FERC actions on data center costs moved from a regional concern to a national tariff issue on June 18, 2026, when the Federal Energy Regulatory Commission issued show cause orders to all six regional grid operators. The orders asked those operators to justify existing tariff rules or propose changes for connecting data centers and other large energy users to the grid. As of August 21, 2026, the filings due on October 15, 2026 had not yet been submitted, so the practical effects remained uncertain. What was clear is that FERC placed cost allocation, speed-to-power, and reliability on the same regulatory track.
The issue is not simply whether data centers should pay higher electric bills. The policy question is whether tariffs can assign the right costs to the right customers before large loads trigger new transmission, generation, and reliability needs. That distinction matters for local ratepayers, because poorly designed rules can either slow useful investment or shift grid upgrade costs onto households and businesses that did not cause the new demand. A related site, SGTT, provides policy tracking across the same network.
How FERC Targets Data Center Costs
Data Center Costs And Cost-Shifting
The central concern in FERC’s June 18 action was cost-shifting. In practical terms, cost-shifting occurs when a data center or other large user pays too little for the grid capacity and upgrades required to serve it, leaving other utility customers to cover part of the bill through higher rates. FERC said the orders were intended to speed large-load integration while protecting consumers and updating market rules, according to the commission’s June 18 announcement.
For data center costs, that framing is significant because it separates two types of expense. One is the cost of using embedded grid capacity that already exists. The other is the incremental cost of new transmission, interconnection work, operating procedures, or market products needed because a large load arrives. If tariffs do not distinguish between those categories, regulators may have difficulty deciding which costs belong to the new customer and which are system-wide investments.
Tariff Reform Areas
The research record identifies five reform areas in the show cause orders. They covered faster transmission service applications and studies, greater transparency around transmission cost allocation, rules for co-located load and generation, flexible transmission service for large time-varying loads, and processes for studying nearby generating facilities that serve large loads. Each category points to a specific stress point in grid planning: study delays, unclear cost assignment, physical proximity between load and generation, variable operating profiles, and uncertainty about how much grid service a co-located project will need.
- Faster study processes may reduce delays, but they do not remove the need for engineering review.
- Transparency provisions may limit cost-shifting, but the effect depends on tariff language and enforcement.
- Co-location rules may reduce some grid impacts, yet they can still raise questions about backup service and reliability obligations.
- Flexible service may help integrate time-varying load, but only if operators can rely on curtailment or operating limits during stressed conditions.
Those limits matter because tariff reform is not the same as new steel in the ground. Regulatory changes can clarify who pays and how projects are studied. They cannot by themselves build transmission lines, add generation, or guarantee that local distribution systems can handle new load at a specific site.
Why PJM Became A Test Case
Capacity Prices And Large Loads
PJM has been central to the policy debate because it serves a large region where data center growth, capacity prices, and grid upgrade concerns have intersected. The research record notes that PJM wholesale power prices rose from $77.78 per megawatt-hour in the first quarter of 2025 to $136.53 per megawatt-hour in the first quarter of 2026, a 76% year-over-year increase, with large loads including data centers identified as a major factor. That figure should be read carefully: it describes a market result, not a complete causal attribution for every customer bill.
Capacity market outcomes can affect customers differently depending on supplier contracts, retail rate design, state policy, and utility procurement timing. Even so, the PJM experience shows why FERC is treating data center costs as a tariff and reliability issue rather than a narrow customer-service matter. Large new loads can change peak demand expectations, which may raise the cost of securing enough capacity for future reliability needs.
Co-Location And Behind-The-Meter Questions
Co-location has become a major point of debate because some data centers seek to locate near generation or use behind-the-meter supply. In theory, that arrangement can reduce the need for some grid deliveries. In practice, it raises several questions that tariffs must answer. Will the data center rely on the grid when the nearby generator is offline? How should backup service be priced? Should a co-located facility receive faster treatment than a conventional interconnection request? What happens if the project’s operating profile changes after approval?
FERC’s earlier December 18, 2025 order involving PJM directed the grid operator to create transparent rules for serving large AI-driven data centers co-located with generation. That prior action made PJM an early venue for testing whether clearer rules can reduce disputes without weakening reliability protections. For readers following the same regulatory thread, prior Illinois Energy coverage of a FERC probe into data center energy costs addressed similar ratepayer questions.
Ratepayer Evidence And Open Questions
What The June 22 Letter Said
A June 22, 2026 letter requesting a GAO review gave lawmakers’ view of the rate impact risk. The letter stated that utilities requested $31 billion in rate increases in 2025, roughly double the 2024 total, and $9.4 billion in the first quarter of 2026, with infrastructure investment tied to data centers and AI load growth identified as a major concern in that record. The request is available in the June 22 letter.
That evidence supports caution, but it does not prove that every rate increase was caused by data centers. Utility rate cases can include many categories of spending, including aging infrastructure, storm hardening, generation changes, distribution upgrades, and financing costs. The stronger claim is narrower: rapid large-load growth can add pressure to capital plans, and regulators are trying to determine whether the resulting costs are assigned fairly.
Limits Of The Evidence
The research record also points to mixed findings on retail rates. A June 2026 preprint reportedly found that data centers slightly lowered retail electricity rates on average in many U.S. regions over 2015 to 2024 by spreading fixed grid costs over more electricity sales. The same research record noted that some regions, including PJM, saw rate pressure from peak demand and capacity charges. Because that finding came from a preprint source rather than a settled regulatory record in the permitted source set, it should be treated as early evidence rather than a final answer.
This distinction is central to interpreting data center costs. Higher load can reduce average fixed costs if the system has spare capacity and the new customer pays its share. The same higher load can raise costs if it arrives in a constrained area, adds to peak demand, requires major upgrades, or receives tariff treatment that does not reflect its grid impact. Both outcomes are plausible under different grid conditions.
FERC Data Center Electricity Costs
What The October 15 Deadline Could Clarify
The October 15, 2026 filing deadline was the next key procedural step as of August 21, 2026. By that date, the six grid operators were required to submit tariff proposals or explain why their existing rules remained just and reasonable. If FERC found the responses inadequate, it could impose mandatory tariff changes under its Federal Power Act authority. That process may clarify how regional operators define large loads, how they study those requests, and how they allocate upgrade costs.
For local communities, the most immediate implication is not a single national price for electricity service. It is the prospect of more formal rules that determine whether data center costs stay with project sponsors, are shared across a broader customer base, or fall into contested categories. The outcome will depend on regional grid conditions, state retail regulation, utility rate filings, and the final tariff language accepted by FERC.
The cautious reading is that FERC’s action was a regulatory attempt to catch up with load growth that had already begun affecting planning and market debates. It may reduce some uncertainty for developers and ratepayers, but it cannot remove the physical constraints of the grid. The evidence available by August 21, 2026 supported concern about cost allocation and reliability, while leaving open how much individual customers will pay after the new tariff filings are reviewed.