ABSTRACT This paper examines the impact of high ambient temperatures on road accidents in Taipei City. Using microdata on all casualty‐related accidents from 2020 to 2021 matched with weather records, we estimate the relationship between daily average temperature and district‐level accident counts using Poisson Pseudo‐Maximum Likelihood (PPML) regression with high‐dimensional fixed effects. Results show that days above 27°C are significantly associated with increased accident incidence, with larger effects for accidents caused by drivers of two‐wheel vehicles than those of four‐wheel vehicles. This difference likely reflects greater adaptive capacity among four‐wheel vehicle drivers, who benefit from features such as air conditioning and insulation that mitigate heat exposure. Cause‐specific analysis indicates that heat primarily affects road safety through psychological channels, as elevated temperatures increase risk‐taking behavior. Heterogeneity analysis reveals that psychological effects are stronger during peak‐hour traffic, among middle‐aged drivers, and males, while biological effects disproportionately affect senior drivers. We also find that heat's impact attenuates over consecutive hot days, particularly for four‐wheel vehicle drivers, reinforcing evidence of short‐run adaptation to persistent heat.
Luo et al. (2026) studied this question.