Local planning rules are the wrong tool for hyperscale data centres: parcel-based municipal oversight misses the regional electricity, water and fiscal effects of large AI compute campuses, shifting the burden to utilities and states. Without coordinated state and regional mechanisms for disclosure, cost allocation and cumulative-impact planning, compute will concentrate in jurisdictions with favorable or opaque regimes, distorting siting and project economics.

Main Finding

The rapid expansion of hyperscale data centers has revealed a governance mismatch: parcel-based municipal planning tools are poorly suited to regulate developments whose primary fiscal, utility, water, and environmental impacts are regional and cross-jurisdictional. Local controls often miss or are bypassed by utility and infrastructure decisions, producing limited public scrutiny until projects are legally and financially entrenched. While municipal reforms (zoning, disclosure, conditional approvals) help, meaningful governance of cumulative regional effects requires coordination with utilities, water authorities, state legislatures, and regional institutions.

Key Points

• Structural mismatch: Municipal land-use authority is parcel-focused, while hyperscale data centers behave like utility-scale infrastructure with effects that cross jurisdictional lines.
• Regulatory gaps:
  • Zoning categories are often outdated or too coarse to capture data-center scale and operations.
  • Utility commitments (power contracts, grid upgrades) and water allocations are frequently opaque or decided outside municipal land-use review.
  • Infrastructure decisions (transmission, substations, water supply) can lock in impacts before local oversight can act.
• Community responses: Local governments and residents have used rezoning, moratoria, negotiation, denial, and litigation to push back or reshape projects.
• State action: States have taken divergent approaches—some increasing oversight and disclosure, others preempting local control—reorganizing the balance of authority over siting decisions.
• Limits of local reform: Municipal reforms can improve regulatory legibility and preserve leverage pre-approval, but municipalities alone cannot manage cumulative regional impacts (grid reliability, water stress, fiscalization).
• Practical planner guidance:
  • Treat large data centers as utility-scale infrastructure in planning and permitting.
  • Create distinct land-use categories and performance thresholds for hyperscale facilities.
  • Require early disclosure, utility consultation, water-capacity findings, interagency referrals, and fiscal impact reviews.
  • Use time-limited moratoria only when tied to specific code-writing and infrastructure studies.
• Role of state/regional agencies: Provide cumulative infrastructure data, coordinate grid and water planning, and set baseline siting and cost-allocation rules to internalize regional externalities.

Data & Methods

• Approach: Interpretive review of local regulatory conflicts involving hyperscale data centers, supplemented by a systematic survey of state legislative activity.
• Temporal scope: State legislative activity surveyed between 2024 and 2026.
• Analytical focus: Tracing disputes across different stages of data-center development (pre-application, permitting, construction, operation) and examining how state-level actions reshape local planning authority.
• Evidence base: Case studies of municipal conflicts and compilation of state statutory and regulatory responses (no randomized or quantitative causal identification claimed; analysis is qualitative and institutional).

Implications for AI Economics

• Location of compute capacity: Governance frictions and local-state regulatory variation will shape geographic patterns of AI compute deployment. Regulatory uncertainty, moratoria, or stricter local requirements raise site-specific costs and may push firms toward jurisdictions with clearer, favorable regimes, concentrating AI compute in particular regions.
• Externalities and local public finance: Data centers generate regionally dispersed externalities (electricity demand peaks, transmission upgrades, water consumption) and fiscal effects (property-tax revenue vs. service demands) that are not internalized by parcel-based permitting. This mismatch can distort investment decisions and produce inefficient spatial allocations of AI infrastructure.
• Cost of capital and project economics: Opaque utility commitments and late-stage infrastructure lock-in increase regulatory and operational risk, which can raise firms’ cost of capital or lead to contingency contracting (e.g., requiring firms to pay for grid upgrades), altering AI project economics and potentially inflation of compute costs.
• Market structure and bargaining power: State preemption or strengthened state oversight shifts bargaining power away from municipalities and can lower local transaction costs for hyperscalers, affecting entry costs and competition among jurisdictions. Conversely, stronger local disclosure and permitting thresholds increase developer search and compliance costs.
• Policy levers to align incentives:
  • Disclosure and performance standards reduce information asymmetries and can improve investment efficiency by clarifying expected grid and water impacts.
  • Regional planning and cost-allocation rules internalize externalities (e.g., allocating transmission upgrade costs across beneficiaries), changing the private optimal site choice and leading to more socially efficient siting.
  • Time-limited moratoria paired with defined study programs can provide space to design evidence-based policies without permanently deterring beneficial investment.
• Research opportunities: Quantitative work could measure how municipal and state policy variation affects data-center siting, electricity demand growth, regional prices, and the spatial concentration of AI compute—key inputs for macro and industrial models of AI adoption, supply constraints, and economic spillovers.