U.S. Large-Load Power Cost Monitor · Preview

What power actually costs a large load in U.S.

5 U.S. markets currently carry a delivered-cost band that can be compared like for like. 6 do not, and each says why.

Every band is the recurring delivered cost for the same standard case: a 100 MW continuous, high-load-factor load, in US dollars per megawatt hour, excluding customer-funded connection capital and refundable taxes.

This is a monitor, not an index. Nothing is aggregated across markets. Each row is one market's own band on its own evidence. Calling it an index would imply a composite value and a versioned aggregation method, neither of which exists. The name changes when they do.

Eligible markets

Each market's own band, on its own evidence. Nothing here is averaged or ranked against anything else; the bar shows where each band sits within the range of the others.

Delivered cost bands in US dollars per megawatt hour for eligible markets, with the evidence state behind each.
Market	Low	Mid	High	Evidence
ERCOT Central (Houston & DFW) rolls into ercot	$55	$72	$95	Applicable, evidence incomplete
ERCOT West (West Texas & Panhandle) rolls into ercot	$42	$55	$80	Applicable, evidence incomplete
MISO (Midwest)	$48	$60	$78	Applicable, evidence incomplete
NYISO (New York)	$70	$95	$140	Applicable, evidence incomplete
SPP (Southwest Power Pool)	$42	$54	$68	Applicable, evidence incomplete

Screen these markets → All compiled cost evidence → Methodology →

Aggregate views

A market whose band is built from its own sub-markets. It is the same market seen once more, not an extra one, so it is not in the comparison above.

Market	Low	Mid	High	Range within the cohort	Evidence
ERCOT (Texas)	$42	—	$95		Applicable, evidence incomplete

A constituent envelope and its constituents are the same market seen twice. Both are eligible and both are publishable, but the default comparison shows each market once, preferring the constituents: ERCOT's $42-$95 band is West's floor beside Central's ceiling, a range no single site faces, and reading it as one market's spread overstates the uncertainty at either location.

Excluded markets, and what each is missing

A market is out because its evidence does not yet support a like-for-like comparison, not because it is expensive or cheap. Structural means no amount of work here changes it. Temporary means evidence we have not finished gathering.

U.S. markets excluded from the comparison, with the reason, the missing evidence and the next action for each.
Market	Exclusion	Status and missing evidence	Next action
CAISO (California)	temporary	Methodology unresolved availability class is contingent_evolving, not standardized_observable Methodology unresolved	Resolve the methodology from first principles, then populate. Not a data-collection problem.
Desert Southwest (Phoenix)	structural	Inapplicable evidence no delivered-cost basis is stated Inapplicable evidence no published band no connection-voltage or serving-utility assumption is declared The published band blended an effective regulated rate with APS's PROPOSED extra-large-load class of roughly +45%, which the Arizona Corporation Commission has not decided: the rate-case hearing closed 7 July 2026 and the Commission's vote is due by 31 December 2026. A pending proceeding may not set a current delivered cost. The band's primary source was also an ACC large-load workshop rather than a tariff. not stated in USD per MWh	Obtain evidence for a service the archetype could actually contract. The existing figures cannot be repaired by recalculation.
ISO New England	structural	Observable in part; complete cost fragmented by jurisdiction no delivered-cost basis is stated availability class is varies_by_jurisdiction, not standardized_observable no published band no connection-voltage or serving-utility assumption is declared A single delivered-cost estimate cannot be defended at ISO-NE-wide scope. The regional transmission layer is uniform and verifiable, but local transmission, distribution and retail demand charges are set by 32 serving utilities across six states and five regulators. The superseded values are recorded in the correction record, not here. not stated in USD per MWh	Narrow to one serving utility and interconnection point, then apply that utility's local transmission and distribution tariff. No region-wide document will close it, because no regional body sets these charges.
Northern Virginia (PJM)	temporary	Publication grade no published band	Use the band, with its stated confidence and boundary.
Pacific Northwest	structural	Inapplicable evidence no delivered-cost basis is stated Inapplicable evidence no published band no connection-voltage or serving-utility assumption is declared The published band was derived from the Bonneville Power Administration priority-firm rate, which a new large load cannot obtain: federal hydro allocations are largely committed and a new customer would pay a market premium above the federal rate. A tariff the standard archetype cannot purchase may not set a delivered-cost band. Cost here is set by a bilateral arrangement rather than a posted rate, so no standardized band can represent it. not stated in USD per MWh	Obtain evidence for a service the archetype could actually contract. The existing figures cannot be repaired by recalculation.
US Southeast (Atlanta)	structural	Structurally unavailable no delivered-cost basis is stated Structurally unavailable no published band no connection-voltage or serving-utility assumption is declared The band was anchored on a regulated bundled industrial tariff that a new 100 MW load cannot take: Georgia closed its standard schedules to new large loads on 1 February 2025 and requires an individually negotiated contract instead. The band's primary source was also EIA industrial monthly pricing, which may serve only as a lagged validation layer and not as the headline large-load price. not stated in USD per MWh	Filed or executed contract terms, or a published contract-based rate. Searching for a tariff schedule will not close it.

4 structural, 2 temporary, out of 12 markets assessed.

How a market gets in

Membership is derived from the evidence every time this page is built. There is no list of preferred markets: a market whose band is withdrawn leaves the comparison on the next build, without anyone deciding.

All eight must hold

The delivered-cost boundary is complete and stated; the standard archetype applies at this scope; the band is current and not under correction; connection-voltage or serving utility assumptions are declared; nothing is withdrawn; the price is not specific to one customer's negotiated terms; no regulatory decision is pending that would invalidate the comparison; and units, currency and method are comparable.

What is not done

It does not compute a composite. It does not rank markets by anything other than their own published bands. It does not fill a missing bound, infer a mid-point, or convert a currency. And it holds no list of preferred markets: remove every market from pgiq-data.json and this module returns an empty monitor rather than an error.

Publication rule

Publish with the eligible set that exists. Do not wait for a target count and do not admit an unsupported market to reach one. An empty eligible set is a publishable result: it says the evidence is not there yet, which is true and useful.

Take the data

The same evidence behind every band, per market.

Cost evidence

Compiled recurring-cost bands and their per-line provenance.

GET /api/v1/markets

Browse the compiled evidence →

Screening

Filter the cohort by cost, tier, connection time and group.

GET /api/v1/screen?group=us

API & MCP →

Spreadsheet

Every market in the cohort, eligible or not, with its bands, failing conditions and next action.

Generated us-monitor-eligibility-1.0 · Full methodology · Cost & connection data