中国地方政府碳减排压力驱动因素的省际差异——基于STIRPAT模型

The Differences of Driving Factors of Local Governments’Pressure on Carbon Emission Reduction in China Based on STIRPAT Model

我国各个地区碳减排压力存在区域差异,不同区域间的碳排放与其驱动因素之间的关系十分复杂,并非能由常系数经典回归分析所解释。本文考虑空间距离和局域空间联系因素,把地理空间效应纳入到STIRPAT模型框架中,通过地理加权回归方法(GWR)对回归系数进行局域分解得到区域差异化的回归系数。结果发现,各个地区人口、人均GDP(富裕度)和能源强度(技术)驱动因素对碳排放的弹性存在明显差异,区域碳减排压力及其驱动因素呈现为一种非均衡的联动的局域性特征。其中,相比人均GDP(富裕度)和能源强度(技术),人口对我国碳减排压力的弹性变化区间最小,仅从0.8768到1.5113;各地区人均GDP(富裕度)对碳排放的影响存在着显著不同,比如人均GDP提高1%时,新疆碳排放总量提高1.6118%,而云南仅提高0.8644%;能源强度(技术)是影响碳排放的关键因素,因而提高能源利用效率是碳减排的关键所在,尤其是在耗能产业相对密集的区域及相对发达地区,能源强度对碳排放的驱动影响更大,这些地区的能源利用效率尚有很大的提升空间。因此,针对人口、人均GDP(富裕度)和能源强度(技术)三个驱动因素制定差异化的区域碳减排调控政策是非常必要的。

Carbon emission reduction pressure shows differences in different regions. The relationship between carbon emission and its driving factors is very complicated, and can not be explained by Ordinary Least Square （OLS） estimation of constant coefficients. Considering the spatial distance and local spatial connection factors, this paper took geographical space effects into the Stochastic Impacts by Regression on P （population）, A （affluence） and T （energy intensity） （STIRPAT） model and used geographical weighted regression （GWR） model to get the regression coefficient of regional differences by decomposing the regression coefficient locally. The results show that there are significant correlations between carbon emission pressure and P （population）, A （affluence） and T （energy intensity） according to the OLS analysis. And according to geographical weighted regression （GWR）, significant regional differences exist among and within regions. As for China, with a large population base, the carbon emission reduction pressure is enormous. The coefficient of elasticity also has significant differences in different regions; the elastic variation region of P is the smallest, which is just from 0.8768 to 1.5113. The impact of A on carbon emissions is different too. For instance, when GDP per capita increases by 1%, the carbon emission in Xinjiang increases by 1.6118% and in Yunnan 0.8644 %. T plays a dominant role in the carbon emission control, so improvement of energy efficiency will be the main approach to carbon emission reduction for some time, especially in the relatively intensive area of energy-buring industry and developed regions. Therefore, it is very necessary to formulate differentiated regulation and control policies of carbon emission reduction according to the three driving factors including P, A and T.