Is the ban leaking? Russian diamonds after the G7/EU Jan-2024 import ban

Trade-flow companion to the price wedge (R72), fourth non-China entry. After crude (energy), aluminium and uranium, this is the luxury/gemstone axis of the Russia chokepoint — and the second clean mode-C transformation-washing case after Russian crude. The question is the same as antimony's: is the control actually holding, or is banned material re-entering the G7 through a laundered origin? Here the launder is not a phantom flag — it is a genuine industrial process (cutting and polishing) that, under substantial-transformation origin rules, legally confers a new country-of-origin on a Russian-mined stone. DUAL-SCORE / alternative-track signal — never folded into any Tier-1 exposure score. Research, not investment advice; origin-washing is inferred from the structural implausibility of the transit country's feedstock plus public trade and ownership records, never asserted against any single stone.

Live method + wedge context: [/situation-room → "Is the ban leaking?"](/situation-room).

Verdict

Russia's Alrosa is the world's largest rough-diamond producer — 31% of global output by volume (34.6 M carats, 2023) — and is 66% state-owned (Russian Federation 33% + Republic of Sakha/Yakutia 25% + Sakha municipal districts 8%; free float 34%). OFAC designated Alrosa and its CEO as SDN on 7 April 2022. The G7 then banned direct imports of Russian-origin non-industrial diamonds from 1 January 2024, and — critically — indirect imports of Russian diamonds processed in third countries, phased by stone size: ≥1.0 ct from 1 March 2024, ≥0.5 ct from 1 September 2024. The EU gave this legal force in its 12th sanctions package (Reg. (EU) 2023/2878, adopted 18 December 2023), and the G7 framed the indirect leg explicitly as eliminating the "substantial-transformation loophole."

The direct channel collapsed on schedule. The laundering channel — Russian rough cut in India (≈90% of the world's stones by piece, and effectively 0% of world mine production) and re-flagged through UAE/Dubai (which mines zero diamonds yet overtook Antwerp as the world's top rough-diamond exporter) — is where the exposure persists, and where free trade data goes structurally blind.

Sources: Belgium — FPS Foreign Affairs (Belgium) + The National (Antwerp figures). India Russian-rough — Interfax / GJEPC & Indian commerce data (2024 vs 2023). Alrosa ownership + 31% share — Alrosa investor relations + Wikipedia (shareholder structure); OFAC SDN — JCK (Apr 2022 designation). UAE rough-export lead — Rapaport ("Dubai Overtakes Antwerp as Top Rough Exporter"), bne IntelliNews ("traders flee Antwerp for Dubai"). Control instruments — Lewis Brisbois (G7 phased schedule), EU Reg. (EU) 2023/2878, EU 12th-package FAQ.

The mechanism: transformation-washing, not a phantom flag (mode C)

The antimony fingerprint is volume-implausibility — a country with ≈0% production capacity posts a large export surge (you cannot ship what you cannot make). Diamonds defeat that test the way Russian crude did: the transit country's process is real. Surat genuinely cuts the stone; the polished gem is a different object from the rough. Under customs origin rules, that substantial transformation makes the stone Indian-origin — so a Russian-mined diamond re-enters the G7 wearing an honest Indian label. The implausibility test therefore shifts from "can India produce this?" to "can India produce the feedstock?" — and the answer is no: India mines a negligible few thousand carats/yr (Panna, Madhya Pradesh) yet is origin-of-record for ~90% of the world's polished stones, importing its rough. In 2024 India still bought $664 M of Russian rough — a line that continues because the cut stone launders clean.

UAE/Dubai adds the second washing stage: a pure entrepôt with zero mine output that became the world's #1 rough exporter, a natural re-routing point for rough not welcome in Antwerp. The G7's response — routing compliant trade through Antwerp as the traceability node and eliminating the substantial-transformation loophole on paper — is an admission that stated origin lies for a controlled gemstone.

The ownership question (why this is GATE 0, not GATE 1)

Unlike antimony (Youngsun → Thai Unipet → Youngsun & Essen, a namesake shell pipe) or Russian crude (Rosneft's 49.13% equity in Nayara's Vadinar refinery), no operating common-ownership pipe is traced here. The wash flows through independent Indian cutters with no Alrosa equity — structurally identical to the Reliance/Jamnagar leg of the crude case (transformation-wash without equity). Alrosa has announced plans for an India polishing hub, which would convert this into a mode-B (capacity-relocation) equity pipe — but that is prospective, not operating, and is flagged as a watch item, not evidence.

What is missing for GATE 1: 1. The washed-polished share is unquantifiable in free data. Once a Russian rough is polished in India it is Indian-origin — no free dataset isolates "Indian polished diamonds of Russian rough origin." The whole point of the loophole is that the number does not exist. India's polished exports to the US fell >45% in 2023, but that is demand/tariff-driven, not a circumvention surge — so there is no clean surger figure on the polished leg. 2. The UAE Russian-rough share is qualitative. Dubai's rough-export lead is documented; the Russian fraction of it is reported as a routing shift, not a hard per-origin figure I can trace to a free customs line. 3. No operating equity pipe. Common-ownership tell N (Alrosa India hub prospective only).

The direct-channel collapse and the India-rough-continuation figures are each named-sourced; the laundered fraction — the actual size of the leak — is not, by the structure of the loophole. Hence GATE 0.

Why it matters for the buyer

1. Official bilateral data lies during a control — and here it lies *legally*. A compliance team reading "we no longer import Russian diamonds" is technically correct on the customs label and wrong on the molecules. Any exposure model keyed on stated country-of-origin is compromised for a controlled gemstone; the feedstock-implausibility cross-check (India/UAE mine ≈0) is the correction. 2. It confirms the control binds where it can be enforced and leaks where it cannot. The direct Antwerp channel severed cleanly; the third-country-cutting channel is exactly the surface the G7 traceability scheme was built to close — which tells the buyer where the enforcement front, and the next disruption, sits. 3. Mode C generalises across the Russia chokepoint. Crude → refined products (energy) and rough → polished diamonds (luxury) are the same trick on two commodities: a genuine transformation that legally relocates origin while the economic origin stays with the sanctioned source. Watch for the same shape on any Russian commodity with a substantial-transformation step abroad.

Method & honesty rails

• Trade data: rough diamonds HS 710231 (unsorted/industrial 710210, 710221);

polished HS 710239. Figures here are from named trade-body / national-agency reporting (GJEPC, Belgian FPS, Rapaport, Interfax) rather than a first-party UN Comtrade pull — a labelled sourcing step below the antimony gold standard, noted for gate honesty.

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

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

• Inference, not accusation: origin-washing is inferred from India/UAE's ≈0

mine output plus Alrosa's dominance and SDN status. No individual stone is asserted to be sanction-evading beyond what the cited public record states; a polished diamond of Indian origin is, under customs law, Indian.

• Anchor caveat: the register's closest filed touchpoint is the **US BIS

luxury-goods EAR rule (2022-03-16), which names diamonds/gemstones (HS 7102) but controls the *export-to-Russia* direction. The operative import-side** instruments (G7 Jan-2024 ban, EU Reg. 2023/2878, OFAC Alrosa SDN) are cited in full above and flagged for filing as a discovery lead.