There has been much negative chatter about how data centers are placing upward pressure on electricity rates. However, filtering away the noise, there are many reasons for rising electricity rates that aren’t related to data centers. The reality is that recent rate increases are primarily the result of past decisions that were intended to transform the electric utility industry by:
- Accelerating the retirement of dispatchable fossil-fuel generation prior to commissioning comparable amounts of new “accredited” capacity; and
- Subsidizing resource additions that are: (1) highly weather-dependent; (2) require backup capacity to maintain reliability; (3) displace (or more accurately, “crowd-out”) existing economical 24×7 capacity; and (4) require significantly more transmission/transformation infrastructure to deliver the same capacity to serve the load in the utility’s service territory. (Fact: Aggregate transmission rate base has experienced double-digit growth, leading to more rapidly escalating transmission rates.)
This transformation occurred during a period when utilities were projecting low or flat load growth. Clearly, growth in data center load was not yet on the planning horizon.
How is this relevant to electricity rates?
A fundamental principle of ratemaking is that a rate is cost divided by usage. Rates rise when costs increase faster than usage. Rates fall when usage increases faster than costs. That is simple math.
For example, when a utility replaces a 1,000 megawatts (MW) dispatchable resource with 1,000 MW of weather-sensitive resources, much of the 1,000 MW of transmission equipment capacity is under-utilized as a result of the intermittency. Thus, integrating weather-sensitive resources into the grid drives higher per unit delivery costs. If 2,000 MW or more of weather-sensitive resources are required to replace the accredited capacity of a 1,000 MW dispatchable resource, electricity would be more expensive on per unit basis. Higher per unit delivery costs and higher per unit electricity costs will inevitably lead to higher rates.
Planning for new resources requires long lead times. Even after a new resource and the accompanying infrastructure are approved, it may be years, even a decade or more, from planning to commercial operation. Thus, fast forward to today, the impact of past policies favoring weather-sensitive (rather than dispatchable) resources is only now coming to fruition; that is, utilities are seeking — and receiving — higher rates to recover their prudent and reasonable costs.
However, today’s noise is all about data centers. Noticeably absent from the noise is any discussion or assessment of the policies that have actually resulted in higher rates/costs, as stated above. What is lost in the discussion is that (1) a utility incurs substantial fixed costs to provide the capacity required for reliable service; and (2) adding high load-factor loads like data centers means that usage is growing faster than cost, thereby putting downward pressure on rates. In other words, with data centers, a utility can spread its fixed costs over more usage. Those who claim otherwise either don’t understand the math, have a different agenda, or are anti-data center.
Further, electricity rate increases have not been uniform across the nation. Published industry studies reveal that the largest increases in electricity rates are mostly confined to specific regions that have mandated policies favoring subsidized resources and discourage usage.
Ultimately, while data centers can help lower costs and rates, this won’t happen without ensuring that the rates, terms, and conditions of service provide revenues that fully recover the costs to serve them.
The circumstances I have described are the premise of a recent Substack post by Alexander Muse. For convenience, I have provided a link to the post. More information about the author is provided below.
Alexander Muse is a Fellow at the John Milton Freedom Foundation and publishes daily political analysis at amuseonx.com. Primary sources cited in this piece are linked inline; campaign finance figures are drawn from FEC filings, polling data from publicly released crosstabs, and legal claims from filed pleadings. Corrections are posted to the original URL with a dated changelog. Readers who identify errors are invited to contact the author directly.