Copper Price Forecast
Driven by Energy Transition

The traditional energy system moves chemical energy around. You burn hydrocarbons. The new energy system moves electrons around. Electrons need conductors. Copper's electrical conductivity is second only to silver among metals, and silver is roughly 80 times more expensive per tonne. Aluminum conducts at about 61% of copper's rate, which means you need a fatter wire to carry the same current. In a motor winding or a transformer coil or the inside of a switchgear cabinet, the physical space is fixed. You cannot just use a bigger aluminum conductor because it will not fit. This is not a cost optimization problem. It is a geometry problem.

A gas plant uses less than 1 tonne of copper per megawatt. An offshore wind farm uses about 15. A combustion engine car has roughly 23 kilograms of copper in it, an EV around 83, and if you trace the copper all the way through the charging infrastructure and the grid reinforcement needed to serve that charger, the systemic number approaches 130 kilograms per vehicle. These figures have circulated through enough industry presentations and sell-side reports that repeating them almost feels like an obligation rather than an insight.

What has not been adequately processed is that the relationship between energy transition penetration rates and copper demand is nonlinear. EV penetration going from 5% to 20% does not multiply copper demand by four. It multiplies it by something larger, because at some threshold the entire distribution grid servicing residential areas has to be rebuilt. Every street-level transformer that was sized for households drawing 3 to 5 kilowatts at peak suddenly needs to handle households drawing 7 to 11 kilowatts because of evening EV charging. That transformer gets replaced. The cable feeding it gets replaced. The upstream feeder may need a parallel circuit. This is a step function, not a slope. Most models treat it as a slope.

This part is going to sound abstract, and it is abstract, but it has concrete pricing implications.

For twenty years copper has been slotted into portfolio models as a cyclical industrial metal. Overweight when the global economy expands, underweight when it contracts. The allocation signal is PMI, GDP growth, dollar strength. This works when copper demand is dominated by construction and traditional manufacturing, which are procyclical.

The energy transition is layering a new demand component on top that does not follow the economic cycle. Renewable energy targets are legislative. They do not get revised downward because Q3 GDP disappointed by half a percent. A government that committed to 50 gigawatts of offshore wind by 2030 does not pull that commitment because of a mild recession. The investment may slow at the margin, but the direction is locked in by law, by subsidy contracts already signed, by turbines already ordered.

So copper's demand base now has two layers. One is cyclical, the old layer. One is structural, the new layer. The structural layer acts as a floor. It makes copper's downside shallower than cyclical models predict. When the next recession comes and copper sells off, the trough will be higher than the model says, because the model does not have a variable for "demand that is legally mandated regardless of GDP."

This is the part of the copper story that people outside the metallurgical supply chain simply do not encounter, and it is the part that, once you see it, changes how you think about the interconnectedness of the entire energy transition metals complex.

Copper smelting, specifically pyrometallurgical smelting of sulfide concentrates, produces sulfuric acid as a byproduct. Not a trivial amount. Copper smelters are one of the major sources of sulfuric acid globally. The exact share varies by region, but in China, which dominates global copper smelting, the linkage is particularly tight.

Sulfuric acid is used to leach lithium from spodumene concentrate. It is a key input in the production of battery-grade lithium hydroxide and lithium carbonate. So here is the chain: if copper concentrate supply tightens and smelters cut throughput, sulfuric acid output falls. If sulfuric acid output falls, lithium processors face higher input costs or physical shortages of acid. Lithium prices get pushed up. Copper and lithium, two metals that sit at the center of the energy transition, are chemically coupled through a smelter byproduct that almost never appears in any analyst's supply-demand model for either metal.

The implications compound. Copper prices spike because of concentrate tightness. Simultaneously, lithium processing costs rise because sulfuric acid supply contracts. The battery industry gets hit from both sides. Solar cell manufacturers who use copper interconnects and battery makers who need lithium are both paying more, at the same time, for reasons that trace back to the same root cause: not enough copper concentrate going through smelters. The cost of the energy transition ratchets up, and the mechanism by which this happens is a chemical byproduct that occupies zero space in the public discourse about energy transition costs.

Shorter story. Global copper concentrate quality is degrading. Arsenic content has been rising because the clean ore bodies were mined first over the past several decades. High-arsenic concentrate requires specialized smelting capacity that exists in limited quantities, mostly in China. The practical effect is that not all copper concentrate is created equal, and the market is splitting into two tiers: low-arsenic material that every smelter wants and will pay a premium for, and high-arsenic material that only a few smelters can handle and that trades at a discount. Headline concentrate production figures overstate effective supply because they do not adjust for this quality split.

New copper discoveries skew increasingly toward porphyry copper-gold deposits. The investment case for developing these mines depends partly on gold byproduct credits. High gold prices make marginal copper-gold projects viable. A gold price collapse shelves them. Copper supply forecasting has an embedded dependency on the gold market that is rarely modeled explicitly.

This is where the bulk of the copper goes, and it is where the bulk of the analytical blind spot sits.

The physical logic for why the grid is the biggest copper consumer in the energy transition is mechanical. Wind and solar generation sites are located where the resource is, not where the load is. The wind is strong offshore and in flat open terrain far from cities. The sun is most reliable in deserts and semi-arid regions. Generation and consumption are geographically separated, often by hundreds or thousands of kilometers. Connecting them requires high-voltage transmission lines. Reinforcing local distribution grids to handle distributed solar and EV charging requires copper at the neighborhood level. Both ends of the grid, transmission and distribution, are being rebuilt simultaneously.

China's planned UHVDC lines and distribution upgrades have copper requirements measured in millions of tonnes. A single UHVDC corridor can consume tens of thousands of tonnes just in conductor material, and procurement is not spread over time. When a project is approved and construction begins, the copper order goes in as a block. That block hits the spot market like a pulse.

The US has a different problem layered on top. Its grid infrastructure is old. Many transmission lines date to the 1960s and 1970s and are approaching or past their intended service life. This replacement demand exists independently of the energy transition. The energy transition adds urgency to it. The combined effect is a wave of grid investment that is both overdue and accelerating, creating copper demand that would be large even in a world where nobody cared about decarbonization.

This creates a temporal displacement of copper demand. Projects that would be consuming copper now are instead sitting in a queue. The copper they would have used has not disappeared from the demand forecast. It has been deferred. Deferred demand accumulates. If the transformer bottleneck eases in 2027 or 2028, due to capacity expansions that are currently under construction at a few manufacturers, a bolus of previously queued projects will move to construction simultaneously. The copper market will experience this as a demand surge, even though the underlying project pipeline has been visible for years. The surge will feel sudden because the market was looking at the queue and seeing stasis, not recognizing that stasis was just compression.

Submarine cables for offshore wind deserve a mention not because the tonnage is the largest among grid segments, but because the supply chain is the most constrained. There are perhaps half a dozen factories in the world that can produce the high-voltage submarine cables needed for offshore wind export. Delivery schedules are booked out years in advance. Each major offshore wind cluster in the North Sea, Baltic, or off the coast of China requires thousands to tens of thousands of tonnes of copper in cable alone. The submarine cable supply chain could become a binding constraint on offshore wind deployment independent of any other factor.

Ore grade decline is the supply side story that everyone already knows and that somehow still has not been fully priced in. The arithmetic: average grades at major copper mines have gone from about 1% at the turn of the century to around 0.6% now. In percentage points, that is 0.4. In physical terms, it means 67% more rock has to be moved, crushed, ground, and floated per tonne of copper produced. That 67% increase cascades into everything. The concentrator plant needs more capacity. The tailings dam needs more volume. Water consumption goes up. Energy consumption goes up. In the Atacama, where some of the world's largest copper mines operate in one of the world's driest environments, miners are building desalination plants on the Pacific coast and pumping water up to mine sites at 3,000 to 4,000 meters altitude. The cost of that water becomes part of the cost of copper.

Escondida's grade trajectory over the past fifteen years is a slow bleed. The open-pit ore body that made it the world's highest-output copper mine is being exhausted. The transition to lower-grade material and eventually underground mining will raise costs and likely reduce output. Grasberg has gone underground already. Oyu Tolgoi's underground block cave has been one of the most troubled major mining projects of the past decade, with repeated delays and cost overruns. The generation of mega-mines that supplied the copper market's growth for twenty years is aging out, and nothing coming online matches their scale or grade. The replacement pipeline is a collection of smaller, more expensive, more politically exposed projects scattered across less developed jurisdictions.

Carbon costs layer on further. Mining and smelting copper produces CO2. Carbon pricing mechanisms, whether taxes or cap-and-trade schemes, increase the cost of that production. The energy transition needs copper. Producing copper generates emissions. Emissions get taxed. The tax raises copper's cost. Higher copper costs raise the cost of the energy transition. Both forces push copper prices up simultaneously.

Something is starting to happen in the market for certified low-carbon copper that could reshape trade flows. Some European buyers are already writing carbon footprint requirements into procurement contracts and paying several hundred dollars per tonne above market for copper that meets their standards. If this practice spreads, the copper market bifurcates. Smelters running on hydroelectric power in Canada or Norway get a premium. Smelters running on coal in parts of Asia face a discount or outright exclusion from certain supply chains. Geography and carbon intensity become pricing factors alongside grade and logistics. The copper market has never had to deal with a quality differentiation axis based on how the metal was produced rather than what is in it.

If copper supply constraints slow the energy transition, the entity that benefits is the fossil fuel system. Every tonne of copper that arrives late means a wind farm that commissions late, a transmission line that completes late, a country that peaks carbon emissions later than planned. Copper scarcity extends the economic life of oil and gas assets.

This is not a speculative claim. It is a mechanical consequence. And it has apparently not been lost on some of the major oil companies, several of which have in recent years made moves toward copper mining assets or expressed strategic interest in the metal. The logic is clean: if you are an oil major and you believe the energy transition will eventually shrink your core business, and you also observe that the energy transition cannot proceed without copper, and you further observe that copper supply takes a decade or more to bring online, then acquiring copper resources is a hedge that pays in every scenario. If the transition accelerates, you own a critical bottleneck input. If the transition slows because of copper scarcity, your existing oil and gas assets produce cash for longer than the market currently expects. Either way, the copper position works.

For copper prices specifically, more capital from oil majors entering the mining sector means more supply eventually. But "eventually" is ten to fifteen years from project acquisition to full production. In the medium term, it changes nothing about the deficit.

This is a number that most people outside the metals trade never look at, and it is one of the most informative indicators of copper market tightness available in near-real time.

TC/RC is the treatment and refining charge that miners pay smelters to convert copper concentrate into refined metal. When concentrate is abundant relative to smelting capacity, smelters have pricing power and TC/RC is high. When concentrate is scarce, smelters compete for feed and TC/RC collapses.

In recent years, TC/RC has been compressed to levels that are historically extreme, at times going negative. Negative TC/RC means the smelter is effectively paying the miner for the privilege of processing their concentrate. This happens because smelters need to keep their plants running to cover fixed costs and to maintain sulfuric acid output (there is the coupling again), and they are willing to accept a loss on copper processing to avoid the greater loss of a shutdown.

This signal is telling the market, in real time, that copper concentrate supply is severely tight relative to processing capacity. The futures curve has not fully absorbed this. Partly because TC/RC is a niche data point that most generalist commodity investors do not monitor. Partly because the relationship between concentrate tightness and refined copper availability is lagged and nonlinear. But the signal is there, sitting in the open, for anyone who looks.

Data centers are a new source of copper demand that barely existed in forecasts five years ago. Each large facility draws enormous amounts of power and the internal electrical infrastructure is copper-intensive. The compounding factor is that data center operators increasingly sign renewable energy PPAs, so each new data center also triggers construction of the renewable generation and grid infrastructure to feed it. The copper pull from a single hyperscale data center extends well beyond its own walls.

Military copper demand is opaque and probably underappreciated. Modern defense systems, communications infrastructure, naval vessels, and armored vehicles are copper-dense. Defense procurement does not appear in standard commodity supply-demand models. As global military spending rises across multiple regions, this is a demand increment that is being systematically excluded from forecasts because the data is classified or unreported.

Aluminum substitution in electrical applications proceeds slowly. It works in some low-voltage distribution scenarios. In China, aluminum alloy cable has gained share in building wiring. In transmission networks, high-voltage equipment, and motors, it does not work. Power utilities are institutionally incapable of accepting material substitution risk at the speed the copper price would need them to. Certification cycles for new materials in grid standards take years. No grid operator wants to be the one whose aluminum conductor failed and caused a blackout. Substitution puts a soft ceiling on how high copper can spike, but it does not change the structural trajectory.

In the next one to two years, macro factors dominate. The dollar, China policy, manufacturing sentiment. The energy transition contribution in this window shows up not as a price driver but as a floor. Corrections will be shallower than historical analogs because there is a demand layer underneath that does not respond to PMI deterioration.

Three to five years out, the deficit becomes hard to ignore. Multiple modeling exercises from different institutions converge on a gap of several million tonnes per year by the late 2020s if current transition targets are maintained. Closing that gap requires copper prices high enough to pull forward mine development, which takes a decade, meaning prices have to rise now to signal a supply response that arrives in the 2030s. There is an inherent awkwardness in a market that needs to price in a problem it cannot solve on the relevant timeframe.

Ten years and beyond is where technology matters. High-temperature superconductors need liquid nitrogen cooling and are nowhere near deployable on overhead transmission lines. Carbon nanotube conductors are laboratory curiosities with astronomical production costs. Perovskite solar cells might eventually reduce per-unit copper consumption versus silicon cells, but commercialization timelines keep slipping. None of these technologies will displace copper at industrial scale before 2035. The transition's appetite for copper over the coming decade is, by any reasonable assessment of technology readiness, locked in.

One more thing, aimed at anyone managing copper exposure or negotiating procurement contracts.

When you add a rigid, non-cyclical demand layer to a commodity that already has a supply side with extremely low short-term elasticity, the volatility profile becomes asymmetric. Upward price moves get bigger. Downward moves get smaller. A demand pulse (a major grid project placing a block order, a new gigafactory beginning wiring) can push prices up sharply because there is no swing supply to absorb it. A demand drop (an economic slowdown) cannot push prices down as far because the policy-anchored transition demand is still there, buying.

This asymmetry matters for options pricing. It matters for procurement strategy. It matters for any industrial consumer deciding how much physical inventory to hold. Running lean on copper inventory was a manageable risk when downside and upside were roughly symmetric. It is a different risk profile when the upside tail is fatter than the downside tail.