Tearing apart the DRC cobalt quota expiry trap

Cobalt.

The financial press is writing about the DRC quota extension like it's a supply unlock. Reuters has a headline about "bearish relief." An analyst on Bloomberg is explaining that Q1 and Q2 allocations combined mean a flood of cobalt could hit the market before June 30. You already know what's coming: they're reading the policy document. They are not reading the logistics.

The actual story is simpler and nastier. The DRC's cobalt regulatory authority, ARECOMS, has given producers until June 30 — fourteen days from today — to ship both their Q1 and Q2 2026 export allocations or lose them. Unused volumes don't roll over. They get transferred to Kinshasa's national strategic reserve. The paper supply looks enormous. The physical throughput on the one truck corridor that actually moves this material says something very different.

This is not a new problem. When the DRC lifted its export ban in October 2025 and replaced it with quotas, producers couldn't clear even half their Q4 allocation by the original deadline. A bridge collapse on the primary trucking route. Customs paperwork that requires a specialist. A 10% royalty prepayment that has to clear within 48 hours per shipment before the truck can move. The system was designed to generate revenue. It was not designed for throughput.

The market has spent six months watching ARECOMS issue extensions and interpreting them as supply bearish. Every extension has been followed by continued tightness. This time the calendar is different. June 30 is a real cliff. And the Durban port schedule doesn't care about analyst consensus.

Numbers. No narrative.

The DRC produces roughly 70% of global cobalt supply. In 2025, CMOC alone produced 117,549 tonnes — its largest output ever. Its 2026 export quota is 31,200 tonnes. That's a production-to-export ratio of nearly 4:1. The company is physically moving four kilograms of cobalt for every one it's allowed to sell internationally. The rest sits in warehouses in Katanga, legally stranded.

Glencore's situation is tighter. Its full 2026 DRC allocation, including the 2025 carryover, totals 22,800 tonnes. Glencore has confirmed it shipped the majority of its 2025 quota in Q1 2026, with the balance going out in April. That means it has already deployed a significant portion of its runway. The Q1+Q2 window closing June 30 applies directly to whatever unused allocation remains.

Here's the physical pipeline that paper tonnes must travel through to become real supply:

From December through February — three full months — only 7,800 tonnes cleared export documentation against an 18,125-tonne Q4 quota. That's 2,600 tonnes per month of actual throughput. The Q1+Q2 window piling up for June 30 represents a theoretical export volume several times larger than what the corridor has demonstrably been able to move. Analysts who are modeling quota volumes as supply are not modeling the Zambia highway.

The Zambia corridor. Three thousand kilometers of trucking that nobody in a London trading office has ever personally sat behind. The cobalt hydroxide goes by road — Katanga to Ndola, Ndola to Livingstone, Livingstone to Durban. There is no rail alternative that works at the required throughput. The bridge collapse earlier this year didn't just delay shipments; it revealed that the corridor runs at near-capacity under normal conditions. There is no surge capacity. You can't ship four months of quota in two weeks because you don't have four months of trucks available in two weeks.

Here's the structural problem the consensus is missing. The market is treating the June 30 deadline as a supply release — quota finally clearing, prices soften, buyers relax. But the deadline is actually a forfeiture mechanism. Anything not shipped by June 30 transfers to ARECOMS's national strategic reserve, at the government's discretion. Kinshasa controls when that reserve re-enters the market. They have demonstrated precisely zero urgency about clearing stockpiles quickly. This is not a deadline that creates supply. It's a deadline that transfers ownership of the supply to the entity least likely to dump it.

I've watched this pattern in other quota regimes — Indonesian nickel, Chinese rare earths, Chilean water rights. Every time there's a paper deadline, the financial layer treats it as a supply event. The physical layer doesn't care about the deadline. It cares about whether a truck is available, whether the paperwork cleared, whether the port has berth availability in the next fourteen days. Almost none of those variables are moving favorably.

There is one more thing worth naming. MMG's Kinsevere operation has its cobalt plant on care and maintenance — they're shipping out of existing inventory rather than active production, because their quota is so small it makes active processing uneconomic. The general manager said publicly that the allocation renders cobalt production economically unviable. When producers are mothballing capacity because the quota doesn't justify running the plant, you don't have a supply glut. You have a policy that's masquerading as one.

June 30 passes. The corridor moved what it could move — call it 40–50% of the window's theoretical volume, consistent with every prior execution rate in this quota regime. The rest goes to ARECOMS's reserve. The financial layer expected a relief event. They get a forfeiture event instead.

The Chinese smelter complex is the mechanical center of this thing. Chinese imports of cobalt intermediates already declined through January and February — that's during a period when the market was supposedly unlocking. Smelters that were running on buffer inventory are now structurally shorter than they were in Q4. They need the June shipments to land. If the June shipments don't land on time — and the corridor math says most of them won't — Q3 feedstock availability tightens hard.

The secondary circuit matters here. Fastmarkets projects 36,000 tonnes of recycled cobalt equivalent in 2026 — up from 30,000 tonnes in 2025. Black mass and MHP from Indonesia provide some relief. But secondary supply runs on a completely different timeline than primary. You can't ramp recycled cobalt in two weeks to cover a quota-expiry miss. The Indonesian MHP ramp helps structurally but it's priced in already. It's not the swing factor in June.

Where does capital go if this plays out? Not the broad diversified miners. Glencore and CMOC are quota-capped — their cobalt upside is politically constrained regardless of spot. The edge, if there is one, is in refiners that source outside the DRC system — recycled cobalt processors, Indonesian MHP off-takers with unhedged spot exposure, and defense-grade alloy suppliers whose cobalt contracts are indexed to spot rather than locked forward. Companies that own the refining step, not the mining step.

The cobalt market has been pricing the quota as a ceiling. It may turn out to be a floor. Fourteen days to find out whether the Zambia corridor can move faster than it ever has before. I wouldn't bet on the corridor.