Market setting
Copper spent the week consolidating after last week’s record spike, when COMEX copper touched an all-time high of US$6.90/lb (US$15.21/kg, US$15,212/t) on 6 August. Trading Economics and wire pricing show the most-active COMEX contract easing back into a US$6.58–6.70/lb band (US$14.51–14.77/kg, roughly US$14,510–14,770/t) through the week to 13 August, with the metal still up about 47% year-on-year. London Metal Exchange three-month copper held the US$14,000/t (US$14.00/kg, US$6.35/lb) level it broke through last week, settling near US$14,195/t (US$14.20/kg, US$6.44/lb); Bloomberg’s 10 August headline, “Copper Holds $14,000,” attributed the resilience mainly to a softer US interest-rate outlook rather than fresh industrial demand. The Financial Times and Reuters have both framed the London-New York divergence as a tariff-arbitrage story rather than a demand story, a reading the inventory data below supports. Cash metal on the LME traded roughly US$138–148/t above the three-month price — the widest backwardation since October — which, cross-checked across both exchanges, points to genuine physical scarcity rather than pure speculation. With prices lurching between fresh records and sharp pullbacks inside the same fortnight, the market looks to be entering a phase of volatile troughs and peaks rather than a smooth climb, a pattern several desks expect to persist as tariff, supply and rate headlines collide.
Macro pressure
The Democratic Republic of Congo’s concentrate export ban, in force since late June, continues to unsettle the market, and two fresh shocks have layered on top of it. First, a boiler leak forced Freeport-Mitsubishi’s Gresik smelter in Indonesia offline on 8 August; estimates of the affected cathode capacity range from roughly 340,000 to 400,000 tonnes a year depending on whether the adjoining Manyar plant is included, a discrepancy worth flagging given how thin published detail still is. Second, the Strait of Hormuz crisis is now hitting the cost side of the industry rather than the price side. With tanker flows through the strait reduced to a trickle, US diesel is trading near US$5/gallon and Australian diesel is up almost 49% to A$2.46/litre. Because energy accounts for roughly 22.7% of copper’s cash costs, and every 10% rise in oil prices is estimated to add about 3.5% to mining costs, a sustained disruption could lift all-in costs by mid-teens percentages at US$100/barrel oil — enough to curtail marginal, diesel-intensive open pits well before it shows up in production data. The Economist’s running “copper supercycle” commentary situates both the Congo ban and the Hormuz shock within a broader deglobalisation of critical-mineral supply chains, while the New York Times has kept its coverage anchored to the US tariff and industrial-policy debate — copper as a test of whether tariffs re-shore refining capacity or simply relocate stockpiles. Morgan Stanley separately points to a weakening US dollar, alongside supply disruptions at Kamoa-Kakula, El Teniente and Grasberg, as the structural reasons its desk remains constructive on price.
Regional signals
US warehouses remain the standout regional story. COMEX inventories have risen roughly eightfold since February 2025, from about 80,000 tonnes to a record ~650,000 tonnes, as traders position ahead of tariff decisions, while LME-registered stock has moved the other way, down to roughly 218,000–223,000 tonnes from over 300,000 tonnes in early July. Cross-checked across Trading Economics, COMEX and LME figures, that divergence means a very large share of global “free” copper stock now sits behind the US tariff wall rather than being available to the rest of the world — arguably a bigger driver of the LME’s physical tightness this week than any underlying demand surge. The Australian Financial Review’s domestic angle has focused on what a stronger-for-longer copper price means for ASX-listed producers and the Australian dollar, historically a reliable “copper proxy” currency. In China, elevated prices have started to bite: the Yangshan import premium has narrowed from US$115/t to US$96/t, and Codelco has walked back its 1.34-million-tonne 2026 production target. More strikingly, there is now measurable evidence of substitution at the margin — Chinese cable and wire manufacturers are reported to be substituting aluminium for copper in low- and medium-voltage power cable and automotive wiring as the copper-aluminium price ratio widens, consistent with the historical pattern of demand destruction above roughly US$13/kg (about US$5.90/lb).
Price structure
Forecasts remain unusually dispersed. UBS’s upgraded 2026 call of about US$6.00/lb (US$13.23/kg, US$13,228/t) already looks conservative against spot. Goldman Sachs has lifted its year-end 2026 LME target to US$13,735/t (US$13.74/kg, US$6.23/lb) from US$12,465/t (US$12.47/kg, US$5.65/lb), contingent on the refined-copper tariff taking full effect, and flags a possible 640,000-tonne deficit outside the US — even as the International Copper Study Group’s own estimates swing between a modest global surplus and a 150,000-tonne deficit for 2026. That dispersion is really a debate about causation. Is copper “high” because the market has concluded we have reached, or are approaching, peak mine supply — as some “peak copper” commentary now argues — or is it a narrower, largely US-based story about tariff positioning and a thin LME free float? Cross-referencing Goldman, UBS and the exchange data suggests both are partly true: near-term pricing is dominated by tariff and inventory mechanics, but banks’ willingness to keep lifting long-run 2030–2035 price and incentive assumptions — Goldman’s US$15,000/t (US$15.00/kg) 2035 case among them — shows real money is also betting that a market broadly balanced or in modest surplus today will face a genuine supply gap in three to five years. That, more than this week’s headlines, looks like the more durable reason long-dated copper futures and equities are being bid.
Supply and demand
Wood Mackenzie’s review of 14 major miners shows combined capital expenditure roughly doubling from US$30 billion in 2017 to a projected US$60 billion in 2025, with copper-focused capex up 40% and partly funded by US$19 billion of fresh debt — a sharp reversal of the “returns over growth” capital discipline that dominated roughly the 2013–2023 decade, and one Wood Mackenzie links directly to today’s thin project pipeline. CRU’s mid-year data reinforces the point from the concentrate side: treatment charges have collapsed from around US$21/mt to effectively US$0/mt, pushing miners and smelters toward index-linked pricing — evidence that concentrate supply, not just refined metal, is structurally tight. None of this changes the medium-term demand case: grid investment, EVs, renewables and AI-driven data-centre build-out remain intact structural drivers, and Goldman still expects grid infrastructure alone to generate over 60% of demand growth to 2030. The read-through on capital discipline is that a decade of investor-enforced restraint, more than any lack of resource, is the more persuasive explanation for why supply cannot yet meet that demand — and why, even with prices at record highs and volatility rising, few in the industry expect the incentive to invest to fade soon.
Sources: Australian Financial Review; Financial Times; The New York Times; Wood Mackenzie; CRU Group; The Economist; London Metal Exchange (LME); COMEX; Trading Economics; Reuters; Bloomberg; Goldman Sachs Research; Morgan Stanley Research.