Four Things No One Will Tell You About ESG Data
Четыре вещи, о которых вам никто не расскажет о данных ESG
2019-06-01
SCID: 54.1/xrr9j673
Discuss with AI
ESG benchmarkingESG dataESG data imputationESG data providersESG measurement
Figures from the paper
Abstract (AI)
As the ESG finance field and the use of ESG data in investment decision‐making continue to grow, the authors seek to shed light on several important aspects of ESG measurement and data. This article is intended to provide a useful guide for the rapidly rising number of people entering the field. The authors focus on the following: The sheer variety, and inconsistency, of the data and measures, and of how companies report them. Listing more than 20 different ways companies report their employee health and safety data, the authors show how such inconsistencies lead to significantly different results when looking at the same group of companies. ‘Benchmarking,’ or how data providers define companies' peer groups, can be crucial in determining the performance ranking of a company. The lack of transparency among data providers about peer group components and observed ranges for ESG metrics creates market‐wide inconsistencies and undermines their reliability. The differences in the imputation methods used by ESG researchers and analysts to deal with vast ‘data gaps’ that span ranges of companies and time periods for different ESG metrics can cause large ‘disagreements’ among the providers, with different gap‐filling approaches leading to big discrepancies. The disagreements among ESG data providers are not only large, but actually increase with the quantity of publicly available information. Citing a recent study showing that companies that provide more ESG disclosure tend to have more variation in their ESG ratings, the authors interpret this finding as clear evidence of the need for ‘a clearer understanding of what different ESG metrics might tell us and how they might best be institutionalized for assessing corporate performance.’ What can be done to address these problems with ESG data? Companies should ‘take control of the ESG data narrative’ by proactively shaping disclosure instead of being overwhelmed by survey requests. To that end, companies should ‘customize’ their metrics to some extent, while at the same time seeking to self‐regulate by reaching agreement with industry peers on a ‘reasonable baseline’ of standardized ESG metrics designed to achieve comparability. Investors are urged to push for more meaningful ESG disclosure by narrowing the demand for ESG data into somewhat more standardized, but still manageable metrics. Stock exchanges should consider issuing—and perhaps even mandating—guidelines for ESG disclosures designed in collaboration with companies, investors, and regulators. And data providers should come to agreement on best practices and become as transparent as possible about their methodologies and the reliability of their data.
Key Findings
1
Data providers’ peer-group benchmarking choices strongly affect company rankings, while opaque peer-group compositions and metric ranges undermine reliability and consistency.
2
Different reporting formats for the same companies can produce significantly different ESG performance results.
3
ESG data and measures exhibit substantial variety and inconsistency, including more than 20 reporting formats for employee health and safety information.
4
ESG rating variation can increase with the amount of corporate ESG disclosure, highlighting the need for clearer interpretation and institutionalization of ESG metrics.
5
Provider disagreements increase when analysts use different imputation methods to fill extensive ESG data gaps across companies and time periods.
Research Object
ESG data and measurement practices used in investment decision-making, including companies’ ESG disclosures and provider ratings
Research Subject
Inconsistency, transparency, benchmarking, imputation, and resulting disagreement and reliability problems in ESG metrics and ratings
Publication Details
Publication Date
2019-06-01
Journal
Publisher
ISSN
Cited by
773
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
Cited by3
Environmental, social, and governance (ESG) and artificial intelligence in finance: State-of-the-art and research takeaways2024
Corporate Environmental, Social, and Governance Performance: The Impacts on Financial Returns, Business Model Innovation, and Social Transformation2025
Divergence of ESG Ratings: Foreign Regulatory Trends2022