Datadory notebook

EIA Form 860 Generator Data: The Files, Fields and Workflows Worth Knowing

Datadory delivers electric utilities and independent power producer data covering the US generator census: EIA Form 860 records every American power plant at or above 1 MW nameplate unit by unit - operable, proposed and retired fleets with roughly 73 columns per generator, joined upward to plant, utility and owner sheets, yearly editions from 2001 and predecessor forms reaching 1990. Delivered daily, weekly, or hourly.

1,744 datasets. Pick your catch.

What is EIA Form 860 generator data?

EIA Form 860 generator data is the U.S. Energy Information Administration's mandatory annual census of the American generating fleet, fielded through the agency's statistical system rather than modeled or estimated. The survey records every unit at or above 1 MW nameplate capacity at generator level, joined upward to plant, utility and ownership levels so a single row can be rolled to whichever grain your analysis needs.

The cataloged record is EIA Form EIA-860 — Annual Electric Generator Inventory, which carries a quality score of 10 — one of only 145 perfect scores across the 1,744 datasets in the Datadory catalog, whose mean sits at 7.81. It reaches all US states and territories, geocoding every plant with county, latitude/longitude, NERC region and balancing authority.

Three file families cover the fleet's full life cycle: operable generators online today, proposed units not yet commissioned, and retired units removed from service. Separate technology-specific sheets add detail the core schedule cannot carry, for wind, solar, storage and multifuel units, plus boiler and environmental-equipment associations.

Which files does a Form EIA-860 cycle contain?

One cycle ships as a family of thirteen workbooks - the 2025 early release runs 23 MB compressed, roughly 25 MB uncompressed - each workbook carrying one slice of the fleet:

  1. Operable generators — the workhorse sheet; tens of thousands of rows covering the entire US fleet above 1 MW, one row per generator.
  2. Proposed generators — units reported under construction or planning, the earliest signal of capacity additions before they appear in generation statistics.
  3. Retired generators — units removed from service, with retirement dates supporting fleet-age and decommissioning analysis.
  4. Plant, utility and ownership sheets — the join tables that let you attribute generator capacity to owners, including independent power producers rather than traditional utilities.
  5. Technology-specific schedules — dedicated wind, solar, storage and multifuel sheets capturing technology attributes (tracker type, storage duration, cofiring capability) that generic columns omit.
  6. Boiler and environmental-equipment associations — links between generators and their boilers and pollution controls.

How far back does the generator history go?

The modern annual files run from 2001 through the 2025 early release, giving a quarter-century of consistent generator-level snapshots. That span is long enough to reconstruct when today's gas, wind and solar fleets actually entered service, year by year.

Two extensions deepen the series. Superseded 860A/860B forms reach back to 1990, predating the current file layout by a decade. And since July 2015, monthly 860M supplements carry interim changes between annual waves — the difference between waiting up to a year for a new unit to surface and seeing it within weeks of its status change.

For demand-side context over comparable horizons, EIA's consumption surveys publish cycles reaching much further back in coarse steps: RECS from 1978, CBECS from 1992 and MECS from 1994. Form 860's advantage is the opposite shape — a shorter but completely unbroken annual series at generator grain.

Why get Form 860 generator data through Datadory?

The census itself is simple; the assembly around it is not. A usable fleet model wants the Operable sheet joined to Retired-and-Canceled for survivor curves, the owner sheet decomposed share by share for IPP attribution, and the wind, solar, storage and multifuel schedules read beside the core generator tab - four pulls whose codes and grains then have to agree. Status codes arrive abbreviated (OP, TS, TN), ownership arrives as single letters, and jointly owned units split across rows that only reconcile through the owner tab.

Datadory handles the assembly upstream of you: generator rows typed with nameplate and seasonal capability as numbers, status and ownership codes delivered beside their labels, owner shares resolved per unit, and the Plant Code held stable as the join key into the Electric Power Annual, the Electric Power Monthly and the consumption surveys.

What can you join Form 860 to for a fuller picture?

Form 860 answers "what exists and who owns it"; three sibling EIA records answer what the fleet then does:

Because all four come from the same agency, joins stay simple — shared plant and utility identifiers, consistent state coding, no vendor crosswalk tables to maintain.

Which generator dataset fits which job? (ranked)

Ranked by how directly each cataloged record answers a generator-capacity question:

If your job starts from consumption rather than supply, invert the order and lead with the RECS/CBECS/MECS bundle — the same six-record pool serves both directions.

How do analysts actually use Form 860 generator data?

Four workflows recur across the ways this record gets used:

  1. Fleet composition and aging studies. Join the operable and retired sheets to build survivor curves by technology and vintage; the retired file's dates make decommissioning waves measurable rather than anecdotal.
  2. Renewables pipeline tracking. Diff consecutive years' proposed-generator sheets against the operable file to see which announced wind, solar and storage projects cleared construction — the technology-specific schedules carry the attributes generic columns miss.
  3. Ownership and IPP market share. Roll generator capacity up through the ownership sheet to quantify how much of the fleet sits outside traditional utilities — the record's own tags (ipp ownership, capacity, wind, solar) mark exactly these fields.
  4. Capacity-versus-demand framing. Aggregate nameplate capacity by state or NERC region, then set it against RECS household loads or CBECS building loads so a consumption trend has a supply-side denominator from the same publisher.

Every one of these draws on the same annual census, which is the quiet argument for the record's quality score of 10: one coherent panel of unit rows supports all four.

Form EIA-860 at a glance (Datadory catalog, as of August 2026)
AttributeValue
Cataloged recordEIA Form EIA-860 — Annual Electric Generator Inventory
PublisherU.S. Energy Information Administration (EIA)
Quality score10 / 10 (one of 145 such scores across 1,744 cataloged datasets)
Coverage thresholdEvery US plant and territory, generators at or above 1 MW nameplate
GeocodingCounty, latitude/longitude, NERC region, balancing authority
File familiesOperable, proposed, retired generators + plant, utility, ownership + wind, solar, storage, multifuel + boiler/environmental associations
HistoryAnnual editions 2001 through the 2025 cycle; retirement history back to 2002; 860A/860B forms reach 1990; 860M supplements monthly since July 2015
Size2025 early release: 23 MB compressed, ~25 MB uncompressed, across 13 workbooks
Grain and keysGenerator-level rows joined upward through Plant Code to plant, utility and owner sheets

Pick up where this leaves off

Every one of these ships with sample rows before you commit to anything.

Independent Power Producers & Energy Traders United States - all states and territories, every plant at or…

EIA Form EIA-860 — Annual Electric Generator Inventory

Electric Utilities United States national estimates

EIA Consumption & Efficiency - Electricity Surveys (RECS/CBECS/MECS)

equipment and weight machinery …+8 more

Independent Power Producers & Energy Traders United States national, all states and census divisions, plus…

EIA Electric Power Annual

Independent Power Producers & Energy Traders United States national, all fifty states and census divisions…

EIA Electric Power Monthly

Want rows instead of a pitch? Name the datasets.

API, files, or your warehouse. Daily, weekly, or hourly.

Get a sample

Questions worth asking

What does one Form EIA-860 generator row look like?

One named machine. Utility 63560 Sand Point Generating, LLC reports plant code 1 (Sand Point) in Aleutians East County, Alaska; generator 2 is a petroleum-liquids internal-combustion unit rated 0.9 MW nameplate against 0.3 MW summer capability, status OP, operating since 1976. Tens of thousands of such rows fill the Operable sheet alone, each geocoded with county, latitude/longitude, NERC region and balancing authority.

Does Form EIA-860 cover every power plant in the United States?

It covers every plant at or above 1 MW nameplate capacity across all US states and territories, recorded at generator level with county, latitude/longitude, NERC region and balancing authority. Units below 1 MW fall outside the survey, and respondents self-report their data subject to EIA verification edits documented in the instructions PDF.

Can a delivery be cut to specific technologies, states or vintages?

Yes. Name the technology sheets you need - wind, solar, storage, multifuel - the geographic cut (state, county, NERC region or balancing authority) and the vintage window, whether 2001-2025 or the predecessor-form years reaching back to 1990, and the sample ships shaped to that scope with the field dictionary attached. Ongoing feeds follow the same shape, daily, weekly, or hourly, via API, files, or your warehouse.

How current is Form EIA-860 generator data?

Yearly editions run from 2001 through the 2025 cycle, with a comprehensive retirement history carried back to 2002. Between annual waves, monthly 860M supplements have tracked interim changes since July 2015, and superseded 860A/860B forms extend the underlying series back to 1990 for long-run historical work.