Two-Thirds of American Homeowners Have Experienced a Power Outage This Year, Majority Worry Power Outages Will Become More Common

A homeowner survey found that 59% expect local power outages to become more common, while 75% are concerned that the expanding number of data centers could contribute to future grid disruptions. The findings highlight rising public concern over electricity-infrastructure capacity as data-center demand grows, but provide no evidence of current outages or immediate regulatory action.

Analysis

This is not a demand signal by itself, but it raises the political cost of permitting large-load interconnections. The near-term risk sits with data-center developers whose contracted capacity assumptions depend on utility delivery dates: local opposition can lengthen transmission approvals, require redundant on-site generation, and raise effective development capex. EQIX and DLR are relatively insulated where power costs are contractually passed through, but secondary markets and pre-leased AI campuses face greater schedule risk than established hubs with secured utility allocations.

The more investable second-order effect is an accelerated grid-capex cycle rather than a broad AI retrenchment. Utilities can generally recover reliability investment through rate base, while transmission engineering and electrical-equipment suppliers monetize both the initial interconnection build and subsequent resilience upgrades. PWR, MYRG, ETN and GEV have cleaner volume exposure than regulated utilities, whose allowed returns can be offset by regulatory scrutiny over reliability and affordability.

Over 1-3 months, watch state utility commission proceedings, interconnection-queue withdrawals, capacity-auction pricing and announced data-center power procurement rather than survey sentiment. Over 6-18 months, a widening gap between data-center lease commencements and delivered grid capacity would favor incumbent powered-land owners and merchant generators such as CEG and VST, but would impair speculative development economics. The thesis is falsified if utilities demonstrate timely large-load connections without meaningful transmission spend or if AI-related load forecasts are revised materially lower.

Key Decisions for Investors

• No standalone directional trade on the survey; treat it as an alert for utility-commission actions and data-center project delays over the next 1-3 months.
• Build a 6-12 month long basket in PWR, ETN and MYRG on broad-market weakness: these names capture transmission, substations and resiliency spend with less exposure to retail-rate political backlash. Reassess if major utilities cut 2026-27 capex plans or interconnection queues materially clear without incremental grid investment.
• Consider a measured pair trade long PWR / short DLR only if evidence emerges of delayed campus energization or rising customer-funded interconnection costs. The risk is that DLR's supply-constrained capacity and long-duration leases outperform despite delays; use project-delay disclosures and lease-commencement guidance as the stop signal.
• Maintain a watchlist long CEG and VST for regional capacity-price upside where data-center load is concentrated, but require confirmation from capacity auctions, bilateral power contracts or upward load forecasts before initiating. Merchant-power downside is a power-price reversal, new generation additions, or regulatory caps on reliability-related pricing.