A Study On Total Factor Energy Efficiency Base On Frontier Analysis

In recent years, due to energy shortages and increasingly serious environmental problems posed by energy consumption,energy efficiency attracts more and more researchers'attention. The evaluation of energy efficiency and related factors to energy efficiency are the most important aspects in energy efficiency studies.In this paper, the evaluation of energy efficiency and analysis of related factors are the main research content. First of all, domestic and international literature of energy efficiency is reviewed in detail. It is found that: the current evaluation model of energy efficiency has taken the environmental pollution as a non-consensual output variable, but it is still immature when processing environmental pollution variable. Data envelopment analysis (DEA) and Stochastic Frontier Analysis (SFA) are two main methods of energy efficiency evaluation, both of which have advantages and disadvantages, and there is no method to integrate them. Previous studies on related factors to energy efficiency are mostly focused on the analysis of external factors, and less can be seen on internal factors. Based on DEA and SFA, a new integrated analysis method (D&S) was proposed. Panel data from 49 countries during the period of 1999-2008 are used to compute energy efficiency in these three methods. Then, relevance, consistency and stability of the results were analyzed. Through the differentiation decomposition of internal factors of energy this paper further illustrates its mechanism and contribution.The empirical results show as follow: By the paired T test, Spearman and Kendall rank correlation coefficient test, it is proved that evaluation results of DEA, SFA and D&S are significantly relevant and orderly consistent. From the point of view of stability, SFA is much better than D&S, and followed by DEA, which indicate that the stability of evaluation results of DEA is improved obviously by removing random factors via SFA. There is a tendency that energy efficiency values of the overall sample is increasing over time, while, a big gap about this value still do exist between China and overall sample and this gap tends to gradually narrow, meanwhile, China's efficiency increases faster than that of over sample. Differences in energy efficiency are decomposed into four factors, that is, energy intensity, capital-energy ratio, human-energy ratio and energy-pollution ratio. All factors are negatively correlated with energy efficiency, and their contribution rate of energy intensity and human-energy ratio to the differences in energy efficiency is greater than capital-energy ratio and energy-pollution ratio.