2026 Data Center Announcements Pass All of 2025 as Local Restrictions Spread to 42 States

LandGate tracks 6,561 U.S. data center projects, with 3,045 in development and 3,191 operating; 819 projects were announced through October 2026, already exceeding the 764 announced in all of 2025. Hyperscalers account for 71% of the development pipeline, but power interconnection and local approvals—not capital—are increasingly determining which projects proceed. The report says local restrictions are spreading to 42 states, while FERC and Texas have taken steps addressing large-load interconnection and ERCOT queue projects.

Analysis

Investment read: The scarce asset is increasingly deliverable power plus local approval—not announced demand. That shifts value toward projects with secured interconnection, transmission access and municipal support, while making headline project counts a poor proxy for future revenue: a queue position can be delayed, resized or stranded. LandGate’s pipeline is useful as a directional signal, not independently verified evidence of funded, permitted megawatts or supplier orders.

Winners/losers: Electrical-equipment and grid-infrastructure suppliers such as Eaton, Vertiv and Quanta Services may capture incremental spending if bottlenecks prompt upgrades, but announcements alone do not establish order conversion; monitor backlog and lead times. Existing powered data-center capacity may gain scarcity value, whereas developers dependent on speculative sites face greater schedule and capital risk. Local restrictions create regional dispersion: a national data-center short is too blunt, and a blanket utility long ignores cost recovery and political scrutiny.

Catalysts and risks: Over the next 1–3 months, FERC and ERCOT implementation details could expose queue projects to new requirements, delays or costs. Over 6–18 months, the key test is whether interconnection approvals and utility investment keep pace with planned load. A downside tail is that delayed projects leave equipment capacity and land spending ahead of realized demand; the upside case is that constraints redirect projects toward already powered sites and extend scarcity rents. The contrarian point: hyperscaler budgets do not guarantee buildout, but local constraints may protect incumbent powered assets even as they impair speculative pipelines.

Key Decisions for Investors

• Do not trade the announcement count as equivalent to executable demand. Treat it as an alert; verify permitted capacity, firm interconnection status and project cancellations by region before sizing exposure.
• Prefer a measured relative-value bias toward grid/electrical suppliers (Eaton, Vertiv, Quanta Services) over speculative, unpowered development exposure only if company disclosures confirm backlog and orders are converting. Reassess if orders weaken or lead times normalize without sustained demand.
• Track FERC rule implementation, ERCOT audit outcomes, utility capex and interconnection approvals over the next quarter. The thesis weakens if queue attrition rises or load forecasts are cut; it strengthens if approvals and funded grid upgrades accelerate.
• Avoid a broad short of established data-center operators on this news alone: powered capacity could become more valuable. Compare operator commentary on power availability, utilization and project delays before taking a directional position.