Highlights

• IEA’s 2026 report reveals China dominates rare earth supply chains with 60% of mining, 91% of refining, and 94% of permanent magnet production—a $6.4 billion market controlling trillion-dollar downstream industries
• The West faces a processing and manufacturing deficit, not a mining shortage: by 2035, non-China projects would cover only 50% of mining, 25% of refining, and under 20% of magnet demand
• Diversification will take decades, not years—tacit industrial knowledge in separation, metallization, and magnet-making cannot be quickly replicated despite capital investment

The International Energy Agency’s new report (opens in a new tab), Rare Earth Elements: Pathways to Secure and Diversified Supply Chains, is less a market study than a strategic warning: the modern industrial economy runs on a supply chain China still overwhelmingly controls. The report, directed by Tae-Yoon Kim and Tim Gould and led analytically by Amrita Dasgupta with contributors across the IEA, finds that in 2024 China accounted for about 60% of magnet rare earth mining, 91% of refined output, and 94% of permanent magnet production. Demand for magnet rare earths—especially neodymium, praseodymium, dysprosium, and terbium—has doubled since 2015 and is projected to rise another third by 2030 under current policies. The IEA’s core message is right, and overdue: the West is not facing a mine shortage so much as a processing, metallization, and magnet-manufacturing deficit.

Small Market, Enormous Leverage

One of the report’s sharpest insights is also its most unsettling. Rare earths are a relatively small market—about $6.4 billion—yet permanent magnets account for roughly 95%-96% of rare earth consumption by value and sit inside EVs, wind turbines, robotics, industrial motors, aerospace systems, medical devices, and defense hardware. In other words, a modest upstream market exerts outsized control over trillion-dollar downstream industries. That asymmetry is the story.

Where the IEA Is Likely Right

The IEA is strongest when it maps the bottlenecks. Outside China, current and planned capacity is still nowhere near enough. By 2035, existing and announced non-China projects would cover only about 50% of mining demand, 25% of refining demand, and well below 20% of magnet demand in diversified regions. The report is also right that recycling matters, potentially cutting primary mining needs by up to 35% by 2050, and that magnet production—not mining—is the real choke point.

Where the IEA Overreaches

But the report also stretches in places. Its headline figure—$6.5 trillion in downstream production at risk—is not a forecast and not the same thing as realized economic loss. It is a modeled exposure scenario based on the suspended October 2025 export controls being fully implemented and licenses not being granted. That makes it useful as a stress test, but easy to overread as a probable outcome.

The report also assumes diversification is largely a matter of capital, policy, and coordination. Those matter. But the harder truth is industrial know-how. The IEA itself documents the shortage of separation data, equipment suppliers, metallization capability, grain-boundary diffusion equipment, and magnet-making expertise outside China. Money alone does not compress tacit knowledge.  Other factors such as the financialization of the effort to rebuild supply chains is beyond the scope of the topic.

The Real Takeaway

What the IEA has produced is not a roadmap to quick independence. It is a sober admission that the West is years (likely decades) behind in the most technical segments of the chain. The report’s eight recommendations—stockpiling, financing, ecosystem building, recycling, innovation, price transparency, and international coordination—are sensible. But Rare Earth Exchanges™ suggests that the deeper implication is harsher: diversification will be slower, costlier, and more politically managed than many policymakers still pretend.  

Citation: International Energy Agency, Rare Earth Elements: Pathways to Secure and Diversified Supply Chains (2026).