FERC Issues Show-Cause Orders on How Large Loads Pay to Connect and When They Must Curtail

On June 18, the Federal Energy Regulatory Commission was asked to write a rule, and declined.

The Energy Department had pressed FERC to open a federal rulemaking that would speed data centers and other large loads onto the grid, including a faster interconnection clock. In a unanimous vote, the commission rejected the rulemaking while keeping the urgency. Rather than a national standard, it issued show-cause orders to the regional grid operators, naming PJM, SPP and others, giving each roughly 60 days to justify or rewrite how it studies, prices, and connects large loads.

The order sets two principles at the wholesale level: who pays for the grid a large load requires, and what that load owes the system in return. Neither principle arrives as a number. The demand charge a commercial building actually pays is not set at FERC. It is set downstream, at state public utility commissions, by how this federal cost-causation principle gets translated into rate design over the next several years. The June 18 order is the top of that chain, not the bottom.

What the order requires. RTO filings must demonstrate that large loads pay for the infrastructure built to serve them. FERC framed this as “cost recovery agreements” that prevent the cost of new interconnection from being socialized onto existing wholesale customers. The filings must also address cost transparency, co-location rules, generation studies, and provisions for flexible load. On the flexibility point, the order requires operators to accommodate large loads that can adjust to grid conditions, through co-location with on-site generation or curtailment during peak hours.

What it declined. FERC did not impose the accelerated interconnection clock the Energy Department wanted. There is no mandatory fast-track timeline; the commission left interconnection timelines to each RTO. That choice signals a framework still being filled in rather than a settled standard.

Two handoffs. The structure of the order matters as much as its content. FERC declined to act as a rulemaker and instead handed the design work to the RTOs, which now have roughly 60 days to file. It then explicitly delegated retail consumer protection back to the states. A wholesale cost-causation principle has to survive two translations before it reaches a commercial electricity bill: first into each RTO’s revised tariff, then into each state commission’s rate case. The agency named the principle and forwarded the invoice.

The cost-causation chain. This is the same mechanism that has been moving through state legislatures and PUCs, now carrying a federal endorsement. When the cost of new transmission and substations is assigned to the large loads that trigger it, rather than spread across all customers, the long-run effect on rate design is to make demand charges more cost-reflective. Utilities recover grid capital from the customers driving it. In territories with heavy data-center growth, that pressure tends to sharpen the commercial demand charge, the per-kilowatt fee on a building’s monthly peak that is the single largest controllable line on most commercial electricity bills. FERC has not changed any building’s bill. It has strengthened the principle that will reshape those bills wherever states choose to apply it.

Flexibility versus defection. The order’s second principle is the more consequential one for distributed resources. By requiring operators to accommodate flexible and curtailable large loads, FERC put federal weight behind a specific bargain. A load that can reduce its draw during constrained hours is cheaper and faster to connect than one that demands firm capacity around the clock.

The alternative, framed in a Latitude Media analysis published this week, is defection: large loads that cannot obtain flexible grid service increasingly pair with dedicated on-site generation and exit the utility planning system entirely. The capital already committed to AI infrastructure makes that outcome likely wherever the grid cannot move fast enough. A facility that accepts limited curtailment can often energize sooner than one waiting for firm service, which is the trade FERC is trying to make the more attractive of the two.

A load becomes flexible in two ways. It can pair with dedicated generation, which is the defection FERC is trying to avoid, or it can store energy on site and discharge it during the hours the grid wants the load to step back. The second path is what makes a building dispatchable without building a power plant next to it. FERC’s flexibility provisions do not name storage, but they reward exactly the capability that stationary storage provides: a peak that bends on command.

The downstream question. A show-cause order is weaker than a rule. It asks RTOs to justify their current practices or change them, and the operators will answer in their own interest. PJM and SPP are among those that must file, and a 60-day deadline relocates the cost-allocation question rather than resolving it. Whether the flexibility provisions become a real revenue channel for curtailable load, or merely a new study requirement, depends on filings that do not yet exist. And because retail protection sits with the states, the same federal order can produce a sharp cost-causation rate design in one state and a diluted one next door.

What FERC settled on June 18 is the direction of travel. Large loads will pay more of their own way, and flexible loads will be treated as cheaper and more welcome than rigid ones. Those are durable principles. The dollar figures attached to them live in the RTO compliance filings due over the following weeks and in the state rate cases that follow, which is where the question of what a flexible commercial building is actually worth will finally get a number.

Sources