Is the licence leaking? Tellurium after China's Feb-2025 export-licence regime

Seventh material in the circumvention track (R72), companion to the antimony, graphite, tungsten, bismuth, rare-earth-magnet and germanium cases. The price wedge measures scarcity (what a controlled material costs ex-China). This case measures the other half: is the control actually holding, or is Chinese-controlled material re-entering the US through a laundered origin? This is a DUAL-SCORE / alternative-track signal — never folded into any Tier-1 exposure score. Research, not investment advice; origin-relabelling is INFERRED from statistical implausibility plus public production records, never asserted as smuggling on any single shipment.

Verdict

On 4 February 2025 China's MOFCOM/GAC (Announcement No. 10 [2025]) placed tellurium — with tungsten, bismuth, molybdenum and indium — under dual-use export-licence control (45-day issuance, end-user/end-use verification). China refines ~80% of the world's tellurium (USGS MCS 2026). By the antimony logic that ought to be a prime laundering candidate. It is not — and the reason is the most useful thing in this case: the detector is *negative* because the US was never directly tellurium-dependent on China in the first place. China's direct US tellurium line did not collapse; it rose. No zero-capacity origin surged. The Feb-2025 licence bit hard on price (Rotterdam Te +84% YoY) while leaving no trans-shipment fingerprint in US import-origin data — because there was no China-direct line to reroute.

Source: UN Comtrade (reporter USA, HS 280450, annual), free public preview API, pulled 2026-07-14; committed artifact `data/intelligence/tellurium-circumvention.json` (annual series 2021–2025). Refining shares and price: USGS Mineral Commodity Summaries 2026 via `lib/iptm-material-country-production.ts`. China's peak direct US line was 4,712 kg (8.6% share) in 2022 — it was never a dominant US supplier.

Why the detector does not fire (the honest read)

This is neither the antimony fingerprint (China collapses, a ~0%-capacity origin surges) nor even the bismuth pattern (China was the big supplier, collapses, a genuine ally-refiner fills the gap). It is a third shape: the buyer was already diversified away from the controller before the control landed, so there is nothing to reroute.

• China's line rose, not collapsed. Under any reading of the (polluted) HS

line, China went from 1,247 kg (2024) to 5,243 kg (2025) and its share nearly tripled to ~20%. A control cannot leave a trans-shipment fingerprint if the controlled origin's direct line to the buyer grows. Either the US was a low-priority destination whose licences were routinely granted, or the moves are inside the noise of a boron-polluted code — both readings kill the China-collapse leg the detector needs.

• Every 2025 riser is a genuine tellurium source. Tellurium is a

copper-anode-slime co-product, so "who can plausibly export it" is "who runs a Te-recovery copper smelter." The risers are exactly those: Canada (5N Plus Saint-Laurent, the strategic non-Chinese Te-ingot feedstock for First Solar's Ohio CdTe fab, ramping ~25→50 t/yr), the Philippines (PASAR copper smelter), Japan (genuine Cu-anode Te), and China itself. None is a ~0%-capacity implausible origin. There is no Thailand/Mexico-style relabel tell.

• The dominant line is probably not even tellurium. HS 280450 pools boron

with tellurium and cannot be split below HS10. Germany's large, stable, high-unit-value line ($199–213/kg — above 2024–25 Te spot of ~$75–150/kg) is most consistent with boron element, not Te, and it fell in 2025 anyway. The levels here are unreliable; only the direction of the China line and the genuineness of the risers are load-bearing.

• No common-ownership tell was traced, and none is expected: with no

implausible surger, there is no shell-relocation chain to look for.

Where the control actually shows up: price, not flow-origin

The Feb-2025 licence is not toothless — it is one of the sharpest single-policy Te price shocks on record. Per USGS MCS 2026, Rotterdam tellurium averaged $81.54/kg (2024) → $150/kg (2025), +84% YoY, and the US warehouse price rose $74.77 → $120/kg, +60%. That is the price wedge showing the scarcity the flow-origin data hides. The bite lands on First Solar's CdTe feedstock economics and on 5N Plus's input costs — a margin and capacity-expansion problem — not as a laundered US import origin. The alternative-track lesson is that for tellurium the price channel is the live signal and the origin channel is silent by construction.

What it means for the buyer

1. A dominant-refiner control does not always produce a laundered origin. The antimony fingerprint needs the buyer to have been directly China-dependent. The US sources primary tellurium from copper smelters in Canada, the Philippines, Japan and (USGS-withheld) domestic ASARCO — China was a marginal ~6–9% direct supplier at most. Diversification before the control is the best defence, and it removes the very fingerprint the detector reads. Absence of a fingerprint here is a sign of prior resilience, not of a missed leak. 2. Read the price channel when the origin channel is silent. For an already-diversified buyer, a Chinese control on a co-product metal will surface as a global price step (First Solar's Te bill, 5N Plus's margins), not as a US-import relabel. A risk model watching only import origins would score tellurium as "unaffected" in 2025 — exactly wrong; the +84% Rotterdam move is the true reading. 3. Concentration risk migrated to a single ally, quietly. The non-Chinese Te spine now leans hard on 5N Plus (Québec) as First Solar's feedstock. That is a genuine, friendly source — but it is single-point. A 5N Plus outage or a copper-market shift in Te-recovery economics now moves the Western CdTe-PV supply chain more than any Chinese licence does.

Method & honesty rails

• Trade data: UN Comtrade, US imports, annual, HS 280450 ("boron; tellurium").

The two elements are pooled at HS6 and cannot be split without HTS-10 access (not exposed by the free preview API). Net weight is gross product weight, not contained Te. 2025 is treated as full-year but may still be revised.

• Production baseline: USGS MCS 2026 refined-tellurium table via

lib/iptm-material-country-production.ts: China ~80%; Russia ~6.7%; Sweden+US+Japan+Canada ~27% combined. Genuine risers (Canada 5N Plus, Philippines PASAR, Japan) are all listed copper-smelter Te sources.

• Alternative-track only: never touches buyerRelativeScore or the base

exposure — it sits beside them, like the China–West price wedge.

• Inference, not accusation: the negative circumvention read is driven by the

data (China's line rose; risers are genuine), not by exoneration of any party. No shipment is asserted legal or illegal beyond the public record.

• What would change the verdict: a future year in which China's direct US Te

line collapses and a ~0%-capacity origin (no copper-smelter Te recovery) surges into the gap, or a traced ownership chain from a Chinese Te producer to a surging non-refiner exporter-of-record. Neither is present now. Until then tellurium is the "already-diversified buyer" case: the control bites on price, the origin channel stays silent, and the detector correctly reports nothing.