Who Pays for Data Center Grid Upgrades? FERC's June 18 Decision Will Set the Answer

The FERC data center cost allocation vote on June 18 is the most consequential equipment procurement signal of 2026. Three separate policy moves in the past 60 days have set the stage, and the question heading into Thursday’s open meeting is not whether utilities will build more transmission for data centers. It is who writes the check.

Here is what the ruling means if you buy or sell grid equipment.

Three Moves, 60 Days

The utility industry, the federal government, and the White House each took a position before the June 18 vote. Read together, they read as a coordinated campaign to lock in the outcome before the open meeting.

The EEI industry frame

At the June 2026 POLITICO Energy Summit, Edison Electric Institute president Drew Maloney and Entergy CEO Drew Marsh made the public case for developer-pays cost sharing. Their core argument: data center load growth puts downward pressure on rates by spreading fixed grid costs over more kilowatt-hours sold, so large-load agreements benefit existing customers rather than burdening them.

EEI put numbers behind the claim. U.S. investor-owned utilities plan to invest $1.4 trillion in the grid from 2026 through 2030, with $238.8 billion in capital spending in 2026 alone (a 17 percent year-over-year increase). Entergy offered the working model: its data center customer agreements are projected to deliver $7 billion in total savings for 2.3 million customers across Arkansas, Louisiana, and Mississippi. The Entergy-Meta agreement in Louisiana commits to $2 billion in customer savings over 20 years.

That deal structure is now what utilities cite when they seek state PUC approval for similar agreements. A model that wins at the POLITICO Summit becomes a model that travels to rate cases.

FERC approves PJM EIT on June 9

Nine days before the RM26-4 vote, FERC approved PJM’s Expedited Interconnection Track for generating resources 250 megawatts and larger. The EIT allows shovel-ready projects to pay a premium (a $500,000 non-refundable study deposit plus $15,000 per megawatt) to move outside the standard five-to-seven-year interconnection queue backlog. Projects must come online within three years and bear 100 percent of network upgrade costs.

That is a deliberate supply-side signal: generators willing to accept full cost responsibility can reach market fast. RM26-4 extends the same logic to the demand side.

For a fuller breakdown of the June 9 approval alongside two other concurrent procurement signals this summer, see our post on three signals compressing equipment timelines before July 4.

FirstEnergy’s pipeline proposal

Around June 5, FirstEnergy filed at FERC proposing direct cost assignment for transmission upgrades driven by large loads. The model borrows from natural gas pipeline ratemaking, which has been in use for more than 25 years without additional legislation.

The proposed structure uses two rates under 15-year contracts. The first is a zonal transmission rate covering the data center’s proportionate share of existing system costs. The second is an expansion rate covering only the new transmission facilities built specifically for that customer’s interconnection. The expansion rate can shift over time as additional customers use the same facilities.

Amazon, Google, Meta, Microsoft, OpenAI, Oracle, and xAI all signed the White House Ratepayer Protection Pledge in March 2026. Those companies broadly support developer-pays cost allocation frameworks. They have calculated that schedule certainty and interconnection speed are worth accepting direct cost responsibility on a per-project basis.

What RM26-4 Actually Decides

Docket RM26-4-000 covers all large loads above 20 megawatts and applies to all eight major RTOs and ISOs under FERC’s direct jurisdiction, not just PJM. The central question is whether a large load customer pays for the transmission capacity it requires (developer-pays) or whether those upgrade costs are spread across the full ratepayer base (socialized costs). Current rules in most regions default to socialization.

Critics have a real point. Maven Solutions argues that FirstEnergy’s proposal concentrates demand-forecast risk, utilization risk, and cancellation risk on a single customer class. A 15-year contract with collateral requirements is a significant obligation, and assumptions about data center energy intensity can shift faster than a 15-year amortization schedule.

A partial rule is possible: covering only new transmission facilities rather than requiring a zonal share for legacy infrastructure. That middle-ground outcome would preserve some socialization for existing grid assets while requiring developer-pays for incremental expansion, which is close to where FirstEnergy’s two-rate structure already lands.

The fight over who pays for data center transmission has been building since at least 2024. Our earlier breakdown of the political and technical dimensions is at Data Center Grid Infrastructure Costs: The $10B Fight Over Who Pays.

The Procurement Signal If Developer-Pays Wins

Utilities that secure cost recovery from large-load customers gain a direct revenue stream for transmission buildout without relying on speculative load forecasts. That removes the primary obstacle to placing equipment orders ahead of confirmed demand.

The sequence from here: FERC issues the RM26-4 rule on June 18, utilities file tariff revisions through the third quarter of 2026, and transmission equipment RFPs from PJM-region transmission owners follow in the fourth quarter of 2026 and the first quarter of 2027. The equipment categories on those RFPs are large power transformers, extra-high-voltage switchgear, underground cable, and protection systems.

PJM territory carries the highest concentration of data center development. The Northern Virginia corridor and the Ohio-Columbus build zone are where equipment demand will show up first. Distributors serving EPC contractors in Pennsylvania, Ohio, New Jersey, Virginia, Maryland, Indiana, and Michigan should expect elevated large-project equipment demand inside that 18-month window.

For smaller distribution buyers, the indirect effect is also worth watching. When utilities secure cost recovery certainty on large-load transmission, capital that was previously tied to speculative projects can shift toward distribution-system hardening. Substation upgrades, distribution transformers, and feeder replacements that compete for the same capital budget as transmission expansion become incrementally easier to approve once large-load costs are ring-fenced. That is the path the FirstEnergy-PJM market design story has been building toward over the past several months.

Watch the Tariff Filings, Not the Headlines

The June 18 vote is the starting gun, not the finish line. If RM26-4 passes, the procurement signal will not appear in press releases. It will appear in tariff revision filings from PJM-region transmission owners, followed by engineering studies and equipment RFPs.

July and August 2026 are the months to watch for those filings. That is when utilities translate FERC direction into procurement commitments. The broader context on utility CapEx this cycle is in our post on the $1.4 trillion utility investment wave and transformer procurement timing.

If the vote slips again or the rule comes out narrower than expected, transmission RFPs shift to Q1 or Q2 of 2027. The equipment lead-time math does not change; the procurement window compresses further.

---

Related Reading: Data Center Grid Infrastructure Costs: The $10B Fight Over Who Pays, PJM at the Inflection: 220 GW Queue, FirstEnergy’s Refusal, and the Hull Street Veto, The $1.4 Trillion Utility CapEx Wave and Transformer Procurement

---

The Feeder is our free monthly digest of the most consequential procurement signals from FERC dockets, utility capital plans, and equipment market data. No noise. Subscribe at distroforge.com/feeder.