Reliability of ESG ratings - A qualitative and quantitative assessment

Abstract

This thesis aims at studying the reliability of ratings through three complementary research questions. The first research question showed that, despite an overall agreement on the SRI definition, ESG rating is a relative concept specific to each agency as reflected by different methodologies and purposes. This indicates a possible complementary effect between agencies. The second research question confirmed a size and a location effect and provided indications of a learning effect. Build upon the result of the second research question, the third research question has shown that portfolios of high and low ESG ratings are neither associated with overperformance nor with underperformance. Besides, this question has shown that ESG ratings could not reproduce findings on specific pillar components (such as employees and customer satisfaction).