Carbon emission efficiency of 284 cities in China based on machine learning approach: Driving factors and regional heterogeneity

Peixue Xing, Yanan Wang, Tao Ye, Ying Sun, Qiao Li, Xiaoyan Li, Meng Li and Wei Chen

Energy Economics, 2024, vol. 129, issue C

Abstract: The rational categorization and assessment of carbon emission efficiency (CEE) and its drivers are crucial for coping with the global climate crisis. To address the bias of univariate modeling and challenge of ignoring the heterogeneity of drivers across cities, this study explores differences between carbon emission drivers across different types of cities and regions to reveal the spatial distribution characteristics of urban CEE and heterogeneity of emission reduction potential. We use a non-radial, non-directional relaxation measure-based directional distance function (SBM-DDF) model to assess the CEE of 284 cities over the period from 2006 to 2020. Machine-learning algorithms are applied to identify city characteristics to determine the effects of city- development types and their characteristic drivers. The results of the driver analysis show that energy consumption, gross regional product, spatial area, and population size are the key factors influencing in the heterogeneity of cities' CEE, with an importance ranking of 0.578, 0.507, 0.432, and 0.418, respectively. The results of for the heterogeneity of the cities' heterogeneity further confirm that energy consumption has the greatest impact on energy-dependent cities (EDCs), economic-development cities (ECDCs), and low-carbon potential cities (LPCs), whereas among the Low-carbon growth cities (LCGs), science, technology, and innovation, urban greening, and electricity consumption play an important roles in promoting greening and low- carbon development, which can help to determine the low- carbon development model for each type of city. Finally, energy consumption affects cities in the central region more than in the eastern and western regions. Based on the results of estimating the heterogeneity of urban carbon- emission rates, we propose customized emission- reduction development pathways to guide urban low-carbon development and formulate carbon- reduction policies.

Keywords: Carbon emission efficiency; Machine learning; Slacks-based measure directional distance function (SBM-DDF); Driving factor; Heterogeneity (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (4)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S014098832300720X
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:eneeco:v:129:y:2024:i:c:s014098832300720x

DOI: 10.1016/j.eneco.2023.107222

Access Statistics for this article

Energy Economics is currently edited by R. S. J. Tol, Beng Ang, Lance Bachmeier, Perry Sadorsky, Ugur Soytas and J. P. Weyant

More articles in Energy Economics from Elsevier
Bibliographic data for series maintained by Catherine Liu ().