The Rule That Was Suspended and the Rule That Was Not

On the tenth of November a Chinese suspension expires. Six export control announcements issued last October, including an extraterritorial rule that would claim jurisdiction over any product anywhere in the world containing as little as a tenth of one percent by value of Chinese rare earth material, were put on hold for twelve months pending negotiation. No extension has been announced.

It is the wrong date to watch. The control that is actually doing the work was never suspended. On the fourth of April last year Beijing made seven medium and heavy rare earths subject to export licence, among them dysprosium, terbium and samarium, along with any permanent magnet containing them. That regime has run continuously for eighteen months, and the effect is measurable. American imports of yttrium ran 333 tonnes in the eight months before it and 17 tonnes in the eight months after, a fall of about ninety five percent. Every tonne of dysprosium China exported in July went to South Korea.

The Historical Echo

This has been run once already, and the whole cycle took six years from embargo to bankruptcy.

In September 2010 a Chinese trawler collided with two Japanese coast guard vessels near the Senkaku islands and Japan detained the captain. Within a fortnight Chinese customs had stopped unprocessed rare earth shipments to Japan. Beijing denied any embargo throughout; the only formal measure on the books was a July quota cut that reduced the second half allocation by seventy two percent. Shipments resumed at the end of November. Japan then relied on China for close to ninety percent of its rare earths.

The price response outlasted the disruption. Dysprosium oxide averaged 129 dollars a kilogram in January 2010 and 2,377 in August 2011, a twenty six fold increase in thirty one months, then fell for five straight years to roughly 180 by 2016. The clearest evidence of what destroyed it is in the quota data: by the end of November 2011 only forty nine percent of China's export quota for that year had been used. The quota was not the binding constraint. The price was.

The Western supply response has an epitaph. Molycorp listed in New York in July 2010, never earned a profit after 2011, and filed for Chapter 11 in June 2015 while the Pentagon said it had no plans to rescue it. The Mountain Pass mine sold out of bankruptcy for about 20.5 million dollars.

Japan's response worked, but not the way the policy documents claim. Tokyo put a hundred billion yen into the problem in October 2010 and lent 250 million dollars to an Australian producer the following March for a guaranteed 8,500 tonnes a year. Dependence on China did fall from roughly ninety percent to about sixty, but it took fifteen years, and the larger part of the gain came from the demand side: Japanese rare earth consumption today is around half its 2010 level.

Where Patient Capital Is Positioning

The structural numbers have barely moved. China holds more than ninety percent of global refining and about ninety five percent of magnet making. The International Energy Agency's own projection is that existing and announced capacity outside China will cover roughly half of mining requirements, a quarter of refining and under a fifth of magnet demand by 2035. The only commercial producer of separated heavy rare earths outside China produced nineteen tonnes of dysprosium and terbium combined in its June quarter, against 1,857 tonnes of the lighter neodymium and praseodymium.

There are now two prices for the same element. Dysprosium oxide trades near 188 dollars a kilogram inside China and around 3,250 outside it. Terbium is about 860 inside and 7,500 delivered to North America. Any quoted rare earth price without a stated basis is close to meaningless, and the economics of a Western separation plant depend entirely on which curve it is allowed to sell into.

Washington's chosen instrument is no longer a tariff. The January proclamation covering all seventeen rare earths imposed nothing and raised the possibility of minimum import prices instead. One American producer already has a floor of 110 dollars a kilogram, and with the Chinese price near 97 that floor is being drawn on now. For long horizon capital the scarce thing here is not ore but separation and magnet capacity, which takes the better part of a decade and has so far been built at a rate of tens of tonnes a year against a problem measured in tens of thousands. The faster lever, as Japan found, is using less.

The Capital Memo