High Temperatures and Rural Household Expenditure: Evidence From Longitudinal Data in Colombia

ABSTRACT

This paper examines the impact of rising temperature on rural household expenditure in Colombia by combining high‐resolution gridded weather data with a rich longitudinal household survey. Our results show that a one‐standard‐deviation increase in average daily temperature during the rainy months significantly reduces rural household expenditure by approximately 13%. Using highly granular satellite imagery, farmers' self‐reported shock data, and local crop yields, we establish that crop losses mediate the effect of temperature on expenditure. Our analysis indicates that although rural households respond to these shocks by engaging in non‐agricultural work and selling livestock, these responses are insufficient to offset temperature‐induced income losses fully. We show that this is attributable to households' limited access to alternative risk‐coping mechanisms. Furthermore, we find that wealth and rainfall can play a crucial role in mitigating the adverse effects of temperature rise. Our estimates are robust across various specifications, alternative definitions of temperature change, and alternative weather datasets.