Governing the Demand Shock: Data Centers, Local Power, and the Distribution of Cost in Virginia
On a Tuesday night in late May 2026, the Hanover County board voted four to three to turn away a 900-megawatt data center, walking away from the largest taxable investment ever offered it, and on water.
On a Tuesday night in late May 2026, the Hanover County board voted four to three to turn away a 900-megawatt data center, walking away from the largest taxable investment ever offered it, and on water. Forty miles south, Chesterfield approved its largest data center campus by right — no public hearing, no disclosure of water use, and a tax rate frozen at twenty-four cents per hundred dollars for thirty years inside a private contract every future board must honor. Two counties, one industry, one state subsidy, one regulated utility, the same eight-year window, opposite distributions of cost. This book explains why. The argument is narrow and holds across the cases: when a capital-intensive, low-employment, utility-intensive land use arrives faster than local review can adapt, what determines who bears the cost is the timing of institutional reform relative to the industry's arrival. Beneath timing sits a single mechanism — leverage that depreciates with every commitment a community makes. From that mechanism the cases sort onto a spectrum of five postures: reform-after, refusal-before, contract-lock, governance-capture, and pre-entitlement-by-ordinance. Grounded in a fifteen-year Mecklenburg audit and the Fredericksburg fall-line natural experiment, the book treats announced investment as a promotional genre rather than an accounting category. Approximately 55,360 words across an introduction, four parts, seventeen chapters, a conclusion, and appendices; Part Three's seven case chapters are built at full depth from the county records. Evidence snapshot: June 12, 2026. Book (1.0.1) update Corrected edition. Restores a sentence at the end of Chapter 3 ("Water, Noise, and the Late Surprise") that was truncated in the prior version's PDF. No changes to the analysis, data, figures, endnotes, or appendices. As a living edition, the concept DOI always resolves to this latest version. Restores text truncated in Chapter 3; fixes Chapters 8–11, which a conversion error had merged into the preceding pages; adds a running header showing the current chapter. No changes to analysis, data, figures, or citations.