The impact of air pollution on consumer spending patterns: evidence from China

While the adverse effects of air pollution are well documented, its impact on high-frequency, essential consumption remains underexplored. Our study aims to empirically examine how air pollution affects daily dining expenditure and choices, thereby quantifying a key, yet often overlooked, economic cost of environmental degradation. We utilize a high-frequency dataset of over 1.66 million dining transactions from a large university student population in Beijing, China. We employ a fixed-effects panel regression model to analyse the intensive margin (expenditure) and a binary logit model to analyse the extensive margin (i.e. the choice to switch to food delivery). Our results reveal a significant negative association between air pollution and in-person dining expenditure, with a 1% increase in the PM2.5 concentration corresponding to a 0.016% decrease in spending. We find that pollution significantly increases the likelihood of students opting for food delivery. The analysis reveals a complex, nonlinear relationship and significant heterogeneity. The negative impact is more pronounced for female students and graduate students and is substantially exacerbated during the high-pollution heating season. Our study highlights the hidden economic costs of pollution beyond direct health impacts and the need for integrated environmental and economic policies that account for these nuanced behavioural shifts.