Here is a sentence worth sitting with: the next big disruption in mining may not start in a mine at all. It may start in a tank of sulphuric acid, or in a smelter running at half speed, or at a port that cannot take another ship. That is the quiet conclusion of the International Energy Agency’s Global Critical Minerals Outlook 2026, and it is the most under-reported shift in the commodities world in years.
For the past several years, the conversation about supply security has been dominated by one question: who controls the mines? Countries have scrambled for mining rights, poured money into exploration, and treated resource ownership as the whole game. The data says the game has moved. The prize is no longer the ore body. It is everything that happens after the ore comes out of the ground.
More ore does not mean more metal
Think of supply security as a long multiplication chain: mine capacity, times chemical inputs, times processing capacity, times power and water, times logistics, times customer certification. Any link that approaches zero collapses the whole chain. The West learned this the hard way during the pandemic-era chip shortages, but the same logic applies to every critical material. Having ore underground is not the same as having metal on the market.
The IEA numbers are blunt. Refining is becoming more concentrated, not less. Excluding rare earths, the top refining country’s average market share rose from 70 percent in 2023 to 72 percent in 2025. Meanwhile, new investment is still flowing disproportionately into upstream mines, while refining and downstream manufacturing lag behind. And here is the oddity the outlook highlights: utilisation at smelters outside China has fallen below 70 percent, while China’s own smelters run at roughly 85 percent. The world may own more mines than ever, yet the usable midstream capacity has never been more concentrated.
You do not need to draw a political conclusion from that. The mechanical conclusion is enough: whoever controls the processing step controls the supply. And the processing step, unlike a mine, cannot be switched on quickly. A mine can sit idle and restart; a smelter that loses its workforce, its acid supply or its power contract takes years to bring back. Midstream capacity is the part of the chain that is hardest to rebuild, which makes it the part worth worrying about.
One barrel of acid, ninety-day stress test
The study’s most striking finding is about a chemical most people have never thought about: sulphuric acid. It is not glamorous, and it is not rare. But it is the workhorse input for processing copper, nickel, cobalt, lithium and rare earths, not to mention phosphate fertiliser. When it gets scarce, mines keep digging while the metal stops flowing.
The scale is larger than you would guess. More than 15 percent of primary copper output — roughly 3.6 million tonnes a year — relies on acid-leaching routes. In a stress test, a 90-day interruption to acid supply would expose about 111,000 tonnes of copper production to risk; stretch that interruption to a year, and the exposure grows to nearly 950,000 tonnes. That is the difference between a wobble and a genuine supply crisis.
2026 already ran a real-world version of this test. The Middle East supplies roughly a quarter of the world’s sulphur, and about half of global seaborne sulphur trade passes through the Strait of Hormuz. With that route disrupted, sulphur and acid markets tightened fast. China tightened its own acid exports in May. The result is a cascading squeeze: copper smelters, lithium processors and phosphate fertiliser plants all facing higher input costs at exactly the moment demand for battery materials stays structurally high. The acid bottleneck is the kind of problem that touches nearly every supply chain at once, because so many processing routes depend on the same humble input.
Prices have already noticed
Markets rarely wait for reports to tell them what matters. Base metal prices rose roughly a third between January 2025 and April 2026. Lithium prices more than doubled. Cobalt climbed around 130 percent — driven substantially by export restrictions in the Democratic Republic of Congo rather than by demand growth alone. Each of those moves is a different story, but they share a common thread: constraints that sit downstream of the mine gate.
What matters for the next few years is not whether the world has enough ore. It probably does. What matters is whether the world has enough refining, chemical supply, electricity, water and logistics to turn that ore into usable metal. Those are harder problems than drilling a new pit, and they take longer to solve. A new mine can be permitted and built in years; a new midstream cluster — with its smelters, chemical plants, power contracts and trained workforce — takes a decade and a half of patient, coordinated investment.
What this means for investment and policy
The implications reach far beyond the commodity desks. Governments courting critical-minerals investment — and several Gulf sovereign funds are actively deploying into African mining and downstream processing — would do well to design fiscal terms around processing-chain dependencies, not just resource access. A royalty regime tied purely to extraction risks leaving value on the table while the bottleneck persists, as it is now expected to, well into 2027.
For industrial buyers, the practical lesson is to stop obsessing over mine supply headlines and start building visibility into the midstream: who supplies the acid, who runs the smelter, who owns the power contract, which port can take the cargo. Resilience in this market is built downstream, one input at a time. Companies that map their exposure beyond the first supplier will be the ones that keep producing when the next disruption arrives.
There is also a policy lesson for governments trying to build domestic processing capacity: subsidising a smelter is not enough. A smelter needs acid, power, water, rail, ports and a workforce, and each of those has its own bottlenecks. The countries that succeed at onshoring processing will be the ones that treat it as a systems problem, not a building problem. The ones that build the plant and nothing else will watch it sit idle waiting for inputs.
For everyone else, the takeaway is simpler and a little unsettling. We spent a decade assuming the critical-minerals race would be won by whoever sat on the biggest ore bodies. The 2026 outlook suggests the race is actually won by whoever can process, refine and deliver — the unglamorous, uncelebrated middle of the chain. The mining shock you should worry about is not the one that starts underground. It is the one that starts in a chemical tank, on a power line, or at a port that quietly became the load-bearing wall of the entire supply chain. That is where the next disruption lives, and it was never in the mine.