Input-Output Efficiency of Chinese Power Generation Enterprises and Its Improvement Direction-Based on Three-Stage DEA Model

This paper uses the three-stage DEA method to measure the input-output efficiency of China's 23 listed power generation companies (mainly thermal power generation) in 2019, and uses the SFA regression model to eliminate environmental elements and random disturbances. The results show that in a non-homogeneous environment, the scale efficiencies of most power generation companies are greater than or equal to their pure technical efficiencies. These companies should first improve management and technical levels, and then optimize the scale of investment. Furthermore, after removing environmental variables, half of the companies should turn to increasing economies of scale instead of diminishing economies of scale. It can be seen that environmental factors, such as the degree of regional development and IPO time, have reduced the economies of scale of enterprises, so they should strengthen the communication between different regions, and the government should provide assistance to companies that are listed late.