On the same day the European Union shortlisted tungsten, rare earths, and gallium for its first coordinated strategic mineral reserve, China published implementation regulations under its revised Mineral Resources Law that will take effect June 15, 2026, empowering Beijing to control total mineral output, restrict mining entities, and mandate five-year in-situ reserves. The simultaneity was not coincidence: it was a precise illustration of the structural trap now confronting Western industrial policy. Both the EU and the United States are racing to warehouse materials they can currently only source from the country they are trying to hedge against.
Introduction
Three policy developments published within days of each other in late May 2026 collectively define the current condition of the global critical minerals order. The European Union has shortlisted tungsten, rare earths, and gallium for its first jointly coordinated strategic mineral stockpile, with logistics discussions underway at the Port of Rotterdam. China's State Council has promulgated implementing regulations under its revised Mineral Resources Law, effective June 15, 2026, that impose total output controls, restrict mining entities, conduct mandatory security reviews on foreign investment in Chinese mining operations, and establish a minimum five-year in-situ reserve system. And the Peterson Institute for International Economics has published a policy brief concluding that Project Vault, the United States' $12 billion public-private stockpiling initiative, is structurally flawed in its current form and risks failing precisely when its insurance function is most needed.
Read in isolation, each development is significant. Read together, they describe a fundamental reorganisation of how states are beginning to treat critical minerals: not as commodities governed by comparative advantage and market pricing, but as strategic assets subject to the same logic of national accumulation that once governed oil, naval coal, and enriched uranium. The policy instruments being deployed on all sides, reserve mandates, export controls, price floors, and coordinated procurement, are not temporary crisis responses. They are the architecture of a new resource order.
The critical analytical point is not that stockpiling is occurring. It is that all three actors are attempting to stockpile from the same constrained supply pool, at the same moment, while the dominant supplier is simultaneously locking down both its export flows and its domestic extraction rates. The contradiction embedded in that sentence is the defining strategic problem of the second half of 2026.
China's June 15 Regulations: Constraining Supply from Both Ends Simultaneously
China's new implementing regulations, signed by Premier Li Qiang and published by the State Council, represent the most comprehensive formalisation of Beijing's mineral resource sovereignty framework since the original Mineral Resources Law was introduced in 1986. The regulations establish a three-tier reserve architecture covering physical product stockpiles, production capacity reserves for rapid emergency scale-up, and designated in-situ resource sites that cannot be mined without State Council approval. The minimum five-year reserve mandate for in-situ sites is particularly consequential: it does not merely restrict exports of processed materials, it removes a portion of extractable ore from the global supply calculation entirely.
This is qualitatively different from the export control measures China has deployed since April 2025. Those controls, covering gallium, germanium, graphite, and seven rare earth elements, operated at the point of outward shipment. The June 15 regulations operate upstream of that, at the level of extraction rights and output authorisation. China is simultaneously constraining both the export tap and the production pipeline, a dual-sided squeeze that Western stockpiling programmes were not designed to address.
The escalation timeline is important context. China introduced export restrictions on gallium and germanium in 2023, expanded to graphite in late 2023, added seven rare earth elements in April 2025, and then in October 2025 published its most comprehensive measures to date, explicitly modelled on the US Foreign Direct Product Rule and extending Chinese regulatory authority to any product manufactured outside China that incorporates Chinese-origin rare earth materials. That October package was suspended at the APEC summit in Busan until November 10, 2026. The June 15 mining regulations, which are not subject to any suspension, arrive as a structural layer beneath those export controls, hardening the supply constraint regardless of whatever diplomatic accommodations are reached at the transactional level.
David Merriman, research director at Project Blue, has stated that the ex-China market will continue to face bottlenecks in the supply of heavy rare earth products over 2026 and 2027 as alternative suppliers are constructed and commissioned. The elements most directly at risk are yttrium, lutetium, terbium, and dysprosium: the building blocks of high-performance permanent magnets used in defence systems, wind turbines, and electric drivetrains. China currently accounts for approximately 90% of global rare earth processing, roughly 80% of tungsten production, and an estimated 98% of global gallium refining capacity. The June 15 regulations do not reduce those shares; they formalise the governance apparatus that makes them durable.
The EU's Rotterdam Strategy: From Policy Architecture to Operational Logistics
The EU's decision to shortlist tungsten, rare earths, and gallium for its first coordinated strategic mineral reserve marks the transition from what the Commission's December 2025 RESourceEU Action Plan described as a pilot scheme to what is now, in operational terms, a procurement and logistics programme. Active discussions with the Port of Rotterdam and other major European facilities represent a meaningful shift from the indicative language that has characterised EU critical minerals policy since the passage of the Critical Raw Materials Act. Warehousing discussions are infrastructure commitments.
The governance structure being assembled is consequential in its design. Ten member states are participating in planning groups led by Italy, France, and Germany. France, holding the G7 presidency, has made supply chain diversification a priority and is pushing for a permanent secretariat to oversee the initiative beyond the bloc's rotating political leadership cycles. That push reflects a hard lesson from the EU's 2022 gas storage coordination exercise: crisis-driven cooperation that depends on political continuity rather than institutional permanence tends to fracture under subsequent administrative transitions. The proposed European Critical Raw Materials Centre, drawing explicitly on the model of Japan's JOGMEC, is intended to provide that permanence, though its expanded mandate requires a second legislative package expected in Q2 2026 and unlikely to be operational before 2027.
The selection logic for the shortlisted materials is rooted in a specific intersection of three criteria used by both NATO and EU criticality assessment frameworks: economic importance to strategic industries, supply concentration risk, and the absence of technically and economically viable substitutes at industrial scale. Tungsten, with the highest melting point of any metal at 3,422 degrees Celsius, is physically irreplaceable in armour-piercing projectiles, cutting tools, and high-performance aerospace components. Gallium is essential to high-frequency semiconductor systems underpinning radar, satellite communications, and electronic warfare. Rare earth elements, particularly heavy rare earths, are the enabling input for permanent magnets in which China now holds 94% of global sintered magnet production, up from approximately 50% two decades ago.
The EU's import dependency figures make the strategic case without requiring embellishment. The bloc relies on external supply for over 95% of its rare earth elements, with the vast majority sourced from China. Import dependence for magnesium approaches 100%. In permanent magnets specifically, 93% of wind turbine magnets installed in the EU are purchased from Chinese suppliers. Executive Vice-President for Industrial Strategy Stéphane Séjourné has explicitly framed the RESourceEU initiative in the terms of REPowerEU: a crisis-driven reorganisation of supply that only state-level coordination can deliver at the required speed. The analogy is imperfect in at least one important respect. Natural gas storage infrastructure existed and could be filled relatively quickly once political will was established. Critical mineral stockpiling at the processed-materials level requires either sourcing from China or waiting for non-Chinese processing capacity that, by most independent assessments, will require a decade or more to construct at meaningful scale.
A Jacques Delors Centre analysis from December 2025 noted that China's October 2025 rare earth controls extended to five additional rare earths and manufacturing equipment, and introduced categorical denials for defence-related end-use applications. That suspension expires in November 2026: the same six-month window in which the EU is attempting to move from planning groups to operational stockpile protocols. The timeline compression is not comfortable.
Project Vault's Structural Vulnerabilities and the July 13 Decision Point
Project Vault was announced on February 2, 2026, at a White House ceremony attended by, among others, General Motors CEO Mary Barra and GE Vernova CEO Scott Strazik. The programme is funded by a $10 billion loan from the Export-Import Bank, the largest in the Bank's 92-year history, supplemented by approximately $2 billion in private capital. It is structured as a public-private partnership in which participating companies submit lists of required minerals, which are then acquired and held in reserve. Participating firms commit to purchase the minerals at a fixed price on drawdown and pay fees covering storage costs and loan interest. The fixed-price commitment feature is designed to provide the price floor stability that, as the Foundation for Defense of Democracies has noted, could make capital-intensive mining and processing projects bankable by neutralising predatory Chinese spot price manipulation.
The PIIE policy brief published in May 2026 identifies three structural weaknesses that could cause the programme to fail at precisely the moment its insurance function is required. First, participation is voluntary, which means large firms with sufficient balance sheets to self-insure and small firms without the sophistication to assess their exposure are both likely to opt out, hollowing the risk pool. Second, the programme covers all 60 minerals on the USGS critical minerals list, a scope that introduces severe storage complexity: unlike petroleum, many processed mineral derivatives degrade over time, require controlled handling conditions, and cannot be used directly by manufacturers without further processing. Third, and most structurally significant, prioritising processed materials over raw ore in the near term means sourcing from China. The PIIE brief is explicit: building reserves of processed materials would likely require dependence on Chinese suppliers, making long-term investment in US and allied processing capacity critical but years away from providing relief.
The Section 232 deadline of July 13, 2026 introduces a further layer of policy uncertainty. President Trump's Critical Minerals Proclamation directed the Department of Commerce and the United States Trade Representative to report on the status of trade negotiations with mineral-rich partner countries within 180 days. The proclamation does not impose tariffs immediately but explicitly reserves the right to do so on processed critical minerals and their derivatives if negotiations prove insufficient. The materials at risk in subsequent actions include processed lithium, cobalt, nickel, manganese, and graphite compounds used in batteries; rare earth oxides and salts; rare earth permanent magnets; and specialty alloys incorporating critical minerals. Building on my analysis of lithium carbonate futures and the structural repricing of battery inputs in May 2026, a tariff layer on processed mineral imports would compound price signals that are already at historically elevated levels for several battery-relevant materials.
Sahar Hafeez of Pillsbury Winthrop Shaw Pittman has identified the benchmark question as the central unresolved implementation problem: most of these commodities, particularly where there is significant concentration in China, are priced off Chinese markets. Constructing an alternative price reference that is credible, liquid, and not immediately gameable by Chinese state enterprises is a prerequisite for any price floor mechanism to function. No such benchmark currently exists for rare earths, tungsten, or gallium at the required granularity and trading depth. The Forum on Resource Geostrategic Engagement (FORGE), announced by Secretary of State Rubio on February 4, 2026, as the successor to the Minerals Security Partnership, is the institutional vehicle through which allied price floor coordination is intended to be developed. As of late May 2026, its operational framework remains under negotiation.
The paradox that both the PIIE brief and independent analysts have identified is structural rather than correctable through programme design alone. The United States is fully import-dependent for 12 critical minerals and relies on imports for more than 50% of an additional 29. China is the leading producer of 30 of the 60 minerals on the USGS list and maintains the leading refining position across 19 of 20 strategic minerals tracked by the IEA, with an average market share of approximately 70%. There is no near-term pathway to stockpiling sufficient quantities of processed critical minerals without transacting with the country those stockpiles are designed to hedge against. That is not a political observation; it is a supply chain arithmetic problem.
The Synchronised Squeeze: Why the Second Half of 2026 Is the Critical Window
The three policy developments described above did not occur in sequence. They occurred simultaneously, which is analytically important. On the same day the EU shortlisted its stockpile minerals, China published the State Council implementing regulations that will take effect June 15. Two weeks earlier, the PIIE brief on Project Vault's vulnerabilities was circulating in Washington. The convergence is not coincidental: each policy is a direct response to the same underlying condition, and each response is, in part, making that condition worse for the other actors.
Western governments are attempting to acquire processed strategic minerals for stockpile purposes at the same moment China is restricting domestic extraction through in-situ reserve mandates and tightening export licensing. Every tonne of processed rare earth oxide, gallium metal, or tungsten carbide that the EU or the US acquires for reserve purposes at current market prices competes directly with industrial demand in a market that China controls at the refining level. The IEA's April 2026 assessment, prepared in support of France's G7 presidency, placed the economic output at risk from full reimplementation of existing Chinese controls at $6.5 trillion annually for countries outside China, with automotive sector exposure alone exceeding $3 trillion. If the November 10, 2026 suspension of the October 2025 extraterritorial measures lapses without renewal, any product manufactured outside China that incorporates Chinese-origin rare earth materials would again require a MOFCOM export license.
The July-to-November 2026 window therefore contains three distinct decision points with compounding consequences. By July 13, Commerce and USTR must report on Section 232 negotiations; tariffs on processed critical minerals remain possible and would immediately affect the economics of Project Vault's procurement operations. By November 10, China's extraterritorial suspension expires; if it is not renewed, the global magnet and semiconductor supply chains face a compliance shock with no near-term workaround. By Autumn 2026, the CRMA Round Two selection process concludes, which will inform the final composition of the EU's stockpile and, critically, the funding mechanisms and drawdown protocols that remain publicly undisclosed. Each of these deadlines interacts with the others, and none can be managed in isolation.
Chris Berry, an independent battery metals analyst, has characterised China's posture as sustained saber-rattling on dual-use and export restrictions that will continue to impede trade flows and elevate prices. The CSIS one-year assessment of the April 2025 controls frames the structural conclusion more directly: China is not a reliable export partner during periods of heightened geopolitical tension, and the pattern of broad diplomatic commitments that are not operationalised into regulatory concessions suggests that export controls should now be treated as a permanent structural feature of global trade rather than a temporary negotiating instrument. That assessment, if accurate, has profound implications for both the EU's Rotterdam stockpile and Project Vault's sourcing strategy: programmes designed as emergency buffers against intermittent disruption must instead be redesigned as permanent structural countermeasures against a durable supply constraint.
What the Data Describes: Processing Capacity Is the Actual Problem
The debate over stockpile design, governance structure, and sourcing logistics is, at its core, a debate about a single underlying deficiency: Western nations do not possess sufficient processing and refining capacity to convert primary ore into the usable mineral derivatives that defence, energy, and technology industries require. China's dominance was not constructed overnight. It reflects three decades of deliberate industrial policy, subsidised refinery development, and tolerance for environmental compliance costs that market-driven Western economies were unwilling to accept. Rebuilding that capacity independently requires, by most independent assessments, between twenty and thirty years under normal investment conditions. The current geopolitical window is measured in months.
The PIIE brief's recommendation to prioritise processed materials over raw ore in Project Vault is correct as industrial logic but confronts this capacity gap directly. The DOE has acknowledged the challenge through its $134 million Rare Earth Elements Demonstration Facility and $355 million Mine of the Future initiative, but demonstration-scale facilities do not constitute industrial-scale processing chains. The EU's RESourceEU framework commits to ensuring that 40% of annual consumption of strategic raw materials is processed within EU borders, but that target is aspirational against a baseline in which virtually all rare earth processing, and essentially all gallium refining, occurs in China. The European Critical Raw Materials Centre, modelled on JOGMEC, is designed to bridge the coordination gap, but its most consequential tools, including joint purchasing, price stabilisation mechanisms, and diversification obligations, remain, in the Jacques Delors Centre's assessment, indicative or politically contingent.
For individual minerals, the capacity gap is quantifiable. Gallium provides a precise illustration. US final demand for gallium in 2024 was approximately 19 metric tonnes, roughly 2.5% of global production. China accounted for about 99% of primary gallium output while operating at only three-quarters of estimated production capacity. Gallium is not mined directly; it is recovered as a trace byproduct of bauxite and zinc processing. Building independent gallium refining capacity therefore requires not just investment in separation technology but structural co-location with aluminium smelting and zinc processing operations, an industrial integration challenge that cannot be resolved through financial commitments alone. Japan's government has explicitly warned that its inability to import required rare earth elements for magnet production would eventually affect all companies downstream in the global supply chain; Japan holds approximately 15% of global advanced rare earth permanent magnet manufacturing, second only to China.
The price signal environment provides the most direct evidence of where the constraint is tightest. China's export controls introduced since 2023 have triggered price spikes of up to sixfold outside China in certain mineral categories, with licensing approvals for European firms falling below 25% in some sectors. Against that backdrop, the EU's €3 billion committed over the next twelve months under RESourceEU represents a meaningful but insufficient down payment on structural diversification. Project Vault's $12 billion is larger in absolute terms but, as the PIIE analysis makes clear, is undermined by voluntary participation, scope complexity, and the near-term sourcing paradox. The combined Western stockpiling commitment across both programmes is, in structural terms, a demand-side intervention in a market where the supply-side constraint is located in processing infrastructure that takes decades to build.
Conclusion: The Reserve Race Is Not the Solution, But It Is Where Policy Is
Strategic mineral stockpiling is, in the most precise analytical sense, a second-best policy. The first-best policy is independent processing capacity, diversified refining infrastructure, and supply chains that do not route through a single jurisdiction. Western governments know this. The language of every major policy document, RESourceEU, Project Vault, FORGE, and the CRMA, acknowledges the long-run imperative to rebuild midstream capacity. The gap between that acknowledged imperative and current operational reality is where the stockpile programmes live: they are not solutions to the structural problem, but they are the only instruments available on the relevant policy timeline.
The three developments analysed here, China's June 15 mining regulations, the EU's Rotterdam stockpile discussions, and the PIIE's critique of Project Vault, collectively describe a market in which the structural supply constraint is hardening faster than Western diversification capacity is building. The June 15 regulations formalise upstream control. The November 2026 extraterritorial suspension expiry threatens to reimpose export licensing across the global supply chain. The July 13 Section 232 deadline introduces tariff risk on the processed imports that both EU and US stockpile programmes currently depend on in the near term. The Autumn 2026 CRMA Round Two process will determine which additional materials enter the EU's priority framework at the moment when the price and availability environment is most adverse.
The data does not support optimism about a rapid resolution of the underlying asymmetry. China's three-tier reserve system, encompassing physical stockpiles, production capacity reserves, and in-situ site designations, is now legally codified and operationally superior to anything the West has assembled or announced. China controls approximately 90% of global rare earth processing, is simultaneously restricting both exports and domestic extraction rates, and is building its own strategic reserve infrastructure under the same June 15 regulatory framework. The IEA's $6.5 trillion annual risk figure is not a projection about a possible future; it is an assessment of the economic exposure that already exists under current control architecture.
What the reserve race does accomplish, even in its current imperfect form, is price floor stabilisation for investment in alternative supply chains. The FDD's analysis of Project Vault's most consequential feature, the fixed-price off-take commitment that protects investors from predatory Chinese spot price manipulation, is correct in identifying this as the mechanism through which emergency stockpiling programmes create long-run structural value. By guaranteeing demand at a specified price, they change the investment calculus for the non-Chinese refining and processing projects that are the actual solution. On that basis, the EU's €3 billion and the US's $12 billion are best understood not as insurance against the next supply shock, but as the capital signal intended to make the next generation of non-Chinese processing projects financeable. Whether that signal is large enough, fast enough, and structured well enough to change behaviour before the November 2026 deadline is the question the data does not yet answer.