Unmasking Climate Risk in Earnings calls: Traversing Storms and Fire with a Taxonomy-GPT prompting approach

Abstract

Understanding how companies manage climate change risks and opportunities is critical for investors, financial institutions, and analysts. We use corporate earnings call transcripts from European

and American corporations over the past two decades to assess how companies are affected by different climate hazards. Our approach involves several steps. First, we develop a classification system

(taxonomy) for each climate hazard by reviewing scientific reports and by identifying semantically similar words using a Word2Vec model. We then identify sentences in the transcripts that contain

these climate-related keywords. Leveraging generative AI techniques, we analyse these sentences to gain insights into how individual companies are exposed to climate change risks. We distinguish

between negative impacts (risks) and potential benefits (opportunities) for business activities. We also identify the channels through which climate risks affect different companies: 1) through direct

damage to the company’s assets and operations, and 2) through disruptions to the company’s supply chain or impacts on the company’s end markets. Our findings show that exposure to physical climate

risk varies widely across sectors, the type of events, and the channels through which firms are affected. We also find a significantly negative relationship between physical risk, as measured through earnings

call transcripts, and stock returns, supporting the view that markets incorporate these information in market prices.

Keywords: Climate change, Physical risks, Text Analysis, Pattern Matching, Conference calls, Word Embedding JEL: C58 Financial Econometrics, C63 Computational Techniques, G32 Financial Risk and Risk Management, Q54 Climate, Natural Disasters and Their Management, Global Warming