The term "ESG," which is short for "environmental, social, and governance," started appearing in investment fund names circa 2010. By 2019, there were hundreds of “ESG funds.” Still, what exactly these funds encompass is unclear. Despite efforts to define and clarify what “ESG funds” are, the term’s ambiguity persists.

“How to Build a Better ESG Fund Classification System” aims to clarify the meaning of “ESG funds” with respect to fund classification, which involves grouping funds based on defined boundaries. The report focuses on funds that consider ESG information to any extent for any purpose. Unlike previous efforts that offer conceptual frameworks or brief definitions, this report addresses practical challenges in evaluating funds against categorical criteria—providing guidelines, examples, and case studies to help overcome definitional challenges.

Our report emphasizes defining groups and boundaries rather than debating terminology. As such, we use generic terms like "Feature 1" and "Group A" to avoid getting caught up in specific labels. The report separates the decision making about group boundaries from the names of the groups, focusing on the former.

The report reviews various ESG fund classification frameworks created by asset managers, industry associations, and regulators, including those from the European Union, United States, and United Kingdom. It concludes that these frameworks are largely inadequate for practical categorization because they lack observable features, rigorous definitions, or necessary logic for sorting funds into useful, mutually exclusive groups.

To address these deficiencies, our report defines three observable features of funds that can be the foundation of a robust ESG fund classification system:

Feature 1: The existence of one or more processes that consider ESG information with the aim of improving risk-adjusted returns.
Feature 2: The existence of one or more policies that control fund investors' exposure and contribution to specific systemic ESG issues.
Feature 3: The existence of an explicit statement of intent, and an action plan, to help bring about a target future state in environmental and/or social conditions and a process to measure progress.

Using these features, the report proposes creating mutually exclusive groups. One group includes funds with only Feature 1, using ESG information to enhance decision making for higher risk-adjusted returns. These funds are popular in some regions but not in others. Another group includes funds with Feature 3, which aim to achieve specific environmental or social outcomes and have plans and processes to measure progress. These funds often also have Features 1 and 2 but are less prevalent.

The remaining group, funds with Feature 2, includes funds with policies that control investors' exposure to, and simultaneously their contribution to, systemic ESG issues. These issues affect the broader economy and multiple companies or assets; they are not idiosyncratic. This group of funds is heterogeneous, but within-group variation is traceable to fund or firm policies. Our report notes that this final group might need further subdivision and suggests it as a future task.

Fund classification is complex and challenging. Simplifying it has not led to a practical system. The report’s technical detail is necessary for designing effective ESG fund classification systems, though it may not appeal to all readers.

We see this report as a starting point for improving ESG fund classification. We invite feedback and testing from organizations, researchers, and fund managers. The paper suggests adding granularity to partition funds that have Feature 2 into subgroups for better distinction and proposes extending the classification system to funds-of-funds and other investment products. We also explore the potential for machine learning to enhance classification efficiency.

We hope this paper will inspire further ideas and collaboration, advance knowledge and practice in this area, and lead to more effective and efficient classification systems in the investment industry.