Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The directional distance function (DDF) with multiple bad outputs allows estimation of conditional marginal abatement cost (MAC) that would be more informative than the unconditional MAC estimated with a single pollutant. Applying a multi-pollutant parametric output DDF approach, we estimate the MACs of CO2, SO2, and NOx emissions for 105 Chinese cities in 2006–2014. It is found that the period medians of the MACs for the three emissions all increase over time, while the medians of the MACs for different cities tend to vary significantly. A goodness-of-fit criterion is developed and used to evaluate the fit of alternative models, which demonstrates the superiority of the multi-pollutant model. We have also shown that there exists a positive relationship between MAC and economic growth. Finally, several policy implications are suggested for reducing the overall emission abatement costs of China.