The Hidden Carbon Cost of Extreme Low Temperatures: Fiscal Support for Agriculture and Agricultural Carbon Emissions in China

Extreme low-temperature events are among the meteorological hazards that affect agricultural production in China, yet their association with agricultural carbon emissions remains insufficiently understood. Using a balanced panel of 31 Chinese provinces from 2010 to 2023, this study employs a two-way fixed-effects model to examine the relationship between extreme low-temperature exposure, measured by the annual number of extreme low-temperature days (LTD), and crop-production-related carbon emissions. The results show that one additional LTD is associated with an increase of 0.298 ten thousand tons of carbon emissions. A one-standard-deviation increase in LTD (17.8 days) corresponds to approximately 5.3 ten thousand tons, equivalent to about 2% of the sample mean. The positive association is statistically significant in major grain-consuming regions, but not in major grain-producing or production–marketing balance regions. Pairwise Wald tests provide limited evidence of differences across groups, with only the difference between marketing and production–marketing balance regions being marginally significant at the 10% level. The interaction between LTD and fiscal support for agriculture is negative, whereas that between LTD and agricultural labor is positive, suggesting that these resources may be associated with different patterns of emission responses to extreme cold. An accounting-based decomposition indicates that the aggregate association is concentrated in fertilizer, pesticide, and diesel-related emissions, while the estimated responses of plastic film, sown area, and irrigated area are statistically insignificant. These component estimates are accounting-based evidence rather than causal mediation effects. Although instrumental-variable estimates provide additional evidence, potential endogeneity of fiscal support, the use of annual temperature indicators that do not distinguish crop growth stages, and the limitations of province-level aggregate data constrain causal interpretation. The findings highlight the potential carbon implications of agricultural adjustment to extreme cold and the importance of coordinating climate adaptation and agricultural emission-reduction policies.