Investors increasingly examine a firm’s environmental, social, and governance (ESG) activities to understand how the firm is managed using a broader lens than simply financials alone. ESG activities and issues include those related to climate change, board compensation, workforce, and tax contributions. 

A recent study, “Does ESG Score and Intangible Asset Improve Financial Performance? Machine Learning Model of KOSPI Enterprises” explored how a company’s ESG rating and other intangible assets might impact its financial market performance. The study, published in the Proceedings of the 2022 IEEE International Conference on Big Data, collected company data from the Korea Composite Stock Price Index (KOSPI). It then applied machine learning to analyze three performance indicators—return on assets (ROA), return on equity (ROE), and Tobin’s-Q—to predict a firm’s financial performance in relation to its ESG and other intangible assets.

Study Overview
Although past studies have measured the relationship between ESG disclosure and firm performance, the findings have been mixed, with some showing positive impact and others showing a negative impact. This study’s goal was to examine how ESG management scores and intangible expenditures—such as costs for advertising and research and development—might impact a firm’s financial performance. 

To achieve this, it derived a list of 684 publicly traded Korean companies on the KOSPI from 2021 to 2022 and examined their ROA, ROE, and Tobin’s-Q as proxies for firm performance and their ESG scores and intangible expenditures as independent variables. The study then used various machine-learning approaches, including linear regression, support vector machine (SVM), and artificial neural networks, to predict financial performance metrics and compare the predictions to outcomes.

Study Findings
The study’s regression analysis results confirm that ESG scores and intangible assets have a significant impact on firm performance (ROA, ROE, Tobin’s-Q). In general, both ROA and ROE metrics were better predicted, with lower mean absolute percentage error (MAPE) than Tobin’s Q. Of the various machine learning models used, both SVM and the Random Forest algorithm offered the lowest MAPE.

Overall, the study’s results suggest that companies that put greater effort into improving their ESG scores and intangible assets can positively impact their financial performance. Further, the study suggests that stakeholders, including investors, should pay more attention to ESG scores, reports, and related company activities.

Learn More
The results of this study highlight the impact of a company’s ESG activities and intangible asset expenditures. For a detailed discussion of the study, including related work, machine-learning approaches, and the study’s findings, download the full research.

Does ESG Score and Intangible Asset Improve Financial Performance?
Sign up for IEEE SustainTech Expo Alerts
Privacy Policy