US critical minerals and rare earth strategy: breaking dependence on China after

The United States is not going to win a mining race against China. What it can do — and is now doing, urgently — is build just enough sovereign and allied supply to keep fighter jets flying, submarines running, and defense production lines open while the broader ecosystem catches up. That is the realistic goal. Everything else is a longer game.

Here is what you need to understand, and what it means for how you think about the next three years.

What to know first

China controls roughly 85% of global rare earth processing and 94% of the permanent magnets that go into F-35 actuators, electric vehicle motors, missile guidance systems, and wind turbines. In April 2025, Beijing began requiring export licenses for seven critical rare earth elements. Shipments of rare earth magnets to the United States fell 93% within weeks. Ford suspended SUV production. European prices hit six times Chinese domestic levels. This was not an embargo — it was something more insidious. Call it "licensed scarcity": Beijing selectively delays or denies export approvals to specific buyers and end-uses without ever formally cutting off supply, staying technically within trade rules while degrading your supply chain in ways that are nearly impossible to challenge legally or attribute publicly. That mechanism is still running. It did not pause when the two governments shook hands on a truce in October 2025.

What the U.S. is doing about it

The response is the largest critical minerals mobilization in American history, and it is moving on four tracks simultaneously.

Domestic onshoring. The Pentagon has taken an equity stake in MP Materials, the only large-scale rare earth mine in the Western Hemisphere, and committed to buying up to 7,000 tonnes of rare earth products per year for ten years at a guaranteed price floor of $110 per kilogram — roughly double the current spot price. That price floor is the key mechanism: it explicitly protects the domestic producer from the scenario that killed the last U.S. rare earth company in 2015, when China flooded the market with below-cost production and drove it into bankruptcy. MP Materials is building a 7,000-tonne-per-year magnet manufacturing campus in Texas, targeting 2028 commissioning. Apple has committed $500 million to source magnets from that facility. Combined with a smaller facility already running and a recycling operation in San Marcos, total U.S. magnet capacity could reach roughly 14,000 tonnes per year by 2028 — enough to cover projected defense needs.

Strategic stockpiling. Project Vault, announced in February 2026, is a $12 billion public-private reserve funded partly through the largest Export-Import Bank loan in U.S. history. It is the bridge: the only tool that can absorb a supply shock between now and 2028, when domestic capacity comes online. Think of it as the buffer that buys time for everything else.

Allied coalition. FORGE — the Forum on Resource Geostrategic Engagement — launched in February 2026 with 54 countries and 11 new bilateral agreements. The strategic logic is to create a "China-light" supply ecosystem across Australia, Canada, Japan, South Korea, and resource-rich partners in Africa and Latin America. The critical test comes around July 2026, when a 180-day negotiation deadline under U.S. trade law forces FORGE members to either commit to binding price floors or face tariffs on their mineral exports to the U.S. That deadline will reveal whether the coalition has real teeth or is a collection of press releases.

Innovation and substitution. BMW is rolling out cars in 2026 with rare-earth-free motors. The Department of Energy has a $500 million funding round for processing, recycling, and alternative materials. Recycling currently covers less than 3% of demand, but the economics improve as end-of-life magnets accumulate. This track matters most for commercial applications — EVs, consumer electronics — where rare-earth-free alternatives are technically feasible and the security stakes are lower.

What you've probably been told wrong

The story is usually framed as a mining problem. It is not. The U.S. has ore in the ground. The chokepoint is processing — specifically, the separation of heavy rare earth elements like dysprosium and terbium, which are essential for magnets that work at high temperatures. China has a near-monopoly on that capability. Even MP Materials' flagship mine produces ore that is almost entirely the wrong type for heavy rare earths, meaning the new separation circuits will depend on feedstocks from third parties whose own supply chains may still run through Chinese processing technology. Digging more holes in California does not solve that problem.

You have also probably heard that the October 2025 truce means the crisis is paused. It is not. The original April 2025 controls on seven rare earth elements remain fully in force. The truce suspended some later expansions — and even that suspension expires sometime in late 2026, with the exact date still publicly ambiguous. Licensed scarcity continues to operate in the background.

What is still genuinely uncertain

The 2028 commissioning date for the Texas magnet campus is aggressive for a greenfield industrial facility; delays would extend the vulnerability window significantly. The heavy rare earth separation problem has no clean solution before 2028 under any realistic scenario. Whether FORGE members will accept binding price commitments when the economic costs become concrete — particularly South Korea and Japan, which have deep commercial ties to China — is unproven. And China retains the option to simply drop rare earth prices to $40 per kilogram, bankrupting unprotected Western producers before they reach scale.

The window of maximum danger is now through 2028. The architecture to survive it exists. Whether it executes on time is the only question that matters.

Full research report with citations, methodology, and evidence →