中国区域外向型经济发展与碳排放关系实证研究
|
摘要
近二十年来，全球气候变暖逐渐成为全球普遍重视的问题。全球气候发生了很明显的改变，大气不断越来越炎热，因气候变迁造成的灾害亦不断增加。全球的热浪、干旱频传；南极的大冰层开始融化，北极也出现没有冰层的海洋；沙漠的面积逐渐扩大；海平面上升，造成海岛国家面临被淹没的危机；整个生态系统也受到严重的影响。
随着中国经济快速增长过程中出口导向型驱动特征愈发明显，其增长代价之一的环境污染问题也愈发严峻。在这一背景下，国际贸易增长与FDI引入的碳排放的影响问题已经成为学术研究焦点。首先，国际贸易的环境效应相关研究框架起始于Grossman and Krueger（1991）的研究。目前学术界将对外贸易增长对环境的影响分别从规模、结构和技术三个角度进行分析1。国内外许多学者运用不同方法从不同角度对国际贸易的环境效应进行了分析和检验，但总结起来存在着两种截然不同的学术观点。一种观点认为国际贸易会带来环境质量的改善，另一种观点则认为贸易增长会导致环境的恶化。比如，Antweiler（2001）运用回归分析方法对贸易的环境效应进行了计量分析，得出国际贸易的增长并非必然带来环境压力；Eliste and Fredriksson（2002）等学者研究认为，贸易增长能够促进环境的改善。
本文拟通过贸易增长与FDI流入环境效应的作用机制来确定其对中国各个地区碳排放的影响路径。传统的理论和经验研究表明，国际贸易增长与FDI流入之间呈替代或互补影响关系。即当贸易壁垒不是主要FDI流入障碍时，作为国际贸易的补充，FDI主要将流入东道国的出口部门。否则，作为国际贸易增长的替代，FDI主要将流入东道国的进口部门。据此逻辑，国际贸易环境效应的作用机制内生于贸易业务增长之中，而相比较而言，FDI流入则发生在贸易业务增长之前，属于微观投资决策主体对于贸易和投资的一种选择性决策。因此，FDI流入环境效应的作用机制决定于其流入动机。基于此，本文将利用现代计量经济学分析方法分析和刻画国际贸易增长与FDI流入对地区碳排放的影响。
论文的框架结构和主要内容如下：
第1章为引言部分，包括问题的提出、研究意义、核心概念的界定、相关问题综述、论文的结构。
第2章外向型经济对环境影响的理论分析。本章将从理论分析的角度，详细阐述外向型经济对环境影响机制和途径，这是本文后续实证检验的理论基础。
第3章外向型经济与地区碳排放的特征及演变路径分析。本章详细分析了各种估计方法的优劣，并通过具体的测算结果比较了两种常用估计方法测算结果的合理性。并在测算的结果基础上分析中国地区碳排放的动态演变特征。本文还考虑到从经济理论上而言，随着经济增长规模的不断扩张，能源消费不断增加，由此导致的碳排放也会因此“发散增长”，微观层面、中观层面和宏观层面的收敛机制使得碳排放出现收敛现象。在收入提高的过程中，随着产业结构向信息化和服务业的演变、清洁技术的应用、环保需求的加强、环境规制的实施以及环保投资的增加等，碳排放量先上升然后逐步下降，理论上会出现收敛的情况，这样碳排放收敛假说成立。在本章中将利用测算得到的中国地区碳排放数据结合面板数据模型方法对上述碳排放收敛假设进行检验。
第4章出口贸易、FDI与地区碳排放的长、短期均衡关系分析。在本章中利用最新的面板协整检验方法——加权对称估计下的异质面板协整检验方法对中国29个地区的出口贸易、FDI和地区碳排放之间的稳定均衡关系进行检验。
第5章出口贸易、FDI与地区碳排放影响的区制效应分析。FDI、出口贸易与地区碳排放影响往往具有结构转变性特征，本章首先对环境Kuznets曲线进行相关的理论分析和阐述。环境Kuznets曲线可能存在着区制效应。但是数据是否支持区制效应的存在，需要利用相关的方法进行检验。本章利用半参数分位数回归理论对环境Kuznets曲线中是否存在着一定的区制效应进行相关的检验。
第6章出口贸易、FDI对地区碳排放的结构效应和技术效应的动态关联性分析。本文利用相关数据从全国加总、东部地区、中部地区、西部地区四个角度分解了到1987-2011碳排放结构效应和技术效应分解结果。通过建立动态面板数据模型分析和检验了FDI、出口贸易对地区碳排放结构效应和技术效应的影响。
第7章是结论和政策建议。
本文的研究发现：
省级二氧化碳排放强度的变化趋势与两种措施相似。在“第八个五年计划”（1991-1995年）和“第十个五年计划”（2001-2005年），系数显示为增加的趋势，并在“九五”计划（1996-2000年）和“十一五”规划（2006-2010年），系数显示为下降趋势。尽管每个时期的变化趋势是稳定的，整体周期的变化趋势并不稳定，而且不会逐渐减少，所以无法得到收敛的一致结论。从估计结果可以看到，在每一种模型设定下，系数都是显著为负。这意味着，省级二氧化碳排放存在着绝对收敛特征。从结果还可以看到，和是显著为正，所以空间效应是显著存在的。回归结果还表明，煤炭消费量达到能源消费总量中，第二产业产值占全行业总产值和能源强度的比值进行统计显著影响二氧化碳排放强度的增长速度，这进一步影响的收敛速度省级二氧化碳排放强度。在本文中，我们分析了中国二氧化碳排放强度的省级收敛。我们建议，空间溢出效应引入到纯动态面板数据模型的空间动态面板数据模型。通过使用此SDPD模型，我们可以避免遗漏变量偏涉及横截面方程和纯动态面板数据方程。我们的实证结果表明：（1）中国省际CO2排放的省级收敛性；（2）有条件的收敛速度比绝对-收敛率较高；（3）省际CO2排放在空间相关性和收敛与考虑空间效应的比率较高；（4）煤炭消费量达到能源消费总量中，第二产业产值占全行业总产值和能源强度的比值进行统计显著影响收敛速度。
通过面板协整模型，可以检验到由于总体而言，中国出口对CO2排量的增加有正向的作用，因此，中国需要采取措施调整出口产业结构和技术两方面。在整体上使出口产业向集约型方向发展，通过制定恰当的产业政策鼓励低能耗出口产业的发展，抑制高能耗出口产业的发展，从而增加低能耗产业在出口产业中的比重，降低高能耗产业在出口产业中的比重。通过碳排放税、碳交易平台等经济手段，积极引导出口产业的生产技术向清洁生产方向发展。从而积极引导出口的结构效应和技术效应在改善环境方面的积极作用，使之逐渐超过规模效应给环境带来的负面影响，从而使经济和环境朝着可持续的方向发展。由于出口对环境的影响对高、中、低排量省份有所差别，因而政策的制定应该具有地区针对性，对高排量省份的政策应较为稳健，而对中、低排量省份的政策力度应该有所加强。在政策制定过程中，应避免过度关注某单一地区而给周围地区带来额外的消极影响，应保证公平性原则。整体上，FDI对CO2排量的增加也有正向的作用，因此，在吸引和接受FDI时应保持更加谨慎的态度，综合考虑FDI对地区的经济、社会影响。考虑到这种正向影响较弱且部分资源大省、沿海、边疆地区的FDI会对环境改善产生积极影响，在这些地区还可以适当鼓励FDI的流入。由于短期中，出口和FDI对CO2排量并没有较强的影响，因而，为了调整出口和CO2排量之间、FDI和CO2排量之间的关系，经济和环境政策以长期政策为宜，短期措施的效果是相当有限的。
本文利用半参数分位数回归理论对环境Kuznets曲线中是否存在着一定的区制效应进行相关的检验。研究发现人均FDI变量系数为正，且在各模型设定下数值大致相同，此意味着该变量在解释人均二氧化碳排放量上有其重要性，但其影响程度较低。而人均对外贸易变量系数值亦为正值，且对人均二氧化碳排放量有着显著的影响。然而，在观察与人均GDP有关的变量后，发现不同模型设定形式下（二次式或者三次式）所得到的系数估计值差异很大，此结果显著有采用半参数回归进行分析的必要。除了分量25%与10%以及分量75%与90%两组彼此之间没有显著差异外，其余则呈现显著差异，及表示在不同分量下，所代表的意义有显著不同。此结果指出分位数回归方法的重要性，也说明不同二氧化碳排放水平下与人均收入两者之间存在着不同的变化。除了分量25%与10%以及分量50%与75%两组彼此间不显著差异之外，其余则存在显著的差异性，同样说明了分位数回归之下，不同程度的二氧化碳排放水平与人均GDP两者间存在着不同的变化。实证结果发现GDP与区域二氧化碳排放正相关，我国各区域高能耗、高污染的增长方式没有改变。2显著为负，说明了区域二氧化碳排放与区域的人均GDP水平之间存在着“倒U”型关系。同时人均FDI对区域二氧化碳排放的影响会由于人均GDP门限值的不同产生不同方向的影响。当区域人均GDP小于门限值时，人均FDI的增加会减少对区域二氧化碳排放，但是一旦区域人均GDP大于门限值时，人均FDI的增加会增加区域二氧化碳排放量。同时人均对外贸易额对区域二氧化碳排放的影响会由于人均GDP门限值的不同产生不同方向的影响。当区域人均GDP小于门限值时，人均对外贸易额的增加会增加对区域二氧化碳排放，但是一旦区域人均GDP大于门限值时，人均对外贸易额的增加会增加区域二氧化碳排放量。在影响作用的门限效应中，人均对外贸易额对区域二氧化碳排放的影响不同于人均FDI对区域二氧化碳排放的影响。
利用相关数据从全国加总、东部地区、中部地区、西部地区四个角度分解了到1987-2011碳排放结构效应和技术效应分解结果。全国分解结果，可以看到一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为26.054%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为12.687%。一次能源消耗的结构效应对于人均二氧化碳排放的影响显著的大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。东部地区一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。东部地区1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为17.618%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为14.818%。一次能源消耗的结构效应对于人均二氧化碳排放的影响并不显著大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。中部地区一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为28.885%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为8.725%。一次能源消耗的结构效应对于人均二氧化碳排放的影响显著的大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。西部地区一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为28.737%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为12.214%。一次能源消耗的结构效应对于人均二氧化碳排放的影响显著的大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。
本文还通过建立动态面板数据模型分析和检验了FDI、出口贸易对地区碳排放结构效应和技术效应的影响。实证结果表明FDI、对外贸易这两个变量对地区碳排放的结构效应均存在显著的正向影响。FDI对地区碳排放结构效应影响最高的五个地区分别是：上海、北京、江苏、山东、广东。FDI对地区碳排放结构效应影响最低的五个地区分别是：新疆、内蒙古、福建、江西、海南。从FDI对地区碳排放结构效应影响特征来看呈现东、中、西部地区差异。对外贸易对地区碳排放结构效应影响最高的五个地区分别是：广东、浙江、江苏、上海、福建。对外贸易对地区碳排放结构效应影响最低的五个地区分别是：宁夏、青海、新疆、海南、江西。从对外贸易对地区碳排放结构效应影响特征来看也呈现东、中、西部地区差异。中、西部地区的碳排放的结构效应大于技术效应，但是中、西部地区FDI、对外贸易这两个变量对地区碳排放的结构效应影响还远远低于东部地区。所以进一步提高FDI、对外贸易这两个变量对地区碳排放的结构效应影响是降低中、西部地区碳排放的一个有利的途径。FDI对地区碳排放技术效应影响最高的五个地区分别是：广东、黑龙江、四川、辽宁、北京。FDI对地区碳排放技术效应影响最低的五个地区分别是：新疆、海南、内蒙古、广西、山西。对外贸易对地区碳排放技术效应影响最高的五个地区分别是：黑龙江、海南、江苏、云南、浙江。对外贸易对地区碳排放结构效应影响最低的五个地区分别是：青海、北京、河南、山东、陕西。相比对地区碳排放结构效应的影响，FDI、外贸易对地区碳排放技术效应的地区差异并不明显。所以为了进一步降低地区碳排放，应该提高各个地区中FDI、外贸易对地区碳排放技术效应的影响。
The past two decades, global warming has become the world universal attention. Globalclimate change has occurred obviously, the atmosphere constantly getting hot, disasters caused byclimate change are increasing. Global heat waves, droughts frequent; great Antarctic ice began tomelt, there is no ice in the Arctic Ocean; desert area gradually expanded; sea-level rise, resultingin an island nation in danger of being overwhelmed by the crisis; entire ecosystem is badlyaffected.
With the rapid economic growth in China’s export-oriented process-driven feature isbecoming increasingly apparent, environmental pollution costs of its growth is also increasinglygrim. In this context, the impact of carbon emissions growth of international trade and theintroduction of FDI has become the focus of academic research. First, the environmental effects ofinternational trade-related research framework Grossman and Krueger (1991) study began in.Current academic impact of growth on the foreign trade environment is analyzed from the size,structure and technology of the three angles. Many scholars from different angles using differentmethods of international trade on the environmental effects were analyzed and tested, but summedup the existence of two distinct academic perspectives. One view is that international trade willbring about improvements in environmental quality, and the other view is that trade growth willlead to the deterioration of the environment. For example, Antweiler (2001) using regressionanalysis of environmental effects of trade were measured analysis, the growth of international tradedoes not necessarily bring pressure on the environment; scholars Eliste and Fredriksson (2002)consider that the environment can promote the growth of trade improvements.
This paper intends to grow through trade and FDI inflows mechanism to determine theenvironmental effects of its impact on the various regions of China the path of carbon emissions.The traditional theory and empirical studies have shown that the growth of international trade andFDI was an alternative or complementary affect the relationship between inflows. That is, whentrade barriers are not the main obstacle to FDI inflows, as a supplement to international trade, FDIinflows will host the main export sector. Otherwise, as an alternative to the growth of internationaltrade, FDI inflows to the host country’s main import sector. Accordingly logic, the mechanism ofaction of the environmental effects of international trade growth among trading business was born,but comparatively speaking, FDI inflows occurred before the trade business growth, investmentdecisions are the subject of a micro-selectivity for trade and investment decisions. Thus, FDI inflows environmental effects of the decision on its mechanism of action inflows motivation. Basedon this, the paper will analyze and characterize the growth of international trade and FDI inflowsimpact on regional carbon emissions using modern econometric analysis.
Thesis framework and main contents are as follows:
Chapter1is an introduction, including issues raised, research significance, the definition ofcore concepts, review questions, the paper structure.
Chapter2analyzes the export-oriented economy on the theory of environmental impact. Thischapter from the theoretical point of view, the detail-oriented economy mechanisms and pathwaysof environmental impact, which is the theoretical basis of this paper subsequent empirical tests.
Chapter3Characteristics of export-oriented economy and regional carbon emissions andevolution path analysis. This chapter analyzes the pros and cons of various estimation methods, andthrough specific estimation results compared two methods commonly used estimate calculationresults are reasonable. And analyze the dynamic evolution characteristics of carbon emissions inChina on the basis of the results of the calculation. The article also take into account the economictheory, with the continuous expansion of the scale of economic growth, increasing energyconsumption, carbon emissions will therefore resulting " divergent growth", the micro-level,meso-level and macro-level convergence carbon emissions occur mechanism makes convergencephenomenon. Revenue increased in the process, as the industrial structure and the evolution ofinformation technology services, implementation and application of increased environmentalinvestment in clean technology, strengthen environmental protection requirements, environmentalregulations, etc., carbon emissions increased at first and then gradually decline, theoreticallythere will be convergence of circumstances, so carbon emissions convergence hypothesis holds. Inthis chapter, the use of estimates resulting carbon emissions data in China combined with paneldata model approach for the carbon emissions convergence hypothesis testing.
Chapter4long export trade, FDI and regional carbon emissions, the analysis of short-termequilibrium relationship. Using the latest panel cointegration methods in this chapter-theheterogeneous panel cointegration estimation method of weighted symmetric equilibriumrelationship export China29regions, FDI and regions between carbon emissions inspection.
Chapter5export trade, FDI and regional district system effects analysis of the impact ofcarbon emissions. FDI, export trade and the impact of carbon emission regions tend to have astructural change characteristics, environmental Kuznets curve first chapter related theoreticalanalysis and elaboration. Environmental Kuznets curve effect may exist system area. However,whether the data support the existence of district system effects, require the use of related methodsfor testing. This chapter uses the semi-parametric quantile regression theory of environmentalKuznets curve, whether there is a certain district system effects associated test.
Chapter6export trade, FDI analysis of dynamic correlation structure effects and technicaleffects of regional carbon emissions. In this paper, data from the National total four angles, eastern, central and western region of the1987-2011decomposition of carbon emission structureeffect and technology effect decomposition. Through the establishment of a dynamic panel datamodel analysis and inspection of the FDI, the impact on export trade and technical effects of carbonemission structure effect region.
Chapter7, conclusions and policy recommendations.
The findings in this article:
Similar trends provincial carbon dioxide emissions intensity and two measures. In the "Eighth Five-Year Plan "(1991-1995) and the " Tenth Five-Year Plan "(2001-2005), thecoefficient is displayed as an increasing trend, and in the "Ninth Five " program (1996-2000) andthe " Eleventh Five-Year plan"(2006-2010), the coefficient is shown as a downward trend.Although the trend in each period is stable, the overall trend cycle is not stable, and does notdecrease gradually, it is not unanimous conclusions convergent. From the estimation result can beseen, each of the models in the set, the coefficient is significantly negative. This means that thereis an absolute carbon dioxide emissions provincial convergence characteristics. You can also seefrom the results, and is significantly positive, so the effect is significant space exists. Theregression results also show that coal consumption reached total energy consumption, the ratio ofthe total industry output value of the second industry and energy intensity were statisticallysignificant impact on the growth rate of carbon dioxide emissions intensity, which further affectthe convergence rate of the provincial carbon dioxide emissions strength. In this paper, we analyzethe convergence of Chinese provincial-level carbon dioxide emissions intensity. We propose tointroduce spatial spillover effects into the space of pure dynamic panel data model dynamic paneldata model. By using this SDPD model, we can avoid omitted variable bias involved in thecross-sectional equation and pure dynamic panel data equation. Our empirical results show that:(1) Chinese provincial CO2emissions provincial convergence;(2) Conditional convergence ratethan the absolute-higher convergence rate;(3) inter-provincial spatial correlation between CO2emissions and convergence with consider the effect of a higher percentage of space;(4) coalconsumption reached total energy consumption, the ratio of the total industry output value of thesecond industry and energy intensity were statistically significant impact on the convergence rate.
Through panel cointegration model can be tested to due Overall, China 's exports to theincrease of CO2emissions has a positive effect, therefore, China needs to take measures to adjustthe industrial structure of both exports and technology. So on the whole export industry tointensive direction, through the development of appropriate industrial policy to encourage thedevelopment of low-power export industry, inhibit the development of export industries with highenergy consumption, thereby increasing the proportion of low-energy industry in the exportindustry, reducing energy-intensive industries the proportion of the export industry. Through acarbon tax, carbon trading platform of economic means, and actively guide the export industryproduction technology to develop in the direction of cleaner production. Structural effects and techniques in order to actively guide the effect of improving the environmental aspects of exportactive role in making gradual scale over the negative impact to the environment, so that theeconomic and environmental direction towards sustainable development. The impact on theenvironment due to the export of high, medium and low emission provinces vary, and thereforeshould have a regional policy targeted policies for high-emission provinces should be more robust,and policies, and low-emission provinces efforts should be strengthened. In the policy-makingprocess, should avoid excessive attention to a single region and surrounding areas bring additionalnegative impact should ensure fairness principle. Overall, FDI is also on the increase in CO2emissions has a positive effect, therefore, should attract FDI and to accept a more cautious attitudein keeping, considering the economic and social impact of FDI on the region. Given this positiveeffect is weak and some resources in the province, FDI on the environment along the coast, borderareas, improvements have a positive impact in these areas can also be appropriate to encourage FDIinflows. Due to the short-term, exports and FDI are not a strong impact on CO2emissions, andtherefore, in order to adjust the relationship between exports and CO2emissions, FDI and CO2emissions between economic and environmental policies to long-term policy is appropriate, theeffect of short-term measures is quite limited.
In this paper, the use of semi-parametric quantile regression theory of environmental Kuznetscurve, whether there is a certain district system effects associated test. Study found that per capitaFDI variable coefficient is positive, and roughly the same value set in each model, this means thatthe variable is in the interpretation of carbon dioxide emissions per capita have their place, but to alesser degree affected. Foreign Trade Department of the per capita value of the variable is alsopositive, and per capita carbon dioxide emissions have a significant impact. However, afterobserving the per capita GDP-related variables, found under different model specification form(quadratic or cubic type) obtained coefficient estimates vary widely, there is a significant result ofthis semi-parametric regression analysis is necessary. In addition there was no significantdifference in weight of25%and10%and75%and90%weight between the two groups with eachother, the rest showed significant difference, and means under the different components,meanings are significantly different. These results indicate the importance of quantile regressionmethod, also shows different levels of carbon dioxide emissions per capita income between thetwo there are different variations. In addition to weight25%and10%and50%and75%of thecomponents of the two groups are not significantly different from each other, the other is there aresignificant differences, also explains under quantile regression, different levels of carbon dioxideemissions per capita GDP between the two there are different variations. The empirical resultsshow positive GDP and regional carbon dioxide emissions related to our various areas of highenergy consumption, high pollution growth pattern has not changed. Significantly negative,indicating the existence of the "inverted U" shaped relationship between regional GDP per capitacarbon dioxide emissions and the region. Meanwhile the impact of FDI on regional per capita carbon dioxide emissions per capita GDP due to different threshold impact in different directions.When the region 's per capita GDP is less than the threshold value, the increase in per capita FDIwill reduce carbon dioxide emissions for the region, but once the region 's per capita GDP isgreater than the threshold value, the increase in per capita FDI in the region will increase carbondioxide emissions. Meanwhile impact on the region 's foreign trade volume per capita carbondioxide emissions per capita GDP due to different threshold impact in different directions. Whenthe region 's per capita GDP is less than the threshold value, the increase in per capita foreign tradevolume will increase on regional carbon dioxide emissions, but once the region 's per capita GDPis greater than the threshold value, the increase in per capita foreign trade volume will increaseregional carbon dioxide emissions. The influence of the threshold effect, the impact on the region's foreign trade volume per capita carbon dioxide emissions per capita differs from the impact ofFDI on regional carbon dioxide emissions.
Use of relevant data from the National total four angles, eastern, central and western regionof the1987-2011decomposition of carbon emission structure effect and technology effectdecomposition. National decomposition results, you can see the effects of structural effectstechnology and energy consumption of primary energy consumption intensity dynamic evolution ofthe contribution of the per capita carbon dioxide emissions.1987-2011annual average energyconsumption structure effect of carbon dioxide emissions per capita contribution of26.054%,while the average energy intensity effects technology for carbon dioxide emissions per capitacontribution of12.687%. Structural effects of primary energy consumption per capita carbondioxide emissions for the impact of significant technical effect is greater than the energy intensityof carbon dioxide emissions per capita impact. For technical effect of carbon dioxide emissions percapita contribution of the dynamic evolution of the eastern region of primary energy consumptionstructure effect and energy consumption intensity. Structural effects of energy consumption1987-2011annual average of the eastern region of carbon dioxide emissions per capita contributionof17.618%, while the average energy intensity effects technology for carbon dioxide emissionsper capita contribution of14.818%. Structural effects of primary energy consumption per capitacarbon dioxide emissions for the impact of technology is not significantly greater than the strengthof the effect of energy consumption per capita carbon dioxide emissions impact. For technicaleffect of carbon dioxide emissions per capita contribution of the dynamic evolution of the centralregion of the structure of primary energy consumption and energy intensity effects.1987-2011annual average energy consumption structure effect of carbon dioxide emissions per capitacontribution of28.885%, while the average energy intensity effects technology for carbon dioxideemissions per capita contribution of8.725%. Structural effects of primary energy consumption percapita carbon dioxide emissions for the impact of significant technical effect is greater than theenergy intensity of carbon dioxide emissions per capita impact. For technical effect of carbondioxide emissions per capita contribution of the dynamic evolution of the western region 's primary energy consumption structure effect and energy consumption intensity.1987-2011annual averageenergy consumption structure effect of carbon dioxide emissions per capita contribution of28.737%, while the average energy intensity effects technology for carbon dioxide emissions percapita contribution of12.214%. Structural effects of primary energy consumption per capitacarbon dioxide emissions for the impact of significant technical effect is greater than the energyintensity of carbon dioxide emissions per capita impact.
In this paper, through the establishment of a dynamic panel data model analysis and inspectionof the FDI, the impact on export trade and technical effects of carbon emission structure effectregion. The empirical results show that FDI, foreign trade structure effects of these two variableson regional carbon emissions, there were significant positive effect. FDI impact on the structure ofthe top five regional effects of carbon emission region are: Shanghai, Beijing, Jiangsu, Shandong,Guangdong. FDI impact on the structural effects of the lowest carbon emissions are five regions:Xinjiang, Inner Mongolia, Fujian, Jiangxi and Hainan. Showing differences in the eastern, centraland western regions from the effects of FDI impact on the structure characteristics of carbonemissions regional perspective. The highest impact on the structure of foreign trade effects ofcarbon emissions are five regions: Guangdong, Zhejiang, Jiangsu, Shanghai, Fujian. Effect onthe structure of foreign trade carbon emissions lowest in five regions affected areas are: Ningxia,Qinghai, Xinjiang, Hainan, Jiangxi. From the effect of foreign trade on the structurecharacteristics of carbon emissions point of view also showed differences in regional eastern,central and western regions., The structural effects of carbon emissions in the western region isgreater than the technical effects, but FDI in the region, the West, foreign trade structure effectsof these two variables affect regional carbon emissions are far lower than the eastern region.Therefore, to further improve FDI, foreign trade structure effects of these two variables affectregional carbon emissions is in a favorable way the western region of carbon emissions reduced.FDI impact of the top five areas of technical effects of carbon emission regions are: Guangdong,Heilongjiang, Sichuan, Liaoning, Beijing. FDI impact of the minimum technical effect of carbonemissions are five regions: Xinjiang, Hainan, Inner Mongolia, Guangxi, Shanxi. Effect of foreigntrade carbon emissions technology areas affected are the top five regions: Heilongjiang, Hainan,Jiangsu, Yunnan and Zhejiang. Effect on the structure of foreign trade carbon emissions lowest infive regions affected areas are: Qinghai, Beijing, Henan, Shandong, Shaanxi. Compared to theeffect of carbon emissions on structural regions, FDI, regional differences in effects of foreigntrade on carbon emissions technology area is not obvious. Therefore, in order to further reduceregional carbon emissions, should be increased in all regions of FDI, the impact effect of foreigntrade on carbon emissions technology areas.
随着中国经济快速增长过程中出口导向型驱动特征愈发明显，其增长代价之一的环境污染问题也愈发严峻。在这一背景下，国际贸易增长与FDI引入的碳排放的影响问题已经成为学术研究焦点。首先，国际贸易的环境效应相关研究框架起始于Grossman and Krueger（1991）的研究。目前学术界将对外贸易增长对环境的影响分别从规模、结构和技术三个角度进行分析1。国内外许多学者运用不同方法从不同角度对国际贸易的环境效应进行了分析和检验，但总结起来存在着两种截然不同的学术观点。一种观点认为国际贸易会带来环境质量的改善，另一种观点则认为贸易增长会导致环境的恶化。比如，Antweiler（2001）运用回归分析方法对贸易的环境效应进行了计量分析，得出国际贸易的增长并非必然带来环境压力；Eliste and Fredriksson（2002）等学者研究认为，贸易增长能够促进环境的改善。
本文拟通过贸易增长与FDI流入环境效应的作用机制来确定其对中国各个地区碳排放的影响路径。传统的理论和经验研究表明，国际贸易增长与FDI流入之间呈替代或互补影响关系。即当贸易壁垒不是主要FDI流入障碍时，作为国际贸易的补充，FDI主要将流入东道国的出口部门。否则，作为国际贸易增长的替代，FDI主要将流入东道国的进口部门。据此逻辑，国际贸易环境效应的作用机制内生于贸易业务增长之中，而相比较而言，FDI流入则发生在贸易业务增长之前，属于微观投资决策主体对于贸易和投资的一种选择性决策。因此，FDI流入环境效应的作用机制决定于其流入动机。基于此，本文将利用现代计量经济学分析方法分析和刻画国际贸易增长与FDI流入对地区碳排放的影响。
论文的框架结构和主要内容如下：
第1章为引言部分，包括问题的提出、研究意义、核心概念的界定、相关问题综述、论文的结构。
第2章外向型经济对环境影响的理论分析。本章将从理论分析的角度，详细阐述外向型经济对环境影响机制和途径，这是本文后续实证检验的理论基础。
第3章外向型经济与地区碳排放的特征及演变路径分析。本章详细分析了各种估计方法的优劣，并通过具体的测算结果比较了两种常用估计方法测算结果的合理性。并在测算的结果基础上分析中国地区碳排放的动态演变特征。本文还考虑到从经济理论上而言，随着经济增长规模的不断扩张，能源消费不断增加，由此导致的碳排放也会因此“发散增长”，微观层面、中观层面和宏观层面的收敛机制使得碳排放出现收敛现象。在收入提高的过程中，随着产业结构向信息化和服务业的演变、清洁技术的应用、环保需求的加强、环境规制的实施以及环保投资的增加等，碳排放量先上升然后逐步下降，理论上会出现收敛的情况，这样碳排放收敛假说成立。在本章中将利用测算得到的中国地区碳排放数据结合面板数据模型方法对上述碳排放收敛假设进行检验。
第4章出口贸易、FDI与地区碳排放的长、短期均衡关系分析。在本章中利用最新的面板协整检验方法——加权对称估计下的异质面板协整检验方法对中国29个地区的出口贸易、FDI和地区碳排放之间的稳定均衡关系进行检验。
第5章出口贸易、FDI与地区碳排放影响的区制效应分析。FDI、出口贸易与地区碳排放影响往往具有结构转变性特征，本章首先对环境Kuznets曲线进行相关的理论分析和阐述。环境Kuznets曲线可能存在着区制效应。但是数据是否支持区制效应的存在，需要利用相关的方法进行检验。本章利用半参数分位数回归理论对环境Kuznets曲线中是否存在着一定的区制效应进行相关的检验。
第6章出口贸易、FDI对地区碳排放的结构效应和技术效应的动态关联性分析。本文利用相关数据从全国加总、东部地区、中部地区、西部地区四个角度分解了到1987-2011碳排放结构效应和技术效应分解结果。通过建立动态面板数据模型分析和检验了FDI、出口贸易对地区碳排放结构效应和技术效应的影响。
第7章是结论和政策建议。
本文的研究发现：
省级二氧化碳排放强度的变化趋势与两种措施相似。在“第八个五年计划”（1991-1995年）和“第十个五年计划”（2001-2005年），系数显示为增加的趋势，并在“九五”计划（1996-2000年）和“十一五”规划（2006-2010年），系数显示为下降趋势。尽管每个时期的变化趋势是稳定的，整体周期的变化趋势并不稳定，而且不会逐渐减少，所以无法得到收敛的一致结论。从估计结果可以看到，在每一种模型设定下，系数都是显著为负。这意味着，省级二氧化碳排放存在着绝对收敛特征。从结果还可以看到，和是显著为正，所以空间效应是显著存在的。回归结果还表明，煤炭消费量达到能源消费总量中，第二产业产值占全行业总产值和能源强度的比值进行统计显著影响二氧化碳排放强度的增长速度，这进一步影响的收敛速度省级二氧化碳排放强度。在本文中，我们分析了中国二氧化碳排放强度的省级收敛。我们建议，空间溢出效应引入到纯动态面板数据模型的空间动态面板数据模型。通过使用此SDPD模型，我们可以避免遗漏变量偏涉及横截面方程和纯动态面板数据方程。我们的实证结果表明：（1）中国省际CO2排放的省级收敛性；（2）有条件的收敛速度比绝对-收敛率较高；（3）省际CO2排放在空间相关性和收敛与考虑空间效应的比率较高；（4）煤炭消费量达到能源消费总量中，第二产业产值占全行业总产值和能源强度的比值进行统计显著影响收敛速度。
通过面板协整模型，可以检验到由于总体而言，中国出口对CO2排量的增加有正向的作用，因此，中国需要采取措施调整出口产业结构和技术两方面。在整体上使出口产业向集约型方向发展，通过制定恰当的产业政策鼓励低能耗出口产业的发展，抑制高能耗出口产业的发展，从而增加低能耗产业在出口产业中的比重，降低高能耗产业在出口产业中的比重。通过碳排放税、碳交易平台等经济手段，积极引导出口产业的生产技术向清洁生产方向发展。从而积极引导出口的结构效应和技术效应在改善环境方面的积极作用，使之逐渐超过规模效应给环境带来的负面影响，从而使经济和环境朝着可持续的方向发展。由于出口对环境的影响对高、中、低排量省份有所差别，因而政策的制定应该具有地区针对性，对高排量省份的政策应较为稳健，而对中、低排量省份的政策力度应该有所加强。在政策制定过程中，应避免过度关注某单一地区而给周围地区带来额外的消极影响，应保证公平性原则。整体上，FDI对CO2排量的增加也有正向的作用，因此，在吸引和接受FDI时应保持更加谨慎的态度，综合考虑FDI对地区的经济、社会影响。考虑到这种正向影响较弱且部分资源大省、沿海、边疆地区的FDI会对环境改善产生积极影响，在这些地区还可以适当鼓励FDI的流入。由于短期中，出口和FDI对CO2排量并没有较强的影响，因而，为了调整出口和CO2排量之间、FDI和CO2排量之间的关系，经济和环境政策以长期政策为宜，短期措施的效果是相当有限的。
本文利用半参数分位数回归理论对环境Kuznets曲线中是否存在着一定的区制效应进行相关的检验。研究发现人均FDI变量系数为正，且在各模型设定下数值大致相同，此意味着该变量在解释人均二氧化碳排放量上有其重要性，但其影响程度较低。而人均对外贸易变量系数值亦为正值，且对人均二氧化碳排放量有着显著的影响。然而，在观察与人均GDP有关的变量后，发现不同模型设定形式下（二次式或者三次式）所得到的系数估计值差异很大，此结果显著有采用半参数回归进行分析的必要。除了分量25%与10%以及分量75%与90%两组彼此之间没有显著差异外，其余则呈现显著差异，及表示在不同分量下，所代表的意义有显著不同。此结果指出分位数回归方法的重要性，也说明不同二氧化碳排放水平下与人均收入两者之间存在着不同的变化。除了分量25%与10%以及分量50%与75%两组彼此间不显著差异之外，其余则存在显著的差异性，同样说明了分位数回归之下，不同程度的二氧化碳排放水平与人均GDP两者间存在着不同的变化。实证结果发现GDP与区域二氧化碳排放正相关，我国各区域高能耗、高污染的增长方式没有改变。2显著为负，说明了区域二氧化碳排放与区域的人均GDP水平之间存在着“倒U”型关系。同时人均FDI对区域二氧化碳排放的影响会由于人均GDP门限值的不同产生不同方向的影响。当区域人均GDP小于门限值时，人均FDI的增加会减少对区域二氧化碳排放，但是一旦区域人均GDP大于门限值时，人均FDI的增加会增加区域二氧化碳排放量。同时人均对外贸易额对区域二氧化碳排放的影响会由于人均GDP门限值的不同产生不同方向的影响。当区域人均GDP小于门限值时，人均对外贸易额的增加会增加对区域二氧化碳排放，但是一旦区域人均GDP大于门限值时，人均对外贸易额的增加会增加区域二氧化碳排放量。在影响作用的门限效应中，人均对外贸易额对区域二氧化碳排放的影响不同于人均FDI对区域二氧化碳排放的影响。
利用相关数据从全国加总、东部地区、中部地区、西部地区四个角度分解了到1987-2011碳排放结构效应和技术效应分解结果。全国分解结果，可以看到一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为26.054%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为12.687%。一次能源消耗的结构效应对于人均二氧化碳排放的影响显著的大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。东部地区一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。东部地区1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为17.618%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为14.818%。一次能源消耗的结构效应对于人均二氧化碳排放的影响并不显著大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。中部地区一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为28.885%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为8.725%。一次能源消耗的结构效应对于人均二氧化碳排放的影响显著的大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。西部地区一次能源消耗的结构效应和能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的动态演化特征。1987-2011年能源消耗的结构效应对人均二氧化碳排放的贡献度的平均值为28.737%，而能源消耗强度的技术效应对于人均二氧化碳排放的贡献度的平均值为12.214%。一次能源消耗的结构效应对于人均二氧化碳排放的影响显著的大于能源消耗强度的技术效应对于人均二氧化碳排放的影响。
本文还通过建立动态面板数据模型分析和检验了FDI、出口贸易对地区碳排放结构效应和技术效应的影响。实证结果表明FDI、对外贸易这两个变量对地区碳排放的结构效应均存在显著的正向影响。FDI对地区碳排放结构效应影响最高的五个地区分别是：上海、北京、江苏、山东、广东。FDI对地区碳排放结构效应影响最低的五个地区分别是：新疆、内蒙古、福建、江西、海南。从FDI对地区碳排放结构效应影响特征来看呈现东、中、西部地区差异。对外贸易对地区碳排放结构效应影响最高的五个地区分别是：广东、浙江、江苏、上海、福建。对外贸易对地区碳排放结构效应影响最低的五个地区分别是：宁夏、青海、新疆、海南、江西。从对外贸易对地区碳排放结构效应影响特征来看也呈现东、中、西部地区差异。中、西部地区的碳排放的结构效应大于技术效应，但是中、西部地区FDI、对外贸易这两个变量对地区碳排放的结构效应影响还远远低于东部地区。所以进一步提高FDI、对外贸易这两个变量对地区碳排放的结构效应影响是降低中、西部地区碳排放的一个有利的途径。FDI对地区碳排放技术效应影响最高的五个地区分别是：广东、黑龙江、四川、辽宁、北京。FDI对地区碳排放技术效应影响最低的五个地区分别是：新疆、海南、内蒙古、广西、山西。对外贸易对地区碳排放技术效应影响最高的五个地区分别是：黑龙江、海南、江苏、云南、浙江。对外贸易对地区碳排放结构效应影响最低的五个地区分别是：青海、北京、河南、山东、陕西。相比对地区碳排放结构效应的影响，FDI、外贸易对地区碳排放技术效应的地区差异并不明显。所以为了进一步降低地区碳排放，应该提高各个地区中FDI、外贸易对地区碳排放技术效应的影响。
The past two decades, global warming has become the world universal attention. Globalclimate change has occurred obviously, the atmosphere constantly getting hot, disasters caused byclimate change are increasing. Global heat waves, droughts frequent; great Antarctic ice began tomelt, there is no ice in the Arctic Ocean; desert area gradually expanded; sea-level rise, resultingin an island nation in danger of being overwhelmed by the crisis; entire ecosystem is badlyaffected.
With the rapid economic growth in China’s export-oriented process-driven feature isbecoming increasingly apparent, environmental pollution costs of its growth is also increasinglygrim. In this context, the impact of carbon emissions growth of international trade and theintroduction of FDI has become the focus of academic research. First, the environmental effects ofinternational trade-related research framework Grossman and Krueger (1991) study began in.Current academic impact of growth on the foreign trade environment is analyzed from the size,structure and technology of the three angles. Many scholars from different angles using differentmethods of international trade on the environmental effects were analyzed and tested, but summedup the existence of two distinct academic perspectives. One view is that international trade willbring about improvements in environmental quality, and the other view is that trade growth willlead to the deterioration of the environment. For example, Antweiler (2001) using regressionanalysis of environmental effects of trade were measured analysis, the growth of international tradedoes not necessarily bring pressure on the environment; scholars Eliste and Fredriksson (2002)consider that the environment can promote the growth of trade improvements.
This paper intends to grow through trade and FDI inflows mechanism to determine theenvironmental effects of its impact on the various regions of China the path of carbon emissions.The traditional theory and empirical studies have shown that the growth of international trade andFDI was an alternative or complementary affect the relationship between inflows. That is, whentrade barriers are not the main obstacle to FDI inflows, as a supplement to international trade, FDIinflows will host the main export sector. Otherwise, as an alternative to the growth of internationaltrade, FDI inflows to the host country’s main import sector. Accordingly logic, the mechanism ofaction of the environmental effects of international trade growth among trading business was born,but comparatively speaking, FDI inflows occurred before the trade business growth, investmentdecisions are the subject of a micro-selectivity for trade and investment decisions. Thus, FDI inflows environmental effects of the decision on its mechanism of action inflows motivation. Basedon this, the paper will analyze and characterize the growth of international trade and FDI inflowsimpact on regional carbon emissions using modern econometric analysis.
Thesis framework and main contents are as follows:
Chapter1is an introduction, including issues raised, research significance, the definition ofcore concepts, review questions, the paper structure.
Chapter2analyzes the export-oriented economy on the theory of environmental impact. Thischapter from the theoretical point of view, the detail-oriented economy mechanisms and pathwaysof environmental impact, which is the theoretical basis of this paper subsequent empirical tests.
Chapter3Characteristics of export-oriented economy and regional carbon emissions andevolution path analysis. This chapter analyzes the pros and cons of various estimation methods, andthrough specific estimation results compared two methods commonly used estimate calculationresults are reasonable. And analyze the dynamic evolution characteristics of carbon emissions inChina on the basis of the results of the calculation. The article also take into account the economictheory, with the continuous expansion of the scale of economic growth, increasing energyconsumption, carbon emissions will therefore resulting " divergent growth", the micro-level,meso-level and macro-level convergence carbon emissions occur mechanism makes convergencephenomenon. Revenue increased in the process, as the industrial structure and the evolution ofinformation technology services, implementation and application of increased environmentalinvestment in clean technology, strengthen environmental protection requirements, environmentalregulations, etc., carbon emissions increased at first and then gradually decline, theoreticallythere will be convergence of circumstances, so carbon emissions convergence hypothesis holds. Inthis chapter, the use of estimates resulting carbon emissions data in China combined with paneldata model approach for the carbon emissions convergence hypothesis testing.
Chapter4long export trade, FDI and regional carbon emissions, the analysis of short-termequilibrium relationship. Using the latest panel cointegration methods in this chapter-theheterogeneous panel cointegration estimation method of weighted symmetric equilibriumrelationship export China29regions, FDI and regions between carbon emissions inspection.
Chapter5export trade, FDI and regional district system effects analysis of the impact ofcarbon emissions. FDI, export trade and the impact of carbon emission regions tend to have astructural change characteristics, environmental Kuznets curve first chapter related theoreticalanalysis and elaboration. Environmental Kuznets curve effect may exist system area. However,whether the data support the existence of district system effects, require the use of related methodsfor testing. This chapter uses the semi-parametric quantile regression theory of environmentalKuznets curve, whether there is a certain district system effects associated test.
Chapter6export trade, FDI analysis of dynamic correlation structure effects and technicaleffects of regional carbon emissions. In this paper, data from the National total four angles, eastern, central and western region of the1987-2011decomposition of carbon emission structureeffect and technology effect decomposition. Through the establishment of a dynamic panel datamodel analysis and inspection of the FDI, the impact on export trade and technical effects of carbonemission structure effect region.
Chapter7, conclusions and policy recommendations.
The findings in this article:
Similar trends provincial carbon dioxide emissions intensity and two measures. In the "Eighth Five-Year Plan "(1991-1995) and the " Tenth Five-Year Plan "(2001-2005), thecoefficient is displayed as an increasing trend, and in the "Ninth Five " program (1996-2000) andthe " Eleventh Five-Year plan"(2006-2010), the coefficient is shown as a downward trend.Although the trend in each period is stable, the overall trend cycle is not stable, and does notdecrease gradually, it is not unanimous conclusions convergent. From the estimation result can beseen, each of the models in the set, the coefficient is significantly negative. This means that thereis an absolute carbon dioxide emissions provincial convergence characteristics. You can also seefrom the results, and is significantly positive, so the effect is significant space exists. Theregression results also show that coal consumption reached total energy consumption, the ratio ofthe total industry output value of the second industry and energy intensity were statisticallysignificant impact on the growth rate of carbon dioxide emissions intensity, which further affectthe convergence rate of the provincial carbon dioxide emissions strength. In this paper, we analyzethe convergence of Chinese provincial-level carbon dioxide emissions intensity. We propose tointroduce spatial spillover effects into the space of pure dynamic panel data model dynamic paneldata model. By using this SDPD model, we can avoid omitted variable bias involved in thecross-sectional equation and pure dynamic panel data equation. Our empirical results show that:(1) Chinese provincial CO2emissions provincial convergence;(2) Conditional convergence ratethan the absolute-higher convergence rate;(3) inter-provincial spatial correlation between CO2emissions and convergence with consider the effect of a higher percentage of space;(4) coalconsumption reached total energy consumption, the ratio of the total industry output value of thesecond industry and energy intensity were statistically significant impact on the convergence rate.
Through panel cointegration model can be tested to due Overall, China 's exports to theincrease of CO2emissions has a positive effect, therefore, China needs to take measures to adjustthe industrial structure of both exports and technology. So on the whole export industry tointensive direction, through the development of appropriate industrial policy to encourage thedevelopment of low-power export industry, inhibit the development of export industries with highenergy consumption, thereby increasing the proportion of low-energy industry in the exportindustry, reducing energy-intensive industries the proportion of the export industry. Through acarbon tax, carbon trading platform of economic means, and actively guide the export industryproduction technology to develop in the direction of cleaner production. Structural effects and techniques in order to actively guide the effect of improving the environmental aspects of exportactive role in making gradual scale over the negative impact to the environment, so that theeconomic and environmental direction towards sustainable development. The impact on theenvironment due to the export of high, medium and low emission provinces vary, and thereforeshould have a regional policy targeted policies for high-emission provinces should be more robust,and policies, and low-emission provinces efforts should be strengthened. In the policy-makingprocess, should avoid excessive attention to a single region and surrounding areas bring additionalnegative impact should ensure fairness principle. Overall, FDI is also on the increase in CO2emissions has a positive effect, therefore, should attract FDI and to accept a more cautious attitudein keeping, considering the economic and social impact of FDI on the region. Given this positiveeffect is weak and some resources in the province, FDI on the environment along the coast, borderareas, improvements have a positive impact in these areas can also be appropriate to encourage FDIinflows. Due to the short-term, exports and FDI are not a strong impact on CO2emissions, andtherefore, in order to adjust the relationship between exports and CO2emissions, FDI and CO2emissions between economic and environmental policies to long-term policy is appropriate, theeffect of short-term measures is quite limited.
In this paper, the use of semi-parametric quantile regression theory of environmental Kuznetscurve, whether there is a certain district system effects associated test. Study found that per capitaFDI variable coefficient is positive, and roughly the same value set in each model, this means thatthe variable is in the interpretation of carbon dioxide emissions per capita have their place, but to alesser degree affected. Foreign Trade Department of the per capita value of the variable is alsopositive, and per capita carbon dioxide emissions have a significant impact. However, afterobserving the per capita GDP-related variables, found under different model specification form(quadratic or cubic type) obtained coefficient estimates vary widely, there is a significant result ofthis semi-parametric regression analysis is necessary. In addition there was no significantdifference in weight of25%and10%and75%and90%weight between the two groups with eachother, the rest showed significant difference, and means under the different components,meanings are significantly different. These results indicate the importance of quantile regressionmethod, also shows different levels of carbon dioxide emissions per capita income between thetwo there are different variations. In addition to weight25%and10%and50%and75%of thecomponents of the two groups are not significantly different from each other, the other is there aresignificant differences, also explains under quantile regression, different levels of carbon dioxideemissions per capita GDP between the two there are different variations. The empirical resultsshow positive GDP and regional carbon dioxide emissions related to our various areas of highenergy consumption, high pollution growth pattern has not changed. Significantly negative,indicating the existence of the "inverted U" shaped relationship between regional GDP per capitacarbon dioxide emissions and the region. Meanwhile the impact of FDI on regional per capita carbon dioxide emissions per capita GDP due to different threshold impact in different directions.When the region 's per capita GDP is less than the threshold value, the increase in per capita FDIwill reduce carbon dioxide emissions for the region, but once the region 's per capita GDP isgreater than the threshold value, the increase in per capita FDI in the region will increase carbondioxide emissions. Meanwhile impact on the region 's foreign trade volume per capita carbondioxide emissions per capita GDP due to different threshold impact in different directions. Whenthe region 's per capita GDP is less than the threshold value, the increase in per capita foreign tradevolume will increase on regional carbon dioxide emissions, but once the region 's per capita GDPis greater than the threshold value, the increase in per capita foreign trade volume will increaseregional carbon dioxide emissions. The influence of the threshold effect, the impact on the region's foreign trade volume per capita carbon dioxide emissions per capita differs from the impact ofFDI on regional carbon dioxide emissions.
Use of relevant data from the National total four angles, eastern, central and western regionof the1987-2011decomposition of carbon emission structure effect and technology effectdecomposition. National decomposition results, you can see the effects of structural effectstechnology and energy consumption of primary energy consumption intensity dynamic evolution ofthe contribution of the per capita carbon dioxide emissions.1987-2011annual average energyconsumption structure effect of carbon dioxide emissions per capita contribution of26.054%,while the average energy intensity effects technology for carbon dioxide emissions per capitacontribution of12.687%. Structural effects of primary energy consumption per capita carbondioxide emissions for the impact of significant technical effect is greater than the energy intensityof carbon dioxide emissions per capita impact. For technical effect of carbon dioxide emissions percapita contribution of the dynamic evolution of the eastern region of primary energy consumptionstructure effect and energy consumption intensity. Structural effects of energy consumption1987-2011annual average of the eastern region of carbon dioxide emissions per capita contributionof17.618%, while the average energy intensity effects technology for carbon dioxide emissionsper capita contribution of14.818%. Structural effects of primary energy consumption per capitacarbon dioxide emissions for the impact of technology is not significantly greater than the strengthof the effect of energy consumption per capita carbon dioxide emissions impact. For technicaleffect of carbon dioxide emissions per capita contribution of the dynamic evolution of the centralregion of the structure of primary energy consumption and energy intensity effects.1987-2011annual average energy consumption structure effect of carbon dioxide emissions per capitacontribution of28.885%, while the average energy intensity effects technology for carbon dioxideemissions per capita contribution of8.725%. Structural effects of primary energy consumption percapita carbon dioxide emissions for the impact of significant technical effect is greater than theenergy intensity of carbon dioxide emissions per capita impact. For technical effect of carbondioxide emissions per capita contribution of the dynamic evolution of the western region 's primary energy consumption structure effect and energy consumption intensity.1987-2011annual averageenergy consumption structure effect of carbon dioxide emissions per capita contribution of28.737%, while the average energy intensity effects technology for carbon dioxide emissions percapita contribution of12.214%. Structural effects of primary energy consumption per capitacarbon dioxide emissions for the impact of significant technical effect is greater than the energyintensity of carbon dioxide emissions per capita impact.
In this paper, through the establishment of a dynamic panel data model analysis and inspectionof the FDI, the impact on export trade and technical effects of carbon emission structure effectregion. The empirical results show that FDI, foreign trade structure effects of these two variableson regional carbon emissions, there were significant positive effect. FDI impact on the structure ofthe top five regional effects of carbon emission region are: Shanghai, Beijing, Jiangsu, Shandong,Guangdong. FDI impact on the structural effects of the lowest carbon emissions are five regions:Xinjiang, Inner Mongolia, Fujian, Jiangxi and Hainan. Showing differences in the eastern, centraland western regions from the effects of FDI impact on the structure characteristics of carbonemissions regional perspective. The highest impact on the structure of foreign trade effects ofcarbon emissions are five regions: Guangdong, Zhejiang, Jiangsu, Shanghai, Fujian. Effect onthe structure of foreign trade carbon emissions lowest in five regions affected areas are: Ningxia,Qinghai, Xinjiang, Hainan, Jiangxi. From the effect of foreign trade on the structurecharacteristics of carbon emissions point of view also showed differences in regional eastern,central and western regions., The structural effects of carbon emissions in the western region isgreater than the technical effects, but FDI in the region, the West, foreign trade structure effectsof these two variables affect regional carbon emissions are far lower than the eastern region.Therefore, to further improve FDI, foreign trade structure effects of these two variables affectregional carbon emissions is in a favorable way the western region of carbon emissions reduced.FDI impact of the top five areas of technical effects of carbon emission regions are: Guangdong,Heilongjiang, Sichuan, Liaoning, Beijing. FDI impact of the minimum technical effect of carbonemissions are five regions: Xinjiang, Hainan, Inner Mongolia, Guangxi, Shanxi. Effect of foreigntrade carbon emissions technology areas affected are the top five regions: Heilongjiang, Hainan,Jiangsu, Yunnan and Zhejiang. Effect on the structure of foreign trade carbon emissions lowest infive regions affected areas are: Qinghai, Beijing, Henan, Shandong, Shaanxi. Compared to theeffect of carbon emissions on structural regions, FDI, regional differences in effects of foreigntrade on carbon emissions technology area is not obvious. Therefore, in order to further reduceregional carbon emissions, should be increased in all regions of FDI, the impact effect of foreigntrade on carbon emissions technology areas.
引文
1Weber, C.L. and G. Peters,2009,.Climate change policy and international trade: Policyconsiderations in the United States,.Energy Policy,37,432-440.
2马欣、李玉娥、仲平，《联合国气候变化框架公约适应委员会职能谈判焦点解析》[J]，《气候变化研究进展》，2012年第8期。
3温室气体主要含二氧化碳、甲烷、氧化亚氮、氟氯碳化物、全氟化物及六氟化物。
4温室气体的生命周期：二氧化碳（50-200年）、甲烷（12-17年）、氧化亚氮（120年）、氟氯碳化物（102年）、全氟化物（260-50000年）及六氟化物（3200年）
5尼古拉斯·斯特恩(Nicholas Stern)，英国人，前世界银行首席经济学家、全球气候变迁政策奠基人、气候经济学之父。
6也称《斯特恩报告》，是尼古拉斯·斯特恩经过一年调研主持完成并于2006年发布的。72007年1月29日-2月1日，政府间气候变化专门委员会第一工作组在巴黎召开了第十次全会，会议通过了第四次评估报告第一工作组报告《气候变化2007：自然科学基础》的决策者摘要，并于2月2日正式发布
8《气候变化2007：自然科学基础》发布[DB/OL].人民网，2007-02-09/2012-03-14http://scitech.people.com.cn/GB/5384994.html
9中国《第二次气候变化国家评估报告》其编制工作于2008年12月启动，于2011年11月15日发布，汇集中国应对气候变化有关科学、技术、经济和社会研究成果，反映中国科学界在气候变化领域的研究进展
10最早可见的政府文件为2003年的英国能源白皮书《我们能源的未来:创建低碳经济》。低碳经济是以低能耗、低污染、低排放为基础的经济模式，是人类社会继农业文明、工业文明之后的又一次重大进步。低碳经济实质是能源高效利用、清洁能源开发、追求绿色GDP的问题，核心是能源技术和减排技术创新、产业结构和制度创新以及人类生存发展观念的根本性转变
11中国于1998年5月29日签署《京都议定书》，于2002年9月3日核准，2005年2月16日正式生效。
12碳交易是《京都议定书》为促进全球减少温室气体排放，以国际公法作为依据的温室气体排减量交易。在6种被要求排减的温室气体中， CO2为最大宗，所以这种交易以每吨二氧化碳当量（tCO2e）为计算单位，所以通称为“碳交易”。其交易市场称为碳市（Carbon Market）。碳交易有配额型交易（Allowance-basedtransactions）和项目型交易（Project-based transactions）两种形式。
13碳关税是指主权国家或地区对高耗能产品进口征收的二氧化碳排放特别关税。主要针对进口产品中的碳排放密集型产品,如铝、钢铁、水泥、玻璃制品等产品而进行的关税税收。
14Weber, C.L. and G. Peters,2009,.Climate change policy and international trade: Policy1c5onsiderations in the United States,.Energy Policy,37,432-440.Dauda, R. O.S.,(2007). The impact of FDI on Nigeria’s economic growth: Trade policymatters.Journal of Business and Policy Research,3,11-26.
16当然有关环境质量的相关研究中还包含了SO2排放等指标。
17有关EKC研究的文献综述可见Stern (1998)、Dinda (2004)等。
18Helmi, H., Sbia, R., Abdelaziz, H. and Hakimi, W. K.,(2013). Multivariate Granger causalitybetween foreign direct investment and economic growth in Tunisia. Economic Bulletin,33,1193-1203.
19邹至庄著《中国经济转型》，北京：中国人民大学出版社，2010年第一版。
20龚曙明编著《宏观经济统计分析——理论、方法与实务》，北京，中国水利水电出版社，2010年第一版。
21IPCC即International Panel on Climate Change （政府间气候变化问题小组），IPCC2006指IPCC2006年国家温室气体清单指南
22CDIAC即Carbon Dioxide Information Analysis Center（美国能源部CO2信息分析中心）
23薛荣久著，《世界贸易组织（WTO）教程》，北京，对外经济贸易大学出版社，2009第二版。
24关于更详细的论述，请参见Koenker和Hallock（2001）与Koenker（2005）。
25钱伯海、黄良文主编《统计学》，四川人民出版社，1992年第一版。
26指数公理化方法的阐述详见Diewert(1999),“Essays in Index Number Theory”, Emerald Group PublishingLimited.
27详见Kawasea, R., Matsuokaa, Y., Fujino, J.,2006. Decomposition analysis of CO2emission in
28巴尔塔基著《面板数据计量经济分析》，北京：机械工业出版社，2010年第1版。
[1]巴里·诺顿.中国经济:转型与增长[M].上海:上海人民出版社,2010.
[2]查尔斯·D·科尔斯塔德.环境经济学[M].北京:中国人民大学出版社,2011.
[3]陈波.碳排放权交易市场的设计原理与实践研究[M].北京:中国经济出版社,2013.
[4]谌伟.上海市工业碳排放与经济增长的关系及影响因素研究[M].上海:同济大学出版社,2012.
[5]高铁梅.计量经济分析方法与建模——EViews应用及实例[M].北京:清华大学出版社,2009.
[6]葛全胜.中国碳排放的历史与现状[M].北京:气象出版社,2011.
[7]宫本宪一.环境经济学[M].北京:生活·读书·新知三联书店出版社,2004.
[8]龚曙明.宏观经济统计分析——理论、方法与实务[M].北京:中国水利水电出版社,2010.
[9]李克国.环境经济学[M].北京:中国环境科学出版社,2007.
[10]钱伯海,黄良文.统计学[M].成都:四川人民出版社,1992.
[11]田立新.能源碳排放系统分析[M].北京:科学出版社,2013.
[12]王锋.中国碳排放增长的驱动因素及减排政策评价[M].北京:经济科学出版社,2011.
[13]王小鲁,樊纲.中国经济增长的可持续性——跨世纪的回顾与展望[M].北京:经济科学出版社,2000.
[14]王毅刚.碳排放交易制度的中国道路[M].北京:经济管理出版社,2011.
[15]王铮.中国碳排放控制策略研究[M].北京:科学出版社,2013.
[16]魏一鸣.中国能源报告碳排放研究[M].北京:科学出版社,2010.
[17]许广月.中国能源消费碳排放与经济增长关系的研究[M].北京:中国书籍出版社,2013.
[18]薛荣久.世界贸易组织（WTO）教程[M].北京:对外经济贸易大学出版社,2009.
[19]张世英,樊智.协整理论与波动模型——金融时间序列分析及应用[M].北京:清华大学出版社,2004.
[20]张晓峒.计量经济分析[M].北京:经济科学出版社,2000.
[21]张云.国际碳排放交易与中国排放权出口规模管理[M].北京:中国经济出版社,2012.
[22]赵成柏.中国经济增长中碳排放强度演化机制[M].北京:经济科学出版社,2013.
[23]中国科学院可持续发展战略研究组.可持续发展战略报告——探索中国特色的低碳道路
[M].北京:科学出版社,2009.
[24]中国气候变化国别研究组.中国气候变化国别研究[M].北京:清华大学出版社,2000.
[25]朱勤.中国人口、消费与碳排放研究[M].上海:复旦大学出版社,2011.
[26]庄贵阳.低碳经济:气候变化背景下中国的发展之路[M].北京:气象出版社,2007.
[27]邹至庄.中国经济转型[M].北京:中国人民大学出版社,2010.中文期刊文献：
[1]鲍健强等.低碳经济:人类经济发展方式的新变革[J].中国工业经济,2008（4）.
[2]蔡防,都阳,王美艳.经济发展方式转变与节能减排内在动力[J].经济研究,2008（6）.
[3]曾贤刚,庞含霜.我国各省区CO2排放状况、趋势及其减排对策[J].中国软科学,2009（11）.
[4]查冬兰.能源效率与二氧化碳排放的差异性:基于Kaya因素分解[J].系统工程,2007（11）.
[5]查冬兰.我国工业CO2排放影响因素差异性研究——基于高耗能行业与中低耗能行业[J].财贸研究,2008（1）.
[6]陈红蕾,陈秋峰.我国贸易自由化环境效应的实证分析[J].国际贸问题,2007（7）.
[7]陈诗一.2009能源消耗、二氧化碳排放与中国工业的可持续发展[J].经济研究,2009（4）.
[8]段红霞.气候变化经济学与气候政策[J].经济学家,2009（8）.
[9]高鹏飞,陈文颖,何建坤.中国的二氧化碳边际减排成本[J].清华大学学报:自然科学版,2004（9）.
[10]何健坤,刘滨.有关全球气候变化问题上的公平性分析[J].中国人口资源与环境,2004（6）.
[11]何建坤,刘滨.作为温室气体排放衡量指标的碳排放强度分析[J].清华大学学报:自然科学版,2004（6）.
[12]胡初枝等.中国碳排放特征与其动态演进分析[J].中国人口资源与环境,2008（3）.
[13]李斌,汤铸,陈开军.贸易自由化对环境污染影响的实证分析[J].商业研究,2006（10）.
[14]李国志,李宗植.中国农业能源消费碳排放因素分解实证分析——基于LMDI模型[J].农业技术经济,2010（10）.
[15]林伯强.中国二氧化碳的环境库兹涅茨曲线预测及影响因素分析[J].管理世界,2009（4）.
[16]刘慧,成升魁.人类经济活动影响碳排放的国际研究动态.[J].地理科学进展,2002（5）.
[17]刘强,庄幸等.中国出口贸易中的载能量及碳排放量分析[J].中国工业经济,2008（8）.
[18]潘仁飞,陈柳钦.能源结构变化与中国碳减排目标实现[J].经济研究参考,2011（59）.
[19]彭海珍.关于贸易自由化对中国环境影响的分析[J].财贸研究,2006（4）.
[20]齐晔,惠民,徐明.中国进出口贸易中的隐含碳估算[J].中国人口资源与环境,2008（3）.
[21]沙文兵,石涛.外商直接投资的环境效应——基于中国省级面板数据的实证分析[J].世界经济研究,2006（6）.
[22]宋帮英,苏方林.我国省域碳排量与经济发展的GWR实证研究[J].财经科学,2010（4）.
[23]宋德勇.碳排放影响因素分解及其周期性波动研究[J].中国人口资源与环境,2009（3）.
[24]孙小羽,臧新.中国出口贸易的能耗效应和环境效应的实证分析——基于混合单位投入产出模型[J].数量经济技术经济研究,2009（4）.
[25]王铮,朱永彬.我国各省区碳排放量状况及减排对策研究[J].中国科学院院刊,2008（2）.
[26]吴献金,邓杰.贸易自由化、经济增长对碳排放的影响[J].中国人口资源与环境,2011（1）.
[27]谢文武.开放经济对碳排放的影响——基于中国地区与行业面板数据的实证检验[J].浙江大学学报,2011（5）.
[28]徐国泉,刘则渊.中国碳排放的因素分解模型及实证分析:1995-2004[J].中国人口资源与环境,2006（6）.
[29]许广月,宋德勇.我国出口贸易、经济增长与碳排放关系的实证研究[J].国际贸易问题,2010（1）.
[30]杨子晖.经济增长、能源消费与二氧化碳排放的动态关系研究[J].世界经济,2011（6）.
[31]袁富华.低碳经济约束下的中国潜在经济增长[J].经济研究,2010（8）.
[32]张彬,姚鄉.基于模糊聚类的中国分省碳排放初歩研究[J].中国人口资源与环境,2011（1）.
[33]张雷.经济发展对碳排放的影响[J].地理学报,2003（4）.
[34]张连众,朱坦,李慕菡.贸易自由化对我国环境污染的影响分析[J].南开经济研究,2003（3）.
[35]赵云君,文启湘.环境库兹涅茨曲线及其在我国的修正[J].经济学家,2004（5）.
[36]政府间气候变化专门委员会第四次评估报告第一工作组.气候变化2007：自然科学基础[J].世界环境,2007（2）.
[37]郑林昌.中国省域低碳发展水平及其空间过程评价[J].中国人口资源及环境,2011（7）.
[38]邹麒,刘辉煌.外商投资和贸易自由化的碳排放效应分析[J].经济与管理研究,2011（4）.
[39]邹秀萍,陈劭锋.中国省级区域碳排放影响因素的实证分析[J].生态经济,2009（3）.
[1] Aldy, J.E.,2006. Per Capita Carbon Dioxide Emissions: Convergence or Divergence?[J].Journal of Econometrics,33(1).
[2] Anselin, L.,1998. Spatial Econometrics: Methods and Models[M]. Kluwer AcademicPublishers.
[3] Antweiler, W., Copeland, B.R., Taylor, M.S.,2001. Is free trade good for the
[4] Baldwin, R. E.1994.The Effects of Trade and Foreign Direct Investment on Employment andRelative Wages, Working paper, OECD.
[5] Barro, R.J., Sala-i-Martin, X.,1992. Convergence[J]. Journal of Political Economics,100(3).
[6] Barro, R.J., Sala-i-Martin, X.,1995. Economic Growth[M]. McGraw Hill, New York.
[7] Bernard, A.B., Durlauf, S.N.,1996. Interpreting tests of the convergence hypothesis.[J].Journal of Economics,71(3).
[8] Birdsall, N.&Wheeler, D.1993.Trade policy and industrial pollution in Latin America:where are the pollution havens?[J]. Journal of Environment&Development,2(1).
[9] Bruneau, J.F. Renzetti, S.J.,2009. Greenhouse Gas Intensity in Canada: A Look at HistoricalTrends[J]. Canadian Public Policy,23(1).
[10] Cole, M.A.2004.Trade, the Pollution Haven Hypothesis and the Environmental KuznetsCurve: Examining the Linkages[J].Ecological Economics,48(1).
[11] Cole, M.A., Elliott, R.J.R, Shimamoto, K.,2005. Inustrial characteristics, environmentalregulations and air pollution: An analysis of the UK manufacturing sector[J].Journal ofEnvironmental Economics and Management,50(1).
[12] Cole, M.A., Elliott, R.J.R., Fredriksson, P.G.,2006. Endogenous pollution havens: Does FDIin.uence environmental regulations?[J].Scandinavian Journal of Economics108(1).
[13] Conley, T.G., Ligon, E.,2002. Economic distance and cross-country spillovers[J]. Journal ofEconnomic Growth,7(2).
[14] Copeland, B.R., Taylor, M.S.,2004. Trade, growth and the environment[J]. Journal ofEconomic Literature,42(1).
[15] Damania, R., Fredriksson, P. G., List, J.,2003. Trade Liberalization, Corruption, andEnvironmental Policy Formation: Theory and Evidence[J]. Journal of EnvironmentalEconomics and Management,46(3).
[16] Dasgupta, S., Laplante, B., Wang, H.&Wheeler, D.2002. Confronting the EnvironmentalKuznets Curve[J], Journal of Economic Perspectives,16(1).
[17] Dean, J.,2002. Does trade liberalization harm the environment? A new test[J]. CanadianJournal of Economics,35(4).
[18] Dinda, S.2004. Environmental Kuznets curve hypothesis: a Survey[J], Ecological Economics,49(1).
[19] Dua, A.&Esty, D. C.1997. Sustaining the Asia Pacific Miracle: Environmental Protectionand Economic Integration.[M] Washington DC: Institute for International Economics.
[20] Ederington, J., Minier, J.,2003. Is environmental policy a secondary trade barrier? Anempirical analysis[J]. Canadian Journal of Economics,36(1).
[21] Eskeland, G.S., Harrison, A.E.,2003.Moving to greener pastures multinationals and thepollution haven hypothesis[J]. Journal of Development Economics,70(1).
[22] Esty, D.C.&Gentry, B.S.1997. Foreign Investment, Globalisation and Environment,Globalisation and Environment: Preliminary Perspectives, OECD, Paris.
[23] Fredriksson, P. G.,Svensson, J.,2003. Political instability, corruption and policy formation:the case of environmental policy[J]. Journal of Public Economics,87(8).
[24] Grossman, G.M.&Krueger, A.B.1991. Environmental Impacts of a North American FreeTrade Agreement', in Garber, P.(ed.) The U.S.-Mexico Free Trade Agreement, Cambridge,MA: MIT Press.
[25] Grossman, G.M.&Krueger, A.B.1993. Environmental Impacts of a North American FreeTrade Agreement', in Garber, P.(ed.) The U.S.-Mexico Free Trade Agreement, Cambridge,MA: MIT Press.
[26] Grossman, G.M.&Krueger, A.B.1995. Economic Growth and the Environment[J].TheQuarterly Journal of Economics,110(2).
[27] Grossman, G.M., Krueger, A.B.,1993. Environmental impacts of a NorthAmerican free tradeagreement,.in: Garber, P.M.(Ed), The Mexico-U.S. FreeTrade Agreement, MIT Press,Cambridge.
[28] He, J.,2006. Pollution Haven Hypothesis and environmental impacts of Foreign DirectInvestment: The Case of Industrial Emissions of Sulphur Dioxide (SO2)in ChineseProvinces[J]. Ecological Economics,60(2).
[29] Jaffe, A. B., Peterson, S. R., Portney, P. R., Stavins, R. N.,1995. Environmental regulationsand the competitiveness of U.S. manufacturing: What does the evidence tell us?[J].Journal ofEconomic Literature,33(1).
[30] Jensen, V.1996. The Pollution Haven Hypothesis and the Industrial Flight Hypothesis: SomePerspectives on Theory and Empirics, Working Paper, Centre for Development and theEnvironment, University of Oslo.
[31] Kao, C.&Chiang, M.2001. On the Estimation and Inference of a Cointegrated Regression inPanel Data[J], Advances in Econometrics,15.
[32] Kao, C.1999. Spurious Regression and Residual-based Tests for Cointegration in PanelData[J], Journal of Econometrics,90(1).
[33] Karaman rsal, D. D.2008. Comparison of Panel Cointegration Tests[J].Economics Bulletin,3(1).
[34] Keller, W., Levinson, A.,2002. Pollution abatement costs and foreign direct investmentin.ows to U.S. States[J].Review of Economics and Statistics,84(4).
[35] Kuznets, S.1955. Economic Growth and Income Inequality[J], The American EconomicReview,45(1).
[36] Levinson, A., Taylor, M.S.,2008. Unmasking the Pollution Haven Effect[J]. InternationalEconomic Review,49(1).
[37] List, J.A., Co, C.,2000. The E¤ects of Environmental Regulations on Foreign DirectInvestment[J]. Journal of Environmental Economics and Management40(1).
[38] Lucas, R., Wheeler, D.&Hettige, H.,1992. Economic Development, EnvironmentalRegulation and the International Migration of Toxic Industrial Pollution'. In Low, P.(ed.)International Trade and the Environment, World Bank Discussion Papers No.159,Washington, DC.
[39] Mani, M.&Wheeler, D.1998. In Search of Pollution Havens? Dirty Industry in the WorldEconomy1960–1995[J], The Journal of Environment Development,7(3).
[40] McCoskey, S.&Kao, C.1998. A Residual-Based Test of the Null of Cointegration in PanelData[J]. Econometric Reviews,17(1).
[41] Mohammed, A,2003. Foreign Direct Investment and the Environment: Pollution HavenHypothesis, The Eight Annual Conference on Global Economic Analysis[J].Journal ofPolitical Economy,32(1).
[42] Nahman, A.&Antrobus, G.2005. The Environmental Kuznets Curve: A LiteratureSurvey[J].South African Journal of Economics,73(1).
[43] Panayotou, T.1993. Empirical Tests and Policy Analysis of Environmental Degradation atDifferent Stages of Economic Development', ILO, Technology and Employment Programme,Geneva.
[44] Panayotou, T.2000.'Economic Growth and the Environment, Working Paper No.56,Center for International Development.
[45] Pedroni, P.1995. Panel Cointegration: Asymptotic and Finite Sample Properties of PooledTime Series Tests, with an Application to the PPP Hypothesis, Working Paper, Series inEconomics95-013, Indiana University.
[46] Pedroni, P.1999. Critical Values for Cointegration Tests in Heterogeneous Panels withMultiple Regressors[J].Oxford Bulletin of Economics and Statistics,61(1).
[47] Pedroni, P.2004. Panel Cointegration: Asymptotic and Finite Sample Properties of PooledTime Series Tests, with an Application to the PPP Hypothesis[J]. Econometric Theory,20(1).
[48] Perkins, R.&Neumayer, E.2009. Transnational Linkages and the Spillover ofEnvironment-efficiency into Developing Countries'[J].Global Environmental Change,19(3).
[49] Selden, T. M.&Song, D.1994.'Environmental Quality and Development: Is there a KuznetsCurve for Air Pollution?[J].Journal of Environmental Economics and Management,27(2).
[50] Shafik, N.&Bandyopadhyay, S.1992. Economic Growth and Environmental Quality: TimeSeries and Crosscountry Evidence', Background Paper for the World Development Report,The World Bank, Washington, DC.
[51] Smarzynska Javorcik, B., Wei, S-J.,2004. Pollution havens and foreign direct investment:Dirty secret or popular myth?[J].Contributions to Economic Analysis and Policy,3(2).
[52] Smarzynska, B.&Wei, S.2001. Pollution Havens and the Location of Foreign Directinvestment: Dirty Secret or Popular Myth?, Washington, D.C., International TradeTeam-Development Economics Research Group, The World Bank. Mimeo.
[53] Stern, D.I.1998. Progress on the environmental Kuznets curve?[J]. Environment andDevelopment Economics,3(2).
[54] Stern, N.2008. The economics of climate change[J].America Economic Review,98(3).
[55] Westerlund, J.&Basher, S.A.2008. Mixed Signals among Tests for Panel Cointegration[J].Economic Modelling,25(1).
[56] Westerlund, J.2005.New Simple Tests for Panel Cointegration[J]. Econometric Reviews,24(2).
[57] Westerlund, J.2006. Reducing the Size Distortions of the Panel LM Test forCointegration[J].Economics Letters,90(3).
[58] Westerlund, J.2007. Testing for Error Correction in Panel Data[J].Oxford Bulletin ofEconomics and Statistics,69(2).
[59] Westerlund, J.2008. Panel Cointegration Tests of the Fisher Hypothesis[J].Journal of AppliedEconometrics,23(3).
[60] Wheeler, D.,2001. Racing to the bottom? Foreign investment and air quality in developingcountries[J]. Journal of Environment and Development,10(3).
[61] Xing, Y., Kolstad, C.,2002. Do lax environmental regulations attract foreign investment?[J].Environmental and Resource Economics,21(1).
2马欣、李玉娥、仲平，《联合国气候变化框架公约适应委员会职能谈判焦点解析》[J]，《气候变化研究进展》，2012年第8期。
3温室气体主要含二氧化碳、甲烷、氧化亚氮、氟氯碳化物、全氟化物及六氟化物。
4温室气体的生命周期：二氧化碳（50-200年）、甲烷（12-17年）、氧化亚氮（120年）、氟氯碳化物（102年）、全氟化物（260-50000年）及六氟化物（3200年）
5尼古拉斯·斯特恩(Nicholas Stern)，英国人，前世界银行首席经济学家、全球气候变迁政策奠基人、气候经济学之父。
6也称《斯特恩报告》，是尼古拉斯·斯特恩经过一年调研主持完成并于2006年发布的。72007年1月29日-2月1日，政府间气候变化专门委员会第一工作组在巴黎召开了第十次全会，会议通过了第四次评估报告第一工作组报告《气候变化2007：自然科学基础》的决策者摘要，并于2月2日正式发布
8《气候变化2007：自然科学基础》发布[DB/OL].人民网，2007-02-09/2012-03-14http://scitech.people.com.cn/GB/5384994.html
9中国《第二次气候变化国家评估报告》其编制工作于2008年12月启动，于2011年11月15日发布，汇集中国应对气候变化有关科学、技术、经济和社会研究成果，反映中国科学界在气候变化领域的研究进展
10最早可见的政府文件为2003年的英国能源白皮书《我们能源的未来:创建低碳经济》。低碳经济是以低能耗、低污染、低排放为基础的经济模式，是人类社会继农业文明、工业文明之后的又一次重大进步。低碳经济实质是能源高效利用、清洁能源开发、追求绿色GDP的问题，核心是能源技术和减排技术创新、产业结构和制度创新以及人类生存发展观念的根本性转变
11中国于1998年5月29日签署《京都议定书》，于2002年9月3日核准，2005年2月16日正式生效。
12碳交易是《京都议定书》为促进全球减少温室气体排放，以国际公法作为依据的温室气体排减量交易。在6种被要求排减的温室气体中， CO2为最大宗，所以这种交易以每吨二氧化碳当量（tCO2e）为计算单位，所以通称为“碳交易”。其交易市场称为碳市（Carbon Market）。碳交易有配额型交易（Allowance-basedtransactions）和项目型交易（Project-based transactions）两种形式。
13碳关税是指主权国家或地区对高耗能产品进口征收的二氧化碳排放特别关税。主要针对进口产品中的碳排放密集型产品,如铝、钢铁、水泥、玻璃制品等产品而进行的关税税收。
14Weber, C.L. and G. Peters,2009,.Climate change policy and international trade: Policy1c5onsiderations in the United States,.Energy Policy,37,432-440.Dauda, R. O.S.,(2007). The impact of FDI on Nigeria’s economic growth: Trade policymatters.Journal of Business and Policy Research,3,11-26.
16当然有关环境质量的相关研究中还包含了SO2排放等指标。
17有关EKC研究的文献综述可见Stern (1998)、Dinda (2004)等。
18Helmi, H., Sbia, R., Abdelaziz, H. and Hakimi, W. K.,(2013). Multivariate Granger causalitybetween foreign direct investment and economic growth in Tunisia. Economic Bulletin,33,1193-1203.
19邹至庄著《中国经济转型》，北京：中国人民大学出版社，2010年第一版。
20龚曙明编著《宏观经济统计分析——理论、方法与实务》，北京，中国水利水电出版社，2010年第一版。
21IPCC即International Panel on Climate Change （政府间气候变化问题小组），IPCC2006指IPCC2006年国家温室气体清单指南
22CDIAC即Carbon Dioxide Information Analysis Center（美国能源部CO2信息分析中心）
23薛荣久著，《世界贸易组织（WTO）教程》，北京，对外经济贸易大学出版社，2009第二版。
24关于更详细的论述，请参见Koenker和Hallock（2001）与Koenker（2005）。
25钱伯海、黄良文主编《统计学》，四川人民出版社，1992年第一版。
26指数公理化方法的阐述详见Diewert(1999),“Essays in Index Number Theory”, Emerald Group PublishingLimited.
27详见Kawasea, R., Matsuokaa, Y., Fujino, J.,2006. Decomposition analysis of CO2emission in
28巴尔塔基著《面板数据计量经济分析》，北京：机械工业出版社，2010年第1版。
[1]巴里·诺顿.中国经济:转型与增长[M].上海:上海人民出版社,2010.
[2]查尔斯·D·科尔斯塔德.环境经济学[M].北京:中国人民大学出版社,2011.
[3]陈波.碳排放权交易市场的设计原理与实践研究[M].北京:中国经济出版社,2013.
[4]谌伟.上海市工业碳排放与经济增长的关系及影响因素研究[M].上海:同济大学出版社,2012.
[5]高铁梅.计量经济分析方法与建模——EViews应用及实例[M].北京:清华大学出版社,2009.
[6]葛全胜.中国碳排放的历史与现状[M].北京:气象出版社,2011.
[7]宫本宪一.环境经济学[M].北京:生活·读书·新知三联书店出版社,2004.
[8]龚曙明.宏观经济统计分析——理论、方法与实务[M].北京:中国水利水电出版社,2010.
[9]李克国.环境经济学[M].北京:中国环境科学出版社,2007.
[10]钱伯海,黄良文.统计学[M].成都:四川人民出版社,1992.
[11]田立新.能源碳排放系统分析[M].北京:科学出版社,2013.
[12]王锋.中国碳排放增长的驱动因素及减排政策评价[M].北京:经济科学出版社,2011.
[13]王小鲁,樊纲.中国经济增长的可持续性——跨世纪的回顾与展望[M].北京:经济科学出版社,2000.
[14]王毅刚.碳排放交易制度的中国道路[M].北京:经济管理出版社,2011.
[15]王铮.中国碳排放控制策略研究[M].北京:科学出版社,2013.
[16]魏一鸣.中国能源报告碳排放研究[M].北京:科学出版社,2010.
[17]许广月.中国能源消费碳排放与经济增长关系的研究[M].北京:中国书籍出版社,2013.
[18]薛荣久.世界贸易组织（WTO）教程[M].北京:对外经济贸易大学出版社,2009.
[19]张世英,樊智.协整理论与波动模型——金融时间序列分析及应用[M].北京:清华大学出版社,2004.
[20]张晓峒.计量经济分析[M].北京:经济科学出版社,2000.
[21]张云.国际碳排放交易与中国排放权出口规模管理[M].北京:中国经济出版社,2012.
[22]赵成柏.中国经济增长中碳排放强度演化机制[M].北京:经济科学出版社,2013.
[23]中国科学院可持续发展战略研究组.可持续发展战略报告——探索中国特色的低碳道路
[M].北京:科学出版社,2009.
[24]中国气候变化国别研究组.中国气候变化国别研究[M].北京:清华大学出版社,2000.
[25]朱勤.中国人口、消费与碳排放研究[M].上海:复旦大学出版社,2011.
[26]庄贵阳.低碳经济:气候变化背景下中国的发展之路[M].北京:气象出版社,2007.
[27]邹至庄.中国经济转型[M].北京:中国人民大学出版社,2010.中文期刊文献：
[1]鲍健强等.低碳经济:人类经济发展方式的新变革[J].中国工业经济,2008（4）.
[2]蔡防,都阳,王美艳.经济发展方式转变与节能减排内在动力[J].经济研究,2008（6）.
[3]曾贤刚,庞含霜.我国各省区CO2排放状况、趋势及其减排对策[J].中国软科学,2009（11）.
[4]查冬兰.能源效率与二氧化碳排放的差异性:基于Kaya因素分解[J].系统工程,2007（11）.
[5]查冬兰.我国工业CO2排放影响因素差异性研究——基于高耗能行业与中低耗能行业[J].财贸研究,2008（1）.
[6]陈红蕾,陈秋峰.我国贸易自由化环境效应的实证分析[J].国际贸问题,2007（7）.
[7]陈诗一.2009能源消耗、二氧化碳排放与中国工业的可持续发展[J].经济研究,2009（4）.
[8]段红霞.气候变化经济学与气候政策[J].经济学家,2009（8）.
[9]高鹏飞,陈文颖,何建坤.中国的二氧化碳边际减排成本[J].清华大学学报:自然科学版,2004（9）.
[10]何健坤,刘滨.有关全球气候变化问题上的公平性分析[J].中国人口资源与环境,2004（6）.
[11]何建坤,刘滨.作为温室气体排放衡量指标的碳排放强度分析[J].清华大学学报:自然科学版,2004（6）.
[12]胡初枝等.中国碳排放特征与其动态演进分析[J].中国人口资源与环境,2008（3）.
[13]李斌,汤铸,陈开军.贸易自由化对环境污染影响的实证分析[J].商业研究,2006（10）.
[14]李国志,李宗植.中国农业能源消费碳排放因素分解实证分析——基于LMDI模型[J].农业技术经济,2010（10）.
[15]林伯强.中国二氧化碳的环境库兹涅茨曲线预测及影响因素分析[J].管理世界,2009（4）.
[16]刘慧,成升魁.人类经济活动影响碳排放的国际研究动态.[J].地理科学进展,2002（5）.
[17]刘强,庄幸等.中国出口贸易中的载能量及碳排放量分析[J].中国工业经济,2008（8）.
[18]潘仁飞,陈柳钦.能源结构变化与中国碳减排目标实现[J].经济研究参考,2011（59）.
[19]彭海珍.关于贸易自由化对中国环境影响的分析[J].财贸研究,2006（4）.
[20]齐晔,惠民,徐明.中国进出口贸易中的隐含碳估算[J].中国人口资源与环境,2008（3）.
[21]沙文兵,石涛.外商直接投资的环境效应——基于中国省级面板数据的实证分析[J].世界经济研究,2006（6）.
[22]宋帮英,苏方林.我国省域碳排量与经济发展的GWR实证研究[J].财经科学,2010（4）.
[23]宋德勇.碳排放影响因素分解及其周期性波动研究[J].中国人口资源与环境,2009（3）.
[24]孙小羽,臧新.中国出口贸易的能耗效应和环境效应的实证分析——基于混合单位投入产出模型[J].数量经济技术经济研究,2009（4）.
[25]王铮,朱永彬.我国各省区碳排放量状况及减排对策研究[J].中国科学院院刊,2008（2）.
[26]吴献金,邓杰.贸易自由化、经济增长对碳排放的影响[J].中国人口资源与环境,2011（1）.
[27]谢文武.开放经济对碳排放的影响——基于中国地区与行业面板数据的实证检验[J].浙江大学学报,2011（5）.
[28]徐国泉,刘则渊.中国碳排放的因素分解模型及实证分析:1995-2004[J].中国人口资源与环境,2006（6）.
[29]许广月,宋德勇.我国出口贸易、经济增长与碳排放关系的实证研究[J].国际贸易问题,2010（1）.
[30]杨子晖.经济增长、能源消费与二氧化碳排放的动态关系研究[J].世界经济,2011（6）.
[31]袁富华.低碳经济约束下的中国潜在经济增长[J].经济研究,2010（8）.
[32]张彬,姚鄉.基于模糊聚类的中国分省碳排放初歩研究[J].中国人口资源与环境,2011（1）.
[33]张雷.经济发展对碳排放的影响[J].地理学报,2003（4）.
[34]张连众,朱坦,李慕菡.贸易自由化对我国环境污染的影响分析[J].南开经济研究,2003（3）.
[35]赵云君,文启湘.环境库兹涅茨曲线及其在我国的修正[J].经济学家,2004（5）.
[36]政府间气候变化专门委员会第四次评估报告第一工作组.气候变化2007：自然科学基础[J].世界环境,2007（2）.
[37]郑林昌.中国省域低碳发展水平及其空间过程评价[J].中国人口资源及环境,2011（7）.
[38]邹麒,刘辉煌.外商投资和贸易自由化的碳排放效应分析[J].经济与管理研究,2011（4）.
[39]邹秀萍,陈劭锋.中国省级区域碳排放影响因素的实证分析[J].生态经济,2009（3）.
[1] Aldy, J.E.,2006. Per Capita Carbon Dioxide Emissions: Convergence or Divergence?[J].Journal of Econometrics,33(1).
[2] Anselin, L.,1998. Spatial Econometrics: Methods and Models[M]. Kluwer AcademicPublishers.
[3] Antweiler, W., Copeland, B.R., Taylor, M.S.,2001. Is free trade good for the
[4] Baldwin, R. E.1994.The Effects of Trade and Foreign Direct Investment on Employment andRelative Wages, Working paper, OECD.
[5] Barro, R.J., Sala-i-Martin, X.,1992. Convergence[J]. Journal of Political Economics,100(3).
[6] Barro, R.J., Sala-i-Martin, X.,1995. Economic Growth[M]. McGraw Hill, New York.
[7] Bernard, A.B., Durlauf, S.N.,1996. Interpreting tests of the convergence hypothesis.[J].Journal of Economics,71(3).
[8] Birdsall, N.&Wheeler, D.1993.Trade policy and industrial pollution in Latin America:where are the pollution havens?[J]. Journal of Environment&Development,2(1).
[9] Bruneau, J.F. Renzetti, S.J.,2009. Greenhouse Gas Intensity in Canada: A Look at HistoricalTrends[J]. Canadian Public Policy,23(1).
[10] Cole, M.A.2004.Trade, the Pollution Haven Hypothesis and the Environmental KuznetsCurve: Examining the Linkages[J].Ecological Economics,48(1).
[11] Cole, M.A., Elliott, R.J.R, Shimamoto, K.,2005. Inustrial characteristics, environmentalregulations and air pollution: An analysis of the UK manufacturing sector[J].Journal ofEnvironmental Economics and Management,50(1).
[12] Cole, M.A., Elliott, R.J.R., Fredriksson, P.G.,2006. Endogenous pollution havens: Does FDIin.uence environmental regulations?[J].Scandinavian Journal of Economics108(1).
[13] Conley, T.G., Ligon, E.,2002. Economic distance and cross-country spillovers[J]. Journal ofEconnomic Growth,7(2).
[14] Copeland, B.R., Taylor, M.S.,2004. Trade, growth and the environment[J]. Journal ofEconomic Literature,42(1).
[15] Damania, R., Fredriksson, P. G., List, J.,2003. Trade Liberalization, Corruption, andEnvironmental Policy Formation: Theory and Evidence[J]. Journal of EnvironmentalEconomics and Management,46(3).
[16] Dasgupta, S., Laplante, B., Wang, H.&Wheeler, D.2002. Confronting the EnvironmentalKuznets Curve[J], Journal of Economic Perspectives,16(1).
[17] Dean, J.,2002. Does trade liberalization harm the environment? A new test[J]. CanadianJournal of Economics,35(4).
[18] Dinda, S.2004. Environmental Kuznets curve hypothesis: a Survey[J], Ecological Economics,49(1).
[19] Dua, A.&Esty, D. C.1997. Sustaining the Asia Pacific Miracle: Environmental Protectionand Economic Integration.[M] Washington DC: Institute for International Economics.
[20] Ederington, J., Minier, J.,2003. Is environmental policy a secondary trade barrier? Anempirical analysis[J]. Canadian Journal of Economics,36(1).
[21] Eskeland, G.S., Harrison, A.E.,2003.Moving to greener pastures multinationals and thepollution haven hypothesis[J]. Journal of Development Economics,70(1).
[22] Esty, D.C.&Gentry, B.S.1997. Foreign Investment, Globalisation and Environment,Globalisation and Environment: Preliminary Perspectives, OECD, Paris.
[23] Fredriksson, P. G.,Svensson, J.,2003. Political instability, corruption and policy formation:the case of environmental policy[J]. Journal of Public Economics,87(8).
[24] Grossman, G.M.&Krueger, A.B.1991. Environmental Impacts of a North American FreeTrade Agreement', in Garber, P.(ed.) The U.S.-Mexico Free Trade Agreement, Cambridge,MA: MIT Press.
[25] Grossman, G.M.&Krueger, A.B.1993. Environmental Impacts of a North American FreeTrade Agreement', in Garber, P.(ed.) The U.S.-Mexico Free Trade Agreement, Cambridge,MA: MIT Press.
[26] Grossman, G.M.&Krueger, A.B.1995. Economic Growth and the Environment[J].TheQuarterly Journal of Economics,110(2).
[27] Grossman, G.M., Krueger, A.B.,1993. Environmental impacts of a NorthAmerican free tradeagreement,.in: Garber, P.M.(Ed), The Mexico-U.S. FreeTrade Agreement, MIT Press,Cambridge.
[28] He, J.,2006. Pollution Haven Hypothesis and environmental impacts of Foreign DirectInvestment: The Case of Industrial Emissions of Sulphur Dioxide (SO2)in ChineseProvinces[J]. Ecological Economics,60(2).
[29] Jaffe, A. B., Peterson, S. R., Portney, P. R., Stavins, R. N.,1995. Environmental regulationsand the competitiveness of U.S. manufacturing: What does the evidence tell us?[J].Journal ofEconomic Literature,33(1).
[30] Jensen, V.1996. The Pollution Haven Hypothesis and the Industrial Flight Hypothesis: SomePerspectives on Theory and Empirics, Working Paper, Centre for Development and theEnvironment, University of Oslo.
[31] Kao, C.&Chiang, M.2001. On the Estimation and Inference of a Cointegrated Regression inPanel Data[J], Advances in Econometrics,15.
[32] Kao, C.1999. Spurious Regression and Residual-based Tests for Cointegration in PanelData[J], Journal of Econometrics,90(1).
[33] Karaman rsal, D. D.2008. Comparison of Panel Cointegration Tests[J].Economics Bulletin,3(1).
[34] Keller, W., Levinson, A.,2002. Pollution abatement costs and foreign direct investmentin.ows to U.S. States[J].Review of Economics and Statistics,84(4).
[35] Kuznets, S.1955. Economic Growth and Income Inequality[J], The American EconomicReview,45(1).
[36] Levinson, A., Taylor, M.S.,2008. Unmasking the Pollution Haven Effect[J]. InternationalEconomic Review,49(1).
[37] List, J.A., Co, C.,2000. The E¤ects of Environmental Regulations on Foreign DirectInvestment[J]. Journal of Environmental Economics and Management40(1).
[38] Lucas, R., Wheeler, D.&Hettige, H.,1992. Economic Development, EnvironmentalRegulation and the International Migration of Toxic Industrial Pollution'. In Low, P.(ed.)International Trade and the Environment, World Bank Discussion Papers No.159,Washington, DC.
[39] Mani, M.&Wheeler, D.1998. In Search of Pollution Havens? Dirty Industry in the WorldEconomy1960–1995[J], The Journal of Environment Development,7(3).
[40] McCoskey, S.&Kao, C.1998. A Residual-Based Test of the Null of Cointegration in PanelData[J]. Econometric Reviews,17(1).
[41] Mohammed, A,2003. Foreign Direct Investment and the Environment: Pollution HavenHypothesis, The Eight Annual Conference on Global Economic Analysis[J].Journal ofPolitical Economy,32(1).
[42] Nahman, A.&Antrobus, G.2005. The Environmental Kuznets Curve: A LiteratureSurvey[J].South African Journal of Economics,73(1).
[43] Panayotou, T.1993. Empirical Tests and Policy Analysis of Environmental Degradation atDifferent Stages of Economic Development', ILO, Technology and Employment Programme,Geneva.
[44] Panayotou, T.2000.'Economic Growth and the Environment, Working Paper No.56,Center for International Development.
[45] Pedroni, P.1995. Panel Cointegration: Asymptotic and Finite Sample Properties of PooledTime Series Tests, with an Application to the PPP Hypothesis, Working Paper, Series inEconomics95-013, Indiana University.
[46] Pedroni, P.1999. Critical Values for Cointegration Tests in Heterogeneous Panels withMultiple Regressors[J].Oxford Bulletin of Economics and Statistics,61(1).
[47] Pedroni, P.2004. Panel Cointegration: Asymptotic and Finite Sample Properties of PooledTime Series Tests, with an Application to the PPP Hypothesis[J]. Econometric Theory,20(1).
[48] Perkins, R.&Neumayer, E.2009. Transnational Linkages and the Spillover ofEnvironment-efficiency into Developing Countries'[J].Global Environmental Change,19(3).
[49] Selden, T. M.&Song, D.1994.'Environmental Quality and Development: Is there a KuznetsCurve for Air Pollution?[J].Journal of Environmental Economics and Management,27(2).
[50] Shafik, N.&Bandyopadhyay, S.1992. Economic Growth and Environmental Quality: TimeSeries and Crosscountry Evidence', Background Paper for the World Development Report,The World Bank, Washington, DC.
[51] Smarzynska Javorcik, B., Wei, S-J.,2004. Pollution havens and foreign direct investment:Dirty secret or popular myth?[J].Contributions to Economic Analysis and Policy,3(2).
[52] Smarzynska, B.&Wei, S.2001. Pollution Havens and the Location of Foreign Directinvestment: Dirty Secret or Popular Myth?, Washington, D.C., International TradeTeam-Development Economics Research Group, The World Bank. Mimeo.
[53] Stern, D.I.1998. Progress on the environmental Kuznets curve?[J]. Environment andDevelopment Economics,3(2).
[54] Stern, N.2008. The economics of climate change[J].America Economic Review,98(3).
[55] Westerlund, J.&Basher, S.A.2008. Mixed Signals among Tests for Panel Cointegration[J].Economic Modelling,25(1).
[56] Westerlund, J.2005.New Simple Tests for Panel Cointegration[J]. Econometric Reviews,24(2).
[57] Westerlund, J.2006. Reducing the Size Distortions of the Panel LM Test forCointegration[J].Economics Letters,90(3).
[58] Westerlund, J.2007. Testing for Error Correction in Panel Data[J].Oxford Bulletin ofEconomics and Statistics,69(2).
[59] Westerlund, J.2008. Panel Cointegration Tests of the Fisher Hypothesis[J].Journal of AppliedEconometrics,23(3).
[60] Wheeler, D.,2001. Racing to the bottom? Foreign investment and air quality in developingcountries[J]. Journal of Environment and Development,10(3).
[61] Xing, Y., Kolstad, C.,2002. Do lax environmental regulations attract foreign investment?[J].Environmental and Resource Economics,21(1).