Tearing apart the DRC cobalt quota machine

Cobalt.

Bloomberg ran a piece yesterday declaring that cobalt's rally is reversing. Quotas flowing, exports normalizing, rally over. If you read the headline and moved on, you'd think the trade is done. That's the point of the headline. It gives you a story so you stop looking at the machine.

Here's the part it leaves out: the DRC produced 226,000 tonnes of cobalt in 2024. It capped exports at 96,600 tonnes for 2026. Less than half. The rest stays in-country — not because there's no demand, but because Kinshasa decided the global market doesn't get it.

This is not a tariff with a sunset clause. This is ARECOMS — the DRC's strategic minerals regulator — running a centrally planned export ceiling through at least 2027. The system launched in October 2025 after an outright ban that doubled prices. The quotas are designed to keep them there.

So when Bloomberg says the rally is "reversing" because some tonnage is leaving under the new quota, it's like saying a drought is over because a few trucks of bottled water arrived. The valve opened a quarter-turn. The reservoir is still behind the dam.

Strip the narrative. Here's the plumbing.

The DRC controls 73% of global mined cobalt. CMOC — the world's single largest producer — guided 100,000 to 120,000 tonnes for 2026 after a record 117,549 last year. Its export quota: 31,200 tonnes. A production-to-export ratio of nearly 4:1.

Glencore is in the same vise. It cut DRC cobalt production 39% in Q1 to 5,800 tonnes — not because the mine is struggling, but because there's no point extracting material you can't ship. Its 2026 quota: 22,800 tonnes against 33,500 produced last year in Congo alone.

On June 29, ARECOMS announced that all first-half quotas not shipped by June 30 would be forfeited — clawed back into the state's strategic reserve. That erased an estimated 15,000 to 20,000 tonnes of annual export allowance. It wasn't a warning. It was enforcement.

Downstream in China, the cupboard is emptying. Wuxi Stainless Steel Exchange cobalt stocks have more than halved since late January — down to roughly 3,934 tonnes. Glencore has been pulling from Wuxi warehouses to honor contracts with Chinese EV battery makers, because it can't get enough fresh material out of the DRC.

The IEA's Global Critical Minerals Outlook 2026: for copper and lithium, projected supply gaps narrowed. For cobalt, the gap widened — from just over 15% to above 25%. The only critical mineral where the outlook got worse. One word explains it: quotas.

Global demand: 276,000 tonnes in 2025, up 48% from 2022. Batteries now eat 75% of it. Indonesia is adding tonnes — 53,000 forecast this year — but that's HPAL nickel-cobalt byproduct, not primary cobalt. It shifts the dependency. It doesn't solve it.

The sell-side read on this pullback is that the DRC opened the tap and the price is normalizing. That framing requires you to ignore everything ARECOMS has actually done since October.

This is a regulator that forfeited 15,000 to 20,000 tonnes of unused quota at the end of June. Not rolled it over. Forfeited it. When Fastmarkets asked if quotas could shrink further, ARECOMS said publicly that reductions remain possible. That's a government telling you the ceiling can get lower.

The stranded stockpile is the hidden accelerant. CMOC alone has roughly 78,800 tonnes sitting in DRC warehouses — mined, processed, legally unexportable. Every month those piles grow, leverage shifts toward Kinshasa. Domestic processing feedstock. Strategic reserve. Bilateral bargaining chip. What the DRC does not have to do is release those tonnes to the global spot market on anyone else's timeline.

I've watched resource-state supply interventions before — tin in the 1980s, OPEC more times than I can count, Russia with palladium in the early 2000s. The pattern is always the same: the market prices the intervention as temporary, the government discovers it likes the revenue, and "temporary" becomes permanent. The DRC looked at OPEC and said, "Why not us?" Nobody on the sell-side has an answer.

S&P Global's own modeling shows the quota system pushes the cobalt market into a near-term deficit — and under a $20/lb price scenario, DRC export value rises roughly 24% by 2027 compared to 2024. If you're Kinshasa, that math works. You're selling less metal for more money. There is exactly zero incentive to loosen the valve.

Follow the pipe downstream. The DRC's quota system is not a commodity story. It's a manufacturing input story, and the transmission mechanism runs through one city: the Chinese refining complex.

China processes 78% of global refined cobalt. The feedstock comes overwhelmingly from the DRC as cobalt hydroxide. When the quota tightens the hose, there's no backup at scale. Indonesia's HPAL output helps at the margin but still routes through Chinese processing. The bottleneck doesn't move. It narrows.

Hydroxide prices surged 263% from Q4 2024 and are still up 77% year-over-year after this week's pullback. The pullback is real. What it isn't is structural. Paper reacting to a headline while physical runs short.

If ARECOMS tightens further — and they've said they can — the next leg isn't gradual. When exchange stocks go to zero, contract enforcement replaces price discovery. If you've seen palladium in 2000 or nickel in 2022, you know the scramble doesn't send a calendar invite.

Where does the capital go? Not into the broad battery-metals ETFs — those are diluted with lithium names and companies whose cobalt exposure rounds to zero. The edge sits with operators outside the DRC's quota wall. Jervois Global, which acquired Freeport Cobalt's Finnish refining operation, processes cobalt outside the Chinese chain entirely. And the recyclers — companies pulling cobalt from spent EV batteries — are the only players adding supply that doesn't cross a single African border.

The consensus sees a chart that went up and came down. The physical market sees warehouse stocks that halved in seven months, a regulator that clawed back 15,000–20,000 tonnes of quota it had already issued, and an IEA outlook that widened the projected supply gap by two-thirds. Kinshasa has the leverage. They're already using it. The only question is whether the market prices it in before or after Wuxi hits triple digits.