Choosing the Right Emission Factors for Your CSRD Disclosure

By Felix Gruber, Head of Climate Science

Of all the technical decisions in a CSRD disclosure, emission factor selection generates the most restatements. Not scope boundary setting. Not consolidation approach. Emission factors — specifically the use of outdated ones, geographically wrong ones, or ones drawn from databases that weren't designed for the disclosure context in question. In our experience analyzing early Wave 1 CSRD filings, this is the single most common technical deficiency auditors flag.

The underlying problem is that emission factors are not a monolithic dataset. There are at least six major factor libraries in common use for CSRD disclosures, each with different geographic coverage, update cadence, methodological assumptions, and appropriate application scope. Choosing the wrong one doesn't just produce inaccurate figures — it produces figures that a CSRD auditor will challenge, because factor selection leaves a paper trail.

This guide walks through the major emission factor databases, when each is appropriate, and the practical rules that determine defensible selection under ESRS E1.

Why Factor Selection Matters for CSRD Specifically

Under CSRD and ESRS E1, companies are required to disclose their GHG inventory using methodology consistent with the GHG Protocol Corporate Standard. The GHG Protocol doesn't mandate a single emission factor database — it requires that factors be appropriate for the activity data type, current, and consistently applied across periods. Auditors reviewing CSRD disclosures will assess all three of those criteria.

"Current" is where most enterprises fall short. A 2019 national grid emission factor was current in 2019. Using it in a 2025 or 2026 CSRD disclosure, without documentation of why a more recent factor wasn't available, creates a restatement risk. Some EU national grid mixes have shifted by 10-20% since 2019 as renewable penetration increased, which means the error on Scope 2 electricity could be material for energy-intensive operations.

The Major Factor Libraries and Their Appropriate Use

Understanding when to reach for each database is the core of defensible factor selection.

ecoinvent (v3.x)

ecoinvent is the most widely used life-cycle inventory database for Scope 3 process emissions. It covers several thousand industrial processes with detailed supply chain modeling and is updated approximately annually. Ecoinvent is appropriate for manufacturing process emissions, material production (steel, cement, chemicals), and Scope 3 Categories 1 (purchased goods and services) and 4 (upstream transportation) where process-level accuracy matters.

The limitation is cost and complexity: ecoinvent requires a license, and matching your activity data to the right ecoinvent process unit requires LCA expertise or a software platform that manages the mapping. For spend-based Scope 3 estimation, ecoinvent is typically not the right tool.

IPCC AR6 (Sixth Assessment Report)

The IPCC AR6 emission factors are the scientific standard for combustion, refrigerant leakage, agricultural process emissions, and similar activity categories. They are appropriate for Scope 1 direct emissions — stationary combustion (natural gas, fuel oil, diesel), mobile combustion (company-owned vehicles), and fugitive emissions (refrigerants, methane leakage). The AR6 global warming potential values — 100-year GWP for methane is 29.8 CO2e under AR6, up from 25 under AR5 — are current and represent the expected standard for CSRD filings going forward.

One practical note: CSRD doesn't yet mandate AR6 GWPs explicitly, but auditors familiar with EFRAG guidance will note if AR5 values are used without explanation. Documenting the GWP source is a two-line addition to your methodology note; it removes a question before it's asked.

IEA Grid Emission Factors

The International Energy Agency publishes annual grid emission factors by country, updated with roughly a one-year lag. These are the standard reference for Scope 2 location-based electricity accounting under GHG Protocol's Scope 2 Guidance. They are freely available, widely recognized, and the expected default for Scope 2 location-based calculations in EU and global CSRD disclosures.

For market-based Scope 2 accounting, IEA factors serve as the residual mix fallback when no supplier-specific or REGO-backed residual mix factor is available. The European residual mix factors published by AIB (Association of Issuing Bodies) are the more precise alternative for European operations when market-based accounting is used.

Spend-Based Coefficient Databases (Exiobase, US EPA USEEIO)

For Scope 3 categories where supplier-specific or process-level data is not available, spend-based estimation using economic input-output models is the standard fallback. Exiobase is the European-standard MRIO database covering 49 countries and 200 industry sectors. US EPA USEEIO covers North American operations with sector-level intensity factors in tCO2e per USD of spend.

Spend-based estimates carry higher uncertainty than process-level calculations — typically ±30-50% relative to primary data — and ESRS E1 expects disclosure of estimation uncertainty for material Scope 3 categories. This isn't a reason to avoid spend-based methods; it's a reason to document confidence levels and set out a roadmap for improving data quality over reporting cycles.

Supplier-Provided Primary Data

Where suppliers provide verified emission intensity figures — either through CDP responses, direct data requests, or product-level environmental footprint data — these take precedence over proxy factors for the relevant supply chain node. The documentation requirement is higher: you need the methodology the supplier used, the verification status, and the data year. But the disclosure quality improvement is significant, and ESRS E1 encourages movement from estimated to primary data over time.

A Practical Decision Framework

Mapping the right factor library to each activity category doesn't need to be complex. We use a three-level hierarchy internally, and our customers find it useful for structuring their methodology notes:

Activity Category	First choice	Fallback
Scope 1 combustion	IPCC AR6 combustion factors	National regulator published factors
Scope 1 refrigerants	IPCC AR6 GWP values	Prior AR5 values with documented rationale
Scope 2 location-based	IEA country factors (current year)	National grid operator annual publication
Scope 2 market-based	AIB residual mix (EU) / supplier EAC	IEA location-based as fallback
Scope 3 Cat 1 (purchased goods)	ecoinvent process data	Exiobase spend-based coefficients
Scope 3 Cat 4/9 (transport)	Carrier-provided fuel intensity	ecoinvent or GLEC Framework factors
Scope 3 Cat 15 (investments)	PCAF standard financial emission factors	IPCC sector defaults with asset-level modeling

The Versioning Obligation

Whatever factors you select, your methodology note must record: the database name, the version or publication year, and the access date. This applies even for freely available government data. An auditor following a Scope 2 calculation back to its grid emission factor will ask which vintage of the IEA publication was used — and "the one in our template" is not an acceptable answer.

We recommend maintaining a factor registry: a single reference document or tab in your calculation workbook that lists every factor used, its source, version, and the Scope or Category to which it applies. Factor registries take about two hours to build from scratch; they save significant time during assurance preparation and make base-year recalculations tractable when factors are updated in subsequent years.

Getting Restatements Right When Factors Change

Under GHG Protocol and ESRS E1, if a significant error is identified in a prior-period inventory — including the use of an incorrect emission factor — the base year must be restated. Materiality thresholds for restatement are typically expressed as a percentage of total inventory; ESRS E1 guidance suggests significant recalculation triggers when a change affects more than 5% of total reported Scope 1-3 emissions.

Factor version updates that change emissions intensity by a few percent are generally below the restatement threshold. Large grid factor movements (some EU countries saw 15-20% changes in grid intensity over 2021-2024 due to rapid renewable deployment) may cross the threshold for energy-intensive industries. Documenting the calculation when you switch factor versions lets you assess restatement necessity quickly, rather than reconstructing prior-period computations from scratch.

Emission factor selection is a technical decision with regulatory consequences. Making it deliberately, documenting it completely, and reviewing it annually is the foundation of a defensible CSRD disclosure.