regime that could be tightened or loosened at the discretion of the Chinese government, on any timeline, against any target country, with no warning and no appeal. The licensing regime did not go away. It is still in effect as of the time of this writing.
On October 9, 2025, Beijing extended the same regulatory architecture to lithium-ion battery supply chains. The signal was unmistakable. China had decided that its dominance of critical mineral processing was no longer an economic asset to be quietly enjoyed. It was a strategic instrument, to be used selectively, with calibrated pressure, on a timeline of its choosing.
The thing Deng Xiaoping had described in 1992 was no longer theoretical.
And on July 9, 2025, in a conference room at the Pentagon, three men sat down to sign papers that would represent the first serious American answer.
The Pentagon writes a check
The deal that MP Materials and the U.S. Department of Defense announced on July 10, 2025, was unprecedented in modern American industrial policy.
The DoD purchased $400 million of newly-created preferred stock in MP Materials, convertible into common stock at $30.03 per share. On a fully-converted basis, the Pentagon became the single largest shareholder of MP Materials, with an effective economic stake of roughly 15%.
That was the equity piece.
The commercial piece was, if anything, more remarkable. The DoD guaranteed a ten-year price floor of $110 per kilogram for MP’s neodymium-praseodymium oxide. If the market price falls below that level — which it does, with some regularity, because China can flood the global market at any time — the U.S. government writes a check to MP for the difference. In exchange, the government takes 30% of any profits MP generates above the floor.
This single provision is, in our view, the most strategically significant clause of the deal. It immunizes MP’s margins against the one weapon China has consistently used against Western producers for thirty years: the sudden release of subsidized inventory that crashes the global spot price. The Pentagon, in effect, has told China: do that to MP, and you are taking money out of our pocket. We will not let it work.
The third piece was the offtake. The DoD committed to purchase 100% of the output of MP’s planned 10X magnet manufacturing facility in Northlake, Texas, for a minimum of ten years.
Five days after the Pentagon deal was announced, Apple signed.
On July 15, 2025, Apple committed to a $500 million long-term supply agreement with MP Materials, including the co-development of a recycled-magnet production line at Mountain Pass and dedicated magnet manufacturing capacity at the Fort Worth Independence facility. The offtake was structured to support “hundreds of millions” of iPhones, iPads, and MacBooks starting in 2027.
The signal, again, was unmistakable. The Pentagon’s check was about national security. Apple’s check was about commercial supply chain resilience. They had arrived at the same answer from opposite directions.
By the end of 2025, MP Materials’ equity had approximately quadrupled. By mid-May 2026, the company trades around $55-60 per share, with a market capitalization of approximately $10 billion. The high in the 52-week range was set during the rally that followed the Pentagon deal. The current price reflects a market that has partly absorbed the implications of what happened in July 2025, but in our view has not yet absorbed the structural endpoint.
What the rest of the West is finally doing
The Pentagon deal was not an isolated event. It was the most visible signal in a much larger pattern.
In February 2026, the Trump administration announced Project Vault, a first-of-its-kind strategic critical minerals reserve pooling $2 billion in private capital with a $10 billion Export-Import Bank loan — a $12 billion stockpile mechanism explicitly designed to address U.S. critical minerals vulnerability. The announcement triggered double-digit single-day moves across the rare earth complex.
Lynas Rare Earths, the Australian operator of the Mount Weld mine and the only commercial-scale rare earth processor outside China, became the only commercial producer of separated heavy rare earth oxides outside China in 2025.
Energy Fuels, primarily known as a uranium producer, achieved pilot-scale production of 99.9% pure dysprosium oxide at its White Mesa Mill in Utah in 2025 and is targeting commercial-scale heavy rare earth production by late 2026.
USA Rare Earth raised $1.5 billion in early 2026 and acquired Serra Verde, building toward a global mine-to-magnet platform.
NioCorp Developments holds one of the largest undeveloped reserves of heavy rare earths in North America at its Elk Creek project in Nebraska, including significant scandium production capacity.
Ucore Rare Metals is commercializing its proprietary RapidSX separation technology, building the Strategic Metals Complex in Louisiana backed by an $18.4 million DoD contract.
The European Union, under its Critical Raw Materials Act, has classified rare earth magnets as a Strategic Raw Material, targeting 40% of EU consumption to be processed domestically by 2030.
Sum the announced government programs. Sum the private capex commitments. Sum the offtake contracts. Sum the strategic stockpile mandates. Sum the parallel build-outs in Korea, Japan, Saudi Arabia, the United Kingdom, France, Australia, and Canada.
Somewhere between $50 billion and $100 billion of Western capital expenditure is going to flow into rebuilding a mine-to-magnet supply chain that has been allowed to atrophy since the 1980s, on a timeline of less than ten years.
The equity market has begun to express the beginning of this rerate — MP at $10 billion, Lynas with its multi-year run, the second-tier names beginning to move on Project Vault headlines. But, in our view, the market is still pricing yesterday’s world, not the structural endpoint.
Why the check writers will keep writing
There is one more thing about this story that, in our view, the market has not fully internalized.
In 2015, when the Pentagon signed the F-35 waiver, it did so because it had no alternative and because senior defense officials assumed — reasonably, at the time — that the strategic dependency on China for rare earth magnets would be solved through normal industrial development over a five-to-ten-year horizon.
Ten years later, the dependency is worse. China’s share of global rare earth refining is approximately 90%. Its share of permanent magnet production is approximately 95%. The 2010 demonstration was a warning. The 2025 export controls were the demonstration becoming policy. And the F-35, the Arleigh Burke destroyer, the Virginia-class submarine, the Tomahawk cruise missile, the Patriot missile defense system — every major American weapons platform deployed in the past twenty years contains, in some form, components dependent on Chinese rare earth supply.
A single Lockheed Martin F-35 Lightning II contains approximately 920 pounds of rare earth elements.
A single Arleigh Burke-class destroyer contains 5,200 pounds.
A single Virginia-class nuclear submarine contains 9,200 pounds.
Pentagon annual consumption of rare earth permanent magnets is currently estimated at 3,000 to 4,000 tonnes. High-performance magnet requirements specifically are projected at 1,200 tonnes by 2030. Until the new American facilities ramp, domestic American production is effectively zero.
This is why the deal MP Materials signed was structured the way it was — not as a subsidy, not as a procurement contract, but as a strategic equity participation, with a guaranteed price floor and a ten-year offtake commitment. The Pentagon is not building a market. It is building a strategic asset, with the same logic that built the interstate highway system in the 1950s and the lunar program in the 1960s.
And that means it is durable in a way that ordinary industrial policy is not.
A future administration may slow the pace. It will not reverse the direction. The structural fact that the United States cannot, in 2026, manufacture its own most advanced weapons system without help from its principal strategic adversary is, simply, too uncomfortable to leave unaddressed. Every appropriations cycle from this point forward will, in our view, extend rather than retract the framework that was put in place in July 2025.
The window
If you have read this far, you understand the setup.
You understand the forty-year arc that led from Deng Xiaoping’s quote in 1992, through the quiet Chinese acquisition of the Magnequench patents in the late 1990s, through the 2010 demonstration against Japan, through the 2015 F-35 waiver that no Pentagon official wanted to sign, to the April 4, 2025 export controls that no major Western OEM was prepared for, to the July 10, 2025 Pentagon equity stake.
You understand who is positioning. You understand why. You understand the numbers — the $400 million Pentagon preferred, the $500 million Apple offtake, the $110 per kilogram floor, the $12 billion Project Vault stockpile, the $50-to-$100 billion capex cycle being committed to in real time across the Western alliance.
What you do not yet have is the operating framework.
MP Materials has already moved from $1.5 billion to $10 billion. The headline trade is mostly done. The question now is different — and harder. Where, in the rest of the rare earth complex, is the next 5x-10x asymmetry? Which of the second-tier names — the Energy Fuels, the USA Rare Earths, the NioCorps, the Ucores — actually deserve capital, and which are momentum chasers that will round-trip on the next news cycle? At what price points? Sized how, against what hedges? What are the catalysts that will move these equities over the next twelve months — and which ones do we already see on the calendar?
What is the bear case — the scenario where US-China relations stabilize, China releases inventory, and the entire complex compresses 40-50%? How do we hedge that scenario?
These are the questions the second half of this letter answers.
Our conviction: the rerate of the second-tier rare earth names will play out over the next 12-18 months, on a sequence of identifiable catalysts — Project Vault deployment, the first commercial-scale heavy rare earth production milestones, additional Pentagon equity deals on the MP Materials template, and the EU Critical Raw Materials Act implementation milestones. When those catalysts have fired, the names will be on the cover of Barron’s. By then, the rerate will already have happened.
Macro Notes Premium
The remainder of this letter, reserved for Macro Notes Premium subscribers, includes:
A four-name basket covering MP Materials at current levels plus three second-tier candidates with multibagger asymmetry, with sizing weights, entry zones, and trim levels. A quarter-by-quarter catalyst calendar through Q1 2027.
Three valuation scenarios — bear, base, bull — for every position.
The terrestrial-mining short basket we use to neutralize sector beta. An options framework that can be layered on top of the common-equity position to magnify the asymmetry on the smallest names.
As always with Macro Notes Premium: not a tipsheet. A constructed thesis, verifiable numbers, and a methodology you can apply to your own portfolio.
In 5 days, the price of Macro Notes Premium rises from $180 to $220 for the first year (with renewals at $300 per year). You can still lock in the current rate — and once subscribed, your price will never change.