Antimony Supply Deficit 2025–2030: Where the Non-Chinese Tonnes Actually Come From

China stopped shipping antimony to the West in December 2024. Not a quota. A ban, effectively — export licences for the US and a few others dried up overnight, and the price of antimony trioxide went from roughly $14,000 a tonne to north of $39,000 inside eight months.

That's the number people should sit with. A metal most buyers couldn't point to on the periodic table, tripling in price because one country controls the taps.

I run GBX Resources out of Gilgit-Baltistan. We hold concessions that carry antimony among other things, so yes — I've got a horse in this. But I'd rather talk straight than sell you a fairy tale about tonnes that don't exist yet. Here's the honest version of the antimony mine supply forecast, and where the non-Chinese material actually shows up between now and 2030.

The deficit is real, but it's not the number everyone quotes

Global antimony demand sits around 140,000 to 170,000 tonnes a year depending on whose figures you trust. China produces roughly 48% of mined supply and controls a much bigger slice of refining — call it 80-something percent of the trioxide and metal that actually reaches a factory floor. So the bottleneck isn't just mines. It's the smelters and the roasters.

That matters because a lot of the "new supply" you read about is ore with no home. You can dig antimony in six countries. Turning stibnite concentrate into military-grade trioxide or high-purity metal is a different business, and most of that capacity still sits in China, or feeds China.

The deficit forecasts I take seriously land somewhere between 10,000 and 25,000 tonnes a year through 2028. Not apocalyptic. But structural — meaning it doesn't fix itself when the price spikes, because you can't stand up a new antimony flotation and roasting circuit in eighteen months. It takes four to six years if the geology cooperates, and the geology usually doesn't.

Look, I got this wrong at first. Early on I assumed high prices would pull tonnes out of the ground fast, the way they do with gold. Antimony doesn't work like that. The deposits are small, the metallurgy is fiddly (arsenic association is a real headache), and the buyers who need it most — defence primes making armour-piercing rounds, flame retardant makers, PV glass clarifiers — can't just swap it out.

So where do the actual non-Chinese tonnes come from?

Let me go through the honest list, not the press-release list.

Tajikistan. Already the second-largest producer, but here's the catch — a big chunk of that runs through a joint venture tied to Chinese processing. So on paper it's non-Chinese mine supply. In practice a lot of it feeds the same system. Worth watching, not banking on.

Russia. Polyus and the Olimpiada gold operation throw off meaningful antimony as a byproduct. Sanctions make that awkward for EU and US buyers, obviously. Japan and others quietly still take some. Call it available but politically radioactive for half the buyers reading this.

Bolivia and Mexico. Legacy producers, mostly small artisanal and mid-scale operations. Real tonnes, but fragmented, and the concentrate quality swings a lot. Hard to build an off-take you'd stake a defence contract on.

Australia. Larvotto's Hillgrove project in New South Wales is probably the most credible new Western hard-rock antimony story, with gold credits that pay for a lot of the mining. That's the model that works — antimony rarely pays for a mine on its own, so you want it riding alongside gold or tungsten. Hillgrove's targeting production later this decade. Real, but not tomorrow.

United States. Perpetua's Stibnite project in Idaho has federal backing and Pentagon money behind it. It's the one everyone points to for antimony production outside China. Permitting has been a decade-long grind, and first metal is a 2028-ish story at best. Big deposit, right country, slow clock.

Gilgit-Baltistan. This is where I sit, so weigh it accordingly. Our region sits in a metallogenic belt that carries stibnite mineralisation alongside gold and, in places, tungsten and molybdenum — the exact byproduct pairing that makes antimony economic. We're at the concession and exploration stage on the antimony side, not shipping concentrate yet, and I won't pretend otherwise. What I can say is the licensing framework in GB now allows large-scale foreign JVs, and we're roughly 40km of built road from the Karakoram Highway on the relevant blocks, which is the difference between a real project and a helicopter dream.

What this means if you're buying

Here's the thing serious buyers keep missing. The antimony supply deficit isn't going to be solved by one giant mine. It gets solved — partially — by a handful of mid-sized operations in politically acceptable countries, each producing 3,000 to 8,000 tonnes, most of them riding on gold economics.

That changes how you should structure an off-take. If you're an EV maker, a PV glass producer, or a defence supply chain buyer trying to lock antimony production outside China, you shouldn't be chasing the one mega-project. You should be tying up two or three of these emerging sources early, before they're producing, with pre-payment or equity, when your money actually moves the needle.

The mistake I see traders make is waiting for a producing mine with a spot cargo. By the time a non-Chinese operation is shipping steady tonnes, it's already committed under long-term contracts to whoever funded it. The window to secure supply is now, during the awkward pre-production phase, not after.

And processing — don't forget processing. A tonne of stibnite concentrate in Idaho or Skardu is useless to a German flame-retardant plant until someone roasts it to trioxide. The West is short on that capacity too. Anyone building an antimony mine supply forecast that ignores where the concentrate gets refined is only telling you half the story.

We're mid-exploration on our own antimony blocks and honestly I'd rather undersell that than have a buyer show up expecting containers next quarter. But if you're mapping where the non-Chinese tonnes come from over the next five years — which countries, which byproduct models, which processing gaps — Gilgit-Baltistan belongs on the long list. Where does your own supply run through China right now, and have you actually traced it?