Factors affecting CO2 emissions in China's agriculture sector: A quantile regression

Rapid development of agriculture mechanization and the agro–industry in China (considered as a large agricultural country) has led to a substantial increase in energy consumption and CO2 emissions. Majority of existing studies usually explore the driving forces of the sector's CO2 emissions using the averaging method. However, data distribution of economic variables is often non–normal, with the tail having hidden important information. According to the average annual CO2 emissions, this paper divides China's 30 provinces into six quantile grades, and uses the quantile regression method to investigate the driving forces of CO2 emissions under high, medium, and low emission levels. The results show that the effects of economic growth on CO2 emissions in the upper 90th and 75th–90th quantile provinces are higher than in the 50th–75th, 25th–50th, 10th–25th and lower 10th quantile provinces due to the differences in fixed–asset investment and agricultural processing. The impact of energy efficiency in the upper 90th, 75th–90th, and 50th–75th quantile provinces are stronger than those in the 25th–50th, 10th–25th, and lower 10th quantile provinces because of the huge difference in R&D funding and R&D personnel investments. The effect of urbanization in the higher 90th quantile provinces is higher than in the other quantile provinces, owing to the differences in the level of agricultural mechanization and human capital accumulation. Similarly, financial capacity has the largest impact on CO2 emissions in the upper 90th quantile provinces in all quantile provinces. However, the impact of industrialization in the upper 90th quantile provinces is lower than in other quantile provinces. Thus, the heterogeneous effects of the driving forces should be taken into consideration when discussing CO2 emissions reduction in China's agriculture sector.