Market Data
Material recovery value drivers
Recovered scrap has value — and that value moves. Commodity prices shape how much of a decommissioning cost can be offset, and the market a project begins in is rarely the one it ends in.
Fifteen years of quarterly data — indexed to Q1 2010 = 100 on a log scale — show how dramatically recovery values shift across a project life. Copper has nearly doubled since 2016. Silver has quintupled from its 2015 trough. Lithium went from irrelevance to a 15× peak and back. Cobalt quadrupled in under three years then collapsed.
Current values — absolute reference
| Commodity | Unit | Q3 2025 | Q4 2025 | Q1 2026 | QoQ chg | Index Q1 2026 |
|---|
Steel scrap — Q1 2026 vs peak
−40%
Steel scrap has fallen ~40% from its Q1 2022 peak (~$633/t to ~$377/t), materially reducing the offset scrap value operators can expect from wind turbine structural steel vs. 2022 provisioning assumptions.
Lithium carbonate — Q4 2022 to Q1 2026
−84%
Battery-grade lithium carbonate has collapsed from ~$75,600/t in Q4 2022 to ~$11,800/t in Q1 2026, sharply worsening the economics of first-generation BESS end-of-life recovery.
Silver — 5-year change to Q1 2026
+209%
Silver stood at ~$16.50/oz in Q1 2010, surged to ~$39/oz in the 2011 commodity boom, then spent a decade below $25/oz before a historic rally in late 2025 pushed it to ~$82/oz — a record of ~$122 in January 2026. Silver scrap is the single most financially material recovered commodity in crystalline silicon solar PV panels.
Important — indicative values only
For copper, aluminium, silver, nickel, lithium and cobalt, prices shown are refined primary metal benchmarks — not scrap. Recovered scrap trades at a discount to these benchmarks: typically 5–15% below LME for clean copper and aluminium, and wider for mixed or contaminated material. Steel scrap is shown as a scrap price directly (derived from the US HMS 1&2 benchmark), since the structural steel market for decommissioned wind towers is traded as scrap rather than primary metal. In addition, not all metal content in a decommissioned asset is economically recoverable: effective recovery rates vary by material, asset type, and recycling process (e.g. lithium recovery ranges from ~40% in pyrometallurgical routes to ~85–90% in hydrometallurgical processes; rare earth content in permanent magnets is often unrecovered commercially). These prices are therefore best read as directional indicators of recovery value, not as net realisable amounts per unit of decommissioned capacity.
Methodology
Data notes
This panel tracks the scrap recovery value dimension of decommissioning economics. Prices shown are indicative quarterly averages derived from public commodity price series. They do not represent actual transaction prices for any specific project or asset.
- Steel Scrap: Derived from FRED series WPU10170503 (US Producer Price Index: Iron and Steel Scrap Manufacturing), converted to approximate USD/metric ton by anchoring the PPI index to the observed US HMS 1&2 spot benchmark of ~$400/t in Q1 2021 and back-applying the same conversion ratio across the full Q1 2010–Q1 2026 series. Values are indicative and reflect directional US domestic scrap market pricing.
- Copper, Aluminium, Nickel, Cobalt: FRED/World Bank commodity price series (PCOPPUSDM, PALUMUSDM, PNICKUSDM, PCOBAUSDM). Three-month averages of monthly LME-derived spot prices in USD per metric ton.
- Silver: FRED series SLVPRUSD. Monthly London Bullion Market Association (LBMA) silver fixing price in USD per troy ounce.
- Lithium carbonate: FRED series PLIUSDM (World Bank commodity price index). Battery-grade lithium carbonate, USD per metric ton. Chinese spot market derived.
- All series are indexed to Q1 2010 = 100 on a logarithmic scale, enabling proportional comparison across commodities with very different price magnitudes and volatility profiles. A log scale ensures that a 50% fall and a 100% rise occupy equal visual distance, which is the appropriate representation for long-run commodity price series.
- Quarterly values are three-month averages of monthly observations. Q1 2026 uses January and February 2026 average where March data was not yet available at time of publication.
- Data is updated by Endenex on a quarterly basis. Last updated: Q1 2026.