Will China ease export restrictions or licensing for critical minerals or related technology?

TL;DR

I put YES at 76%, conditional on the September 24 meeting taking place. The main path is a new extension of China’s rare-earth export-control suspension beyond November 10, which U.S. officials reportedly requested for announcement after Xi’s visit. The main failure path is a successful summit centered on agriculture and tariffs, with mineral language too vague to qualify.

Context

China’s April 4, 2025 controls still require licenses for seven medium and heavy rare-earth categories and related products. A later package covering more rare earths, magnets, equipment, foreign-made products and processing technology is currently suspended through November 10, 2026, under MOFCOM/GACC Announcement No. 70. China also suspended a U.S.-specific restriction on gallium, germanium, antimony, superhard materials and graphite through November 27, 2026.

A new extension beyond either expiry would qualify even if China keeps the underlying licensing system. Rare earths are in the final preparatory talks, while broader truce-extension prospects were reported as strengthening on September 16; the planned Bessent–He Lifeng meetings also cover rare earths, trade and AI.

Evidence

The closest exact reference class is tiny and splits one for two. After the October 30, 2025 Busan summit, China announced a one-year suspension of the October controls, while a November 1 White House fact sheet also named general licenses for U.S. end users and their suppliers. The May 2026 Beijing summit produced only promises to study concerns and review compliant civilian applications, despite broader White House language. Under this question’s strict test, Busan was YES and Beijing was NO: 50%, with N=2. (marketscreener.com)

The strongest edge is a September 17 NPR report. Three former U.S. officials with direct knowledge said China had offered to extend the rare-earth moratorium through January 2029. They said Washington instead requested a six-month extension, timed for announcement after Xi’s September visit. That is close to the exact event being forecast. China’s May 20 official statement had already said extending the joint arrangements served both countries’ interests and that negotiations would continue toward an extension. (marketscreener.com)

The adverse evidence is real. In July, USTR Jamieson Greer described the summit as stock-taking rather than a venue for major new deals. In September, he publicly highlighted agriculture and the Board of Trade, not a mineral-control extension. Reuters reporting from September 4 found continuing license delays and supplier refusals, while both governments accuse the other of violating the existing bargain. This raises the value of an extension but also the chance that reciprocal terms remain unresolved.

My event tree assigns 80% to an extension of at least one covered Chinese mineral-control concession being agreed at the summit. Conditional on agreement, I assign 93% to an official source describing the affected controls and new duration clearly enough by the deadline. If no extension is agreed, I assign 8% to another qualifying easing, with an 85% chance of timely, specific documentation. The calculation is:

$$P(\mathrm{YES})=0.80\times0.93+(1-0.80)\times0.08\times0.85=0.7576.$$

The high documentation estimate reflects the same-day Chinese policy summary after Busan and the detailed White House fact sheet two days later. A U.S. readout can resolve this question without waiting for a later Chinese legal order.

What's non-obvious

A YES would probably not mean that China has relinquished its mineral leverage. Extending Announcement No. 70 preserves the suspension of the October expansion while leaving the April 2025 licensing regime intact. China could therefore qualify for YES while sensitive U.S. aerospace, semiconductor and defense users still face slow or denied access.

The timing rule also cuts both ways. A specific extension announced during the pre-summit ministerial talks would enter the baseline and would not count, while a generic leaders’ statement that the “trade truce” continues may fail the item-and-policy specificity test. The positive signal is that the reported U.S. plan explicitly called for announcement after Xi’s visit.

Uncertainties

The central report uses anonymous former officials and describes proposals first made before the May summit, so its terms may have changed over four months. The parties could reserve the extension for November negotiations, issue only language about facilitating civilian exports, or let formal details slip beyond September 26. China’s Mid-Autumn Festival holiday runs September 25–27, which adds some risk to rapid MOFCOM legal publication, though a specific White House or joint readout would still qualify. I found no authoritative Chinese confirmation of a finalized extension by the September 18 cutoff; that absence shows the agreement is not public, not that no private agreement exists.