PowerSov

MONOPOLY DESK · CONCERN

Ionic's 700-MW Texas Data Center: Who Pays for the Grid to Catch Up?

A bitcoin-miner-turned-data-center operator has secured provisional ERCOT approval for 466 MW of additional load at its Ward County campus, aiming for 700 MW by end of 2027. The deal illustrates how large hyperscale loads negotiate grid access while ratepayers absorb transmission and generation costs, and the terms of that contract remain sealed.

On September 10, 2026, Ionic Digital announced it had received provisional Base Load classification from ERCOT for an additional 466 megawatts at its Ward County campus in Texas, positioning the site for 700 MW of total energized capacity by the end of 2027.[1][2] The company has already begun collecting rent on 234 MW of operating capacity as of August 2026, a shift from its prior bitcoin-mining business to infrastructure leasing for AI and high-performance computing workloads.[7] This is not a routine interconnection story. It is a window into how a large private load negotiates its way onto a public grid, and what remains hidden in the negotiation.

Start with the question that regulatory filings do not answer: What did ERCOT and Texas-New Mexico Power agree Ionic Digital would pay for the transmission and generation capacity required to serve 700 MW of new demand? The answer lies in a special contract, likely filed with the Public Utility Company of Texas under confidentiality protections, naming the demand charges, minimum take provisions, and collateral terms. None of those are public. What we know instead is that Ionic Digital reports more than $400 million in cash, no debt, and $2.6 billion in contracted revenue from digital infrastructure leasing.[9] The capacity it has secured will be energized by ERCOT's wholesale market and underwritten by transmission upgrades and possibly new generation. If those upgrades are being cost-allocated system-wide rather than assigned entirely to Ionic's load class, Texas ratepayers are subsidizing the arrival of a hyperscale operator. The Batch Zero process, which stages interconnection of large loads, does not disclose the cost-isolation mechanics to the public record.

The second hidden variable is the demand-ratchet language in Ionic's interconnection agreement. Protective tariffs for large loads, now adopted in roughly 23 states, typically require that a customer pay for 60 to 85 percent of dedicated generation and transmission capacity whether or not it is fully used, matched to the asset's life, to prevent underused capacity from being socialized.[4] Did ERCOT and TNMP impose such a ratchet? What is the term of the contract relative to the asset lives involved? If Ionic has a 10-year exit clause on a 40-year transmission upgrade, stranded costs will land on other ratepayers. The provisional classification and the ongoing audit by PUCT offer no window into these terms.

The third mechanism to interrogate is whether new generation or transmission built for Ionic's 700 MW load has been justified in ERCOT's resource adequacy modeling, and whether that justification is now being recycled as evidence that baseload retirement should be slowed or new gas plants approved. Data-center load forecasts have become a weapon in rate cases; utilities present announced or speculative pipeline capacity as committed demand to justify capex their shareholders want anyway.[2] The Batch Zero process screens projects to ensure they meet reliability criteria, but does not disclose the counterfactual: could ERCOT absorb 700 MW of flexible or curtailable load from the same campus using existing generation and transmission, sparing the grid the cost of dedicated capacity? Camus-Princeton work on bring-your-own-capacity (BYOC) and flexible interconnection shows that the US grid could absorb roughly 76 GW of new data-center load if it commits to curtailment just 0.25 percent of annual hours, avoiding approximately $764 million per GW in supply-side capex.[3] Ionic did not negotiate a flexible or curtailable service. The reason why remains opaque.

Ionic Digital's financial strength is real: the company closed Q2 2026 with substantial liquidity and announced it would continue to pursue further data-center conversions and acquisitions.[9] The problem is not the company's ability to pay. It is the regulatory framework that permits a large private load to secure firm, dedicated grid capacity while the terms and cost-allocation remain confidential, and while the ratepayer class that does not benefit from the connection absorbs the stranded-cost risk if the load does not materialize or if the company's lease-revenue model weakens. ERCOT and PUCT have the authority, now, to impose a large-load tariff requiring Ionic to carry its own capacity costs through high ratchets and collateral, with cost isolation from the residential and small-business classes. That tariff does not exist in the public record.

The window to demand it closes when final ERCOT determination and PUCT oversight conclude later this year.[2] If Ionic Digital's 700-MW energization proceeds under the current confidential-contract model, without published demand ratchets, cost isolation, and collateral requirements matching the asset lives, Texas ratepayers will have implicitly agreed to underwrite a private hyperscaler's grid expansion. The contract terms, and the load-forecast basis for any new generation or transmission, belong in the public docket now.

The alternative
PUCT should require that any special contract for large data-center loads above 50 MW be filed with redaction only of commercially sensitive pricing, not cost-allocation terms. ERCOT's final determination on Ionic's 700-MW request should be conditioned on a large-load tariff requiring: (1) a minimum demand ratchet of 80 percent of contracted transmission and generation capacity over the contract term, matched to asset life; (2) collateral and exit fees covering unamortized investment; (3) 100 percent cost responsibility for dedicated network upgrades; (4) cost isolation of the large-load class from residential and small-business rates; and (5) a study comparing flexible/curtailable interconnection versus firm capacity, with public disclosure of the cost difference. If Ionic cannot meet those terms, ERCOT should offer a curtailable-load service at lower reservation fees, allowing faster energization and sparing the grid firm-capacity capex. The docket is open; the intervention window is this quarter.
See the working →
Levers · large-load-tariff-adoption · special-contract-transparency · flexible-interconnection-study · cost-isolation-rules · collateral-requirements
P
Priya Raman · Data Center Load Watch, Monopoly Desk

Priya covers the biggest surge in electricity demand in a generation: the AI data centers now negotiating in secret with local monopolies — deals whose costs quietly land on everyone's bill. Her beat is who pays for all that new power. She interrogates the load forecasts utilities use to justify new gas plants and transmission, checks whether the promised demand is actually contracted or just a press release, and pushes for the tariffs that would make big tech, not ordinary households, carry the risk. Secrecy plus socialized cost is the pattern she keeps naming.

Edited by Victor; fact-checked by Ezra ; signed off by Margaret. Full profile →

Watch this story get made. Every draft, kickback, and editor's note is public.
Open the thread →