I. The Central Fact: A Deficit Turned Surplus, in Plain Sight
The structural deficit that developed quietly in copper between 2018 and 2023 has, equally quietly, reversed. Combined inventories on the Shanghai, LME and COMEX exchanges surged almost sevenfold to a peak of 1.3 million tonnes at the end of March 2026, a level last exceeded in the early 2000s when copper traded near 65 cents a pound. Adjusted for the doubling of consumption since then, exchange stocks stand at approximately 18 days of demand. The only recent comparisons are 2013 (16 days) and 2018 (14 days). Both preceded roughly 50 percent price declines over the following years.
The surplus is not a forecast. It is what the WBMS numbers already show through May: refined demand of 11.1 million tonnes against refined supply of 11.6 million. The tape is doing what tapes eventually do — it is confirming the arithmetic. Copper still trades near three-year highs, and the consensus remains overwhelmingly bullish. That combination is our concern. When inventories at cyclical extremes coincide with sentiment at cyclical extremes, price and reality begin to argue with one another. Reality usually wins the second round.
II. Analysis: China Was Everything, and China Has Stalled
There is no way to hold a view on global copper without holding a view on China. Between 2010 and 2025, world copper consumption rose from 19.2 to 27.1 million tonnes — a 7.9-million-tonne gain. Chinese consumption alone rose 8.2 million tonnes over the same period. In arithmetic terms, China accounted for more than 100 percent of the increase in world copper demand for fifteen years running. Over the past three years, Chinese growth has slowed from its previous decade's average of roughly 700,000 tonnes a year to about 300,000. In the first five months of 2026, year-on-year Chinese copper consumption turned outright negative.
The prevailing bull thesis holds that copper-intensive investment will keep coming: renewables, electrification, data centres, the whole familiar list. Our work suggests the more interesting comparison is per-capita consumption relative to per-capita GDP. On that measure, China has moved from being an under-consumer of copper to what increasingly looks like an over-consumer: the copper-intensive investment already embedded in the economy appears more than sufficient to support materially higher living standards. If we are right, Chinese demand growth continues to slow through the rest of the decade. If Chinese demand slows, global demand slows with it. The remaining growth engines — India, Southeast Asia, the West's electrification build-out — are simply not large enough, fast enough, to fill the hole.
III. Implications: The Supply Losses Everyone Talks About Have Not Been Enough
Bulls point to the litany of supply disruptions and treat them as vindication. They are the opposite. Chilean production fell more than 200,000 tonnes year-on-year on severe weather at Caserones and Los Pelambres; Indonesian output dropped 125,000 tonnes as Freeport's Grasberg remained curtailed after the September 2025 flooding accident; Ivanhoe's Kamoa-Kakula has lost close to 150,000 tonnes since the May 2025 seismic event. Add it up and roughly a million tonnes of mine supply has been removed over the past twelve months. The market is still in surplus.
What happens when those disruptions reverse? Chilean weather passes. Grasberg is guided back to 65 percent of capacity by end-2026, full capacity by end-2027; full capacity is 800,000 tonnes annually. Kamoa-Kakula's second-half 2026 production is expected up 30 percent, with 400,000 tonnes projected in 2027. Cobre Panama, the 330,000-tonne mine shut since end-2023, may reopen: the Panamanian government has flagged a decision at year-end 2026 on a state co-ownership with First Quantum. Nearly a million tonnes of currently absent supply is scheduled to return over the next two years, into a market where demand has already turned negative in its dominant consumer. This is the classic setup for a multi-year price disappointment.
IV. The Position: Bullish the Complex, Cautious the Metal
We remain firmly bullish on commodities as an asset class. The 1968-1980 and 1999-2011 cycles teach that once capital begins migrating into raw materials as an asset class, individual commodities with deteriorating fundamentals can nevertheless continue rising for surprisingly long stretches simply because the money is looking for somewhere to go. Copper could well make new highs from here on that flow alone. It is not the same as saying the fundamentals justify them. They do not.
The historical parallel we keep returning to is 2006. Copper experienced a violent short squeeze that spring, four-fifths of the way through a bull market that had another four years to run — but roughly 95 percent of the total gains were already behind it. The 2024 short squeeze rhymes uncomfortably. Copper has risen another 25 percent since then, and a real bear market may still be several years away. But the arithmetic that matters for a European investor sizing a position today is that surplus is here, exchange stocks are at 18 days of consumption, China has stalled, and disrupted supply is coming back. The easy money in copper has been made. What remains is late-cycle sentiment against loosening fundamentals — a trade whose risk-reward, at this price and this consensus, no longer favours the buyer. Own commodities. Own energy. Own PGMs and coal. Do not own the last leg of a copper bull whose base has quietly given way.