【作者】 张晨曦；

【导师】 王连芬； 潘菁；

【作者基本信息】 湖南大学 ， 国际商务， 2022， 硕士

【摘要】 各国经济依存度的提升使得对外直接投资成为经济发展的重要助力,但目前国内碳排放量的持续增长已成为制约可持续发展的关键因素。在同时面临“对外开放”与“减排降碳”两大任务的背景下,研究OFDI对我国碳排放的具体影响方向和程度对如期实现经济与气候领域目标具有现实意义。首先,本文基于2003-2019年我国30个省市区的面板数据,在测算国内二氧化碳排放的基础上,同时选取外商直接投资、对外开放程度、能源消费结构等作为控制变量,建立双向固定效应模型,从全国和地区两个层面对我国OFDI的碳排放效应进行实证分析。结果表明在全国层面每增加一单位的投资额将会使我国碳排放量增长0.063%。区域层面结果与全国层面一致,OFDI增多均会显著使碳排放量上升,但不同区域的碳排放效应存在差异,东部和西部的碳排放效应显著高于中部区域。其次,建立中介效应模型,选择非参数百分位Bootstrap法,检验并分析经济规模增长、产业结构优化和绿色技术进步在OFDI的碳排放效应中的具体传导机制。分析结果得出,OFDI会通过助推产业结构优化和促进绿色技术进步这两条渠道对国内碳排放产生显著的间接抑制作用,但通过经济规模增长会产生相反的显著加剧碳排放的作用。进一步检验发现产业结构优化与绿色技术进步在对外直接投资对母国碳排放的影响机制中存在链式中介效应。再次,进一步研究OFDI碳排放效应的时滞影响和投资目的地异质性。借助系统GMM方法对动态面板模型进行回归,结果得到滞后一期和两期的OFDI均会对碳排放产生抑制作用,且该作用强度随时间增长而增加。将投资目的地进行顺、逆梯度划分并回归后发现,顺梯度投资会显著加剧国内的碳排放量,逆梯度投资对碳排放则具有不显著的促进作用。最后,利用变量替换方法和变量滞后方法证明了模型的稳健性,并基于现状及实证分析的结果给出了针对性的政策建议。更多还原

【Abstract】 The increase in the economic dependence of different countries makes outward foreign direct investment an important boost to economic development,but at present,the continuous growth of domestic carbon emissions has become a key restricting factor of sustainable development.In the context of facing the two major tasks of “opening up” and “reducing carbon emissions” at the same time,it is of practical significance to study the specific direction and extent of OFDI’s impact on China’s carbon emissions to achieve the goals in the economic and climate fields as scheduled.First,based on the panel data of 30 provinces in China from 2003 to 2019,and on the basis of measuring domestic emissions of carbon dioxide,foreign direct investment,degree of opening up,energy consumption structure,etc.are selected as control variables and establish a fixed effect model.This dissertation conducts an empirical analysis on the direct effect of China’s OFDI on carbon emissions from the national and regional levels.The results show that each additional unit of investment at the national level will increase China’s carbon emissions by 0.063%.The results at the regional level are consistent with those at the national level.The increase of OFDI will significantly increase the carbon emission,but there are differences in the carbon emission effects in different regions.The carbon emission effect of the eastern and western regions is significantly higher than that of the central region.Secondly,the mediation effect model is established,and the non-parametric percentile Bootstrap method is selected to test and analyze the specific transmission mechanism of economic scale growth,industrial structure optimization and green technology progress in the carbon emission effect of OFDI.The analysis results show that OFDI has a significant indirect inhibitory effect on the carbon emissions of the home country mainly through promoting industrial structure upgrading and promoting green technology progress,but it will have the opposite effect of significantly increasing carbon emissions through economic scale growth.Further examination shows that industrial structure optimization and green technology progress have a chain-based intermediary effect in the impact mechanism of OFDI on carbon emissions in home countries.Thirdly,the time-lag effect of OFDI carbon emission and the heterogeneity of investment destination are further studied.The system GMM method was used to regress the dynamic panel model.Using the system GMM method to regress the dynamic panel model,the results show that OFDI with a lag of one and two periods will have an inhibitory effect on carbon emissions,and the strength of this effect will increase with time.After dividing the investment destinations into the forward and reverse gradients and regressing the model,it is found that investment along the gradient will significantly increase domestic carbon emissions,while investment in the reverse gradient will not significantly promote carbon emissions.Finally,the robustness of the model is proved by the variable substitution method and the variable lag method,and corresponding policy suggestions are given based on the current situation and the results of empiric al analysis.更多还原