ENVIRONMENTAL STRESSORS AND FINANCIAL MARKET RISK: EVIDENCE FROM WILDFIRE SMOKE AND TEMPERATURE ACROSS U.S. CITIES

Abstract

Wildfire smoke increasingly contributes to poor air quality in the United States (US) and adversely affects public health, visibility, and ecosystems. Previous work has shown that poor air quality is linked to lower stock returns in financial markets. In this study, we examine how wildfire smoke, PM2.5 and temperature affect the yields and volatility of major indices on US stock exchanges. To do this, we used smoke data products from the NOAA Hazard Mapping System (HMS), PM2.5 data from the EPA Air Quality System (AQS), and financial data from leading US stock market indices. A classic Ordinary Least Square (OLS) regression was used with economic and weather indicators as controls. The analysis covered the time-period of 2010 to 2024 for 8 largest cities by population with 22 different model combinations for smoke variables over 44 different indexes. Further, we developed nonlinear machine learning models, including Random Forest Regressor (RF) and eXtreme Gradient Boosting Regressor (XGB), to understand the nonlinear relationship. We find that on high-smoke days, New York City (NYC) and some other large cities (e.g., Los Angeles) experienced higher returns, contrary to the negative or neutral returns observed in some other cities. In addition, lower volatility was observed on high-smoke days. On the other hand, temperature provided the strongest relationship of weather and air quality variables. However, the results are not consistent across all models, which shows more as a simple coincidence rather than an actual causal relationship.