Is China's energy policy effective for power plants? Evidence from the 12th Five-Year Plan energy saving targets

We propose the meta-frontier hybrid data envelopment analysis (MHDEA) model, a non-parametric methodology, to assess the total factor carbon performance index (TCPI) of China's fossil fuel power plants from 2005 to 2015. The model allows for both proportional and non-proportional variables within the production technology. In contrast to conventional DEA models, the MHDEA model simultaneously incorporates the characteristics of group heterogeneity, undesirable outputs as well as hybrid measures. In accordance with the definition of the Malmquist-Luenberger (ML) index, we further explore the driving factors of productivity progress, including efficiency change (EC), best-practice gap change (BPC) and technology gap change (TGC). We employ the difference-in-differences (DID) strategy by combining the variations across regions and years to identify the effect of the 12th Five-Year Plan (FYP) energy saving targets on the TCPI of power plants. Frist, we find that the average TCPI of power plants is 0.50 and the average growth rate is 3.58% per year. Second, the productivity progress of the central group is driven by the innovation effect, while the local group stems from the catch-up effect. Third, the energy intensity reduction target is 1% above the average value and the TCPI is expected to increase by 0.32%. We suggest that the Chinese government design different policies for the central and local power plants based on this heterogenicity.