Firms’ Profitability and ESG Score: A Machine Learning Approach

Can AI inform us of the links between ESG and profitability?

Valeria D'Amato, Rita D'Ecclesia and Susanna Levantesi explore this question in their « Firms' profitability and ESG score: A machine learning approach » paper.

They use Refinitiv ESG scores to show that:

• ESG scores can have a significant impact on a firm's profitability, measured by the Earnings Before Interest and Taxes (EBIT) of the company.
• To have a substantial impact on EBIT, a company needs to be highly active towards sustainability and invest to change its business model to comply with ESG targets.
• This translates into higher ESG scores, usually higher than 60 according to Refinitiv ESG score.
• Companies with low ESG scores, which are considered less committed towards the sustainability goal, do not create an extra profit margin.

Even though this article suggests that companies with high ESG scores are more likely to have higher profitability, practitioners may argue that the study does not provide specific strategies for companies to improve their ESG scores and thereby increase their return.

Moreover, Refinitiv ESG data is notoriously biased for measurements of pricing effects, as demonstrated by Florian Berg and his co-authors in their « Is History Repeating Itself? The (Un)Predictable Past of ESG Ratings » article.