This paper aims to detect the use of label adoption practices by early adopters of double materiality reporting.
A gradient boosting machine was used to analyse variables related to financial performance, corporate governance, sustainability reports and control variables of 162 European companies listed on the Dow Jones Sustainability Index (DJSI), which claim to apply double materiality reporting.
The model achieved an overall accuracy of 81.8%, correctly detecting label adopters at 78.6% and committed adopters at 84.2%. These results demonstrate the high accuracy of the model in distinguishing between label and committed adopters, reducing the risk of type I and II errors.
This study is limited to large DJSI companies in Europe that explicitly state the use of double materiality in their sustainability reports. The results may differ if companies of various sizes or in different regions were analysed.
The results highlight challenges for regulators in identifying companies falsely claiming compliance with double materiality reporting in their sustainability reports. Furthermore, this research has implications for assurers, who are responsible for endorsing the provision of sustainability information and may be assisted in this task by the use of detecting tools. Additionally, capital providers can use these tools to ensure companies genuinely implement sustainability practices, not just adopt labels.
This research integrates empirical evidence with the use of machine learning techniques to offer fresh insights into the phenomenon of greenwashing, focusing specifically on the early adoption of the double materiality principle.
