Tearing apart the Western zinc smelter bottleneck

Zinc.

Bloomberg called it "balanced." CNBC ran a segment last week about weak Chinese auto demand dragging zinc lower. The sell-side is still treating this as a demand story — car sales down 22%, die-cast alloy operating rates at 38%. Bearish, right? Case closed.

Except the price just hit a four-year high. Up 25% this year. And nobody on the demand-weakness panel can explain why.

Here's what's actually happening: the zinc market isn't breaking because of demand. It's breaking because the refining layer between the mine and the end user is collapsing in the West. Smelters aren't negotiating fees anymore. They're paying miners for the privilege of processing ore. That sentence is not rhetorical. Spot treatment charges for imported zinc concentrate in China hit negative $117.50 per dry tonne in August — an all-time low. The smelter is literally writing the miner a check to get feedstock through the door.

That's not a pricing quirk. That's a midstream system running backwards. The old model said smelters convert concentrate, charge a fee, ship refined metal. The new model says smelters beg for concentrate, waive the fee, pray the silver content and acid sales cover their electricity bill. I've watched base metals for long enough to know that when the middle of the supply chain starts operating on hope, something downstream is about to snap.

The demand bears are reading the dashboard. The engine is in the smelter yard.

Story off. Numbers on.

When the International Lead and Zinc Study Group met last October, it forecast an 85,000-tonne surplus for 2025 and a comfortable 271,000-tonne cushion for 2026. Both numbers are now wrong. The world recorded a small deficit last year. This year's projection flipped to a 19,000-tonne shortfall — in a 14-million-tonne market, that's a rounding error on paper, but the physical market isn't rounding anything.

The supply side is hemorrhaging. Global mine output fell 2.6% year-on-year in H1 2026 — after ILZSG had forecast 2.4% growth. Zoom out further and the decline is structural: global mine production dropped 8.6% between 2015 and 2025. Smelter capacity, meanwhile, was broadly unchanged. That mismatch is the fault line.

The casualty list reads like a greatest-hits of zinc mining. Glencore's H1 zinc output fell 21% to 365,600 tonnes after its Lady Loretta mine reached end of life in December, removing roughly 110,000 tonnes of annual supply permanently. Teck's Red Dog in Alaska — once the world's largest zinc mine — produced 106,200 tonnes of zinc in Q1, down 9%, working through lower-grade ore as it staggers toward end-of-life by 2032. Boliden's Garpenberg in Sweden, Europe's most important polymetallic underground mine, got hit by a seismic event in March. Production dropped to 30% of guided capacity. Milled-volume guidance was slashed from 3.7 million to 1.5 million tonnes. Then in May, Glencore's Kazzinc smelter in Kazakhstan exploded — three dead, a building partially collapsed, output disrupted.

None of these are temporary. Grade depletion doesn't reverse. Seismic damage doesn't heal in a quarter. End-of-life is end-of-life.

LME three-month zinc touched $3,949.50 on August 26 — the strongest print since June 2022. The cash-to-three-month backwardation reached $231.75 the next day. That's not speculative froth. That's buyers paying a premium because they need metal now and there isn't enough in the warehouse to go around.

LME inventory sits at 95,000 tonnes. At one point on-warrant stocks fell to 22,850 tonnes — the lowest since February 2023. For context, total LME zinc stocks peaked near 276,100 tonnes in 2024. Nearly two-thirds of the buffer is gone.

Here's the part nobody on the sell-side is modeling: China is about to stop sharing.

Chinese refined zinc output rose 5.9% in H1 2026 while the rest of the world contracted 3.4%. By May, SHFE warehouse stocks had nearly doubled to 146,766 tonnes. China isn't short of metal. The West is. And for the first time, Citi analysts say China is reaching "self-sufficiency" in refined zinc — the tipping point where it no longer needs to import, and may start exporting only when the arbitrage window is wide enough to bother.

That's already happening. Chinese refined zinc imports collapsed 57% year-on-year in Q1. China briefly turned net exporter in Q4 2025, shipping metal into LME warehouses in Singapore and Taiwan to arbitrage the London squeeze. Then the window closed. Q1 outbound shipments totaled 10,600 tonnes. A trickle.

Meanwhile, the Western smelters that are supposed to supply LME-deliverable metal are dying on the table. Trafigura's Nyrstar — which runs the Hobart zinc smelter in Tasmania and Port Pirie lead smelter in South Australia — put both operations under strategic review after losing tens of millions per month. The Australian government stepped in with an emergency financial aid package. Trafigura's CEO called it publicly: "Markets alone will not solve this."

He's right. When your core revenue line — treatment charges — has gone negative, you aren't running a smelter. You're running a charity that happens to produce germanium and sulfuric acid. I've seen versions of this in other metals. The smelter doesn't gradually recover. It gets a government check, limps along for a year, then closes when the check runs out. The zinc is still in the ground. It just doesn't become metal fast enough to meet demand.

And here's the dry irony: Chinese smelters are only surviving because they grabbed a larger share of the shrinking concentrate pool. Chinese imports of zinc concentrate surged 30% in 2025 and rose another 5% in the first seven months of 2026. They're hoovering up the feedstock the Western smelters can't afford to bid for. Every tonne China locks up is a tonne that doesn't flow west. The squeeze feeds itself.

The sequence from here is mechanical. I've watched variations of it play out in zinc twice in the last decade, and in copper, nickel, and palladium before that. The choreography barely changes. Only the catalyst does.

On-warrant LME stocks already touched 22,850 tonnes. At current draw rates, the next leg below 20,000 is weeks away, not months. When that happens, the backwardation — already at $231.75, the widest since the October squeeze subsided — accelerates. The LME zinc squeeze last October hit $323 per tonne over the three-month contract, the highest spread since at least 1997. If another Western smelter wobbles — and Nyrstar is one bad quarter away from exactly that — the next squeeze will be worse.

China will export some metal. It already has. But only when the LME premium is fat enough to justify the logistics, and only in bursts — enough to alleviate but not end the tightness, as Reuters put it. China isn't going to bail out the Western zinc market for free. That's not how self-sufficient producers behave.

The second-order effects are what the macro desks are missing entirely. Zinc galvanizes steel. Every transmission tower, every highway guardrail, every solar-panel mounting frame gets dipped in it. Infrastructure bills worldwide are creating demand for galvanized steel at exactly the moment the zinc to coat it is getting rationed. It's like funding a highway project and forgetting to budget for the road surface.

Where does capital go? Not into the broad base-metals ETFs — those are diluted across a dozen metals with different supply dynamics. The edge, if there is one, sits with concentrate-leveraged miners whose feedstock is the bottleneck. Companies like Teck, whose Red Dog output is declining but whose remaining concentrate commands a seller's market with negative TCs. Or the polymetallic producers where zinc byproduct credits — germanium, indium, silver — turn a smelter's loss into a mine's windfall.

The risk is the one Trafigura already named out loud: markets alone won't solve this. If governments start subsidizing smelters at scale — not just one emergency package in Tasmania, but a coordinated Western industrial policy response — the squeeze moderates. I'll believe that when I see the appropriations bill.

Until then, the mechanism is simple. The mines are depleting. The smelters are drowning. The warehouses are emptying. And the only country with surplus metal has no obligation to share it. When a system loses its middle, the ends fly apart. That's where we are.