Threshold Effects Of Energy Efficiency And Output Elasticity In China’s Industrious Sector

Energy is necessary for human survival and development, in the21st century, China’s economic is growing rapidly, followed by a sharp increase in energy demand, China’s steady economic growth requires matching continued safety of energy supply, in addition to opening up alternative green energy sources, It can ease the increasingly tight energy supply and demand through improving energy efficiency, and to be an effective way to solve the economic problems caused by energy effective way. The other hand, China’s energy efficiency is unevenly distributed in different industries because of their different characteristics and technical level of production, while those with low energy efficiency in the industry will inevitably produce high energy consumption, high pollution problem. Therefore, the research of differences in energy efficiency has become an urgent task.In recent years, though many scholars have used stochastic frontier approach to research China’s energy efficiency, but previous literatures in the study of energy efficiency all have a problem, their research is premised on the assumption of linear, they did not consider whether energy efficiency and output elasticity of industries are different or not due to different production characteristics and technological level in industries, that is they ignored the differences between industries. Therefore, in the analysis of energy efficiency in the industrial sector can not only consider the linear model, the use of nonlinear models can better respond to the differences between the industries. Threshold regression model can explain why the model slope coefficient changes arising from non-linear characteristics when some explanatory variables achieve a certain critical value, which is closer to the real economy, so this paper use Hansen’ panel threshold model which is developed in1999to analyze differences of energy efficiency and the output elasticity in the industrial sector between the industries, the core idea of this model is to capture the potential critical point or critical region of economic, the specific method is that divides model into two or more regimes in accordance with the threshold value, each regime has a different regression equation.The empirical application is based on the Cobb-Douglas production function, this paper use linear panel stochastic frontier model and nonlinear panel stochastic frontier model respectively to estimate the date of China’s36industries from1998to2011on fixed effect threshold stochastic frontier model, and further analyze the differences of energy efficiency, the conclusion is summarized as follows:1. Multi-factor inputs have a behavior of nonlinear conversion on economic growth to energy intensity, it will help improve the output elasticity of relevant elements if we reduce energy intensity. Determined existing three threshold effect using the threshold test, and the sample data is divided into four regimes.Through three sets of panel threshold stochastic frontier model regression results, we know that the output elasticity of capital, labor and energy have changed when the energy intensity transforms in different regimes, and the higher energy intensity, the lower the output elasticity, thus differences in energy intensity cause differences of output elasticity.2. Returns to scale of production function varies with the changing of energy intensity, it can increase coefficient of returns to scale if reduce energy intensity, and chang the share of contribution of different elements on the output.Through three sets of panel threshold stochastic frontier model regression results, we know that returns to scale of production function has changed when the energy intensity transforms in different regimes, and the higher energy intensity, the lower coefficient of returns to scale.3. China’s industry average energy intensity of industrial sector decreased year by year, many industries in regime3and regime4made the transition to regime1and regime2gradually, which shows that the output elasticity of input factors in many industries is increasing year by year, and energy efficiency is also increased every year. The reason why China’s industry average energy intensity of industrial sector decreased year by year, which is mainly driven by technical progress, changes in the structure of energy and improvement of energy institutional mechanisms.As can be seen from the conclusion, China’s energy intensity of various industries have big differences, manufacturing, textiles, agro-food processing industry and the printing and copying industry have lower energy intensity, and energy efficiency is higher, while the mining industry, the production and supply industry and the processing industry have higher energy intensity, and energy efficiency has big differences. Therefore, in order to further improve energy efficiency in all sectors of China’s industrial sector so as to promote the optimization and upgrading of industrial structure, the relevant departments should consider the different characteristics of different industries in the policies, and treat them differently.