Analysis On Regional Carbon Emission Efficiency Measurement And Influence Factors In China

In recent years, as a developing country with rapid economic growth, China isalso facing pressures and challenges of resources and ecological environmentcarrying shortage. Especially the rapid growth of greenhouse gas emissions hasbecome the attention focus at home and abroad. In2012, Chinese carbon emissionsaccounted for28.8%of global proportion which made China as the world’s largestcarbon emitter. In the same year Chinese total economy amount accounted for onlyabout10percent of global economy proportion. Chinese economic developmentlevel is far from a developed country, but the pressure to reduce emissions is evenmore drama. In2009China announced that carbon intensity must be reduced by40-45%by2020compared to2005. Based on this back ground, we analyze carbonemission efficiency and influence factors from two perspectives which are totalfactor carbon emission efficiency based on data envelopment analysis and carbonemission intensity. The main contents are:(1) Total factor carbon emission efficiency analysis based on DEA method. Weselect human capital investment, capital stock, carbon dioxide emissions as inputvariables, GDP as output variable. The result shows that regional carbon emissionefficiency is an upward trend which arranged in descending order of east, central andwest. We decompose the carbon emissions efficiency and find out that technicalefficiency in the eastern part of the country is in effective level. The scale efficiencyin the western and central region shows a rising trend. We decompose carbonemissions productivity and find out that science and technology in the eastern regionis in leading level. The enhancement of central and western TFP indices is moredependent on technology efficiency. Technology in these regions appearsdegradation.The analysis impact factors results of total factor carbon emission efficiencywith Tobit model shows that: there is a significant positive correlation between theeconomic development level and industrial structure while a significant negativecorrelation between the level of urbanization, energy structure and energy intensity.Scientific and technological progress and open policy is positive but not verysignificant. There is a significant spatial correlation of carbon emissions total factorproductivity.(2) Cointegration Analysis of regional carbon emission intensity and influencing factors. We use panel cointegration analysis to explore the long-runequilibrium relationship of regional carbon emission intensity&energy intensity,energy structure, urbanization level in2002-2011, and then we use the estimatedcointegration and error correction model. Test results show that: in the country andthe eastern, central and western regions, variables are in a long-standingco-integration relationship. Long-run equilibrium co-integration equation estimationresults show that: there are positive effects of the energy intensity in the national andregional carbon intensity and decreasing impact order is eastern, central and western.The energy structure impacts are all positive and the impact from large to small arethe central, west and the east. There are positive relationships between urbanizationlevel and carbon emission intensity in the country, western and central regionalswhile a negative relationship in the eastern region. In the western region, thecorrelation is not very significant. Positive impact degree turns to the center and west.Error correction model results show that: the western region is the most fast of therapid of adjustment to balance, the eastern region is in the second place and thecentral is the slowest.(3) Spatial econometric analysis of provincial carbon emission intensity and itsimpact factors. We use spatial econometric model to empirical test domain carbonintensity&energy intensity, energy structure, urbanization level in2007-2011. Spatialautocorrelation test results show that:29provincial carbon emissions intensityspatial correlation is remarkable, which means that there are spatial distributionspillover effects between regions. There are similar characteristics in adjacent areas.Spatial econometric regression results show that: energy intensity elasticitycoefficient is1.06with spatial error model, the energy structure elasticity coefficientis0.63, and the population level of urbanization elasticity coefficient is0.21. Energyintensity is the most important factor of carbon emissions intensity. Energy structureand urbanization level also have significant impact on carbon emission intensity.This conclusion coincides with the result of cointegration test.