Open-access journal article · 2024
Evaluating the Anti-Corruption Factor in Environmental, Social, and Governance Indices by Sampling Large Financial Asset Management Firms
Sustainability, 16(23), Article 10240 · Published
Summary
Strang and Vajjhala (2024) tested whether ESG ratings from three providers reflect the anti-corruption record of 14 large asset management firms, using AI-assisted collection of legal, misconduct and arbitration records and Bayesian correlation with bootstrapping. For these firms, each with USD 1 trillion or more in assets under management, none of the ESG scores from MSCI, Sustainalytics or CSRhub was significantly correlated with misconduct or arbitration case counts. MSCI and Sustainalytics scores for the same firms were uncorrelated (r = −0.069), while MSCI and CSRhub scores agreed (r = +0.891).
Practical implication. For investors and regulators, the ESG ratings studied did not capture the anti-corruption record of 14 large asset management firms: no ESG score correlated significantly with misconduct or arbitration counts.
- The Problem
- Investors rely on ESG ratings to judge corporate governance, but it is untested whether those ratings reflect the anti-corruption record — misconduct and arbitration cases — of the largest asset management firms.
- The Methodology
- Quantitative archival study of 14 large asset management firms combining AI-assisted collection of legal, misconduct and arbitration records with ESG ratings from three providers, analysed with Bayesian correlation and bootstrapping.
- The Core Finding
- For 14 asset managers each holding more than USD 1 trillion, ESG scores from MSCI, Sustainalytics and CSRhub were unrelated to misconduct or arbitration case counts (all Bayes factors below 1), and the providers disagreed with each other (MSCI versus Sustainalytics r = −0.069). The ratings studied do not capture this governance evidence.
- The Citation
- Strang, K. D., & Vajjhala, N. R. (2024). Evaluating the Anti-Corruption Factor in Environmental, Social, and Governance Indices by Sampling Large Financial Asset Management Firms. Sustainability, 16(23), Article 10240. https://doi.org/10.3390/su162310240
The summary, key findings, methods, implications and relevance notes on this page are this website’s own description of the paper, written from the published abstract and text. The official abstract and citation details are given further down.
Key findings
- Finding 1. The study sampled 14 asset management firms with USD 1 trillion to USD 10.8 trillion in assets under management and used AI to collect undisclosed legal decisions as a measure of anti-corruption governance (GRI 206-1).
- Finding 2. None of the ESG scores from MSCI, Sustainalytics, or CSRhub was significantly correlated with the firms’ misconduct or arbitration legal case counts; the Bayes factors for these pairs were all below 1.
- Finding 3. ESG rating providers disagreed: MSCI and Sustainalytics scores for the same firms were uncorrelated (r = −0.069), while MSCI and CSRhub scores agreed (r = +0.891, p < 0.001).
- Finding 4. Misconduct and arbitration case counts were strongly correlated for the same firms (r = +0.897, p < 0.001; Vovk-Sellke maximum p-ratio 4411; bootstrapped Bayesian BF+0 = 3005, 99% CI 0.617–0.965).
- Finding 5. Three corroborating studies identified specific firms in the sample as unethical.
- Finding 6. For finance and insurance firms, governance issues such as money laundering are more material than the CO₂ emissions reporting that regulators emphasize.
Source: Strang & Vajjhala (2024), Sustainability, 16(23), Article 10240. DOI: 10.3390/su162310240
The study in detail
What question does this paper answer?
Do the ESG ratings of the world’s largest asset management firms reflect their anti-corruption record — the misconduct and arbitration legal cases they face — and do ESG rating providers agree with each other?
What did the study find, in detail?
No. For 14 asset managers each holding more than USD 1 trillion, none of the ESG scores from MSCI, Sustainalytics, or CSRhub was significantly correlated with the firms’ misconduct or arbitration case counts, and the providers disagreed with each other (MSCI vs Sustainalytics r = −0.069). The two legal measures, by contrast, were strongly related (r = +0.897, p < 0.001; Bayesian BF+0 = 3005).
Why does it matter?
Regulators ask financial firms to report CO₂ emissions, yet for firms that mainly rent offices, governance issues such as money laundering and corruption are far more material. The study shows that ESG indices can miss the governance risks that matter most in finance, which matters for investors, regulators, and anyone relying on ESG scores as a proxy for ethical conduct.
This study is part of the evidence overview Can ESG compliance be measured reliably? Evidence from ESG ratings, project-level instruments, machine learning and supply chain records
Methods and data
| Research question | Do ESG ratings capture the anti-corruption governance factor (GRI 206-1) for large financial firms? |
|---|---|
| Sample | 14 asset management firms with USD 1–10.8 trillion in assets under management |
| Data | AI-collected, previously undisclosed legal decisions; misconduct fines; arbitration cases; ESG ratings |
| Methods | Bayesian correlation with bootstrapping |
| Main result | No ESG score correlated significantly with misconduct or arbitration counts; misconduct and arbitration correlated at r = +0.897 (BF+0 = 3005, 99% CI 0.617–0.965; VS-MPR 4411); MSCI and Sustainalytics scores diverged (r = −0.069) |
| Implication | ESG indices under-weight material governance risk in financial services |
| Citation | Strang & Vajjhala (2024) · DOI 10.3390/su162310240 |
Definitions and key terms
- GRI 206-1
- The Global Reporting Initiative disclosure on legal actions for anti-competitive behavior, anti-trust, and monopoly practices — used here as an anti-corruption governance indicator. — as used in Strang & Vajjhala (2024)
- ESG rating divergence
- When different rating agencies give the same company substantially different ESG scores. — as used in Strang & Vajjhala (2024)
- Bayes factor (BF)
- The ratio of evidence for one hypothesis over another; values above 100 are conventionally treated as extreme evidence. — as used in Strang & Vajjhala (2024)
Abstract
Current ESG indices suffer from incomplete and inconsistent data, with some factors irrelevant to specific industries. Regulators emphasize CO2 emissions reporting for finance/insurance firms, yet rented offices and governance issues like money laundering prove more material. This study examined USD 1+ trillion asset management firms using AI to collect undisclosed legal decisions measuring anti-corruption governance (GRI 206-1). Bayesian correlation with bootstrapping revealed ESG ratings failed reflecting legal cases (BF+0 odds ratio: 3005, 99% CI: 0.617–0.965). Misconduct fines correlated significantly with arbitration cases (Vovk-Selke p-ratio: 4411), yet most ESG scores diverged for identical firms. Three corroborating studies identified specific sample firms as unethical. The authors recommend deeper investigation of implications regarding public interest and stakeholder theory.
Abstract as published in Sustainability under a Creative Commons licence.
Limitations and open questions
- The sample is small (14 firms), and ESG scores were missing for four firms from two providers.
- The authors call for deeper investigation of the implications for public interest and stakeholder theory.
Open questions
- Would ESG ratings reflect anti-corruption records in larger, non-US and non-financial samples, and why are misconduct and arbitration counts concentrated in a few firms?
Open questions from the evidence overview ESG and sustainability measurement.
Practical implications
- For investors, regulators and anyone relying on ESG scores as a proxy for ethical conduct, the study shows that ESG indices can miss the governance risks that matter most in finance.
- Investors who rely on ESG ratings to judge corporate governance face disagreement between providers: for the same firms, MSCI and Sustainalytics scores were uncorrelated (r = −0.069), while MSCI and CSRhub scores agreed (r = +0.891).
- Regulators that emphasize CO₂ emissions reporting for finance and insurance firms are emphasizing an issue less material to those firms than governance issues such as money laundering.
- Researchers examining ESG rating divergence and validity should read the results as those of a small-sample study of very large firms: 14 firms, with ESG scores missing for four firms from two providers.
Drawn from Strang & Vajjhala (2024), Sustainability, 16(23), Article 10240.
When this research may be relevant
This article may be relevant to researchers examining ESG rating divergence and validity, the governance and anti-corruption pillar of ESG (GRI 206-1), misconduct in financial services and asset management, or Bayesian correlation methods with small samples. It is a small-sample study of very large firms and should be read as such.
Research topics addressed: ESG ratings; ESG rating divergence; anti-corruption; corporate governance; GRI 206-1; asset management firms; financial services; money laundering; Bayesian correlation; AI data collection; stakeholder theory
How to cite
- APA 7
- Strang, K. D., & Vajjhala, N. R. (2024). Evaluating the Anti-Corruption Factor in Environmental, Social, and Governance Indices by Sampling Large Financial Asset Management Firms. Sustainability, 16(23), Article 10240. https://doi.org/10.3390/su162310240
- MLA 9
- Strang, Kenneth David, and Narasimha Rao Vajjhala. “Evaluating the Anti-Corruption Factor in Environmental, Social, and Governance Indices by Sampling Large Financial Asset Management Firms.” Sustainability, vol. 16, no. 23, 2024, article 10240. https://doi.org/10.3390/su162310240.
- IEEE
- K. D. Strang and N. R. Vajjhala, “Evaluating the Anti-Corruption Factor in Environmental, Social, and Governance Indices by Sampling Large Financial Asset Management Firms,” Sustainability, vol. 16, no. 23, Art. no. 10240, Nov. 2024, doi: 10.3390/su162310240.
BibTeX
@article{strang2024esg,
title = {Evaluating the Anti-Corruption Factor in Environmental, Social, and Governance Indices by Sampling Large Financial Asset Management Firms},
author = {Strang, Kenneth David and Vajjhala, Narasimha Rao},
journal = {Sustainability},
volume = {16},
number = {23},
pages = {10240},
year = {2024},
publisher = {MDPI},
doi = {10.3390/su162310240},
url = {https://doi.org/10.3390/su162310240}
}RIS
TY - JOUR TI - Evaluating the Anti-Corruption Factor in Environmental, Social, and Governance Indices by Sampling Large Financial Asset Management Firms AU - Strang, Kenneth David AU - Vajjhala, Narasimha Rao JO - Sustainability PY - 2024 DA - 2024/11/22 VL - 16 IS - 23 SP - 10240 PB - MDPI SN - 2071-1050 DO - 10.3390/su162310240 UR - https://www.narasimharao.net/research/esg-ratings-anti-corruption-asset-management-firms/ LA - en ER -
Links and files
- DOI: https://doi.org/10.3390/su162310240
- Publisher’s page: https://www.mdpi.com/2071-1050/16/23/10240
- Full text (PDF): https://www.mdpi.com/2071-1050/16/23/10240/pdf
- This page as text: https://www.narasimharao.net/research/esg-ratings-anti-corruption-asset-management-firms/paper.md
- Permanent address of this page: https://www.narasimharao.net/research/esg-ratings-anti-corruption-asset-management-firms/
Related work
Related work: the literature this study cites
The literature cited by this study documents persistent problems with firm-level ESG ratings: Berg and colleagues attributed 56% of rating divergence across six agencies to measurement differences, 38% to scope and 6% to weights[1], Chatterji and colleagues linked divergent interpretations to a lack of standardized methodologies[2], and Hess reported that many firms disclose anti-corruption policies without detail on implementation[3]. Connecting SDGs to firm-level ESG factors has been argued to require a double materiality perspective[4]. The rating divergence found among asset managers is consistent with the divergence documented by Berg and colleagues and Chatterji and colleagues[1][2]. To explain the extreme case counts, the study draws on Egan, Matvos and Seru, who reported that some firms have far more adviser misconduct relative to their size[5], and on Honigsberg and Jacob, who found that brokers with prior expungements were 3.3 times as likely to engage in new misconduct[6].
- [1] Berg, F.; Kolbel, J.F.; Rigobon, R. Aggregate Confusion: The Divergence of ESG Ratings. Rev. Financ. 2022, 26, 1315–1344.
- [2] Chatterji, A.K.; Durand, R.; Levine, D.I.; Touboul, S. Do ratings of firms converge? Implications for managers, investors and strategy researchers. Strat. Manag. J. 2016, 37, 1597–1614.
- [3] Hess, D. The Transparency Trap: Non-Financial Disclosure and the Responsibility of Business to Respect Human Rights. Am. Bus. Law J. 2019, 56, 5–53.
- [4] Delgado-Ceballos, J.; Ortiz-De-Mandojana, N.; Antolin-Lopez, R.; Montiel, I. Connecting the Sustainable Development Goals to firm-level sustainability and ESG factors: The need for double materiality. BRQ Bus. Res. Q. 2023, 26, 2–10.
- [5] Egan, M.; Matvos, G.; Seru, A. The Market for Financial Adviser Misconduct. J. Polit. Econ. 2019, 127, 233–295.
- [6] Honigsberg, C.; Jacob, M. Deleting misconduct: The expungement of BrokerCheck records. J. Financ. Econ. 2021, 139, 800–831.
From the evidence overview Can ESG compliance be measured reliably? Evidence from ESG ratings, project-level instruments, machine learning and supply chain records, which cites only works that appear in the studies’ own reference lists. This is a selection, not the full reference list of Strang & Vajjhala (2024).
Related research
Related publications on ESG and sustainability measurement
In 207 financial software project records, stakeholder-rated social and governance factors tracked the overall project score almost perfectly (r = +0.995 and +0.966) while the environmental factor was unrelated to it, and machine learning classifiers performed weakly (kNN AUC = 0.497; SVM accuracy 61.8%).
A survey of 2,231 North American manufacturing project sponsors produced a validated seven-item, two-factor instrument — ESG planning and ESG monitoring and controlling — with excellent fit (CFI = 0.99, TLI = 0.98, RMSEA = 0.052, SRMR < 0.035).
Perceived regulatory-volatility risks do not form a single construct (KMO = 0.470), and labor and environmental risks were rated lowest of all (medians 0 and 1 on 0–5 scales) during a period of record forced-labor enforcement.