Environmental efficiency evaluation of China's power system considering factor's spatial dependence: A dynamic spatial network DEA approach

Abstract

Evaluating and improving the environmental efficiency of the power system is the key to achieving sustainable development. However, the classic DEA approach is susceptible to producing biased results due to its inability to address spatial dependence across factors. Considering the spatial dependence of inputs and outputs, this paper proposes a dynamic spatial network DEA model with slacks-based measure (SNSBM) to examine the environmental efficiency of China's power systems, with a focus on the undesirable outputs of CO2, NOx, and SO2. The investigation incorporates the Moran's I test to account for the spatial interactions of patent counts, coal consumption, CO2 emissions, and renewable power generation. Compared to the classic network DEA, the SNSBM results reveal distinct spatial characteristics and a heightened level of differentiation, which highlights the importance of eliminating spatial dependence. From a dynamic perspective, environmental efficiency shows a visible surge in transmission and distribution divisions after 2015, confirming the success of China's market-oriented reform. In addition, CO2 is the largest contributor to the environmental inefficiency of the power systems, followed by NOx and SO2. This study highlights the importance of spatial dependence and provides many insightful policy recommendations for improving the environmental efficiency of China's power systems.