Weathering Exchange Rates: Estimating the Effect of Climate Change Vulnerability on Foreign Currency Hedging Using a Text-Based Approach
Tanakorn Likitapiwat,Sirimon Treepongkaruna,Pornsit Jiraporn
DOI: https://doi.org/10.2139/ssrn.4549570
2023-01-01
SSRN Electronic Journal
Abstract:Purpose: We investigate whether firm-specific vulnerability to climate change influences foreign exchange hedging, using a novel text-based measure of firm-level climate change exposure generated by state-of-the-art machine learning algorithms. Design/methodology/approach: Our empirical analysis includes firm-fixed-effects, random effects regressions, propensity score matching, entropy balancing, an instrumental-variable analysis, and using an exogenous shock as a quasi-natural experiment.Findings: Our findings suggest that greater climate change exposure brings about a significant reduction in exchange rate hedging. Companies more exposed to climate change may invest significant resources to address climate change risk such that they have fewer resources available for currency risk management. Additionally, firms seriously coping with climate change risk may view exchange rate risk as relatively less important in comparison to the risk posed by climate change. Notably, we also find that the negative effect of climate change exposure on currency hedging can be specifically attributed to the regulatory aspect of climate change risk, rather than the physical dimension, suggesting that companies view the regulatory dimension of climate change as more critical.Originality/value: Recent studies have demonstrated that climatic fluctuations represent one of the most recent sources of unpredictability, thereby impacting the economy and financial markets (Barnett, Brock, and Hansen, 2020; Bolton and Kacperczyk, 2020; Engle, Giglio, Kelly, Lee, and Stroebel, 2020). Our study advances this field of research by revealing that company-specific exposure to climate change serves as a significant determinant of corporate currency hedging, thus expanding the existing knowledge base.