Is the ban leaking? Coltan/tantalum laundered DRC → Rwanda
Trade-flow companion to the price wedge (R72). This is a DUAL-SCORE / alternative-track signal — never folded into any Tier-1 exposure score. The divergence between "which flag the customs paperwork carries" and "which ground the ore actually came out of" IS the signal. Research, not investment advice; origin-relabelling is INFERRED from statistical implausibility, the UN's forensic tracing, and public ownership records — never asserted as smuggling on any single shipment.
This case is the conflict-mineral variant of the pattern. It combines two of the hardest modes in the register: the co-producer blind spot first seen on Belarusian potash (the trans-shipment partner, Rwanda, is itself the world's #2 tantalum origin, so a partner-flagged surge trips no naïve implausibility alarm), plus a rare named common-ownership pipe (like the antimony Youngsun→Thai-Unipet tell) — the Rwandan exporter and its offshore buyer share owners. It is also the sharpest illustration of the detector's deepest failure: here the reference production statistic itself is trade-derived, so the contamination reaches into the denominator you would measure the anomaly against.
Verdict
The controlled origin is the DRC, not China. On 12 February 2025 the DRC Ministry of Mines classified 38 coltan/cassiterite/wolframite concessions in Masisi (North Kivu) and Kalehe (South Kivu) — including the Rubaya artisanal zone — as "Red Zones," a total prohibition on extraction, transport and commercialisation, after the December 2024 UN Group of Experts report (S/2024/900) documented M23/AFC seizing and taxing Rubaya (see anchor action). The ban was extended in November 2025 through 11 February 2026. Two Western controls followed within days: the European Parliament (13 Feb 2025) called to suspend the EU–Rwanda critical-raw-materials MoU (Commission put it under review), and US OFAC (20 Feb 2025) sanctioned Rwandan minister-of-state James Kabarebe and M23 figure Lawrence Kanyuka.
The material does not disappear when the ground is red-zoned. It changes flag. The long-run customs record shows the textbook implausible-origin fingerprint, at the country a Tier-1 score would read as clean:
Table A — US tantalum imports by declared origin, 2013–2022 (Oakland Institute "Shafted", Oct 2025, from US customs/Census trade data)
| Declared origin | Volume | Value | note |
|---|---|---|---|
| Rwanda (limited domestic capacity) | >2,000 t | >$135 M | more than double the DRC's, despite far smaller geology |
| DRC (the actual dominant producer) | 878 t | $53.6 M | the real ore basin, under-declared as US origin |
| Rwanda's peak share of US tantalum imports | >50% | — | a ~zero-industrial-mining country as top US supplier |
| Rwanda→US tantalum export growth, 2013→2018 | 15× | — | steps up right after the 2012 M23 invasion + a US sanction waiver |
A country cannot legitimately export more than double the tantalum of the basin that holds the ore. The surge tracks conflict, not geology — it 15×'d after the first M23 invasion, not after any mine opening.
Table B — The 2024–25 smuggling flow the ban was written against (UN Group of Experts on the DRC — public S/2024/900 (Dec-2024) and the confidential mid-2025 report as reported by Reuters / Washington Post / The EastAfrican)
| Metric | Figure | Source |
|---|---|---|
| Rwanda reported coltan-ore exports, 2024 | ≥2,300 t | UN GoE 2025 (via Reuters) |
| Coltan smuggled Rubaya → Rwanda | 120–150 t/month (2024–25); 195 t in the last week of March 2025 alone | UN GoE S/2024/900 + 2025 report |
| Rubaya share of world coltan | ~15% | UN GoE / Reuters |
| M23 monthly revenue taxing Rubaya coltan | ~$800,000/month | UN GoE S/2024/900 |
For scale, USGS puts DRC mine output at 980 t (2023, ~41% of world) and DRC+Rwanda combined at ~58% of global tantalum — but note the trap in the next section: Rwanda's own USGS figure is partly derived from the very export data this case says is contaminated.
The corporate pipe (the common-ownership tell)
Unlike potash — where the co-producer blind spot left no ownership pipe to trace — the 2025 UN Group of Experts report named one, the first time the UN has named a company complicit since M23 took Rubaya:
- Boss Mining Solution (Rwanda) — exported **≥150 t of coltan worth $6.6 M in
2024 = 6.5% of all Rwandan coltan exports, the country's sixth-largest exporter**.
- Its managing director and ~one-third owner is Eddy Habimana, a Rwandan
businessman UN investigators had already identified a decade earlier as a minerals smuggler tied to eastern-Congo rebels.
- Boss Mining sources coltan from Speck Minerals, where Habimana is also a
listed representative — an intra-network supply leg.
- Boss Mining's other owners, Yuriy Tolmatchev and Alexander Konovalchik (dual
British-Russian citizens), also own the downstream companies that buy up Boss Mining's coltan. Specifically, Novacore FZE (UAE), managed by Tolmatchev, purchases all of Boss Mining's coltan.
The exporter and its sole offshore buyer share owners; the seller and its supplier share a principal. That closed loop — Rubaya ore → Boss Mining/Speck (Habimana) → Novacore FZE (Tolmatchev) → smelters — is the coltan analogue of the antimony Youngsun → Thai-Unipet → Youngsun-&-Essen pipe: a controlled-origin flow moving through an affiliated chain that lets the same interest sit on both sides of the "arm's-length" export.
The laundering document is ITSCI: UN and NGO reporting (Global Witness) find smuggled Congolese coltan leaves Rwanda carrying ITSCI traceability tags labelling it Rwandan — the clean paperwork that satisfies Dodd-Frank §1502 / SEC 13p-1 and the EU Conflict Minerals Regulation on its face.
Transmission chain
Rubaya artisanal coltan (M23-taxed, ~$0.8 M/month) → trucked across the border to Rwanda, often at night "to avoid detection" (195 t in one late-March-2025 week) → blended with genuine Rwandan production in Kigali by comptoirs such as Boss Mining → ITSCI-tagged as Rwandan origin → exported (Novacore FZE, UAE) → global tantalum smelters/processors → capacitors, jet engines, missiles, medical devices → US/EU OEMs. The origin label flips from a red-zoned DRC concession to a non-sanctioned, EU-MoU-partner flag at the border crossing.
What it implies for the Tier-1 score's blind spot
A Tier-1 exposure score keyed to declared country-of-origin sees "Rwanda" — a country that is not under a mineral export ban, that holds an EU critical-raw- materials MoU, and whose shipments arrive ITSCI-tagged — and scores the tantalum chain clean. The dual-score reads three divergences the Tier-1 view cannot:
1. Volume implausibility — Rwanda's US-import share (>50% at peak; >2× the DRC over 2013–22) exceeds anything its geology supports. 2. Co-producer blind spot — because Rwanda is a genuine top-2 tantalum origin, a Rwandan-flagged surge trips no simple "zero-capacity country" alarm (the potash failure mode); the signal has to come from the production-vs-export gap + the UN's forensic border tracing, not a naïve customs-share test. 3. Contaminated denominator — the deepest layer: USGS's Rwanda production line is itself substantially inferred from export/trade data. So the reference statistic you would use to size the anomaly already embeds the smuggled Congolese ore. The blind spot is not just in the numerator (Rwanda's exports) but in the yardstick.
Caveats
- Inference, not proof. Artisanal supply chains legitimately mix ores; ITSCI
also certifies genuine Rwandan production; no claim is made that any single shipment is smuggled. The pattern is a statistical-plus-forensic fingerprint.
- Unit heterogeneity. Sources mix coltan-ore/concentrate tonnes with contained-
tantalum tonnes; Table A (US tantalum imports) and Table B (coltan-ore exports) are on different bases and comparisons within each table, not across, are the valid ones.
- Source tiering. The Dec-2024 UN report (S/2024/900) and USGS, EU Parliament,
and OFAC figures are public/primary. The granular 2025 figures — Rwanda's 2,300 t export total and the Boss Mining ownership chain — come from a confidential UN Group of Experts report as reported by Reuters, the Washington Post, and The EastAfrican, not from a primary PDF; treated as named-public-source but a notch below a customs pull.
- Producer-side control. Unlike the China-consumer-ban cases, the control here
red-zones the producer's own ground; the leak is cross-border laundering into a co-dominant neighbour, so the "banned line collapses" test does not apply cleanly — the DRC's formal export line was never the leak.
Source note
Anchor and control: DRC Ministry of Mines Red Zone arrêté (12 Feb 2025; ext. 3 Nov 2025) — see the filed action 2025-02-12-drc-red-zone-masisi-kalehe-coltan- cassiterite, which cites the mines.gouv.cd prorogation and UN S/2024/900. Western controls: European Parliament press release (20250206IPR26752, resolution 13 Feb 2025); US Treasury/OFAC designation of Kabarebe & Kanyuka (20 Feb 2025, Federal Register notice 2025-03166). Trade tables: Oakland Institute, Shafted: The Scramble for Critical Minerals in the DRC (Oct 2025), analysing US customs/Census data (Table A); UN Group of Experts S/2024/900 and 2025 report via Reuters ("Major Rwandan coltan exporter bought smuggled Congolese minerals", 3 Jul 2025), Washington Post, The EastAfrican (Table B + ownership chain). Production shares: USGS Mineral Commodity Summaries 2025 (tantalum). ITSCI/laundering mechanism: Global Witness ("Who buys Rwanda's smuggled coltan").
Open questions
- Whether a further DRC prorogation past 11 February 2026 was issued (the anchor
action's expiry has passed).
- Whether the US-brokered DRC–Rwanda process (2025) conditions any sanctions relief
or MoU restart on verifiable de-linking of Rubaya ore from the Rwandan export stream — and whether ITSCI tightens tagging at the border.
- Novacore FZE's downstream smelter customers — the last unnamed hop between the
UAE trading arm and the ITSCI-clean tantalum reaching US/EU OEMs.
- Whether US import-by-origin data for 2023–2025 (post-Goma) shows the Rwanda
share moving again — a live test of whether the sanctions bit or merely re-routed the pipe (e.g. via the UAE leg).