why everyone in industry is suddenly talking about 1 0 45334
why everyone in industry is suddenly talking about 1 0 45334

Why Everyone in Industry Is Suddenly Talking About Rare Earths

Industry

The answer to “why now” is not a discovery or a shortage. It is a sequence of administrative decisions taken in Beijing between April 2025 and June 2026, each of which turned a specialist procurement question into something chief executives are asked about on earnings calls.

Rare earths dominate industrial conversation because China, which the USGS puts at roughly 70% of world mine production, converted its position into an export licensing instrument. Controls introduced in April 2025 were widened in October 2025, partially suspended in a trade truce, then applied directly to named US companies in June 2026. The bottleneck was never geology. It is separation and magnet-making capacity, and that takes years to rebuild.

Key takeaways

  • USGS Mineral Commodity Summaries 2026: world mine production reached 390,000 tonnes in 2025, with China at about 70%.
  • China holds roughly one third of known reserves, so its dominance is industrial rather than geological.
  • The October 2025 controls were suspended until 10 November 2026; the April 2025 licensing regime was not.
  • The US Department of Defense took a stake in MP Materials in July 2025 with a $110 per kg NdPr price floor.

Not rare, and that is exactly the point

Rare earth elements are a group of 17 chemically similar metals, and the name is a historical accident rather than a description. Several are more abundant in the crust than copper. What makes them awkward is that they occur mixed together in ore bodies and behave almost identically in solution, so separating one from another takes long cascades of solvent extraction, considerable chemistry, and a tolerance for the waste streams that come with it.

That is why the market concentrates. The scarce asset is not the deposit, it is the plant that can turn concentrate into a separated oxide, then into metal, then into a sintered magnet with the right coercivity at operating temperature. The USGS puts China at about 70% of mine production but roughly a third of reserves, and the gap between those two figures is the entire story.

The elements that matter commercially are narrower than the group of 17 suggests. Neodymium and praseodymium make the permanent magnets in electric traction motors and direct-drive wind turbines; dysprosium and terbium are added in small quantities so those magnets keep working hot. The heavy elements are the scarce ones, and they are the ones the export controls targeted first.

The sequence that made this a boardroom topic

  1. April 2025. China adds export controls covering metals, oxides, alloys and compounds of several medium and heavy rare earths, including terbium, dysprosium and lutetium. Shipments now require a licence, and licences take time.
  2. July 2025. The US Department of Defense agrees a $400 million convertible preferred equity investment in MP Materials, the only integrated US mine-to-magnet producer, alongside a $150 million loan for heavy rare earth separation at Mountain Pass.
  3. 9 October 2025. The controls are widened to cover all heavy rare earth elements.
  4. November 2025. Following a US and China agreement, implementation of the October measures is suspended for one year, to 10 November 2026. The April licensing regime stays in force.
  5. 22 June 2026. China adds ten US entities to its export control list, among them MP Materials and USA Rare Earth, moving from licensing to an outright prohibition on dual-use exports to those companies.

Read as a whole, the pattern is a state learning to use a chokepoint with increasing precision: first a category, then a broader category, then named counterparties. That last step is the one that changed the risk calculation for anyone with a single-source magnet supply.

What the 2026 figures actually show

The USGS released its Mineral Commodity Summaries 2026 in February, covering 2025. World mine production rose to about 390,000 tonnes from 380,000 tonnes the year before, so supply grew. The concentration did not meaningfully change.

The dependency numbers are the more uncomfortable part. The USGS reports the United States as 100% import-reliant for 13 critical minerals, scandium and yttrium among them, and more than 50% reliant for a further 20. Mining more ore in Australia, Brazil or the United States shifts the first link in the chain. It does nothing about the second one unless separation capacity is built alongside, which is the part that carries permitting risk, capital cost and a decade of operating know-how.

The scarce asset is not the deposit. It is the plant that turns concentrate into a usable magnet.

Two policy responses, two different bets

Washington and Brussels have taken visibly different routes to the same objective, and the contrast is instructive for anyone modelling where supply will sit in 2030.

  United States European Union
Instrument Direct equity, loans and offtake via the Department of Defense Critical Raw Materials Act, adopted March 2024
Core mechanism $110 per kg NdPr price floor over ten years 2030 benchmarks plus faster permitting for strategic projects
Targets 10,000 tonnes magnet capacity at the planned 10X facility, commissioning from 2028 10% extraction, 40% processing, 25% recycling of annual EU consumption
Concentration rule None stated; addressed through offtake No more than 65% of consumption from any single third country

The American approach removes price risk, which is the specific thing that killed previous Western rare earth ventures when Chinese output pushed prices below the cost of production. The European approach sets quantities and leaves financing to the market. Both are open bets, and the EU benchmarks in particular are widely read as ambitious relative to current project pipelines.

What it changes for an industrial buyer

For most manufacturers the exposure is indirect and therefore easy to miss. You may not buy dysprosium, but you buy motors, actuators, sensors, speakers and generators that contain magnets that contain it. The practical work is mapping that exposure before a licence delay does it for you.

Three things worth doing this quarter. Establish which of your assemblies contain sintered NdFeB magnets and where those magnets are made, not merely where the component is assembled. Ask suppliers whether any part of their chain runs through an entity on a control list, since the June 2026 measures apply to counterparties rather than to materials. And treat design substitution as a real option: ferrite motors, reduced-dysprosium formulations and induction machines all trade efficiency or mass for supply security, and the trade is now worth pricing rather than dismissing. That calculation sits close to the one we examined in whether industry can go carbon-neutral without going broke, because the same magnets sit inside most of the decarbonisation hardware.

The questions that keep coming up

Is recycling a realistic alternative?

It is a genuine and growing stream, and the EU has written a 25% recycling benchmark into law for 2030. It is not a short-term substitute, because the magnets available to recycle today were installed years ago and the collection and separation infrastructure is still small relative to demand.

Could the controls simply be lifted?

Partially, and that has already happened once. The October 2025 measures are suspended to 10 November 2026 under the trade truce. The April 2025 licensing regime was never suspended, and the June 2026 entity listings run in the opposite direction, so planning on relief would be optimistic.

How long does new Western capacity take?

Longer than the news cycle. MP Materials expects to begin commissioning its expanded magnet facility in 2028, and that project has federal equity, a price floor and an existing mine behind it. Projects without those advantages should be modelled on a longer horizon.

Are prices the right thing to watch?

Only partly. Availability and licence turnaround are the operational constraints; price is where the pressure eventually shows up. A stable quoted price with a twelve-week licence queue behind it is not a functioning market.

None of this is a story about a mineral running out. It is a story about where the middle of a supply chain is allowed to sit, and about how quickly an industrial advantage built over three decades can be converted into leverage. The talking will continue for as long as the separation capacity gap does, which on current schedules means well past 2028.

📈

Curious how far back these dependencies run?

Another everyday component with a supply chain most buyers have never traced end to end.

Follow the secret life of a microchip

Sources: US Geological Survey, Mineral Commodity Summaries 2026 (released February 2026, 2025 data); MP Materials and US Department of Defense partnership announcement, July 2025; European Commission, Critical Raw Materials Act (adopted March 2024) and its 2030 benchmarks; reporting on China’s April 2025 and 9 October 2025 export control measures, the November 2025 suspension to 10 November 2026, and the 22 June 2026 entity listings. Updated August 2026.

Comments

No comments yet. Why don’t you start the discussion?

Leave a Reply

Your email address will not be published. Required fields are marked *