These two sectors have been boosted by AI hopes. Why investors should buy one, and trim exposure to the other.
AI euphoria hasn’t just lifted chipmakers and cloud platforms. It has also set off a scramble for electricity and the hardware that moves it. As data center developers race to add capacity and power-hungry AI workloads proliferate, Wall Street has rewarded two adjacent sectors: electrical equipment makers and regulated utilities. Both have credible AI narratives. Only one, however, offers a cleaner path to compounding returns over the next cycle.
The short version
– Buy: electrical equipment and grid technology (transformers, switchgear, power management, high-voltage gear, grid automation).
– Trim: regulated utilities that have rerated as “AI power plays.”
Why AI is a power story
Training frontier models and serving inference at scale concentrates vast loads into clusters of new or expanded data centers. That strains a grid built for more distributed, predictable demand. The result is a multi-year capex wave across:
– Generation: new gas, nuclear life extensions, renewables plus storage.
– Transmission and distribution: high-voltage lines, substations, transformers, undergrounding, protection and control.
– On-site infrastructure: backup generation, switchgear, cooling, power quality and efficiency inside data centers.
Two ways to invest in this buildout emerged:
– Electrical equipment manufacturers, grid automation and power management specialists selling directly to utilities, EPCs, and data center developers.
– Regulated utilities allowed a return on invested capital in exchange for service obligations and regulatory oversight.
Why buy electrical equipment
– Visibility and duration: Order books in medium/high-voltage equipment and large transformers have extended materially as delivery times lengthened and capacity expansions lag demand. That tends to lock in multi-year revenue visibility beyond a single AI cycle.
– Pricing power from bottlenecks: Scarcity in specialized components (e.g., large power transformers, switchgear, insulated busways) has supported price increases and mix upgrades. Even as new capacity comes online, the combination of AI, electrification, and re-shoring keeps the backlog robust.
– Better capital efficiency than utilities: Many power equipment leaders are asset-light relative to the scale of demand, with high incremental margins, strong free cash flow conversion, and the ability to reinvest in capacity and R&D without heavy balance-sheet leverage.
– Multiple end-markets, not just AI: AI-driven data center demand layers on top of secular themes—grid hardening, renewable interconnection, EV charging infrastructure, industrial automation, and building efficiency. That diversification cushions any single end-market slowdown.
– Beneficiaries of standards and software: Grid modernization is as much about intelligence as hardware. Companies with protection relays, SCADA/ADMS, analytics, and power quality solutions capture higher-margin, stickier revenue and service contracts as utilities digitize.
What to own within equipment
– Exposure to medium/high-voltage gear, transformers, switchgear, breakers, power distribution units, busways, UPS, thermal management, and grid automation.
– Global leaders in electrification, power management, grid hardware, and industrial software tend to have the breadth to navigate cycles.
– For diversified access, consider industrials or infrastructure funds with meaningful weights to electrification and grid components. For purer exposure, smart grid or power equipment–focused vehicles can be more targeted.
Note: Concentration risk is real. Individual names can be volatile on cycle turns or policy shifts; a basket approach reduces idiosyncratic risk.
Risks to the buy case
– Capacity catch-up compresses pricing and lead times faster than expected.
– Input cost or supply-chain shocks pressure margins.
– A pause in hyperscaler capex, model architecture changes that improve compute efficiency, or slower permitting for data centers reduces near-term orders.
– Policy changes on domestic content or tariffs alter cost curves.
Why trim regulated utilities
– The timing mismatch: Utilities do benefit from AI-driven load growth, but earnings follow only after spend is placed into rate base and approved. That regulatory lag can be years, while stocks have already priced in faster growth.
– Financing and dilution risk: The capex required for generation, transmission, and distribution upgrades is enormous. In a world of higher base rates than the prior decade, that means more expensive debt and, often, incremental equity issuance—dilutive to current shareholders.
– Valuation drift: A sector long treated as a bond proxy has been re-rated as a growth proxy. Paying a premium multiple for regulated earnings with capped allowed ROEs and rate-case risk leaves little cushion if long yields back up or project timelines slip.
– Regulatory and political friction: Even with constructive jurisdictions, allowed returns can lag actual cost of capital in volatile markets. Siting and permitting for long-distance transmission remain slow; cost recovery for large projects can be contentious.
– Project execution risk: Nuclear life extensions, gas additions for reliability backstops, and large transmission lines carry schedule, cost, and community-opposition risks. Any misstep can delay recovery and invite scrutiny.
How to own utilities if you keep some exposure
– Favor service territories with constructive regulation and proven rate-case execution.
– Prefer lower-leverage balance sheets, diversified generation, and companies with less need for near-term equity issuance.
– Be selective with “AI adjacency”: Not all service territories will see the same data center concentration or load growth; prioritize those with clear, contracted demand and available interconnection capacity.
– Consider a smaller, core defensive allocation rather than a growth bet.
Scenario analysis framing
– Base case: AI/data center buildouts remain strong, but permitting and grid upgrades proceed at a measured pace. Equipment makers compound on backlog and pricing; utilities deliver steady but delayed earnings uplift with periodic equity issuance. Equipment outperforms.
– Bull case: Policy tailwinds accelerate transmission build, transformer capacity remains tight, and efficiency gains don’t materially dent power intensity. Equipment enjoys both volumes and margins; select utilities with the best jurisdictions and least dilution also do well.
– Bear case: AI workloads become far more efficient, hyperscaler capex pauses, or rates rise meaningfully. Equipment demand moderates and multiples compress, but diversified electrification exposure helps. Utilities see valuation pressure as rate sensitivity reasserts itself and financing costs bite.
A practical portfolio tilt
– Overweight electrical equipment and grid technology given superior near-term operating leverage, pricing power, and multi-theme demand.
– Underweight or trim regulated utilities where AI growth optimism has pulled forward returns, leaving investors to absorb financing and regulatory lags.
– Rebalance periodically: If lead times normalize and pricing fades, reduce equipment overweight; if rates fall and regulatory outcomes improve, selectively add back higher-quality utilities.
What would change this view
– A credible, accelerated policy breakthrough on transmission permitting and allowed ROEs that improves utilities’ cost recovery and reduces dilution.
– A sharp, durable decline in long-term interest rates that re-expands utilities’ valuation without compressing allowed returns.
– A supply glut in transformers and switchgear that erodes equipment pricing faster than expected.
Bottom line
The AI buildout is real, but profits accrue unevenly across the value chain. The companies selling the “picks and shovels” of electrification—the gear that makes AI possible—have clearer line of sight, better capital efficiency, and stronger pricing power today. Regulated utilities will ultimately participate, but after the spend, the hearings, and the lag. In a market that has already awarded both the growth premium, investors should buy the equipment makers and trim exposure to the utilities.
