Title:Quantile estimation of CO2 marginal abatement cost across emission-generating technologies

Abstract:Marginal abatement cost (MAC) is a critical metric for designing efficient and cost-effective mitigation policies. However, existing MAC estimates are typically derived under different assumptions about emission-generating technologies, yet few studies have systematically compared these technologies. Moreover, conventional estimators often exhibit biases arising from limited abatement options, production inefficiencies, and data noise. To address these limitations, this paper analyzes the abatement behavior of three emission-generating technologies: by-production, joint disposability, and weak G-disposability, each consistent with the material balance principle. We employ both full and quantile frontier estimation methods to identify optimal abatement strategies. Using data from U.S. coal-fired power plants in 2022, the empirical results suggest that reducing electricity output, rather than cutting emission-generating inputs such as fossil fuels, provides a more cost-effective mitigation pathway. Furthermore, Monte Carlo simulations demonstrate that the quantile estimator consistently delivers more accurate results than the full frontier estimator.