The impact of fuel standard on sulfur dioxide emissions: Evidence from machine learning technique

Green transport is increasingly important in improving urban air quality. Using daily air quality monitoring data at the station level, this paper employs a staggered difference-in-differences approach to examine the impact of high-quality fuel on urban air quality. After applying machine learning techniques to filter out nonlinear, cyclical, and volatile patterns in the high-frequency pollution data, we find that stricter fuel standard can significantly reduces sulfur dioxide, a direct pollutant linked to fuel composition. Compared with the control group, the stricter fuel standard reduces sulfur dioxide concentrations by about 4.89% relative to the sample mean. The green effect of this policy also continues to strengthen over time. Moreover, the enhancement in air quality resulting from high-quality fuel is more pronounced in eastern regions, in cities with stricter environmental regulations, in non-industrial cities, and in cities with higher vehicle stocks. Additional health benefit estimates suggest that fuel standard upgrade reduces losses in the value of a statistical life by approximately 8.629 billion USD. Overall, our findings indicate that stricter fuel standard constitutes a feasible and effective policy option for improving urban air quality.