By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panel is divided on the 'HALO trade' thesis, with concerns about demand volatility, permitting delays, and the potential impact of captive power generation on copper demand. While some panelists are bullish on copper due to electrification and AI data centers, others caution about overhyped near-term demand and cyclical nature of hard assets.
Risk: Permitting delays for new mines and the potential shift towards captive power generation could mute copper demand and derail the 'HALO trade' thesis.
Opportunity: Structurally higher copper and aluminum demand driven by AI data-center power surge and EV-driven electrification.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Global electricity demand growth is set to outstrip GDP growth for the first time, signaling a major boon for metals key to the energy transition.
After a decade of stagnant power demand, the economy is undergoing a massive wave of electrification, Bank of America (BAC) metals strategists led by Michael Widmer wrote in a recent note to clients. Electricity demand is now expected to grow significantly faster than GDP for the next two and a half years.
While the drive toward electrification originally kicked off with the mid-2010s push toward green technology, a series of new developments — namely EVs and the boom in AI data centers — has taken over. Global power usage by data centers is expected to grow from a current level of around 55 gigawatts to 84 gigawatts — equivalent to the power usage of roughly 70 million homes — in only the next two years, according to research from Goldman Sachs.
It's all part of what economists have deemed the "HALO" trade: heavy assets, low obsolescence. The idea is that as the economy shifts forward, the physical manufacturing capacity of the AI industry and the push toward electrification will create a boom time for hard assets and the companies producing them.
While metals have experienced some whipsawing motion in their prices under the Trump administration's shifting tariff policy, prices on copper (HG=F) — the leading industrial metal in terms of usage — are up 21% over the past year.
Prices on aluminum (ALI=F) are up 12%, while those on lithium (LTH=F) and cobalt, which are in smaller supply, have soared by roughly 150% and 85%, respectively.
New technologies such as data centers hit metals demand in two ways, according to the Bank of America strategists. Not only do data centers themselves require mined commodities such as copper, but the electricity needs of the infrastructure will force further build-outs of global power grids, all of which require substantial amounts of metals for wiring, transformers, and other components.
Governments have also begun to turn increasingly toward energy independence in the wake of the US tariff regime on metals and the war in Iran, which threatened global energy supply chains. As countries build out their own energy generation and storage capacity, BofA said, that too will boost metals demand.
Take copper, for example. Global demand for copper is now expected to surge from 28 million tons in 2025 to 42 million tons by 2040. Manufacturing EVs requires nearly three times as much copper as gasoline-powered cars, and solar and wind electricity generation equipment, which accounted for roughly 90% of the new generation capacity installed, also requires large amounts of copper, according to S&P Global.
Four leading AI models discuss this article
"Copper’s 21% rally already prices in much of the near-term AI tailwind, leaving the metal vulnerable to tariff volatility and supply responses by 2027."
The AI data-center power surge and EV-driven electrification do support structurally higher copper and aluminum demand, with global copper needs projected to rise from 28M tons in 2025 to 42M tons by 2040. Yet the article downplays how quickly high prices can trigger new mine supply and how Trump-era tariffs have already produced sharp whipsaws that raise financing costs for grid projects. Goldman’s 55-to-84 GW forecast also assumes uninterrupted hyperscaler capex; any pause in AI ROI would quickly trim that trajectory. BofA’s HALO thesis therefore hinges on policy stability and execution speed that remain untested.
Even if mining supply ramps, multi-year structural deficits are already embedded in S&P Global and other models, so the demand shock could still produce sustained price strength rather than a classic boom-bust pattern.
"The structural deficit in copper supply is inevitable, but the timeline for demand realization is likely slower than the current market enthusiasm suggests."
The 'HALO' trade is structurally sound but dangerously overhyped in the near term. While the electrification of data centers is an undeniable secular tailwind for copper and aluminum, the market is pricing in a linear demand curve that ignores cyclical reality. We are looking at a massive infrastructure bottleneck; grid permitting and localized energy constraints will likely delay the actual deployment of these data centers, creating a 'hurry up and wait' scenario for miners. Copper (HG=F) is the primary beneficiary, but investors should be wary of the 21% run-up, as it prices in perfect execution of grid expansion, which rarely happens in a fragmented regulatory environment.
The thesis assumes AI power demand will remain unconstrained, ignoring the possibility that energy costs could force a pivot toward more efficient, smaller-scale compute or a regulatory crackdown on data center power consumption.
"Electricity demand outpacing GDP is real, but the metals demand thesis assumes capex cycles and supply constraints persist unchanged—both historically unreliable over 2.5-year horizons."
The article conflates two separate demand drivers—AI data centers and EV/renewables electrification—as if both are equally durable. Data center power demand is real but faces three headwinds: (1) efficiency gains in chip design and cooling could flatten per-compute power needs, (2) hyperscalers are building captive nuclear/renewable capacity, reducing grid-dependent demand, and (3) the current 55→84 GW projection assumes zero slowdown in AI capex, which is cyclical. Meanwhile, copper demand math assumes 2040 demand at 42M tons, but supply-side responses (new mines, recycling tech, substitution) typically lag by 5-7 years. The 'HALO trade' framing obscures that hard assets are cyclical; peak capex cycles precede demand maturation. Metals are up 12-150% YoY partly due to tariff hedging and supply disruption fears, not fundamental demand certainty.
If AI efficiency improves faster than expected and hyperscalers achieve their renewable/nuclear targets, grid-dependent electricity demand growth flatlines—and with it, the metals supercycle thesis. Additionally, a recession would crater both data center capex and EV sales simultaneously, leaving miners with stranded capacity.
"Copper’s structural demand growth from electrification and data-center expansion should support FCX upside, provided grid/EV investment remains on track."
Macro view: The article paints a bullish picture for copper and related metals on an electricity demand boom tied to electrification, AI data centers, and energy independence. While copper demand rising from 28 Mt in 2025 to 42 Mt by 2040 fits a structural story, the near‑term path is not guaranteed. Key risks include a potential slowdown in GDP or energy-intensity reductions through efficiency, lagging capex in mines, potential substitutions (e.g., aluminum or advanced materials), and policy shocks that could derail grid or EV pipelines. If the demand signal weakens or prices re-rate, miners like FCX could underperform.
The rally hinges on optimistic buildouts that may falter; efficiency gains, recycling, and substitution could cap copper demand, while policy shifts or a macro downturn could mute the supposed structural upside.
"Hyperscaler captive power shifts could reduce grid copper demand faster than projected if policy accelerates SMRs."
Claude correctly flags hyperscalers building captive power as a demand headwind, but this also links to Grok's policy stability concern: accelerated SMR approvals under new administrations could make on-site generation viable faster than grid upgrades. That would mute copper demand from the 55-84 GW projection more than efficiency gains alone. No one has modeled how quickly this pivot alters the 28M-to-42M ton trajectory.
"The structural supply deficit is guaranteed by multi-decade permitting timelines, making the copper price floor much higher than demand-side models suggest."
Grok and Claude are missing the primary bottleneck: domestic copper supply isn't just about demand, it's about the 'permitting wall.' Even with captive nuclear or SMRs, the distribution infrastructure requires massive copper intensity. The real risk isn't just demand volatility or efficiency; it's that U.S. mining projects take 10-15 years to permit. We are pricing in a 2040 demand spike while ignoring that the physical supply chain is structurally incapable of scaling to meet it, regardless of AI capex fluctuations.
"Permitting delays matter only if grid-dependent demand is real; captive power renders the entire 28M-42M ton thesis contingent on execution that bypasses regulated infrastructure entirely."
Gemini's permitting wall is real, but it conflates two supply chains. Copper for *distribution* faces 10-15 year delays; copper for *data center infrastructure* (cabling, switching) moves faster and is already embedded in capex budgets. The 28M-to-42M ton gap assumes grid-connected demand; if hyperscalers truly pivot to captive SMRs (Grok's point), that demand never materializes. Gemini is solving for the wrong bottleneck—the constraint isn't permitting copper mines, it's whether the demand signal itself survives the captive-power shift.
"Permitting isn't the sole bottleneck; recycling, substitution, and captive-power demand can reshape copper supply/demand before new mines scale."
Gemini overstates permitting as the sole bottleneck. Mine permits matter, but recycling, substitution (alternative alloys), and captive-power demand can reshape copper needs without waiting on new mines. If AI capex slows or grid expansion accelerates under policy, the HALO thesis could fade even with long permitting timelines. Financing costs and grid readiness are the real swing factors to watch for markets.
The panel is divided on the 'HALO trade' thesis, with concerns about demand volatility, permitting delays, and the potential impact of captive power generation on copper demand. While some panelists are bullish on copper due to electrification and AI data centers, others caution about overhyped near-term demand and cyclical nature of hard assets.
Structurally higher copper and aluminum demand driven by AI data-center power surge and EV-driven electrification.
Permitting delays for new mines and the potential shift towards captive power generation could mute copper demand and derail the 'HALO trade' thesis.