What data centers might mean for your electric bill

Data center development is intensifying debate over utility bills and infrastructure spending, a topic now highly politicized as seen in calls from President Trump for tech companies to bear full energy costs and East Coast Democrats' focus on energy affordability. Experts project a significant boost in U.S. electric load, comparable to adding 15 New York Citys in peak demand over five years, with much of this growth concentrated from Texas to Pennsylvania. The non-profit overseeing the grid in Minnesota and central states anticipates more than three New York Citys' worth of demand by 2035 within its territory. Electric companies are responding by undertaking massive infrastructure projects to support these new industrial loads, harden grids, integrate clean energy, and prepare for increased demand from electrified vehicles and buildings. Investor-owned utilities, including Xcel Energy, plan at least $1.4 trillion in capital spending over the next five years. Xcel Energy alone intends to spend $60 billion on projects such as a controversial transmission line connecting Minnesota and Wisconsin. Similarly, Minnesota Power is planning a high-voltage line between Grand Rapids and Hermantown, Minnesota, where Google is set to construct a large data center. These capital expenditures typically lead to utilities requesting "ratebasing" from state regulators, allowing them to recover costs and a regulated profit margin from customers. For instance, the Minnesota Public Utilities Commission is currently reviewing an Xcel request to raise Minnesota electric rates by 13.2% over two years. Charles Hua, founder of PowerLines, notes that utilities increase profits through capital expenditures, attracting institutional investors like BlackRock, which recently acquired Minnesota Power's parent company. This surge in spending is creating a backlash from consumers and voters concerned about rising utility rates. The connection between data centers and these rate increases is increasingly explicit. For example, DTE Energy in Michigan offered a two-year rate freeze if an Oracle and OpenAI data center project proceeds, implying that large new customers could help stabilize rates by utilizing underutilized grid capacity, especially if they can manage their load through onsite batteries or flexible operations during peak times. However, Hua's group questions whether all proposed capital spending is truly necessary for consumer benefit, suggesting that lower-cost solutions improving grid efficiency are often overlooked.