Automated Identification of Climate Risk Disclosures in Annual Corporate\n Reports

Abstract

It is important for policymakers to understand which financial policies are\neffective in increasing climate risk disclosure in corporate reporting. We use\nmachine learning to automatically identify disclosures of five different types\nof climate-related risks. For this purpose, we have created a dataset of over\n120 manually-annotated annual reports by European firms. Applying our approach\nto reporting of 337 firms over the last 20 years, we find that risk disclosure\nis increasing. Disclosure of transition risks grows more dynamically than\nphysical risks, and there are marked differences across industries.\nCountry-specific dynamics indicate that regulatory environments potentially\nhave an important role to play for increasing disclosure.\n