Three converging data points tell the same uncomfortable story: the United States can extract enough raw copper to meet 146% of its demand, hosts rare earth deposits at Mountain Pass, and sits atop vast bauxite reserves, yet remains structurally dependent on foreign refiners to turn those resources into usable industrial inputs. With China's rare earth export control suspension expiring on November 10, 2026, and aerospace manufacturers already rationing yttrium, the processing bottleneck has moved from long-term policy concern to immediate industrial emergency.
Introduction
Three stories landed on my desk this week that, at first glance, look like separate chapters: a Chinese rare earth price index sitting at 2.5 times its 2010 baseline, aerospace manufacturers rationing a thermal coating material called yttrium, and U.S. aluminum smelting running at levels not seen since the Carter administration. Read together, they are not separate chapters at all. They are the same chapter, told through three different commodities.
The connecting thread is not scarcity in the ground. It is the near-total absence of the infrastructure needed to turn what is in the ground into what manufacturers actually need. The United States has spent decades exporting raw materials, shuttering processing plants, and implicitly outsourcing the most capital-intensive steps of industrial production to China. The bill for those decisions is now arriving, and it is arriving faster than most policy timelines anticipated.
This matters to anyone who flies on commercial aircraft, drives an electric vehicle, works in defense manufacturing, or simply uses electricity transmitted through copper wire. The vulnerability is not theoretical. In the case of yttrium, it is already costing production capacity. In the case of aluminum, it is already costing jobs and tax revenue. In the case of rare earth magnets, it may cost defense readiness. What follows is an attempt to explain how we got here, what the data actually show, and what the narrowing window for action looks like.
The Same Problem, Three Commodities
Start with copper, because the arithmetic is so stark it almost defies belief. Benchmark Mineral Intelligence data show the United States can source 146% of its domestic copper demand through a combination of mine output and scrap. Yet nearly half of that mined copper concentrate gets exported, largely because domestic smelting and refining capacity has been allowed to erode. The number of U.S. copper refineries fell from nine in 2000 to just five in 2023, and the country now operates only two primary smelters. The result is that manufacturers end up importing refined copper that was, not long ago, American ore.
Aluminum tells a similar story with even starker numbers. In 2025, the United States smelted 660,000 tonnes of primary aluminum against consumption of 5.7 million tonnes, a gap so wide it required importing 4.4 million tonnes of crude and semi-fabricated product to keep industry running. China, for comparison, smelted 46 million tonnes. The last new U.S. aluminum smelter was built in 1980. American industries now rely on imports for roughly 85% of their aluminum needs, a dependency that 50% Section 232 tariffs have done more to raise costs than to rebuild capacity.
Rare earths follow the same logic but with an additional layer of geopolitical complexity. The United States has Mountain Pass in California, one of the richest rare earth deposits outside China. MP Materials is producing at record rates, hitting 917 metric tons of NdPr in the first quarter of 2026. But for years, even MP's output was shipped to China for the separation and refining steps that turn ore into usable material. The problem was never what was in the ground. The problem was always what happens next.
Building on my earlier analysis of the copper refining crisis in June 2026, this processing bottleneck is not a new observation. What is new is the urgency created by a hard diplomatic deadline that is now less than five months away.
The November Clock and What It Actually Means
On November 10, 2026, China's suspension of its expanded rare earth export controls expires. That deadline has been well-reported, but the framing around it has sometimes been misleading. The suspension does not mean controls were lifted. It means a second, more aggressive wave of controls, announced in October 2025, was paused as part of a diplomatic arrangement following the Xi-Trump meeting at the APEC summit in Busan. The core April 2025 licensing regime covering seven rare earth categories, including yttrium, dysprosium, terbium, and scandium, has never been suspended. It is active right now.
Those April 2025 controls require special export licenses for the seven affected materials and their derivatives. There is no statutory approval timeline for licenses, but reviews have routinely taken two to four months in practice. When the October 2025 suspension lapses, if it lapses, the extraterritorial provisions return: any product manufactured outside China that incorporates Chinese-origin rare earth materials would require licensing from China's Ministry of Commerce before it could be re-exported to third countries. The reach of that provision is enormous, touching supply chains in Japan, South Korea, Germany, and the United Kingdom.
China's Rare Earth Price Index, published by CREIA using 2010 as a baseline of 100, reached 249.4 on June 1, 2026. That is a meaningful easing from the spike above 300 seen earlier this year, but it masks significant divergence within the complex. NdPr metal is approaching the $110 per kilogram floor established by the MP Materials-DoD agreement. Dysprosium and terbium, both subject to the April 2025 controls, are up more than 100% year to date. The FOB China premium for neodymium oxide is running at $184 per kilogram against a domestic benchmark of $113 per kilogram, a 63% spread that represents, in concrete terms, the price of supply chain insecurity.
Gracelin Baskaran, who directs the Critical Minerals Security Program at CSIS, made the structural problem plain: "The U.S. still has to tread carefully in its relationship with China to avoid those disruptions, given how long it takes to transform rare-earth announcements, funding, and partnerships into actual supply." That sentence deserves to sit with you for a moment. The gap between announcement and supply is not measured in months. It is measured in years.
Yttrium: When the Bottleneck Becomes a Production Halt
The yttrium story is worth dwelling on because it is the clearest documented case we have of licensing friction translating into actual downstream industrial harm. CSIS analysis of Chinese customs data shows that China exported just 17 tonnes of yttrium to the United States across the eight months from April to December 2025, compared with 333 tonnes in the comparable prior period. That is a 95% collapse. By February 2026, exports had recovered to approximately 20 tonnes per month, still less than a third of the pre-restriction baseline of 66 tonnes per month.
Yttrium is not an exotic laboratory curiosity. It is the thermal barrier coating applied to jet engine turbine blades to prevent them from melting under operating temperatures. Without it, you cannot certify jet engines. Without certified jet engines, you cannot build or maintain commercial aircraft or military platforms. Aerospace manufacturers have publicly warned of material rationing and potential production pauses. One executive at a major aerospace components manufacturer described "immediate bottlenecks" and warned of "potential production slowdowns by early next year." That warning was issued before the November deadline had come back into focus.
Prices outside China for yttrium oxide have surged by several thousand percent since early 2025, while Chinese domestic prices remain at a fraction of those levels, a dual-price system separated not by production costs but by export controls. Benchmark Mineral Intelligence data show that shipments of yttrium, dysprosium, and terbium to the United States are running at 42%, 41%, and 49% of pre-restriction volumes respectively, despite the formal suspension of the October 2025 controls. This is the key finding that deserves broader attention: the suspension of the headline policy has not normalized trade flows. The licensing friction, the informal quota management, and the administrative burden of end-user documentation are operating as supply constraints independent of what the headline policy status says.
Think of it this way. Suppose your bank announced a temporary suspension of withdrawal limits. You go to the ATM and find it still takes two months to process your request, still requires paperwork that reviewers routinely lose, and still dispenses far less than you asked for. The headline suspension is real. The practical reality is something different.
Concentration Is Getting Worse, Not Better
The IEA's data on global refining concentration should be read alongside each of these commodity stories, because it provides the structural backdrop against which individual disruptions occur. The average market share of the top three refining nations across key energy minerals rose from 82% in 2020 to 86% in 2024. Roughly 90% of supply growth came from the single top supplier in each category: China for cobalt, graphite, and rare earths; Indonesia for nickel. The United States accounts for approximately 1% of global refined rare earth production.
The IEA is explicit that this trend is not self-correcting. Market forces alone will not drive diversification because high capital costs in new jurisdictions, typically 50% higher than in established hubs, and relatively low commodity prices in 2024 are deterring investment from new entrants. Looking forward to 2035, the IEA projects that the average share of the top three refined material suppliers will decline only marginally, to 82%, effectively returning to 2020 levels after a decade of announced diversification strategies and billions in public investment. That is a damning projection.
The IEA's recommended policy tools, contracts-for-difference, cap-and-floor pricing, and volume guarantees, are worth taking seriously here. They are, in essence, mechanisms to make refinery investment bankable in places where it is currently not, by reducing the revenue risk that deters private capital. The DoD's $110 per kilogram price floor agreement with MP Materials is a rough approximation of exactly this kind of instrument, applied to a single company for a single material. The question is whether that model can be scaled and replicated across the processing bottleneck more broadly.
The IEA's additional finding that three-quarters of the 20 minerals it tracks are now more volatile than crude oil adds another dimension to the investment problem. Commodity price volatility is itself a barrier to refinery construction, because lenders price in the risk that prices will collapse before a new facility recoups its capital costs. Without mechanisms to manage that volatility, private capital will remain reluctant, and public capital alone will not be sufficient to close a gap of this magnitude.
Policy Announcements Are Not Processing Capacity
The honest accounting of where U.S. policy stands requires distinguishing between what has been announced and what has been built. The DoD holds a 15% stake in MP Materials, worth $400 million, and has committed $1.6 billion to USA Rare Earth. MP Materials is commissioning its own dysprosium and terbium separation capability at Mountain Pass, targeted for mid-2026. These are meaningful steps. But Lynas Rare Earths, widely cited as the Western alternative to Chinese rare earth processing, still ships intermediate materials to China for final processing. Real supply chain independence remains, as the research notes put it plainly, aspirational through at least 2026 if not beyond.
The aluminum situation illustrates the timeline problem with particular clarity. Emirates Global Aluminium and Century Aluminum announced in January 2026 a joint development agreement to build a new primary aluminum smelter in Inola, Oklahoma, the first such facility in the United States since 1980. The plant would produce up to 750,000 tonnes per year and more than double current U.S. production. Engineering firm Bechtel is conducting preparatory work. A final investment decision is targeted for the end of 2026. Even in an optimistic scenario, the facility would not begin producing aluminum until the late 2020s at the earliest. Against a November 2026 rare earth deadline, a 2027 DoD sourcing mandate barring Chinese rare earth materials from defense supply chains, and yttrium rationing happening right now, that timeline is deeply uncomfortable.
The DoD sourcing mandate is worth examining directly. The department has established a requirement barring the use of rare earth materials and magnets from China, Russia, Iran, and North Korea by January 1, 2027. The mandate is real and the deadline is firm on paper. But analysts are skeptical that sufficient capacity will be online in the next eight months to make compliance feasible across the defense industrial base. A defense supply chain that cannot actually comply with its own sourcing requirements is not a supply chain that has been secured. It is a supply chain that has been documented.
The energy dimension adds one more complication. Heavy industrial processing, whether of aluminum, copper, or rare earth oxides, is enormously electricity-intensive. Growing competition for power from AI data centers is reshaping the cost calculus for industrial facilities. The EGA smelter in Oklahoma is contingent, in part, on securing a competitive power deal with Public Service Company of Oklahoma. That negotiation is ongoing. The broader point is that the infrastructure problem in critical minerals refining is not just about capital costs and processing technology; it is also about whether the United States can provide industrial-scale electricity at prices that make domestic refining economically viable.
What the November Deadline Should Force Us to Decide
The November 10, 2026 expiration of China's export control suspension is not primarily a threat to be managed. It is a deadline that forces a clarification of strategy. The United States and its allies have spent the past 18 months announcing frameworks, signing partnerships, and making equity investments in mining and processing companies. As I noted in my June analysis of the November countdown, the West's critical minerals problem is fundamentally a processing problem, not an extraction problem. That diagnosis remains accurate. What has changed is that the time horizon for acting on it has compressed sharply.
Three things need to happen before the deadline becomes a crisis rather than a catalyst. First, inventory building needs to accelerate immediately. Ex-China buyers have thin liquidity, most transactions remain bilateral and confidential, and the market structure does not currently support the kind of transparent spot trading that would allow efficient stockpiling. Procurement urgency is only beginning to show up in inventory-building strategies, according to market observers, which suggests that many downstream manufacturers are still operating on optimistic assumptions about the suspension being extended.
Second, the policy tools recommended by the IEA need to move from recommendation to implementation. Contracts-for-difference for critical mineral refiners, volume purchase agreements modeled on the DoD-MP Materials deal, and cap-and-floor pricing mechanisms are the instruments most likely to make new refining investment bankable. The IEA is explicit that without mechanisms to stabilize markets and reduce investor risk, the geographic concentration of refining will remain essentially unchanged through 2035.
Third, the honest conversation about timeframes needs to happen in public, not just in policy documents. The DoD sourcing mandate, the EGA smelter, the MP Materials dysprosium and terbium separation facility, the Lynas expansion in Malaysia: all of these are important. None of them closes the processing gap by November 2026. Baskaran's warning at CSIS captures the stakes precisely. Transforming announcements, funding, and partnerships into actual supply takes years. The November deadline is months away. That gap is the most important number in critical minerals policy right now, and it is not getting nearly enough attention.