China's Export Controls Rewrote the Playbook

On April 4, 2025, China introduced an export licensing regime covering seven medium and heavy rare earths (samarium, gadolinium, terbium, dysprosium, lutetium, scandium, yttrium), including metals, oxides, alloys, and permanent magnets containing Dy/Tb (FACT: Chinese Ministry of Commerce, April 2025). October 2025 decrees extended controls extraterritorially, restricting re-exports of Chinese-origin rare earths by foreign entities.

The impact has been dramatic. NdPr oxide surged from approximately $49/kg in December 2024 to $103/kg by March 2026 (FACT: SMM, March 2026). Dysprosium oxide reached $890-895/kg and terbium oxide hit $730-760/kg in April 2026 (FACT: CRIA, April 2026). BMI revised its 2026 NdPr forecast up to $90,000/t (FACT: BMI/Fitch, 2026).

The controls are not a temporary trade measure — they represent a structural shift in China's rare earth policy, treating rare earths as strategic national assets. Export licenses are granted on a case-by-case basis, and approval rates have been low for Western buyers since the controls took effect.

Where the Consensus Is Wrong: Substitution Is Not a Near-Term Solution

The consensus narrative holds that permanent magnet alternatives (ferrites, induction motors) will reduce rare earth demand. This ignores two realities: first, EV traction motors and direct-drive wind turbines require NdFeB magnets for efficiency and power density that ferrites cannot match. Second, the substitute technologies have 5-7 year development cycles before commercial deployment at scale.

The market also underestimates the magnet supply chain bottleneck. China controls approximately 90% of permanent magnet manufacturing capacity (FACT: Adamas Intelligence, 2026). Western magnet projects (MP Materials in Texas, Lynas in Kalgoorlie, Neo Performance in Estonia) are progressing but face 2-4 year construction and qualification timelines.

The IRA's 30D tax credit for EV battery components requires critical minerals processing in US FTA partners. Rare earth magnets are explicitly covered, creating a compliance imperative for automakers that no substitute can resolve in the near term.

The Supply Gap: Mining Is Not Processing

MP Materials (Mountain Pass, California) restarted rare earth mining and now produces approximately 40,000 tonnes of REO annually, but its processing capacity lags significantly. MP ships rare earth concentrate to China for separation — the same dependency the policy seeks to eliminate. Its downstream separation facility in Texas is expected to start production in 2027 (FACT: MP Materials annual report, 2025).

Lynas Rare Earths (Mt Weld, Australia) operates the only significant non-Chinese rare earth processing facility at its Kalgoorlie plant, with capacity to produce approximately 9,000 tonnes of NdPr annually (FACT: Lynas quarterly, Q1 2026). Lynas's expansion to meet 25% of global NdPr demand by 2028 is on track but faces cost pressures.

The fundamental issue is separation chemistry. Rare earth ores contain 14+ elements that must be separated individually through solvent extraction — a process that requires hundreds of mixer-settler stages. Building and commissioning this capacity takes 3-5 years and $200-500 million per facility.

Regional Breakdown: The Fragmented Rare Earth Supply Chain

China (70% mining, 90% processing): Dominant across the entire value chain. Baotou Steel Rare Earth and China Northern Rare Earth control most light REE production. Southern China (Jiangxi, Fujian) produces ion-adsorption clays rich in heavy REEs. Export controls are binding and enforcement is strict.

United States: Growing. MP Materials is the sole domestic rare earth miner (40,000 t REO/yr). DOE has allocated $700 million for domestic rare earth processing and magnet manufacturing under the IRA, but operational capacity remains 2-4 years away.

Australia: Lynas is the largest non-Chinese producer. Mt Weld mine produces approximately 25,000 t REO/yr. Kalgoorlie processing plant is operating but below nameplate capacity. Rare earths are classified as strategic minerals with expedited permitting.

Southeast Asia: Myanmar was a significant heavy REE supplier but political instability and Chinese border controls have disrupted supply. Vietnam has rare earth reserves but limited processing. Illegal mining in Indonesia is growing.

What We Do Not Know

Whether China will expand export controls to cover light rare earths (Nd, Pr) beyond the current medium/heavy focus. Such expansion would impact 90%+ of global NdPr supply (ESTIMATE: Adamas Intelligence, 2026).

The pace of Western magnet manufacturing scale-up. Current announced projects total approximately 30,000 tonnes of NdFeB magnet capacity by 2030, but only 20% is fully financed (ESTIMATE: Project Blue, 2026).

The potential for rare earth recycling (urban mining) to reduce primary demand. Current recycling rates for REEs are below 1% (FACT: IEA, 2025). Technology improvements could raise this to 10-15% by 2035.

What this means for buyers

Procurement teams purchasing rare earths in 2026 should prioritize supplier diversification, lock in annual volumes where possible, and monitor the shifting trade policy landscape. The structural themes outlined above will play out over 12-24 months, creating windows for renegotiation and hedging alike.