Revealing stock liquidity determinants by means of explainable AI: The role of ESG before and during the COVID-19 pandemic

Abstract

The purpose of the paper is to reveal the impact of different environment, social, and corporate governance (ESG) indicators on stock liquidity in the emerging Russian market, which is heavily dominated by the natural resources sector. We first apply Explainable Artificial Intelligence (AI) to identify and rank determining factors for stock liquidity on a sample of Russian companies whose stocks are traded on the Moscow Exchange is examined over the period from 2013 to 2020. The main focus of our research is on environmental performance, including using of natural resources. We use a novel methodology based on a three-stage approach: 1) the principal components analysis is used to construct integral indices of stock liquidity, 2) neural networks with dense layers help to account for nonlinear effects, and 3) the Shapley values from the game theory help to interpret empirical results. We obtain several new conclusions on the influence of ESG and macrofactors influence. In the pre-pandemic period, environmental factors (implementation of environmental innovations, the use of «green» technologies in a company's supply chain management and the reduction of emissions) were important for stock liquidity, but their influence was negative. During the pandemic, environmental factors changed in their direction of influence. We attribute this to the fact that investors changed their preferences during the pandemic and started to view stocks of eco-friendly companies as defensive assets. The social responsibility score was not important during the COVID-19 pandemic.