China's Magnet Exports Hit Record High While U.S. Supply Remains Restricted
0xBroomberg
China exported 31,507 tonnes of rare-earth permanent magnets in H1 2026, annualizing above the all-time record — yet the U.S. is the only major market that never recovered. Beijing's export-licence regime was never lifted, and America's share fell from 14% to 9.1%.
Exports at a record — so where is the "cut-off"?
H1 2026 shipments hit 31,507 tonnes worth $1.71 billion, annualizing at roughly 63,000 tonnes — above the 2024 full-year record of 58,142 tonnes.
The average export price rose to $54.3 per kilogram, a three-year high, driven by a higher share of premium-grade magnets.
This means → China's total magnet output never actually shrank. The country still makes over 90% of global rare-earth permanent magnets, with no viable alternative source.
What actually happened during the 2025 "cut-off"?
On April 4, 2025 — two days after Trump's "Liberation Day" tariffs — Beijing imposed case-by-case export licensing on seven heavy rare earths and magnets containing them.
Global shipments plunged from 5,325 tonnes in March to 1,239 tonnes in May, a 75% drop. That month, only 46 tonnes went to the U.S. and 26 tonnes to Japan.
After licences were gradually issued, July rebounded to 5,575 tonnes and stayed at pre-control levels. In plain terms = the real disruption lasted about two months, but the licensing regime itself was never revoked — it survived the October 2025 U.S.–China trade truce intact.
Why is the U.S. the only market that never recovered?
U.S. imports fell from 7,446 tonnes in 2024 to 5,934 tonnes in 2025; H1 2026 totalled just 2,874 tonnes, averaging roughly 479 tonnes per month — about 20% below the 2022–2024 mean.
America's share of China's magnet exports dropped from 13%–14% pre-controls to 9.1% in early 2026, yet U.S. buyers paid an average of $53 per kilogram.
Over the same period, EU imports rose 28% year-on-year while U.S. imports fell 22%. This reflects a shortage that is destination-specific, not a global supply squeeze.
Where exactly is Beijing's valve set?
Supply side: licence applications involving foreign military end-use face a presumption of denial; the U.S. is explicitly listed. Heavy rare earths are the tightest choke — U.S. yttrium imports were just 17 tonnes from April to December 2025, versus 333 tonnes in the prior eight months.
Demand side: after the $7,500 U.S. EV tax credit expired, Q4 2025 U.S. EV sales fell over 30% and Q1 2026 fell 28%; Europe kept its subsidies and held demand steady.
This means → both sides of the equation work against the U.S. — supply is selectively restricted, domestic demand is shrinking on its own, and the two effects compound each other.
Can America make its own magnets?
MP Materials began commercial magnet production at its Fort Worth, Texas plant in late 2025, with initial capacity of roughly 1,000 tonnes per year, backed by a U.S. Department of Defense agreement that includes equity investment, a $110-per-kilogram price floor, and offtake commitments.
In plain terms = the U.S. currently produces 1,000 tonnes against annual demand north of 7,000 tonnes. The gap is still filled by the same Chinese supply now under restriction. A 10,000-tonne expansion — and rival projects — are not expected to reach scale until around 2028.
Japan spent a decade "de-risking" — did it work?
After the 2010 Senkaku/Diaoyu boat-collision incident triggered an informal Chinese embargo, Japan meaningfully reduced its dependence on Chinese rare-earth raw materials — but finished magnets still rely heavily on Chinese processing.
From January 2026, Japan's magnet imports face new restrictions too. This reflects a hard lesson: even an economy that invested over a decade in diversification remains exposed at the finished-magnet stage — raw materials and finished goods are two different gates.
Content is for reference only, not financial advice.