Many investors and lenders assume that the emissions reported by companies are directly measured using real-time sensors and constant monitoring. However, the reality can be quite different.

Despite technological advancements, even the most advanced reporting systems often rely on estimates. And legally, they’re allowed to. This article explores the history of reported emissions and what this means for financial institutions relying on these figures for everything from regulatory reporting to risk management and portfolio allocation.

Why does modelling still dominate emissions reporting?

Under emissions trading systems worldwide, for example the EU Emissions Trading System (EU ETS), Europe’s flagship climate policy, companies subject to it are explicitly allowed to choose between two main methodologies for reporting their emissions:

• A measurement-based methodology requiring continuous emissions monitoring systems (CEMS) to track greenhouse gas concentrations and flows;

• A calculation-based methodology that estimates greenhouse gas emissions based on activity data (e.g., fuel consumption) and standardized emissions factors drawn from the internationally recognized GHG Protocol.  

“The operator shall choose to apply either a calculation-based methodology or a measurement-based methodology, subject to specific provisions…” – EU Regulation 2018/2066, Article 21

This provision, hidden in plain sight, has significant implications.  

Even under one of the world’s most stringent emissions trading schemes, companies are not required to measure emissions directly. Instead, they can legally rely on modelled estimates derived from standard protocols, activity metrics, and default emissions factors when continuous measurement is not economically or technically feasible.  

Under the EU regulation (2018/2066, Art. 21 & Annex II-IV), operators must submit a Monitoring Plan justifying their chosen methodology, but they are not required to prove that continuous measurement is infeasible in all cases, especially for smaller sources where estimation via standard emission factors is considered proportionate and acceptable.

Implications for emissions data reliability

Examining the “letter of the law” in emissions disclosure regulations helps to expose a common misconception: companies are measuring their emissions rather than relying on modelled estimates.  

In practice, few companies have emissions monitoring systems installed across all their assets. Instead, they rely on a combination of fuel input and operational throughput data, combined with standardized emissions factors from external sources. Adding to the challenge, companies rarely disclose which specific emissions factors they use, and with multiple valid options available (e.g. IPCC defaults, national inventories, supplier-specific values), this lack of transparency makes it difficult to ensure consistency or comparability across disclosures.

In other words, corporate disclosures are often built on the same foundational logic as the modelled emissions data produced by external analytics providers: companies frequently model their company-level emissions based on the output of their assets rather than measuring them directly. This insight, which remains largely unarticulated in debates that privilege corporate disclosures over external sources, should reshape how financial institutions understand the quality of disclosed emissions data, especially when comparing company reports with estimates made by ESG modelling or financial climate risk assessment tools.

Case in point

Even companies recognized for environmental transparency explicitly acknowledge the hybrid nature of emissions reporting in their official documentation. For example, according to their 2023 Carbon Footprint Report, leading European utility, EDF, determines emissions from thermal power plants “either through direct measurements or based on analyses of fuels or standard emission factors.” This blended methodology, common even within critical sectors like power generation, highlights that many company reported emissions are modelled.  

This observation challenges the common assumption that corporate disclosures represent a definitive “ground truth”. Instead, external platforms employing a consistent methodology across multiple companies offer a more robust comparative framework than corporate self-reporting, which often involves varying assumptions, regional differences, and accounting scopes.

To illustrate this clearly, we compared Asset Impact’s asset-based company indicators (Q1 2023) with EDF’s own asset-level disclosures for their power sector activities in France and China. We standardized EDF’s data by applying our Equity Ownership and Financial Control consolidation methods, which follow GHG Protocol recommendations.  

The analysis reveals that Asset Impact’s asset-based data provides slightly broader coverage — 3% under Equity Ownership and 5% under Financial Control — compared to EDF’s own asset-level disclosures. The primary driver of this difference is Asset Impact’s inclusion of recently operational renewable energy projects in France.

Our approach provides extra information not available from EDF’s public reports, for example by:  

• providing plant-level granularity, while the company reports only group-wide data;

• using transparent assumptions and system boundaries;

• reporting emissions per ton of steel, and not including other products (e.g. scrap, alloys) in its denominator.

Why this matters  

For financial institutions, including banks, asset managers, and institutional investors, emissions data is no longer a box-ticking exercise. It informs credit decisions, risk models, engagement strategies, and regulatory disclosures. That’s why the quality, comparability, and consistency of emissions data is critical.

Asset Impact’s methodology offers a robust alternative to relying solely on corporate-reported figures, which are often modelled using inconsistent assumptions, aggregated across scopes, or based on undisclosed emissions factors.

Our approach provides:

• Consistent methodologies applied globally across all assets and companies

• Transparent assumptions, with emissions factors and calculation logic clearly documented

• Asset-level granularity, enabling precise exposure analysis without scope blending or geographic masking

This matters because what drives financial exposure isn’t just what companies emit, but how clearly and comparably those emissions can be understood across a portfolio.

In a world moving toward climate accountability and regulatory scrutiny, transparent and consistent modelled data can be just as critical and often more useful than fragmented corporate disclosures. For financial institutions managing systemic climate risk, it’s not about choosing between reported or modelled data. It’s about choosing what delivers the most clear, comparable, and actionable insights

‍