The Pentagon Has 60 Days of Rare Earths Left

Mike Crabtree keeps a number on a whiteboard in his Saskatoon office: 60 days. That’s the U.S. defense establishment’s rare earth inventory for manufacturing F-35s, Tomahawk guidance systems, and permanent magnets in military drone motors. Crabtree, CEO of the Saskatchewan Research Council, is building the largest heavy rare earth metallization plant outside China. The deadline driving him is a U.S. ban on Chinese-sourced rare earths for defense applications, effective 2027.

100 Generals and a Shrinking Toolkit

Xi Jinping has removed more than 100 senior PLA leaders since 2022. In January 2026, China’s top general Zhang Youxia was removed. Nine more officers followed in March. The conventional read is that this instability makes China less dangerous. But that analysis confuses military capacity with coercive capacity.

The purges have degraded command relationships required for complex joint operations. Defense News reported reduced Chinese fighter activity near Taiwan — consistent with operational disruption, not strategic restraint.

Here is the counterintuitive turn: a regime that has gutted its own military coordination doesn’t lose the desire to coerce. It retains multiple tools — cyber operations, grey-zone maritime pressure, economic sanctions — but rare earth export controls stand out as uniquely low-cost and high-impact. China controls 60% of global rare earth production and 90% of refining. The EU sources 100% of its heavy rare earth elements from China. Export restrictions require no joint operations, no command chain integrity, no trust between generals. They require a signature.

One defense supply chain analyst noted that Chinese air defense systems have failed conspicuously in real combat deployments in Iran and Venezuela, eroding confidence in Chinese military exports. When hardware disappoints, resource leverage becomes more valuable.

Could this reading be wrong? If the purges are genuinely about anti-corruption and the resulting leadership is more competent, the narrowing-toolkit thesis collapses. The test is whether PLA exercises resume at scale by mid-2026. If they do, the purges were renovation, not demolition.

A military that can’t coordinate a strait crossing can still shut off a pipeline.

From Saskatoon to the Pentagon’s Calendar

The 2027 ban created a hard deadline on a procurement scheduler’s calendar. But the purges may be compressing the timeline by raising the probability that China tests the export weapon before alternatives are ready.

Three facilities are advancing simultaneously. In Saskatchewan, Crabtree’s SRC plant targets heavy rare earth metallization — the process that turns refined oxides into usable metal for defense applications. In Ohio, a processing facility is scaling toward commercial output. In Virginia, a Virginia Tech pilot project is developing domestic extraction techniques. All three announced major milestones in March 2026.

The binding number is China’s 90% share of global rare earth refining. That is not a market position. It is a permission gate — a chokepoint where one actor’s decision determines whether others can operate.

Japan’s government is funding rare earth mining and refining in Malaysia, tapping into approximately 16 million tons of reserves. Tokyo is not waiting for North American capacity. It is building a parallel bypass through Southeast Asia, hedging against the 2027 timeline.

China has strong economic incentives not to weaponize rare earths aggressively, since full restrictions would accelerate the diversification that erodes its leverage. But this assumes Beijing optimizes for long-term market share rather than short-term coercive impact. The purges suggest a leadership prioritizing political control over institutional rationality.

Sixty days of inventory is not a buffer. It is a countdown someone forgot to start.

The Window Between the Purge and the Plant

First the command relationships broke. Then the coercive toolkit narrowed. By the time the processing plants in Ohio and Saskatchewan reach commissioning, the window of maximum vulnerability — where the export weapon is most tempting and alternatives aren’t ready — may have closed. Or it may not.

Crabtree framed the stakes directly: “If China said we’re not going to give you rare earths, that means no F-35s, no missiles.” The purges have made that hypothetical more probable because they’ve degraded coercive tools requiring institutional trust while leaving intact the one that doesn’t.

Reuters has reported that U.S. aerospace and semiconductor suppliers are already turning away some commercial customers due to rare earth material tightness — an informal rationing system where defense buyers receive priority. The grey market is in access to existing supply, allocated by phone calls rather than published policy.

The Pentagon has known about the 60-day buffer for years. The U.S. maintained that buffer while Chinese processing dominance reached 95%. Diversification is now happening on a sprint timeline, with facilities announced in March 2026 racing against a 2027 Chinese ban enforcement date.

The cost of revision was institutional — admitting dependency, requesting budget, disrupting procurement relationships. So the buffer stayed at 60 days while dependency deepened, and now the correction is a sprint.

If China restricts rare earth exports before 2027, you face a choice between waiving your own ban on Chinese sourcing — politically humiliating — or accepting production gaps in frontline weapons systems. Both are consequences of paying to avoid earlier revision.

What to Watch

SRC and Ohio facility commissioning milestones, Q3–Q4 2026. Any delay past September pushes the 2027 deadline from tight to unworkable.

PLA exercise resumption near Taiwan. If large-scale joint exercises return by July 2026, the purges were disruptive but not structurally degrading.

Chinese rare earth export license processing times, April–June 2026. Watch for slower processing, additional paperwork, selective delays — early signals of restriction.

Snap point: U.S. rare earth spot prices exceeding 3x their March 2026 levels. Past that threshold, the cost of maintaining the 2027 ban without ready alternatives exceeds the cost of a politically embarrassing waiver.

I predict China will impose selective rare earth export delays — not a full ban, but targeted slowdowns on heavy rare earth oxides — by Q4 2026. If not, the April summit produced a tacit agreement to keep the resource lever holstered, and the purges haven’t shifted Beijing’s coercive calculus.

If This Thesis Is Wrong

The weakest link: the step from “purges degrade military options” to “export restrictions become more likely” is the inferential leap with the least direct evidence. It’s possible rare earth policy is set by the Ministry of Commerce on purely economic grounds, entirely disconnected from PLA institutional health.

The question worth sitting with: if the 60-day buffer has been known for years and the correction only started twelve months before the deadline, what other known vulnerabilities are sitting on whiteboards right now, waiting for their own countdown to begin?

The whiteboard in Saskatoon counts down whether Washington reads it or not.

Word count: 1,186

Sources