With China's sweeping rare earth export controls set to expire on November 10, Washington has responded with a flurry of executive action: a Defense Production Act rule locking black mass inside U.S. borders, half a billion dollars in DOE processing grants, and a $12 billion strategic reserve. But across each of these initiatives runs the same structural fault line: the United States is better at controlling what it has than building the refining and separation infrastructure needed to make any of it usable. The September 24 Trump-Xi summit may buy time. It cannot buy capacity.
Introduction
The loading dock at a battery recycling facility outside Columbus, Ohio, does not look like a front line in a global supply chain war. Forklifts move pallets of shredded lithium-ion scrap, a dark, granular material known in the industry as black mass, toward sorting conveyors that hum under fluorescent light. The facility manager, who asked not to be identified by name while his company navigates new federal compliance requirements, points to a corner of the warehouse where several hundred metric tons of processed scrap have sat for three weeks. His buyers in South Korea and Japan are no longer permitted to receive it. His domestic customers do not yet have the refining capacity to absorb it. The material is, for now, going nowhere.
This is the paradox at the center of American critical mineral policy in the autumn of 2026. Washington has, in the span of roughly eight weeks, invoked the Defense Production Act to mandate that all black mass and tungsten scrap stay inside U.S. borders, pledged $500 million to expand domestic processing, and unveiled Project Vault, a $12 billion strategic reserve modeled loosely on the Strategic Petroleum Reserve. President Trump will sit across from Xi Jinping in Washington on September 24 knowing that China's sweeping rare earth export controls, suspended as part of the Busan summit deal last October, expire on November 10, with a second deadline for gallium, germanium, and antimony falling on November 27.
The diplomatic stakes are real. The IEA estimates that full reimposition of China's controls could put $6.5 trillion in annual downstream production at risk for countries outside China, with the United States and Europe together accounting for nearly half that exposure. But a harder question has been surfacing in industry analyses and government accountability reviews alike: even if the summit produces another suspension, even if the deadlines roll over, has Washington built anything structural enough to matter? The answer, traced across three converging policy developments this summer, is uncomfortable. The United States is getting better at controlling what it has. It remains deeply underprepared to refine, separate, and manufacture what it needs.
The Truce That Kept Moving
When Trump emerged from the October 2025 Busan summit declaring rare earths were "settled" and there was "no roadblock at all," the rare earth market briefly exhaled. Beijing had agreed to suspend its October 9, 2025 expansion of export controls, which would have broadened restrictions to cover samarium, gadolinium, lutetium, europium, and ytterbium, and extended jurisdiction to any foreign-made product containing at least 0.1% by value of Chinese-origin rare earth inputs. That extraterritorial provision alone was, as I noted in my September analysis "Forty-Seven Days," an ecosystem-level lever capable of touching semiconductor fabs, aerospace suppliers, and wind turbine manufacturers far beyond U.S. borders.
The suspension, however, never meant withdrawal. The export controls China imposed in April 2025 on heavy rare earth elements, including dysprosium, terbium, and yttrium, remained fully active throughout the truce period. BMI data show exports of yttrium, dysprosium, and terbium running at just 42%, 41%, and 49% respectively of volumes recorded in the twelve months before the restrictions. CSIS research published in May 2026 found that yttrium exports to the United States fell from more than 333 metric tons in the eight months before the April controls to just 17 metric tons in the eight months after, with aerospace manufacturers reporting shortages of a material critical for turbine blade coatings.
China did not merely hold the line during the truce. In June 2026, Beijing added ten U.S. entities to its export control list, including MP Materials and USA Rare Earth, the two companies at the core of Washington's domestic rare earth ambitions. A month later, on July 24, MOFCOM placed 14 EU entities on the same list, among them Rheinmetall AG and Lafert S.p.A. The pattern is unmistakable: even as diplomatic channels remained technically open, China was narrowing the list of permitted trading partners with precision, using mineral access as a scalpel rather than a sledgehammer. Analysts at BMI concluded bluntly that China's stranglehold on global rare earth supply is expected to "stay firmly in place" following any summit outcome.
Now the clock is running openly. The September 24 summit arrives with 47 days until the November 10 deadline and 64 days until the November 27 gallium and germanium expiry. Neil Thomas, among others, has described China's posture as a continuing "supply chain war," and the market pricing reflects it: terbium reached $4,150 per kilogram by April 2026, yttrium climbed to $1,000 per kilogram, driven by defensive procurement from Japanese and North American aerospace buyers who have learned, the hard way, not to trust diplomatic assurances about material availability.
Locking the Barn Door: The Black Mass Rule
Washington's most striking domestic response arrived not from Congress but from the Bureau of Industry and Security, moving with unusual speed. On July 30, 2026, President Trump signed a Presidential Determination under Section 101 of the Defense Production Act, finding that recoverable critical minerals and materials were scarce and essential to national defense. Within two business days, a pre-publication copy of the implementing rule appeared in the Federal Register. By August 6 it was final. By August 27 it was in force.
The rule does something without clear recent precedent in American trade law. It requires U.S. persons engaged in the sale of black mass, defined as shredded lithium-ion battery scrap containing cathode materials such as lithium, cobalt, nickel, and manganese, as well as anode materials and other residual cell components, to allocate 100% of their monthly sales to U.S. buyers. Tungsten waste and scrap falls under the same mandate. Any exception requires affirmative BIS authorization, with the agency promising 14-day response windows. The rule expires August 27, 2027, though BIS has signaled it may issue additional allocation orders as conditions warrant.
The logic is straightforward and, from a national security standpoint, defensible: if the United States is generating 60,000 metric tons of black mass annually but domestic refining capacity can only process 25,000 metric tons of that material, then the remaining 35,000 metric tons, which had been flowing largely to South Korean and Japanese processors, represents recoverable strategic value being surrendered to foreign supply chains. Keeping it inside U.S. borders, the argument goes, creates the feedstock certainty that domestic refiners need before committing to capacity expansion. Jeet Basi of Modern Mining captured the sequencing logic precisely: "Financing and building the refining capacity to process it is the necessary second step."
Industry reaction has been broadly supportive among domestic processors. Amermin, a tungsten carbide manufacturer, called the rule a necessary step toward giving domestic processors "reliable and consistent sources" of scrap, and projected that its planned full-campus buildout could supply roughly 18% of estimated annual U.S. tungsten consumption. But the enthusiasm is not universal. Recyclers that had built export revenue into their business models now face a compliance scramble, and the Holland and Knight analysis of the rule notes a significant regulatory convergence: materials long treated as waste streams under RCRA and TSCA are now simultaneously subject to defense production allocation requirements, creating overlapping compliance obligations that smaller operators may struggle to navigate.
The deeper problem is what the rule does not solve. Mandating domestic retention of black mass does not conjure the hydrometallurgical plants, solvent extraction circuits, and cathode-active material facilities needed to transform that scrap into battery-grade lithium carbonate, cobalt sulfate, or nickel sulfate. The 35,000-metric-ton gap between generation and refining capacity is not a legal problem. It is an infrastructure problem, and no allocation order builds a plant.
Half a Billion Dollars and a Feedstock Wall
On August 20, the Department of Energy announced $500 million in awards to seven processing and recycling projects, the latest in a sequence of funding rounds that has also included $162 million for nine critical mineral recovery projects, $134 million for two rare earth refining facilities, $75 million for coal-feedstock recovery, and $45.7 million across nineteen processing technology projects. Since 2020, the Department of Defense has separately committed approximately $439 million to rebuilding domestic rare earth supply chains, including a $400 million Pentagon equity stake in MP Materials and a $1.6 billion conditional commitment to USA Rare Earth.
Three of the seven DOE August projects are battery recycling or reprocessing facilities, placing a significant share of this funding round in the circular economy rather than primary mining. Lilac Solutions received $100 million to develop a commercial lithium extraction facility at the Great Salt Lake in Utah. DOE Assistant Secretary Audrey Robertson told Reuters that hundreds of applications had been received and the selected projects were those offering the highest potential return in supply chain priority areas. Energy Secretary Chris Wright offered the animating philosophy: "For too long, America has depended on foreign actors for critical materials essential to modern life."
But the GAO's July 2026 assessment of battery recycling introduced a qualification that has not received adequate attention in Washington's policy conversation. Battery recycling is, the GAO found, the strongest near-term option for reducing U.S. dependence on imported lithium, cobalt, nickel, and copper, with established technologies capable of reducing imports within two to three years. The limiting factors, however, include inadequate domestic refining capacity and the continued loss of usable feedstock to landfills and overseas processors. The same structural problem identified in the BIS black mass rule reappears here: collection and generation are outpacing the ability to do anything useful with the material.
Moreover, the GAO's 11-mineral scope excludes rare earth magnets entirely, leaving the segment most exposed to Chinese leverage, and most critical for defense, EV motors, and wind turbines, outside the near-term recycling calculus. The IEA's 2026 minerals outlook made the broader investment skew explicit: project pipelines outside dominant supplier regions are heavily weighted toward mining development, with refining and downstream processing capacity lagging significantly behind. Cumulative planned production of metals, alloys, and finished magnets from projects announced as of early 2026 amounts to around 18 kilotonnes on a rare earth element content basis, roughly one-third of anticipated needs. As Jack Lifton, co-chair of the Critical Minerals Institute, argued in a widely circulated InvestorNews essay this summer, U.S. policy continues to over-index on mining and under-invest in the downstream refining and manufacturing stages where the real strategic value, and China's dominant advantage, actually resides.
Project Vault and the Stockpile Illusion
The most ambitious single instrument in Washington's critical mineral toolkit is Project Vault, announced on February 2 and backed by up to $10 billion in EXIM financing, the largest direct loan in the agency's 92-year history, combined with approximately $2 billion in private capital. Modeled on the Strategic Petroleum Reserve established after the 1973 Arab oil embargo, Project Vault is designed as a demand-driven reserve spanning all 60 minerals on the USGS 2025 Critical Minerals List, with procurement guided by long-term purchase commitments from manufacturers rather than centralized government forecasting. Commodity traders Mercuria, Traxys, and Hartree Partners serve as core procurement managers; General Motors and other major consumers anchor the demand side.
The case for a reserve is not difficult to make. In 2025, Ford was forced to halt production of its Explorer model due to a rare earth shortage, a vivid illustration that supply disruptions have migrated from defense procurement planning into civilian manufacturing. The IEA's stockpiling recommendation, released alongside its $6.5 trillion risk estimate, called for an initial multilateral purchase of $9.2 billion in 11 high-risk materials, at a net annual carrying cost of roughly $900 million, arguing that those figures are "quickly dwarfed" by the potential impact of supply disruptions.
Yet critics of Project Vault have identified a tension that its architects have not fully resolved. A stockpile is a buffer against supply shocks; it is not a substitute for the separation, metallization, alloying, and magnet-manufacturing capacity that the United States currently lacks. China's ability to exercise leverage does not rest solely on its control of raw ore in the ground. As of mid-2026, China continues to process approximately 99% of the world's heavy rare earths. A reserve stocked with separated oxides or metals can cushion a short-term disruption. It cannot replace a missing domestic industry if controls are sustained for a year or more, which is precisely the scenario the November 10 deadline now forces planners to contemplate.
The processing gap also raises sourcing questions for the reserve itself. Project Vault is designed to hold processed and refined materials, not raw ore. But with China controlling approximately 90% of global rare earth processing capacity, acquiring meaningful volumes of specification-grade separated rare earth materials from non-Chinese sources, at the scale and speed the reserve's ambitions require, runs directly into the same midstream bottleneck constraining every other element of U.S. policy. The reserve needs the processing capacity to exist. The processing capacity needs the reserve's purchase commitments to justify construction. The circularity is not lost on the project's own advisors.
The Architecture of a Structural Problem
Building on my analysis in "The Price of Security" from earlier this month, the emerging Western policy architecture, price floors, quota tools, strategic reserves, export mandates, and bilateral diplomacy, is coherent in its parts but incomplete in its logic. Each instrument addresses a symptom of Chinese market dominance. None yet addresses the underlying cause: the absence of a competitive non-Chinese refining and separation sector capable of operating at scale, at cost, and at commercial quality.
The BIS black mass rule, the DOE funding rounds, Project Vault, and the diplomatic calendar around the Trump-Xi summit are all responding, in different registers, to the same structural reality that Beijing spent two decades building deliberately. China's dominance did not emerge from mining supremacy alone. It was constructed through sustained state investment in the unglamorous middle stages of the supply chain: hydrometallurgical processing, solvent extraction, rare earth separation, alloy production, and magnet sintering. Individual rare earth processing facilities can require hundreds of millions of dollars. They generate waste containing naturally occurring radioactive material. They consume substantial water. They take years to permit and build. And they face a market in which Chinese producers can influence benchmark prices through output policy, as they demonstrated repeatedly between 2014 and 2023 in ways that made Western investment economics precarious.
The November deadlines create pressure but not capacity. Even the most optimistic projections from the IEA suggest that mining capacity outside China could cross 50 kilotonnes of rare earth element content by 2035, led by Australia and the United States. But refining and separation capacity amounts to less than 40 kilotonnes on the same horizon, concentrated in Malaysia, the United States, Australia, Vietnam, Japan, the United Kingdom, France, and Estonia. A decade from now, on the most favorable assumptions, the West will still have a processing gap.
Washington's hard-retention mandate for black mass, its funding rounds for recycling facilities, and its diplomatic push for a renewed mineral truce in Washington this week are all necessary steps. But they are first steps toward a destination that requires the kind of sustained, decade-long industrial policy commitment that democracies find structurally difficult to maintain across electoral cycles. The EU's ReSourceEU experience, as I documented in "Paper Ambitions," offers a cautionary parallel: ambitious targets, financing gaps, permitting delays, and an auditors' verdict of "serious risk." The instruments Washington has assembled this summer are more muscular than Brussels managed. Whether they are durable enough to outlast the next summit, the next suspension, and the next deadline is a different question.
Conclusion: The Dock in Columbus
The facility manager outside Columbus checks his compliance calendar against the BIS allocation rule and finds that it answers one problem while exposing another. His black mass stays in the United States, as the law now requires. But the refiner two states away that might process it is operating at 60% of nameplate capacity because its feedstock contracts were designed for a world in which South Korean competition set the floor price. The investment committee that would approve a capacity expansion is waiting on a longer-term policy signal than a one-year allocation order provides. The material sits.
On September 24, Trump and Xi will meet in Washington with a diplomatic agenda spanning trade, technology, Taiwan, and what the State Department calls "strategic stability." Rare earths and critical minerals will be somewhere in that agenda, described, almost certainly, in the language of cooperation and mutual benefit. If history is a guide, a new suspension may be announced, a new window opened, another set of deadlines extended. Markets will briefly exhale.
But the harder accounting will continue regardless of what emerges from that meeting. China controls approximately 99% of global heavy rare earth processing. U.S. companies generate 60,000 metric tons of black mass annually and can refine fewer than half of them. Planned non-Chinese magnet production meets roughly one-third of projected needs by the mid-2030s. A diplomatic truce, as the past twelve months have demonstrated with uncomfortable clarity, cannot substitute for separation plants, metallization capacity, alloying infrastructure, or qualified magnet production that remains, in any meaningful commercial quantity, absent from the Western supply chain.
The pallet in Columbus is a small emblem of a large problem. America has gotten better, in this season of executive action, at keeping what it has inside its borders. What it has not yet built is the industrial chain capable of turning what it keeps into what it needs. That gap, measured not in deadlines but in decades, is the real terrain on which this competition will be decided.