Data centers and local benefits ^{CEFR B2}

Artificial intelligence has driven a substantial wave of data center construction across the United States. These facilities often cover large land areas and represent major capital expenditures, prompting local leaders to question whether the projects deliver meaningful local benefits.

Researchers Daniel Yue and Yiyang Zeng at Georgia Tech Scheller College of Business combined detailed facility-level records with county-level economic data to assess impacts on jobs, wages, business establishments, household income, building permits and electricity prices. Their method links openings of individual centers to subsequent county outcomes.

The study finds measurable but modest average gains. In the first three years after activation employment rises about 0.9%, wages 1.1%, business establishments 1.0% and household income 0.7%. Over a longer horizon these effects increase to approximately 3.5% for employment, 5.0% for wages, 4.7% for establishments and 1.9% for household income. Building permits jump sharply during construction, showing short-term local activity.

Gains are uneven. Metropolitan areas capture most spillovers because dense labor markets and supplier networks allow services to scale up quickly (a process the authors call agglomeration). Rural counties typically see far fewer effects: facilities there often employ relatively few permanent workers and rely on outside suppliers. A single large facility can use roughly as much power as 80,000 homes, and where price effects are measurable electricity prices rise by about 5% after a center arrives. Yue and Zeng recommend evidence-based decisions: communities should examine tax abatement terms, electricity tariff arrangements and which parties will fund infrastructure upgrades before courting or approving new data centers.