ESG Rating Disagreement : Implications and Aggregation Approaches

This study explores environmental, social and governance (ESG) rating disagreement among different providers, given the variation and lack of consensus in ESG ratings. We utilise three prominent ESG ratings (Sustainalytics, MSCI and Asset4) and employ four aggregation methods (equal-weighted average, principal component analysis (PCA), voting average and optimised ESG score). Our analysis reveals relatively high correlations between Asset4 and Sustainalytics, while MSCI exhibits negative correlations with both. Additionally, we examine the correlations among individual ESG pillars and find that the aggregated ESG ratings exhibit overall high correlations. However, the voting average and PCA methods demonstrate lower correlations. Further, European countries display lower ESG rating disagreement, as evidenced by higher average aggregated ranks and smaller range differences. This study enhances our understanding of ESG rating disagreement and emphasises the importance of considering multiple ratings when assessing firms’ ESG performance