Scotiabank remains bullish on copper miners, saying operating margins are still relatively robust even after raising Q2-Q4 C1 copper cash cost estimates by 10% to $2.13/lb and lifting the 2026 forecast to $2.09/lb. RBC said oil and gas insiders have been net sellers by $17.6 million year-to-date, while Morgan Stanley argued the agentic AI cycle is accelerating, with AMD now seeing server CPU TAM growth above 35% annually to more than $120 billion by 2030 and Arm’s data center royalties more than doubling year over year.
The key read-through is that the copper complex is now trading more on earnings durability than on headline metal prices. Higher diesel, energy, and acid costs are real, but they are being absorbed by an even stronger pricing backdrop, which keeps marginal cash generation intact and pushes the pain curve out to Q4 when contract resets start to bite. That creates an asymmetric setup: near-term estimates may be nudged down, but the bigger move is likely in second-derivative winners that have clean balance sheets, lower jurisdictional risk, and less direct exposure to the worst input inflation. The more interesting implication is that the market is likely underestimating how much supply discipline is being imposed by operational setbacks at large incumbents. When major mines stumble, the effect is not just tighter balances today; it forces downstream restocking, lifts forward strip confidence, and supports higher medium-term contract pricing for the entire complex. That also widens the gap between names that can self-fund growth and those that need to spend heavily just to stand still, which should show up in relative performance once formal guidance revisions start landing. In semis, the AI trade is broadening from GPU capex to the less appreciated infrastructure layers that make agentic workloads possible. The second-order winners are not just CPUs, but the whole ecosystem that monetizes socket density, memory interfaces, substrates, and storage persistence, because agentic architectures increase coordination overhead per rack rather than merely adding more accelerators. The risk is that this is becoming consensus before the full earnings revision cycle has played out; the right way to own it is through suppliers with visible content gains and constrained supply, not through the most obvious benchmark beta. The insider flow in energy is a softer signal than the commodity setup, but it matters for timing. Broad-based buying tends to matter more than one-off sales, so the current net selling read is less a bearish tell than a reminder that the group is no longer uniformly cheap enough to attract indiscriminate insider accumulation. The contrarian issue across all three themes is valuation: when markets are already pricing perfection, you want to lean into the names with the best operating leverage and the least exposure to next-quarter disappointment, not the ones with the loudest narrative.
AI-powered research, real-time alerts, and portfolio analytics for institutional investors.
Request a DemoOverall Sentiment
moderately positive
Sentiment Score
0.45
Ticker Sentiment
