The Facts on Data Centers and Electricity Prices
American data centers are the physical backbone of the digital economy — the facilities that run the cloud services, payment systems, hospital records, and AI tools tens of millions of people use every day. As that infrastructure expands, data centers and electricity prices have become a single story in the public conversation, and the numbers driving that story deserve precision. In June 2026, PolitiFact examined one of the most-cited claims about that link and rated it "Mostly False." The correction matters, because energy policy built on a misread statistic produces the wrong solutions.
What the Data Shows About Data Centers and Electricity Prices
The claim under review held that residents near large data centers had seen their electricity bills rise "by as much as 267%" over five years. PolitiFact found the 267% figure describes wholesale power prices — the amount utilities pay generators — which account for only 30% to 50% of a residential bill. The remaining cost is transmission, distribution, grid maintenance, and taxes.
The actual figure for consumers tells a clearer story. Residential electricity prices rose about 42% on average nationally over five years, with regional highs of 94% in Washington, D.C., 74% in Maryland, and 73% in Maine. Those increases track aging grid infrastructure, transmission constraints, extreme-weather resilience spending, and fuel costs — factors that were pushing bills upward before the AI buildout began.
Do data centers raise electricity bills? In specific regional markets, large computing loads do add measurable pressure. The Independent Market Monitor for the PJM Interconnection found data-center demand contributed to a 174% rise in capacity costs for the 2025–26 delivery year. That strain is real and worth addressing directly. The honest accounting shows a regional capacity issue inside a national grid-cost picture — a problem with targeted engineering and rate-design answers, sized correctly.
Precision Produces Better Energy Policy
Large-load customers give utilities something residential demand alone rarely does: a creditworthy, long-horizon commitment that finances new generation and transmission the whole grid uses. Regulators in several states are already pairing that demand with large-load tariffs designed so data centers pay for the capacity they draw. That design work depends on accurate inputs. A 267% headline points toward blanket restrictions on computing infrastructure; a correct 174%-capacity-in-one-market figure points toward interconnection reform and rate design that protect households while the grid grows.
The scale of the underlying investment is worth stating plainly. America's largest technology companies are on track to spend roughly $725 billion on AI and computing infrastructure in 2026 — a 77% jump over the prior year. Capital at that scale, directed toward American soil, is a generational opportunity to modernize the grid. The policy question is how to channel it well, and that requires starting from verified numbers.
The Standard for Energy Claims
Every serious energy debate should meet one test: does the number describe what it claims to describe? Wholesale prices and residential bills are different measures. Capacity costs in one regional market and national average rates are different measures. Treating them as interchangeable produces conclusions that fail on contact with the data. Households facing higher bills deserve solutions built on the real drivers of those bills, and American infrastructure investment deserves an evidence standard as rigorous as the engineering behind it.