中国产业结构调整的碳排放效应——基于STIRPAT扩展模型及空间面板数据的实证研究

Carbon Emission Effect of Industrial Structure Adjustment in China: Empirical Research Based on Extended STIRPAT Model and Spatial Panel Data

利用1997—2010年我国省际面板数据,基于STIRPAT的扩展形式,运用空间面板模型实证分析了我国产业结构调整的碳排放效应。与以往研究不同的是,将空间自相关性纳入传统模型中,并且未采用简单的三次产业划分法描述产业结构,而是将生产部门细分为五类行业。结果表明:我国各省区的二氧化碳排放量存在空间维度上的依赖性和异质性,产业结构高级化进程受经济规律和发展实际的限制,这影响了减排效果的发挥;农业机械化的发展步伐不断加快,但工业发展的高能耗、高污染特征仍非常明显,农业和工业增加值占GDP比重分别与二氧化碳排放量显著正相关,其他行业比重调整的降碳作用有待更充分的激发。指出:我国应推进农业机械化与产业结构战略性调整相结合,促进以技术创新与进步为基础的产业内部结构升级;各区域要力求与周边地区协调、互助,共建以低碳排放为特征的产业体系。

Using the provincial panel data of China from 1997 to 2010, this paper constructs the spatial panel data model based on extended STIRPAT model to analyze empirically the carbon emission effect of industrial structure adjustment in China. It is different from the former related research that it introduces the spatial autocorrelation into conventional model,and subdivides the production department into five industries but three industries to describe industrial structure. The results show as follows ; there are the spatial autoeorrelation and the heterogeneity a- mong provincial carbon emission, and industrial structure upgrading is controlled by economic law and current situation, which weakens its effect of emission reduction;the development of agricultural mechanization is speeding up, while the characteristics of high energy consumption and serious pollution of industrial development are distinct, and the proportions of agriculturial and industrial added values all have the remarkable pos- itive correlation with carbon emission; the carbon reduction which results from other industries; proportions adjustment should be inspired more. Finally,it points out as follows：agricultural mechanization should be combined with the strategic adjustment of industrial structure,internal industrial structure should be upgraded based on technology innovation and progress;each area should strive to coordinate with its surrounding areas so as to build an industrial system with low carbon emission.