The shutdown of Sherritt's Fort Saskatchewan cobalt-nickel refinery, record typhoon-driven port congestion across Chinese gateways, and China's multi-month rare earth export licensing queues have arrived simultaneously in the second half of 2026. Taken together, they reveal a structural pattern: North American critical mineral supply chains are being stressed at every node at once, from the upstream feedstock source through the midstream processing link to the final logistics leg. The common thread is single-source dependency, and the policy window to address it is narrowing.
Introduction
Three separate events unfolded across the critical mineral supply chain in the summer and early autumn of 2026. A Canadian refinery ran out of feedstock and went dark. Consecutive typhoons locked up the world's busiest container ports for weeks on end. And China's export licensing regime continued to impose approval queues measured in months on the rare earth magnets that power electric motors, wind turbines, and defense guidance systems.
Each event, examined in isolation, looks like a discrete operational problem with a specific cause and a manageable fix. Examined together, they reveal something more troubling: the supply chains that North American industry is counting on to deliver the energy transition and satisfy Pentagon procurement requirements are fragile at nearly every node simultaneously. The feedstock layer, the logistics layer, and the regulatory compliance layer are all under stress at once.
This matters most to three overlapping audiences. Battery and EV manufacturers who assumed that building domestic processing capacity was sufficient to secure supply. Defense contractors facing a January 2027 deadline to eliminate Chinese-sourced rare earths from their supply chains. And policymakers who have spent the last two years treating supply chain security as primarily a capital deployment problem, when it is increasingly a geopolitical and logistics management problem.
Building on my earlier analysis of the feedstock gap facing newly built US processing facilities, the picture that emerges in October 2026 is starker than even that assessment suggested. It is not just that plants lack raw material. It is that the pathways to obtain that material are being cut, delayed, or regulated into dysfunction from multiple directions at once.
When Sanctions Strand an Allied Refinery
The clearest illustration of what single-source dependency actually costs came on June 22, 2026, when Sherritt International announced the controlled shutdown of its Fort Saskatchewan refinery in Alberta. The facility, capable of producing roughly 38,200 tonnes of nickel and cobalt per year, had run through its last stockpile of mixed sulfide precipitate from Cuba's Moa joint venture. With no feedstock and no alternative source, there was nothing left to refine.
The proximate cause was a set of US Treasury OFAC sanctions announced on May 7, 2026, which targeted GAESA, the Cuban military conglomerate that controls large portions of the island's economy, including Moa Nickel S.A., Sherritt's 50% joint venture partner. The sanctions included broad secondary provisions, which meant that Sherritt's Canadian refinery, sitting in a politically secure allied jurisdiction, was effectively cut off from its only feedstock source overnight.
The numbers tell the story of how fast a single upstream interruption propagates downstream. Mixed sulfide precipitate production at Moa had already fallen 43% year on year in the first quarter of 2026, after fuel shortages forced a mining suspension in February. Finished nickel output at Fort Saskatchewan dropped 36% to 1,885 tonnes; cobalt fell 34% to 213 tonnes. Metals revenue fell from $113.7 million to $93.8 million in a single quarter. By mid-June, the buffer inventory was gone.
What makes this case strategically significant is not just the scale of the loss, though losing Canada's only domestic source of refined cobalt and one of just three nickel refineries in North America is not a minor disruption. It is the structural lesson embedded in the story. Sherritt operated a 32-year supply chain partnership that worked flawlessly as long as the political relationship between the United States and Cuba remained stable enough not to trigger secondary sanctions. The moment that condition changed, the entire downstream asset became stranded. The refinery's technical capability, its workforce, its environmental compliance record, none of it provided any protection against a feedstock cut that originated in Washington, not in Fort Saskatchewan.
For procurement teams designing battery-materials supply chains, this is the cautionary template. Physical security of a downstream refining asset provides no guarantee of supply continuity if the upstream feedstock pathway runs through a sanctions-sensitive jurisdiction. The lesson is not that Canadian infrastructure is unreliable. It is that single-source upstream dependencies are structural liabilities regardless of where the refinery sits.
Typhoons, Blank Sailings, and the Compounding Cost of Congestion
While the Sherritt shutdown illustrated how geopolitical decisions can sever a supply chain at the feedstock level, a separate but simultaneous disruption was degrading the logistics layer across the Pacific. Consecutive typhoons, including Bavi, Noul, Dolphin, and Saudel, struck China's eastern coastline across July, August, and early September 2026, forcing repeated temporary closures of the ports of Shanghai and Ningbo, two of the world's largest container gateways.
The congestion numbers are worth holding in mind. Following Typhoon Saudel in late August, vessel waiting times at Shanghai climbed to between six and eleven days depending on the terminal, with the longest delays at Yangshan exceeding twelve days. Ningbo's yard occupancy reached 92 to 95 percent. Linerlytica estimated that the combined logjam was holding up 3.92 million TEU of capacity, equivalent to roughly 11 percent of the entire global container shipping fleet, at peak congestion. Shanghai recorded a 70 percent drop in port-call movements in the 24 hours before Saudel made landfall.
Carriers responded by executing blank sailings at an accelerating pace. Approximately 20 percent of transpacific capacity was pulled from the market through blank sailings in early September, pushing Shanghai-to-Los Angeles spot rates to $7,185 per FEU as of September 3, rising to $7,712 by September 17. The Drewry World Container Index held near $4,465 to $4,500 per 40-foot container through the month, reflecting the rate support that capacity withdrawal provided even as underlying demand remained strong.
For mineral cargo specifically, the typhoon disruptions compound a pre-existing problem. Mineral supply chains are not like retail consumer goods shipments. They typically require multiple processing stages across different countries before a finished material reaches its end user. A delay at the Chinese port stage can cascade through the entire downstream schedule, because the next processing facility may have booked slot time or blended its own inventory assumptions around an expected arrival date. Adding ten to fourteen days of waiting time to routes that are already operating on stretched lead times does not produce a linear ten-to-fourteen-day delay at the destination. It can produce months of schedule disruption.
The demand-side picture adds further pressure. The Port of Los Angeles completed its busiest three-month period on record between June and August 2026, handling more than 2.9 million TEUs. The National Retail Federation forecast September as the single busiest import month of the year at 2.31 million TEUs, up 9.6 percent year on year. Port of Los Angeles Executive Director Gene Seroka attributed the unusual volume pattern to importers stepping away from seasonal shipping norms in response to tariff uncertainty, Middle East instability, and fuel cost volatility. The same structural forces that are disrupting mineral supply chains are producing record demand at the very port infrastructure that mineral shippers depend on to clear cargo.
China's Licensing Regime: When Compliance Is the Bottleneck
The third stress point is the one that operates most quietly but may prove most durable. Since April 2025, China has maintained an export licensing regime covering seven heavy and medium rare earth elements, including dysprosium, terbium, and yttrium, that are essential inputs for the sintered neodymium-iron-boron magnets used in electric motors, wind turbines, data center cooling systems, and military guidance equipment. Every shipment requires a non-automatic export license from MOFCOM. The statutory review target is 45 working days. In practice, approvals routinely take two to four months, and applications can sit without a clear decision indefinitely.
The documentation requirements have escalated well beyond a standard end-user certificate. MOFCOM now requires exporters to disclose the complete manufacturing value chain, the final application, authenticated customer data, composition certificates, and detailed technical descriptions of the goods. Incomplete documentation is the most common avoidable cause of delay. There is no general license pathway available in practice; every shipment is a case-by-case filing.
On June 22, 2026, the regime became sharply more targeted. China added ten US entities to its export control list, including MP Materials and USA Rare Earth, the two firms that represent the structural backbone of America's domestic rare earth independence strategy. MP Materials operates Mountain Pass, the only rare earth mine of meaningful scale currently active in the United States, and receives Department of Defense funding specifically to reduce reliance on Chinese-controlled supply chains. Blacklisting these firms approached a near-total prohibition on their ability to source controlled rare earth materials from China, effectively targeting the firms the US government had identified as its primary solution to rare earth dependency.
China's escalation pattern is now well established. Export controls were tightened in April 2025, expanded extraterritorially in October 2025, partially suspended following the Busan summit in November 2025, and then sharpened again with entity-specific blacklisting in June 2026. Each move is calibrated, reversible in form but durable in effect, and targeted at the specific firms and materials where Western supply chains are least able to substitute quickly. David Merriman, research director at Project Blue metals consultancy, summarized the medium-term outlook clearly: the ex-China market will continue to face bottlenecks in heavy rare earth products through 2026 and 2027 as alternative suppliers are constructed and commissioned.
One regulatory date deserves particular attention. The suspension of MOFCOM Announcement 61, the extraterritorial measure that would require Chinese export licenses for products manufactured outside China that contain more than 0.1 percent Chinese-origin rare earth content, expires on November 10, 2026. That is five weeks from today. If the suspension lapses, manufacturers in Europe, North America, and Southeast Asia that use Chinese-origin rare earths in finished products could find themselves subject to Chinese licensing requirements even for shipments that never touch Chinese territory. The compliance implications for global magnet and motor manufacturers would be immediate and severe.
The Pattern Beneath the Events
Taken individually, each of these three developments has a specific cause: a sanctions decision in Washington, a sequence of Pacific typhoons, a licensing architecture in Beijing. Taken together, they trace the outline of a structural vulnerability that capital investment alone cannot resolve.
All three cases share a single underlying condition: single-source dependency at a critical node. Sherritt's refinery had one feedstock source and no fallback. Western manufacturers of high-performance magnets have, in practice, no short-term substitute for Chinese-processed heavy rare earth elements. And the typhoon congestion revealed that transpacific mineral cargo lanes, already operating without redundancy or meaningful buffer inventory, buckle quickly under weather events that have historically been managed as routine seasonal disruptions.
The policy response to date has concentrated heavily on building new domestic capacity. The DOE's $162 million tailings recovery program, the Defense Production Act determinations, the funding flowing to MP Materials and USA Rare Earth: all of these are supply-side interventions aimed at creating alternative sources of material. They are necessary. But as the Sherritt case illustrates, the ability to refine a material domestically provides no protection if the feedstock pipeline can be severed by a third-country sanctions decision. And as the rare earth blacklisting illustrates, building a domestic supply chain that China has specifically identified as a strategic threat will attract targeted countermeasures, not passive tolerance.
What the three events in aggregate suggest is that supply chain security for critical minerals requires redundancy and regulatory resilience at every layer of the chain simultaneously, not just at the processing stage. That means multiple feedstock sources, buffer inventory strategies sized for multi-month disruptions, logistics routing that does not concentrate exposure on a single port complex or trade lane, and compliance infrastructure capable of managing a documentation-intensive licensing environment across multiple jurisdictions.
What Happens Next
The immediate horizon is dominated by the November 10 date. If China reinstates the extraterritorial provisions of Announcement 61, the compliance burden on non-Chinese manufacturers using Chinese-origin rare earths will expand significantly and rapidly. Logistics planners advising clients on magnet supply chains should already be stress-testing their documentation readiness and building ten-to-twelve-week license windows into forward scheduling, as the 45-working-day statutory target has proven unreliable in practice.
For Sherritt, the path back to production depends on conditions that the company itself has said it cannot provide guidance on: the resumption of Moa joint venture operations, the reconstruction of the mixed sulfide precipitate feed pipeline, and, implicitly, some shift in the sanctions environment that severed the supply relationship in the first place. Fertilizer and sulfuric acid production continue at Fort Saskatchewan, preserving part of the site's industrial activity, but cobalt and nickel refining remain suspended with no disclosed restart timeline. North America will remain without domestic cobalt refining capacity for the foreseeable future.
On the logistics side, the typhoon season will eventually end, but the structural congestion problem will not. Hapag-Lloyd's CEO has characterized the underlying congestion dynamic as structural and likely to persist for years. Average global vessel waiting times through the first seven months of 2026 were already nearly double comparable 2019 levels before the September typhoons added their incremental disruption. Mineral shippers who have not yet built multi-week schedule buffers and carrier redundancy into their transpacific routing plans are operating on assumptions that the current environment does not support.
The broader policy question, and it is genuinely contested, is whether the United States and its allies can build enough alternative supply, processing capacity, and logistics resilience to reduce their exposure to these concentrated chokepoints before the current disruption cycle forces procurement failures in defense or clean energy sectors. The capital commitments are large and growing. But the three events of mid-2026 are a precise illustration of how quickly a supply chain designed around a single source, a single route, or a single regulatory relationship can be disrupted by decisions made in Washington, Beijing, or the Western Pacific weather system. The timeline for building genuine resilience is longer than the timeline on which those disruptions are arriving.