Copper Value-Chain Deep Research

Copper: a record high forged by concentrate famine, green demand taking the baton, low inventory amplifying elasticity

The direction of copper is set by concentrate scarcity, guided by green demand, and amplified by China's low inventory. With the SMM Imported Copper Concentrate Index (weekly spot TC) sunk to −$120.82/dmt (18 Jun 2026) and the long-term treatment charge at zero, the “mine pillar” of smelting profit has been hollowed out; meanwhile copper cathode spot has climbed to a record ¥104,885/t and China's social inventory has drawn from a peak of roughly 570kt in March to 194kt — a structure of “supply capped by the mine, demand reshaped by green, price made highly elastic by low inventory” has already taken shape.

The copper market at mid-2026 is a structural bull run driven by a famine at the raw-material end, validated by green demand, and amplified by low inventory. The SMM #1 copper cathode spot average printed ¥104,885/t on 18 June (+4.3% YTD), in the record-high region; the SHFE copper front-month was ¥104,780/t (+3.4%) and LME copper 3M $13,690.5/t (+9.8%). The “height” of the price is not a sentiment spike but the realisation of a long-running imbalance on both sides of supply and demand.

This report is built on SMM's own price, TC, output, inventory and customs data (cut-off 18 June 2026), cross-checked against the exchanges (SHFE / LME), ICSG and public customs figures, and advances one core judgement: copper has entered a new regime in which “the mine sets life-or-death, green sets the direction, and low inventory sets the elasticity.” On the supply side, the SMM Imported Copper Concentrate Index (weekly spot TC) has sunk to −$120.82/dmt as of 18 June 2026 (around −45 at the start of the year, −78 in early April, deepening weekly) and the 2026 long-term TC benchmark has gone to zero (collapsing from 80 in 2024 and 21.25 in H2 2025), hollowing out the treatment-charge income on which smelters depend; on the demand side, grid investment, solar/wind, storage, AI data centres and electric vehicles form a structural pull that offsets the weakening of the legacy property chain; on the inventory side, China's social inventory (ex-bonded) has been drawn sharply from a peak of roughly 570kt in early March 2026 to 194,300 t (19.43 ×10kt) (−24.6% YTD) — but this is a China-only reading: over the same period LME stock built counter-trend to roughly 355,725 t (35.6 ×10kt) (+149.5% YTD), and global visible inventory is not low because the US has built large tariff-driven (Section 232) stockpiles into COMEX; the genuinely tight metric — the one that amplifies price elasticity — is global available inventory-days excluding the US anomaly.

This judgement rests on three mutually independent yet aligned evidence chains. (1) The mine: incremental global mine supply is limited — Chilean and Peruvian legacy mines face declining ore grades and frequent disruptions, while the DRC (Kamoa and others) is the fastest-growing but still modestly sized source; concentrate scarcity transmits directly into a smelter profit crisis through a negative TC. (2) The smelting paradox: despite a zeroed/negative TC, China's monthly refined copper output still holds near its record high at 116.94 ×10kt (2026-05) — sustained by by-product (sulphuric acid) income, locked-in long-term volumes and the running-down of earlier concentrate stocks, with some lines pivoting to scrap/anode feed (the operating split: concentrate lines 88.5% vs scrap/anode lines 70.4%). We judge this “anti-TC high output” to be unsustainable — the key risk point for future production cuts and de-stocking. (3) The inventory divergence: China's SHFE stock 143,875 t (−20.3% YTD) keeps drawing down, while LME stock 355,725 t (+149.5%) builds markedly — the same metal, two markets, opposite directions, reflecting a regional mismatch of a tighter China against a looser overseas.

For the miners, smelters, fabricators, traders and financial institutions along the chain, the report's value is concrete and actionable. Using a three-factor framework of “supply-demand balance anchor + cost/ inventory floor + macro and sentiment premium,” we give a month-by-month price range for copper cathode over the next 12 months and quantify probabilities and trigger conditions across bull/base/bear scenarios (Chapter 7). Readers who need finer segment-level data can refer to Chapters 3–6; those who care only about price can go straight to the executive summary and Chapter 7.

The copper chain at a glance — the mine caps, smelting is squeezed, green pulls

The copper value chain is a long chain in which “the mine sets the ceiling, smelting is the squeezed middle, and demand is reshaped by green.” Upstream concentrate is highly concentrated in Chile / Peru / DRC; it is converted in China's smelters (about 55% of the world) into copper cathode, then through copper rod / copper products into grid cables, home appliances, transport and EVs, construction and new energy (solar/wind/storage). Running in parallel with the main chain is a scrap loop driven by the refined-scrap spread, which provides feedstock flexibility to smelting.

1. A record high: the resonance of macro and fundamentals

Understanding the 2026 copper price is not about any single day's level but about “why it can hold at a record high.” This chapter lets price behaviour itself tell the story — domestic and overseas markets strengthening in tandem, inventory drawdown and a negative TC corroborating one another; the “height” of the price is the realisation of imbalance, not a sentiment spike.

In H1 2026 the copper price ran in the record-high region and kept setting fresh records. The SMM #1 copper cathode spot average started the year around ¥100,575/t, surged to a year-to-date high of ¥108,500/t on 13 May, and printed ¥104,885/t on 18 June (+4.3% YTD). Over the same period LME copper 3M strengthened from the start of the year, touched $14,153/t in May, and closed at $13,690.5/t on 18 June (+9.8% YTD), with the LME cash close at $13,624/t. The synchronous strength at home and abroad confirms this is a trend driven by global supply and demand, not a local move in a single market.

The reason the price can “hold at a high” is the mutual corroboration of three fundamental signals. First, a negative TC: the SMM Imported Copper Concentrate Index (weekly spot TC) has sunk to −$120.82/dmt as of 18 June (around −45 at the start of the year, −78 in early April, deepening weekly) and the 2026 long-term TC benchmark has gone to zero — scarcity at the raw-material end has been written directly into the price. Second, inventory drawdown: copper cathode social inventory (ex-bonded) has been drawn sharply from a peak of roughly 570kt in early March to 194,300 t, deliverable resource is tightening, and this provides spot-level support for the high price. Third, demand validation: peak-season downstream operating rates have rebounded, with the wire & cable weekly op-rate +17.5% YTD — the high price has not broken through green-related rigid demand. Together the three form an “iron triangle” underpinning the high-level price.

It is clearer to separate macro from fundamentals. The macro side (the USD rate cycle, global liquidity expectations) sets copper's valuation elasticity and sentiment rhythm — the “amplifier” for whether the price can rise further in 2026; the fundamental side (concentrate scarcity, green demand, low inventory) sets the centre and the direction of the price — the “engine.” The hallmark of H1 2026 is precisely this: the fundamental engine continuously supplying power while the macro amplifier intermittently boosts — the two in resonance pushing copper to a record high. This is essentially different from a “valuation bubble” blown up by macro liquidity alone and lacking fundamental support: the high price of copper has real concentrate scarcity, real inventory drawdown and real green demand as its floor.

3. Concentrate famine: TC zeroed/negative, limited mine increment

Copper's supply story can be summed up in one sentence: incremental mine supply is limited, scarcity is realised through a negative TC, and the smelting treatment charge is hollowed out. This forms the fundamental constraint that makes copper hard to fall and easy to rise, and plants the seed for future smelter production cuts and de-stocking.

Copper concentrate scarcity is the most solid supply foundation of this bull run.

The SMM Imported Copper Concentrate Index (weekly spot TC) has gone from the positive zone of H2 2024 (around +$1 to +$11/dmt), turned negative in early 2025, and plateaued near −$40/dmt in H2 2025; entering 2026 it accelerated its deterioration — around −45 at the start of the year, −56 in March, −78 in early April, and to −$120.82/dmt by 18 June, deepening weekly — a trend deterioration, not a one-off disturbance. More symbolically, on long-term contracts: the 2026 copper concentrate long-term TC benchmark has gone to zero, collapsing from 80 in 2024 and 21.25 in H2 2025. The TC is the treatment charge a smelter earns per tonne of concentrate; a negative TC means the smelter is “paying to buy ore,” and the treatment-charge income on which it survives has been thoroughly hollowed out.

The root of the negative TC is a chronic shortfall in incremental mine supply. Global copper-mine increment is constrained by several structural factors: the legacy mines of traditional producing regions such as Chile and Peru face continually declining ore grades and rising capital and energy intensity per unit of output, with frequent disruptions from water, communities and strikes; the exploration-permitting-construction cycle for new large mines runs more than a decade, making it hard to respond quickly to price signals. The fastest-growing marginal source is the DRC (Kamoa-Kakula and other projects), but its scale and supporting infrastructure (power, logistics) remain insufficient to fill the global gap.

By flow, concentrate from Chile (largest), Peru and the DRC goes mainly to China — whose smelting capacity is about 55% of the world, making it the largest buyer and processing hub for copper concentrate. This also puts Chinese smelters in the passive position of “excess capacity scrambling for limited ore” in TC negotiations.

Customs data corroborate this pattern.

In April 2026 China's copper concentrate net imports were about 235.2 ×10kt (−10.3% YTD), down YoY; at the same time the smelter operating rate stayed elevated (total op-rate 85.44% in 2026-05). The surface contradiction of “concentrate imports down YoY but smelter op-rate not falling” is precisely the micro-mirror of the negative TC: in an environment of tight feed and a negative treatment charge, smelters still maintain high output by running down earlier inventory and pivoting to scrap/anode feed (see Chapter 4). High output in this state, built on the fragile foundation of “eating the stock, relying on by-products,” is inherently unsustainable.

It must be clarified that concentrate scarcity and the supply-demand of copper cathode (the refined end) are not the same thing. Concentrate scarcity constrains the ceiling of refined supply — when concentrate is insufficient, the TC is deeply negative, and the by-product and inventory dividends are exhausted, smelters will ultimately be forced to cut output, and refined copper supply growth will be “locked” by the mine bottleneck. This is also the fundamental reason the global refined balance deficit widens year by year in Chapter 6. In other words, today's negative TC is the “leading indicator” of tomorrow's refined production cuts; the current high level of cathode output masks the structural constraint being accumulated on the supply side.

It is worth dwelling on why a negative treatment charge is such a powerful signal — far more so than the copper price itself. The TC is not a forecast or a sentiment gauge; it is the price at which two parties with opposite interests, the miner and the smelter, actually transact for the right to process a tonne of concentrate. When that price goes negative, it means smelters are collectively willing to surrender their entire processing margin and then some, simply to secure feed. No survey of mine expansion plans, no rig-count proxy, no inventory snapshot conveys scarcity with the same finality. A miner with ore to sell holds the whip hand in the negotiation precisely because the buyer — the smelter — has fixed capital sunk in furnaces that bleed cash when idle. The negative TC is therefore the market's own confession that, at the raw-material end, demand for concentrate (the smelter's installed capacity) has decisively outrun supply (the mine's output). This is the cleanest, hardest evidence in the entire copper complex, and it is why we elevate it to the status of the supply storyline's master signal.

The geological backdrop reinforces the point. The copper industry has been mining its highest-grade, most accessible deposits for over a century, and the ore now coming out of the ground is progressively leaner: the average head grade at many flagship South American operations has drifted from well above 1% toward the 0.6–0.7% range, which means more rock must be moved, crushed and floated to yield the same metal. Declining grade is not a temporary disruption that reverses when a strike ends or a drought breaks; it is a one-way ratchet that raises unit costs and caps the rate at which existing mines can grow. Layered on top are the discrete disruptions — water-use restrictions in the Atacama, community blockades and royalty disputes in Peru, power and logistics constraints in central Africa — each of which subtracts a slice of expected output in any given quarter. The combination of a structurally declining grade trend and a steady drumbeat of disruptions is why mine supply elasticity is so low that even a record copper price cannot conjure meaningful new tonnes within the relevant horizon.

The DRC deserves a closer look because it is simultaneously the source of the only material upside and the reason that upside is bounded. New African projects have lifted the country into the top tier of copper producers, and their grades are enviable by global standards. Yet the very factors that make the DRC the fastest-growing source also limit how fast it can scale: chronic power deficits force operators to import generation or curtail output, the export route to ocean ports runs thousands of kilometres over congested road and rail, and the institutional environment carries its own risk premium. A barrel of incremental DRC concentrate is real, but it arrives slowly and unevenly, and it must be netted against grade decline and disruption losses elsewhere before it reaches the global balance. The arithmetic is unforgiving: the fastest-growing source in the world is still not growing fast enough to refill the trough that the negative TC is signalling.

5. Green takes the baton: grid/new energy/EV reshape copper demand

Copper demand is undergoing a structural gear-change: the legacy property chain weakens, offset by the structural pull of grid investment, solar/wind, storage, AI data centres and electric vehicles. High prices suppress some fabrication demand and stimulate scrap substitution, but the narrowing refined-scrap spread weakens this cushion.

China's refined-copper end-use demand is absolutely dominated by grid and cables, about 46%; home appliances about 14%, transport and EVs about 13%, new energy (solar/wind/storage) about 10%, construction about 9%, and other about 8%. The core implication of this structure is that the “baseplate” of copper demand rests on power-system investment, and the expansion of the power system is increasingly driven by green transition — precisely the fundamental reason copper is called “the metal of electrification.”

The pull of green demand is multi-dimensional and certain. Grid investment: large-scale grid connection of new energy, ultra-high-voltage and distribution-network upgrades, and cross-regional transmission corridors carry a higher copper intensity per unit of investment than the traditional grid. Solar and wind: the copper use per megawatt of installed capacity is significantly higher than for thermal power, and offshore wind is more copper-intensive because of submarine cables. Storage: the scaled build-out of battery energy storage systems (BESS) brings incremental copper use. AI & data centres: AI compute expansion lifts copper use in data-centre power distribution, power supplies and liquid cooling — a newly material structural increment. Electric vehicles (EV): the copper use per vehicle is about three to four times that of a traditional internal-combustion car (motor, battery, wiring harness, charging infrastructure). Together these scenarios push up the copper intensity of GDP — i.e. each unit of economic growth now requires more copper than before. This is a structural, cross-cycle demand upgrade, in sharp offset to the cyclical weakening of the legacy property chain.

The phrase “copper intensity of GDP rising” deserves to be unpacked because it overturns a long-standing assumption. For decades, the conventional wisdom held that copper intensity declines as economies mature — that services displace heavy industry and each additional unit of output requires less metal. The green transition breaks that relationship. Electrification is, at its core, the substitution of copper-bearing electrical infrastructure for hydrocarbon-bearing thermal infrastructure: every electric motor, every kilometre of cable, every transformer and every charging point embeds copper that its fossil predecessor did not. An economy that decarbonises is therefore an economy that re-intensifies its copper use, even as it grows more efficient in other respects. This is the deepest reason the demand story is structural rather than cyclical: it is not merely that green sectors are growing fast, but that the very process of growth is becoming more copper-hungry. A given rate of global GDP growth now pulls more tonnes of copper than the same rate would have a decade ago, and that ratchet runs for as long as the energy transition does.

The transport channel illustrates the magnitude with particular clarity. A conventional internal-combustion vehicle carries on the order of 20–25 kg of copper, mostly in its wiring harness and small motors. A battery-electric vehicle carries roughly three to four times that, once the traction motor, battery interconnects, expanded harness and on-board power electronics are counted — and that figure excludes the copper embedded in the charging infrastructure that must be built out alongside the fleet. Every percentage point of EV penetration therefore lifts the copper content of the vehicle stock, and the charging network adds a second, parallel layer of demand. The same multiplier logic applies on the generation side: a megawatt of solar or wind capacity carries several times the copper of a megawatt of thermal capacity, and offshore wind, with its long submarine export cables, is more copper-intensive still. These are not speculative future demands; they are the mechanical consequence of installation programmes already under way, which is why we regard the green pull as the most certain element in the entire balance.

High-frequency data validate the resilience of demand.

Peak-season downstream operating rates have broadly rebounded: the wire & cable weekly op-rate is 69.27% (+17.5% YTD), the refined-copper rod weekly op-rate is 67.35%, enamelled wire monthly 72.32%, and copper-product blended 63.34%. Among these, the sharp rebound in the wire & cable op-rate directly corresponds to the landing of grid- and new-energy-related cable demand — high-frequency evidence that green demand is “observable and verifiable,” and support for the judgement that demand has not collapsed under high prices.

That said, the “+17.5% YTD” figure is only a within-year change and is easily confused with the seasonal peak-season effect; it needs to be calibrated against a cross-year seasonal comparison. Overlaying the wire & cable op-rate for 2024 / 2025 / 2026 by ISO week shows that the 2026 rebound follows the typical “Spring-Festival trough → peak-season recovery” rhythm, but in absolute level the 2026 peak season (weeks 20–25 averaging about 68%) is clearly below both 2024 (about 77%) and 2025 (about 78%) in the same period — roughly 9–10 percentage points lower year-on-year. This gives a more balanced read: on one hand, demand has genuinely recovered in peak season and the green-related rigid floor is intact (+17.5% YTD); on the other, copper above ¥100k/t is already exerting an observable marginal drag on the downstream fabrication links, leaving the op-rate “recovered, but below the prior two years.” This corroborates the scrap substitution and the suppression of price-sensitive demand discussed below — and signals that demand resilience is structural, but by no means indifferent to price.

The high-frequency operating-rate data carry particular weight in this report's logic because they are the antidote to a tempting bearish argument — that a record price must, sooner or later, destroy the demand that created it. Demand destruction is real and visible at the margin, as the next paragraph discusses, but the wire & cable op-rate running up +17.5% year-to-date in the teeth of ¥100k-plus copper is direct evidence that the core of demand is not price-elastic in the way a textbook commodity would be. A utility building out a transmission corridor or a developer wiring a new substation cannot substitute away from copper at any plausible price within the project's timeline; the copper is a small share of total project cost and an irreplaceable input. This is the rigidity that the bears underestimate, and it is why the operating-rate panel is not a footnote but a load-bearing piece of the bullish case: it shows the demand floor holding precisely where the green narrative predicts it should.

Of course, high prices are not without a suppressive effect on demand. On one hand, some price-sensitive legacy fabrication demand (such as low-end copper tube and some construction copper) is markedly suppressed above ¥100k; on the other, high prices stimulate scrap substitution — replacing refined-copper rod with cheaper scrap (recycled-copper rod). Yet in 2026 the effectiveness of this substitution is declining: the refined-scrap spread has narrowed sharply from roughly ¥5,126/t at the start of the year to ¥2,207.77/t (−56.9% YTD), with #1 bright copper at ¥91,650/t — when scrap's discount to refined copper shrinks sharply, the cost advantage of using scrap weakens and the incentive to substitute follows. This means the effectiveness of scrap as a “demand release valve” is dulling, more demand is forced back to refined copper, and in turn this supports the refined-copper price.

7. Multi-factor price outlook: a three-factor framework and three scenarios

This chapter is the report's landing point. Using a three-factor framework of “supply-demand balance anchor + cost/inventory floor + macro and sentiment premium,” we give a month-by-month price range for copper cathode over the next 12 months and quantify probabilities and trigger conditions across bull/base/bear scenarios.

Multi-Factor Bull/Bear Scorecard (End-2026 Price)

Each of eight price-driving dimensions is scored from −3 (strongly bearish) to +3 (strongly bullish), weighted into a composite, converted to a 0–100 score (50 = neutral), and mapped to an end-2026 price by interpolation within the scenario range. Scores are SMM's own assessment — comprehensive in coverage, objective in calibration, transparent in weighting.

Scoring the eight driver dimensions (green = bullish, red = bearish) shows the bullish forces (concentrate scarcity, green demand, low inventory, macro) dominate in both number and strength; the bearish forces (scrap substitution at high prices, high-price demand suppression) genuinely exist, but are dulled by the narrowing refined-scrap spread and partly offset by rigid green demand, insufficient to reverse the direction. This forms the quantitative basis for our “base-case bullish, amplified volatility” view.

The discipline of a three-factor decomposition is that it separates what each force can actually do to the price. The balance anchor sets where the price gravitates over the cycle but says little about its day-to-day path; it is a statement about the centre of the distribution. The cost/inventory floor truncates the downside — it explains why a retreat tends to find support rather than cascade, because every leg lower triggers restocking and tips marginal smelters into cuts that tighten the balance and arrest the fall. The macro and sentiment premium, finally, governs the width and the tails: it is the channel through which a weaker dollar, a rate-cut cycle or a convergence of the inventory divergence can push the price to the upper edge of the range, and through which a recession scare can drag it to the lower edge. Read together, the three factors imply a distribution that is anchored high, floored not far below the current level, and skewed to the upside — which is precisely the shape the three scenarios below are built to express.

Three scenarios: probability, target and trigger conditions

Mapping the three scenarios onto a month-by-month price path: in the base case, copper cathode starts from roughly ¥104.9k/t in June, strengthens month by month through H2 2026 to about ¥114k by year-end, and rises into 2027 to a ¥113k–119k range on green demand and the concentrate bottleneck. The bull case corresponds to the upper edge of the range (¥130k–140k) and the bear case to the lower edge (¥98k–103k) — consistent with Chapter 6's “deficit widening year by year” balance judgement.

Price sensitivity matrix

The price centre is highly sensitive to the two most critical variables — the trajectory of the concentrate TC (which sets the intensity of smelter supply contraction / the supply floor) and the growth of green demand (grid + new energy + EV, which sets the demand pull). The table below gives a quantified estimate of the 2026 copper cathode annual centre (¥10k/t) under different combinations of the two variables, as a quick-reference for scenario tracking:

Two patterns can be read from the matrix: first, only under the bottom-left combination of “TC recovers to positive + green demand weakens” does the centre fall back below ¥100k — and a short-term recovery of the TC to positive requires a material release of mine supply, which is unlikely; second, the price is roughly equally sensitive to the TC (supply side) and to green demand (demand side), corroborating the dual-storyline judgement that “the mine sets life-or-death, green sets the direction.” With the concentrate bottleneck unchanged, the base and bull scenarios occupy the majority of cells in the matrix, consistent with the bullish tilt of the driver scorecard.