On January 1, 2027, a provision buried in the National Defense Authorization Act expands from a narrow restriction into a full mine-to-magnet ban. Through the end of 2026, defense contractors cannot deliver magnets that were “melted or produced” in China. Starting New Year’s Day, that prohibition covers every stage — mined, refined, separated, melted, or produced. If the neodymium in a magnet was pulled from the ground in Baotou, the magnet fails compliance. It does not matter where it was sintered or who shipped it.
That single sentence is why critical minerals became one of the most heavily promoted investment themes of 2026. Multiple newsletter publishers — Paradigm Press, Brownstone Research, Oxford Club, TradeSmith — have pointed at the same structural fact from different angles. Some frame it as a defense play. Some as an AI infrastructure play. Some as a deep-sea mining or geothermal energy angle. The underlying thesis is identical: the United States is 90 percent import-reliant on roughly 25 critical minerals, and Washington has decided that ends now.
The scale of the dependency
China controls approximately 69 percent of global rare earth mining and an estimated 90 percent of processing capacity, according to USGS data. For neodymium-iron-boron magnets — the components inside F-35 actuators, Tomahawk guidance systems, EV traction motors, wind turbines, and industrial robots — China’s share runs above 90 percent. For heavy rare earths like dysprosium and terbium, which keep magnets from demagnetizing at the temperatures inside a jet engine or a missile seeker, the concentration is even more extreme. There is no commercial-scale substitute.
The Department of Defense’s own supply-chain analysis estimates an F-35 production line could halt within six to eleven months if magnet supply were cut. That timeline is measured in months, not years. It is the strategic anxiety behind every dollar Washington has deployed since 2023.
The U.S. government maintains a formal critical minerals list, updated periodically by the USGS, covering materials deemed essential to the economy and national security where supply chains are concentrated in few countries and substitutes are limited. The 2025 list runs 50 entries deep — from rare earths to antimony, tungsten, gallium, germanium, graphite, and uranium. The U.S. is over 50 percent import-reliant for more than half of them.
How Washington is spending to fix it
The federal response has moved from policy statements to capital deployment through multiple channels. The Department of Defense has invoked Defense Production Act Title III to fund domestic processing. The Department of Energy’s Loan Programs Office has backed mine-to-manufacturing projects. The Export-Import Bank has become a major financing engine. The Inflation Reduction Act added tax credits for domestically sourced battery materials.
The most aggressive intervention came in July 2025, when the Pentagon bought $400 million in newly created preferred stock in MP Materials — convertible to roughly 15 percent of the company. That made the Department of Defense the largest shareholder in America’s only operational rare earth mine. Layered on top: a 10-year contract-for-difference pinning neodymium-praseodymium at $110 per kilogram, with the Pentagon paying any shortfall below that floor and capturing 30 percent of the upside above it. A $150 million loan for heavy rare-earth separation at Mountain Pass, California. A 10-year commitment to buy 100 percent of the magnets produced at a planned Texas campus.
That package is the financial spine of “10X,” a roughly $1.25 billion magnet manufacturing campus in Northlake, Texas. The Pentagon became MP’s largest shareholder, its price insurer, and its guaranteed customer in a single transaction.
The FY2027 defense budget request pre-commits $48.675 billion to Critical Minerals as a single appropriation line. That breaks into $24.3 billion for the Industrial Base Analysis and Sustainment program, $6.4 billion in Defense Production Act purchases for mining and processing, and $18 billion to rebuild the National Defense Stockpile. An additional $17.5 billion Golden Dome program — covering interceptor and radar production across Patriot, SM-6, THAAD, and a new space-based layer — draws from the same magnet stack.
The historical parallel that matters
In September 2010, a Chinese fishing trawler collided with two Japanese Coast Guard vessels near the Senkaku Islands. Japan detained the captain. China responded by cutting off rare earth exports to Japan. The embargo lasted roughly two months. It was the first time a major economy used rare earth access as a geopolitical weapon.
The episode terrified Tokyo and Washington. Japan’s Ministry of Economy, Trade and Industry scrambled to diversify supplies, funding Australian processor Lynas and investing in recycling. The U.S. Congress held hearings. The Department of Defense began studying its supply chain vulnerability. Fifteen years later, those studies became the NDAA provisions taking effect in January 2027.
The deeper parallel runs to the Strategic Materials Stockpile created after World War II. The U.S. government maintained reserves of critical raw materials — manganese, chromium, cobalt, tin — to sustain the nation through a prolonged conflict. That stockpile was drawn down and neglected through the post-Cold War era. The $18 billion rebuild in the FY2027 budget is Washington acknowledging that the stockpile strategy was correct and the neglect was a mistake.
The companies in the crosshairs
MP Materials is the closest thing to a national champion. The Mountain Pass mine in California is the only operational rare earth mine in the United States and the second-largest outside China. Q1 2026 NdPr production hit 917 tonnes, up 63 percent year-over-year. The Independence magnet facility in Fort Worth reached commercial production in late 2025 at roughly 1,000 tons per year. The 10X campus targets 10,000 tons annually — but commissioning begins in 2028, leaving a multi-year gap between the policy deadline and the industrial capacity it demands.
USA Rare Earth received up to $1.6 billion in federal funding through the Department of Commerce in June 2026, with the U.S. government taking 16.1 million shares and 17.6 million warrants. The Round Top project in Texas and the Stillwater magnet plant in Montana form the build plan. The company carries a formal going concern warning and has produced zero commercial output to date.
Lynas Rare Earths, the Australian processor, operates one of the few significant non-Chinese processing facilities. Energy Fuels has added mine-to-magnet vertical through its VAC acquisition. NioCorp is developing the Elk Creek project in Nebraska for niobium, scandium, and titanium, with a Defense Production Act Title III award for scandium supply chain work and a pending Export-Import Bank loan. Perpetua Resources holds the Stibnite gold-antimony project in Idaho, targeting domestic antimony production for defense energetic materials.
Each company sits at a different stage of the same build sequence. Some have binding government contracts. Most have non-binding term sheets. The 10-Q filings tell a story the press releases don’t: NioCorp is 27 percent funded for a $1.14 billion project. Perpetua is 32 percent funded for a $2.2 billion project facing two federal lawsuits. The dilution across the sector has been severe.
What the convergence tells you
When four or five publishers, operating independently, all point their subscribers at the same sector — critical minerals, rare earths, strategic materials — they are reading the same forcing function. The NDAA deadline is statutory law. The Pentagon’s equity stake in MP Materials is a matter of public record. The $48.675 billion in the FY2027 budget request is a line item Congress can read. The supply gap between U.S. capacity and defense-grade demand is measurable in the SEC filings of every company attempting to close it.
The investment thesis is not subtle. Washington has written into law a requirement that the defense supply chain be de-risked from China, the deadline is January 1, 2027, and the domestic production base does not exist at the scale required. The government is spending tens of billions of dollars to close that gap, and it is taking equity stakes in the companies building the solution.
The 10-Q filings, the NDAA text, and the Senate testimony from February 2026 confirm the structural problem. What remains open is which companies can execute on the timeline the law demands — and how much of the upside the government captures through its price floors, offtake guarantees, and equity stakes. The SEC filings tell a different story than the press releases — the filings show what the companies have actually done, not what they hope to do.