Conference Agenda

Using the meteorological data from the European Centre for Medium-Range Weather Forecasts (ECMWF), and the workplace accident records from China Stock Market and Accounting Research (CSMAR) database, we investigate the impact of extreme temperatures on workplace accidents via fixed effect model, Poisson Pseudo Maximum Likelihood (PPML) estimation and Causal Forest-based DML. By referencing historical temperature distributions and constructing biometeorological indicators such as Heat Index (HI) and Wind Chill Index (WCI), we capture the extreme temperature events at both daily and hourly levels. Our findings indicate that both extreme heat and extreme cold increase the occurrence of workplace accidents, and the impacts of extreme temperature shocks on workers exhibit different patterns in the short term and long term. The potential mechanism through which extreme temperatures affect workplace accidents is by undermining workers’ positive mindset and reducing their work quality. Additionally, by employing a Causal Forest-based DML model, we investigate the non-linear heterogeneous variations in extreme temperature effects under different levels of air pollution, meteorological conditions, and urban development. Finally, we estimate the additional workplace injury and wage premium caused by extreme temperature events, and confirm the necessity of heat allowances. The findings of this study also provide some enlightenment into adapting to extreme temperatures from perspectives such as early warning systems, infrastructure development, and individual-level behavior.