Explore the scholarly record.

Abstract

Extreme cold exposure may constrain the diffusion of low-carbon transport technologies by reducing vehicle usability and increasing consumers’ perceived mobility risk. This study examines how extreme cold affects electric vehicle (EV) sales in Chinese cities from 2017 to 2023. Using daily meteorological observations, we construct a standardized city-level extreme cold index based on historical temperature thresholds and match it with EV sales, charging infrastructure, socioeconomic characteristics, policy indicators, and online consumer reviews. A two-way fixed effects model estimates the average effect, while an instrumental variable based on the interaction between the initial atmospheric water vapor state and city latitude addresses potential endogeneity. The results show that a one-unit increase in the extreme cold index is associated with an approximately 0.76% decline in EV sales. This finding remains robust across alternative measures, specifications, and identification tests. Mechanism analysis based on LLM-assisted processing of online reviews indicates that extreme cold increases consumers’ range anxiety, economic concerns, and resale-value concerns. The adverse effect is concentrated among battery electric vehicles and is stronger in cities with weaker charging infrastructure and lower power-supply resilience, while supportive policy environments partly mitigate the effect. Brand-level results further reveal substantial differences in sensitivity to extreme cold.The effect is also weaker under drier extreme cold conditions. These findings suggest that EV policies should place greater emphasis on low-temperature technology adaptation, charging reliability under extreme cold, transparent low-temperature performance information, and geographically differentiated transport policy design.