Corporate disclosure of greenhouse gas (GHG) emissions is becoming more common around the world. Companies across sectors are facing pressure from investors — and in some markets such as the EU and US, requirements from regulators — to collect and report their emissions. As an example of this trend, the percentage of companies disclosing Scope 1 and Scope 2 emissions in the S&P Global Broad Market Index (BMI)increased from 42% in 2020 to 58% in 2022, according to S&P Global Sustainable1 data. The BMI includes more than 14,000 listed companies globally.

Widespread emissions disclosure is a boon for banks, asset managers, insurers and other financial institutions looking to track their portfolios’ emissions, progress on decarbonization and alignment with the Paris Agreement on climate change. A robust database of portfolio companies’ emissions is the basis for a financial institution to calculate its own Scope 3 financed, facilitated or insurance-associated emissions. 

With the proliferation of as-reported disclosures comes a new problem, however: the potential for incompleteness or inaccuracy in the emissions data that companies provide. 

Emissions data is often more complex than it seems at first glance. Taking a nuanced view of a company’s as-reported Scope 1 emissions, or the emissions from its direct operations, we might ask: Does this data include all direct operations globally? Does it include other GHGs aside from CO2 such as methane, which accounted for 12% of US emissions in 2022? Does it align with the GHG Protocol’s standard?

An accurate accounting of a company’s emissions requires these questions — and many others — to be answered, especially for financial institutions using portfolio companies’ emissions data to inform their investment decisions, their calculations of Scope 3 emissions and their net-zero strategies. Prioritizing disclosed emissions without adjustments to ensure data quality and comparability can create misleading results. 

In an analysis of S&P Global Sustainable1 Trucost Environmental data, we found that in 2022, only 32% of companies reported their Scope 1 emissions comprehensively — meaning they needed no supplementing to reflect the company’s actual emissions. About 46% of companies reported emissions data that was partial, requiring some adjustment to have a comparable scope, while the remaining 22% of companies reported emissions data that did not reflect their total emissions because it omitted significant portions of their global operations. 

In the category of disclosures that needed at least some level of adjustment, company-reported data can also miss the mark by a large margin: about one in four company-disclosed emissions values in 2022 were at least 50% larger or smaller than their S&P Global-adjusted figures.

The quality of emissions disclosure can also vary widely among sectors and regions. Even in parts of the world where climate reporting is relatively advanced, the percentage of companies making sufficiently accurate disclosures is still low. Latin American financial sector companies, for example, had the highest rate of reporting emissions figures that did not need adjustment. But that percentage was only 65% — meaning that 35% of these companies’ figures did need adjusting to accurately represent their real emissions. In the US and Canada, at least two-thirds of companies in every sector reported insufficient emissions data. 

How do we gauge whether a company-reported emissions figure requires adjustment? 

1. As-reported data is gathered from a company’s environmental, sustainability and corporate social responsibility reports. Data is also collected from CDP, a third-party emissions database.

2. Collected data is standardized and verified for accuracy, robustness and comparability. At this stage, we might discover that a company’s as-reported data only reflects the impact of its headquarters or the operations in some locations but not others. Where necessary, we adjust the data using scaling or partial data disclosure calculations to make sure it represents the entire company’s environmental impact.

3. S&P Global’s Environmentally-Extended Input-Output model accounts for any gaps in non-disclosed environmental fields. The model uses industry-specific environmental performance data with quantitative macroeconomic data on the flow of goods and services among different sectors in the economy to estimate environmental impacts for a company’s own operations and across its entire global supply chain. This process refines the cleaned data to help generate the final adjusted values.

4. In a final step, we apply our validation checks, or “alarms,” that verify data consistency against thresholds and business rules. If discrepancies arise, an exception report is generated for analysts to review. They either correct the data or document comments to clarify any issues.

These steps ensure that GHG data meets a higher standard of accuracy and reliability.

A company collecting and reporting its own emissions is still the foundation of building a robust database of corporate climate disclosures. But emissions standards that do not prioritize quality checks and adjustments in favor of as-reported corporate data can obscure a company’s actual environmental profile. This raw, as-reported data is the beginning, rather than the end, of a process for accurately measuring a company’s direct emissions for use in net-zero planning, capital investing decisions, and long-term climate action.

Thank you to Stenin Joshi for his contribution to this blog.