T&D Europe Commodity Indices & Reports data

The input-cost side of the European transformer industry, cut into rows instead of association releases. T&D Europe - the Brussels-based association standing for Europe's grid technology providers, 600+ companies including Siemens Energy, Hitachi Energy, ABB, Schneider Electric and GE Vernova - runs a monthly commodity index program through its Transformers Working Group, tracking the materials that dominate a transformer's bill of materials: grain-oriented electrical steel (GOES) in Conventional and Super High Grade variants (split by core-loss thresholds), mineral insulating oil per IEC 60296, compressed paper and cardboard for electrotechnical use per IEC 60641-1-2-3, and hot and cold steel.

Each reading is a price-variation index, not a spot quote. Every commodity is pinned to 1.0 at a reference date, so movement reads directly as cumulative inflation since that point. A mineral-oil print of 178.159 for August 2022 says the input stands 78 index points above its April 2017 base - arithmetic a cost engineer can drop into an escalation clause unchanged. Compilation sits with an independent service provider that blends publicly available national statistical office data with non-public figures supplied by participating manufacturers, a lineage that traces back to the COTREL project before T&D Europe absorbed it. No single supplier's price list leaks into the series, which is precisely why the industry trusts it enough to reference.

Seven series share one schema, six of them currently live: the GOES High Grade line is suspended because too few companies report - a deliberate guard against inferring any one firm's commercially sensitive position. The archive reaches back to 2020. Get a sample of this dataset scoped to the commodities you model.

One row per commodity per reference month - flat enough to paste beside a cost model:

# one row per commodity per reference month
commodity        : mineral insulating oil (IEC 60296)
reference_month  : 2022-08
index_value      : 178.159     # 78 points above the April 2017 base
publication_date : <stamped on the release, normally the month-end after the reference month>
currency         : EUR         # reporting is almost entirely euros

commodity        : GOES conventional
reference_month  : 2022-08
index_value      : <same release sheet, own series level>

commodity        : GOES high grade
status           : SUSPENDED   # too few reporting companies - no current prints

# base mechanics: every series reads exactly 1.000 at its reference date
commodity        : GOES super high grade
reference_month  : 2017-04
index_value      : 1.000

Two things worth noticing before you model. First, the base row is part of the data: because every series equals 1.0 at its reference date, percentage moves need no rebasing work on your side - subtract and divide. Second, stamps can disagree with themselves in the wild (one December 2024 release circulates stamped 31 Jan 2024), so delivered rows carry the reference_month resolved from file contents, not from listing labels. That cleaning pass ships done.

Five fields define the record: the commodity identity, the index_value that carries the economics, and three time-and-context keys (reference_month, publication_date, currency). The split is deliberate - one column identifies what moved, one says by how much relative to base, and the rest make the row joinable across months, releases and currencies without guesswork.

Geography - Europe, EU-focused: readings are compiled from European contributors and national statistical offices, so the series describe the input-cost environment facing European transformer manufacturing rather than a global composite. If your cost base buys GOES or insulating oil on European terms, these are the right coordinates; if you build in North America or Asia, treat the series as a correlated benchmark, not your local price.

Temporal - an archive of monthly readings running from 2020 to the present, roughly seventy-plus releases deep, spanning the entire GOES repricing cycle that upended transformer economics. Each row carries its own reference_month and publication_date, so lag between the month being measured and the release landing is visible in the data rather than assumed.

Granularity - one row per commodity per reference month, seven commodity series, currency almost entirely euros. That is the resolution escalation clauses and cost models consume, and it stacks cleanly against production and trade statistics when a study needs both input costs and shipped volumes.

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

Pick the channel your stack already speaks: endpoints for live lookups against a named commodity set, flat files sized for overnight loads, or a direct pipe into Snowflake, BigQuery or Redshift beside your cost sheets. Cadence is yours to set - and to change when tender season changes.

Every delivery ships with the full field dictionary, sample rows for validation, and a schema held steady between loads.

• Transformer tender pricing and escalation clauses - anchor price-adjustment formulas to a published, industry-governed index instead of a bilateral hunch, with the reference-month arithmetic already resolved.
• Procurement cost modeling - track GOES, insulating oil and electrotechnical paper movements month over month, and see which input is doing the damage before the supplier meeting.
• Margin and earnings forecasting - map index turns onto OEM gross-margin bridges; the 2021-2022 GOES repricing visible in the archive is the stress case every model needs.
• Utility and grid-operator capex planning - translate commodity drift into equipment-cost expectations for multi-year connection and replacement programs.
• Supply-chain risk monitoring - a suspended series or a stalled release is itself signal about reporting participation in a concentrated market.
• Financial analysis of grid-equipment makers - benchmark input-cost pressure across the same vendors that sit inside the contributing association, on one consistent schema.

Procurement and sourcing leads in grid equipment get the sharpest fit: an industry-governed reference series for exactly the materials their tenders escalate on. FP&A and equity analysts covering electrical equipment get a monthly input-cost curve to hang margin narratives on, back to 2020. Strategy and market research teams get defensible numbers for transformer-industry cost structures instead of interview recollection. Regulatory and capex teams at utilities get a third-party yardstick when equipment price increases need justifying. All of it delivered daily, weekly, or hourly, on the channel that fits the workflow - API, files, or warehouse.

Three things worth stating up front. These are indices of price variation, not transaction prices - they describe movement against a base date, so absolute euro-per-tonne levels come from elsewhere. The series reflect a range of contributor sources rather than any single supplier's prices, which makes them ideal benchmarks and useless as a competitor's invoice. And one series is suspended by design: absent prints for GOES High Grade are a governance decision, not missing data.

Notes that pair well with this page:

• The heavy electrical equipment data hub collects every dataset in the industry slice, this one included.
• Best-of ranking: heavy electrical equipment datasets puts the alternatives in context.
• Field guides: transformer commodity price index, power transformer and distribution transformer fix the vocabulary before you model the escalators.
• Neighbors: Data.gov US federal transformer datasets, UK ONS PRODCOM and EPRI power grid equipment research cover the production and engineering sides of the same industry.