ResponsibleSteel, a not-for-profit organization developing sustainability performance standards and an independent third-party certification program for the steel value chain, has stated that Global Warming Potential (GWP) values reported in Environmental Product Declarations (EPDs) cannot currently be used reliably to determine its site-level Decarbonisation Progress Levels (DPLs) due to differences in system boundaries, emissions accounting methodologies and scrap definitions.
In its September 2026 report, From Data to Decisions: Comparable GHG Emissions Data for Steel Procurement, ResponsibleSteel examined whether GWP values and scrap-related information in EPDs could be converted into the data required for its DPL system. The study was carried out with four ResponsibleSteel member steelmakers operating in different regions and producing various steel products. According to the report, the global iron and steel industry accounts for approximately eight percent of greenhouse gas emissions, while the construction sector consumes 52 percent of steel produced globally. More than 130,000 EPDs for construction products have been published worldwide, with the declarations increasingly being incorporated into procurement requirements, green building certification systems and regulatory frameworks.
EPDs and DPLs use different system boundaries
ResponsibleSteel stated that EPDs and DPLs serve different purposes. Steel EPDs generally measure the environmental footprint of individual products on a cradle-to-gate basis, including downstream processes such as rolling and coating. By contrast, ResponsibleSteel's DPL system assesses greenhouse gas emissions from raw material extraction through crude steel production immediately after casting, excluding downstream processing and evaluating emissions intensity against the share of scrap in metallic inputs.
The study found that system boundary adjustments could change emissions intensity by as much as 35 percent upwards or 26 percent downwards. Different co-product allocation methodologies affected results by up to 27 percent, while differences in secondary background datasets for upstream emissions resulted in variations of up to 18 percent.
No common conversion factor between EPD values and DPL data
Even after adjustments were made for the three most significant variables, differences of 7-16 percent remained between EPD GWP values and site-level crude steel emissions intensity. ResponsibleSteel stated that no common conversion factor could therefore be established. Even using 15 additional data points, the highest conversion accuracy achieved was 94 percent, which the organization considered insufficient for reliably determining a DPL.
Scrap definitions create further inconsistencies
The report also highlighted inconsistencies in the treatment of steel scrap. ResponsibleSteel noted that recycled content, secondary material and scrap are related but distinct concepts and are calculated differently under various standards. ResponsibleSteel calculates scrap share as qualifying scrap divided by total metallic inputs to crude steel production and excludes internal scrap. According to the report, global steel scrap recovery and recycling rates are already as high as 85 percent, while available scrap currently satisfies only about one-third of global steel demand.
The organization warned that selecting steel solely according to a low EPD GWP value could redirect existing scrap toward products requiring low declared emissions without necessarily increasing overall scrap availability or reducing emissions from primary steel production.
ResponsibleSteel calls for greater transparency in steel EPDs
ResponsibleSteel recommended that steel EPDs disclose crude steel emissions intensity and scrap share, together with information on production sites, upstream emissions databases and co-product allocation methodologies. It also called for A1-A3 emissions to be reported separately for raw material supply, transport and manufacturing rather than solely as an aggregated figure.
The organization recommended the adoption of the Steel Standards Principles' Criteria for Transparency in GHG Reporting and greater methodological harmonization between EPD programs. According to ResponsibleSteel, every steel product EPD should disclose both scrap share and emissions intensity on a cradle-to-crude-steel basis.
EPDs and DPLs should be used together rather than interchangeably
ResponsibleSteel concluded that interoperability between EPDs and its DPL system is theoretically possible but remains resource-intensive and insufficiently accurate under existing methodologies. The organization therefore recommended using the two metrics together rather than interchangeably, with DPLs indicating the decarbonization performance of the steelmaking site and EPDs showing the cradle-to-gate environmental footprint of specific steel products.
The report explicitly stated that A1-A3 EPD values should not be used to calculate or replace a ResponsibleSteel DPL. It also recommended that low-emissions steel specifications use a scrap-variable scale based on both emissions intensity and scrap share instead of relying solely on product-level GWP values.
