The Role of Air Pollution in the ESG Model at the World Level

Abstract

In the following chapter, the authors analyse the role of air pollution (AP) in the context of ESG (environmental, social, and governance) model. They use data from the World Banks' ESG Database for 193 countries during the period 2011-2020. They perform panel data with random effects, panel data with fixed effects, pooled OLS, and WLS. Results show that the level of PA is positively associated, among others, to “cooling degree days,” “CO2 emissions,” and “agriculture, forestry, and fishing, value added,” and negatively associated, among others, to “terrestrial and marine protected areas,” “proportion of seats held by women in national parliaments,” and “mammal species threatened.” Furthermore, they confront eight different machine learning algorithms for the prediction of the future value of AP. Polynomial regression is the best predictive algorithm in the sense either maximization of R-squared either minimization of MAE, MSR, and RMSE. The future value of AP is expected to reduce on average of -0,060% for the analysed countries.