Fundamental ratios as predictors of ESG scores: a machine learning approach

Valeria D’Amato, Rita D’Ecclesia and Susanna Levantesi ()
Additional contact information
Valeria D’Amato: University of Salerno
Rita D’Ecclesia: Sapienza University of Rome
Susanna Levantesi: Sapienza University of Rome

Decisions in Economics and Finance, 2021, vol. 44, issue 2, No 27, 1087-1110

Abstract: Abstract Sustainable and responsible finance incorporates Environmental, Social, and Governance (ESG) principles into business decisions and investment strategies. In recent years, investors have rushed to Sustainable and Responsible Investments in response to growing concerns about the risks of climate change. Asset managers look for some assessment of sustainability for guidance and benchmarking, for instance, $30 trillion of assets are invested using some ESG ratings. Several studies argue that good ESG ratings helped to prop up stock returns during the 2008 Global Financial Crisis (Lins et al. J Finance 72(4):1785–1824, 2017). The ESG score represents a benchmark of disclosures on public and private firms, it is based on different characteristics which are not directly related to the financial performance (Harvard Law School Forum on Corporate Governance, ESG reports and ratings:what they are, why they matter. https://corpgov.law.harvard.edu/2017/07/27/esg-reports-and-ratings-what-they-are-why-they-matter/ , 2017). The role of ESG ratings and their reliability have been widely discussed (Berg et al. Aggregate confusion: the divergence of ESG ratings, MIT Sloan Research Paper No. 5822-19, 2019). Sustainable investment professionals are unsatisfied with publicly traded companies’ climate-related disclosure. This negative sentiment is particularly strong in the USA, and within asset managers who do not believe that markets are consistently and correctly pricing climate risks into company and sector valuations. We believe that ESG ratings, when available, still affect business and finance strategies and may represent a crucial element in the company’s fundraising process and on shares returns. We aim to assess how structural data as balance sheet items and income statements items for traded companies affect ESG scores. Using the Bloomberg ESG scores, we investigate the role of structural variables adopting a machine learning approach, in particular, the Random Forest algorithm. We use balance sheet data for a sample of the constituents of the Euro Stoxx 600 index, referred to the last decade, and investigate how these explain the ESG Bloomberg ratings. We find that financial statements items represent a powerful tool to explain the ESG score.

Keywords: Machine learning; ESG investments; Firm performance (search for similar items in EconPapers)
JEL-codes: D8 L25 M14 (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (5)

Downloads: (external link)
http://link.springer.com/10.1007/s10203-021-00364-5 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:decfin:v:44:y:2021:i:2:d:10.1007_s10203-021-00364-5

Ordering information: This journal article can be ordered from
http://www.springer. ... ry/journal/10203/PS2

DOI: 10.1007/s10203-021-00364-5

Access Statistics for this article

Decisions in Economics and Finance is currently edited by Paolo Ghirardato

More articles in Decisions in Economics and Finance from Springer, Associazione per la Matematica
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().