EN中文

地理研究 >

2016 , Vol. 35 >Issue 1: 82 - 94

DOI: https://doi.org/10.11821/dlyj201601008

产业结构对区域碳排放的影响——基于多国数据的实证分析

  • 原嫄 , 1 ,
  • 席强敏 2 ,
  • 孙铁山 3 ,
  • 李国平 , 3
展开
  • 1. 西北工业大学人文与经法学院,西安 710072
  • 2. 南开大学经济学院,天津 300031
  • 3. 北京大学政府管理学院,北京 100871
李国平（1961- ）,男,黑龙江拜泉人,教授,博士生导师,主要从事经济地理学、区域经济学、城市与区域规划等方面研究。E-mail:

作者简介：原嫄（1986- ）,女,陕西西安人,讲师,主要从事区域经济学研究。E-mail:

收稿日期: 2015-06-23

  要求修回日期: 2015-11-05

  网络出版日期: 2016-01-23

基金资助

国家重点基础研究发展计划（973计划）项目（2012CB955802）

国家自然科学基金项目（41171099）

收起

The impact of the industrial structure on regional carbon emission: Empirical evidence across countries

  • YUAN Yuan , 1 ,
  • XI Qiangmin 2 ,
  • SUN Tieshan 3 ,
  • LI Guoping , 3
Expand
  • 1. School of Humanities, Economics and Laws, Northwestern Polytechnical University, Xi'an 710072, China
  • 2. School of Economics, Nankai University, Tianjin 300031, China
  • 3. School of Government, Peking University, Beijing 100871, China

Received date: 2015-06-23

  Request revised date: 2015-11-05

  Online published: 2016-01-23

Copyright

《地理研究》编辑部
Fold

摘要

人类行为所引起的全球气候变暖趋势已经无可争议,所带来的影响可能对全球发展方向和生产方式产生重大的作用。在建立产业结构对区域碳排放的影响模型基础上,在全球尺度下进行计量分析,主要结论：第一,理论模型证明区域碳排放随经济发展推进具有先上升后下降的不可抗的基本客观规律,故减排应从降低峰值高度、促进峰值提前等方向入手;第二,实证结果说明第二产业份额对碳排放的影响强度为恒正值,而服务业的影响强度逐步降低,促使第二产业向服务业的份额流动最终将带来整体影响强度的下降;第三,产业结构调整所引起的碳排放变动强度具有明显差异,产业升级对于中高等发展水平国家的减排效率明显高于极高发展水平国家,且中等发展水平国家将在更早的发展阶段迎来碳排放高峰。

关键词： 碳排放; 产业结构; 面板回归分析; 影响强度

本文引用格式

原嫄 , 席强敏 , 孙铁山 , 李国平 . 产业结构对区域碳排放的影响——基于多国数据的实证分析[J]. 地理研究, 2016 , 35(1) : 82 -94 . DOI: 10.11821/dlyj201601008

Abstract

Global warming is a direct consequence of the increasing CO2 concentration in the atmosphere, which is caused by the abnormal increase in carbon emission levels. Such phenomenon has become a threat to the safety of living conditions. Many studies had proven that the increase in carbon emissions over the past century was mainly caused by human activities, but the factors currently known that contribute to carbon emissions are difficult to be mitigated. Therefore, further studies on the effect of economic development on carbon emissions might provide more feasible and efficient techniques for reducing carbon emissions. First, based on the framework of the effect of industrial structure on carbon emissions, the industrial structure determines the convergence of the equilibrium path of the regional economy and the final output. The final output and industrial structure influence carbon emissions simultaneously. The dynamic model shows that when higher energy intensity has a low share, its growth will dominate the overall regional carbon emissions. By contrast, when the lower energy intensity has a high share, its growth will lead the whole region to reduce carbon emissions. Second, an empirical analysis is performed to investigate the influence of industrial structure on global carbon emissions. Both the shares of the manufacturing and service industries positively affect carbon emissions. However, the influencing intensity of the service industry decreases along with an increasing share. Therefore, in the early stages of economic development, the rapid growth in the share of the manufacturing sector will increase the amount of carbon emissions; however, in the matured stages of economic development, the increasing share of the service sector and the declining share of the manufacturing sector will decrease the overall influencing intensity of these sectors. Third, an empirical analysis is conducted under different groups of countries according to the developing levels. All in all, compared with very-high-class group of countries, upgrading the industrial structure is a more efficient mitigating path in high-class and middle-class groups of countries. Meanwhile, adjusting the internal structure of their manufacturing and service sectors can inhibit the influencing intensities of different industries as well.

Key words： carbon emission; industrial structure; panel regression; influencing intensity

1 引言

全球气候变暖已经成为亟需世界各国共同应对和解决的热点问题,而人类经济行为所引起的包括CO2在内的温室气体大量排放是造成气候变暖的关键因素[1]。目前,有关气候变化的物理学和政治经济学研究已经证明,全球气候变化问题是人类迄今为止面临的规模最大、范围最广、影响最为深远的挑战之一,也将是影响未来世界经济和社会发展、重构全球政治和经济格局的最重要因素之一[2-4]。因此,在保证环境安全与可持续发展的前提下进行合理的碳约束必须作为世界各国的共同责任[5]。
已有大量研究着眼于经济发展过程对区域碳排放的影响,无论是基于理论建模还是实证分析,其本质均以寻求可能的碳减排手段或路径为目的[6]。当前研究中已获得公认的区域碳排放影响因素主要包含人口规模、经济发展水平和技术进步水平等。首先,人口规模对碳排放所具有的决定性控制作用最先引起学术界的关注[7,8],人口数量增长所带来的粮食安全、水污染、能源和交通压力等问题将会对环境产生多重性的负面影响[9]。第二,以区域人均GDP指代的经济发展水平与区域碳排放在部分地区呈现倒U型关系,即环境库兹涅茨曲线（Environmental Kuznets Curves: EKC）[10-13],在一定程度上说明随着经济的发展,碳排放的增长速度将逐步降低。但是,仍有包括发达国家在内的不同空间尺度实证结果未能显示EKC的特征趋势[10,14]。这可能是由于CO2属于特殊污染物,短期内无毒无害,治理代价高昂,导致环境治理政策的重心很难快速向降低碳排放的方向倾斜,为EKC拐点的出现制造了客观困难[15,16]。第三,技术进步水平是抑制碳排放的最重要因素[17,18]。大量研究已经证明,地区间不同的技术水平是造成能源利用效率迥异的关键原因[19]。因而,大力促进技术进步一直以来就是提升区域能源利用效率、降低排放增速的关键内容。以上受到广泛认可的碳排放影响因素均可作为抑制排放的重要路径。但是,由于各影响要素的自身属性,其减排潜力也存在差异。首先,全球范围内的人口数量增速虽然在逐步降低,但基本趋势仍处于持续性增长阶段,考虑到人口因素对碳排放的恒正影响,该因素很难作为有效的减排路径;第二,经济发展水平对碳排放影响的研究局限在现象的相关性特征分析,未能从机制上提出解释和动因,故很难作为制定减排政策的依据;第三,技术进步水平高度依赖科技创新过程,具有较高的偶然性与突发性,故不应作为唯一的减排方式。
基于上述研究梳理,有必要进一步挖掘经济发展水平这一影响因素的深层内涵。经济发展水平的提升,实质上受到区域经济发展阶段不断推进的控制,而经济发展阶段正是对地区经济、社会、制度和技术等多个方面的综合体现。大量的实证研究已经从现象角度说明随着经济发展水平的提升,区域产业结构的特征会发生具有规律性的演化[20]。换言之,经济发展过程可以被归纳为国民经济结构出现的一系列变化,其变化型式（Pattern）与国民收入水平的增长有密切联系[21]。自20世纪50年代,美国经济学家钱纳里对跨国家比较研究基础上的经济发展过程及其特征进行了一系列探索和总结[22]。理论上的经济发展型式在某种程度上的一致性建立在几乎所有国家供给和需求状态存在某种相似特征这一假设之上,即均受到“共同因素”的影响,例如技术和知识、人类需求、进出口市场的相似可达性、随收入水平增长的资本累积等[23]。钱纳里运用这种方法,通过国际比较表明,经济增长是不同产业和经济活动部门生产结构中的一组变化[24],在跨国家比较中具有较为明显的一致性特征。长期以来,产业结构作为经济结构中的一个最基本的核心结构形式被许多经济学家所重视[25]。这是由于经济增长与产业结构之间具备双向循环和反馈的累积性联动机制[26],产业结构从外生和内生两个角度共同对经济增长方式产生重要影响[27]。换言之,经济增长方式实质上受到高级生产要素在不同产业及其环节间的流动导致一系列生产函数对于要素配置产生作用的改变而逐步发生,其外在表现即为产业结构的不断调整和优化[27]。
同时,有关产业结构对区域碳排放影响的相关研究在近十年来亦逐步涌现。一方面,部分研究沿袭传统的因素分解方法,测算不同空间尺度下产业结构对区域碳排放的贡献量,并探讨其变化规律[28-31],其中有证据显示产业结构对区域碳排放增长具有负向贡献。另一方面,将产业结构作为碳排放或能源消费的影响因素之一的实证研究已成为热点。大部分研究结果显示第二产业份额在各个区域尺度上均对碳排放和能源消费具有显著的正向影响[32,33],其中部分实证分析发现几乎所有产业的规模扩大均会对碳排放产生正向效应,但第三产业的强度明显低于第二产业[34]。同时,产业结构变迁存在显著的潜在减排贡献[35,36]。
总体而言,产业结构对碳排放的影响研究已经相当丰富。本文将继续从产业结构的指示性意义出发,在考虑产业结构与人口、技术等的内部关联性的同时,进一步挖掘经济发展过程中碳排放的变化规律。因此,首先建立产业结构对碳排放影响的理论模型,在全球空间尺度下对这一关系进行实证分析,并从经济发展阶段的角度对结果进行讨论。力图在现有研究的基础上将产业结构的内涵更加具化和深化,为减排政策的制定提供操作性更强的参考。

2 研究方法

区域碳排放所受到的直接影响来源于区域能源消费量,且能源消费量的增加必定会带动区域碳排放的上升。实质上,区域各产业份额通过两个层级对碳排放产生影响,且这种影响的方向和强度随产业自身属性、区域技术水平以及产业原始份额等因素发生变化。
在第一层级中,区域经济在人口规模、技术水平和资本投入等多要素的影响下不断发展。根据新古典主义宏观经济理论,在各生产要素增长速度保持稳定的背景下,区域经济的发展最终会收敛于以单位有效劳动的投资增长率等于零为标志的平衡增长路径上。平衡增长路径由人均产出所表达,而人均产出则能够指征区域发展阶段。区域经济平衡增长路径的位置受到不同产业所占份额的特征所影响,因而产业结构能够作为区域经济发展阶段的直接替代性指标。在第二层级中,将经济发展按照不同产业展开,能源消费总量将是各产业能源消费之和。对于任意产业类型而言,其能源消费量与其产出量、能源相关技术水平等紧密联系,而这些因素则直接受到第一层级中由平衡增长路径所决定的区域经济总产出和区域产业结构的影响。因此,本文主要对第二层级,即直接与产业结构对碳排放影响相关的理论模型进行推导和阐述。

2.1 基本假设

根据新古典主义经济学的经典模型,在任何时刻,经济拥有一定量的资本、劳动和知识,而这些均属于生产要素并结合起来方能生产产品,如式（1）所示：
Y ( t ) = F ( K ( t ) , A ( t ) L ( t ) ) （1）
式中：K为资本投入;L为劳动力规模（为简便起见,令人口规模等同于劳动力规模）;A为技术水平,则AL为有效劳动;各生产要素通过函数F对区域经济产出Y产生共同影响,且服从规模报酬不变原则。定义劳动力和技术进步的存量随时间的增长率不变;而资本投入K实质上是从产出中按照一定比率储蓄的部分,可假定该比率s为外生且为恒常数（假定折旧率为0）,因此有下式：
L ˙ ( t ) = nL ( t ) （ 2 ） A ˙ ( t ) = gA ( t ) （ 3 ） K ˙ ( t ) = sY ( t ) （ 4 ）
定义K/AL为单位有效劳动的资本投入量,Y/AL为单位有效劳动的经济产出量（简称人均产出）。令k=K/AL,y=Y/AL,且f(k)=F(k,1),则有下式：
y = f ( k ) （5）
但是,新古典主义宏观经济学一般反映的经济增长是区域整体的综合情况,当以产业类型对区域经济进行分解时,不同产业占经济规模的份额成为新的平衡增长路径位置确定的影响因素。为了便于分析,假设区域仅存在两个产业类型;其中,产业1产出占区域经济整体规模的份额为 θ 1 ( t ) ,产业2产出的份额为 1 - θ 1 ( t ) 。那么,则有下式：
Y ( t ) = Y 1 ( t ) + Y 2 ( t ) Y 1 ( t ) = θ 1 ( t ) Y ( t ) Y 2 ( t ) = 1 - θ 1 ( t ) Y ( t ) （6）
where 0 < φ 1 , φ 2 < 1
式中：产业1和产业2各自的劳动力增长率分别恒为n1和n2;知识水平增长率分别为 g 1 和 g 2 ;储蓄率均为 s 。

2.2 模型动态学分析

区域碳排放直接受到区域能源消费量的控制,其影响函数为P,该函数具有单调递增的基本特征。因此,任意区域经济过程产生的碳排放可以如下表达：
Car ( t ) = P ( E ( t ) ) （7）
式中：Car为经济过程产生的区域碳排放;E为区域能源消费量;t表示时间。能源消费量受到各产业产出和能源相关技术水平的共同控制。对于任意产业类型,能源消费量与其产出呈现正相关,与其技术水平呈现负相关。同时,任意产业部门的能源相关技术水平与该产业研发投入占总产出的比例显著正相关,假设这一比例为B。结合Cobb-Douglas形式的生产函数,本文认为区域经济产出和技术水平进步共同通过单调递增函数R对能源消费量产生正向作用,且能源消费量的增速应在产出增长和技术提升的作用下逐步降低,即符合稻田条件[37]。能源消费量可表达为：
E ( t ) = ∑ i = 1 n E i ( t ) = ∑ i = 1 n R i ( Y i ( t ) , E A i ( t ) ) E i ( t ) = Y i ( t ) γ i B i Y i ( t ) β i = θ i ( t ) Y ( t ) γ i B i θ i ( t ) Y ( t ) β i ( 0 < γ i , β i < 1 ) ∑ i = 1 n θ i ( t ) = 1 ( 0 < θ i < 1 ) （8）
式中：i为第i个产业部门;γ为对应产业产出对该产业能源消耗量的正向影响弹性;β为对应产业研发投入水平对该产业能源消耗量的负向影响弹性;B为研发投入占总产出的比例。为了简化问题的分析难度,可考虑两个产业部门的基本情景,区域能源消费总量即可根据式（8）转化为：
E ( t ) = E 1 ( t ) + E 2 ( t ) = σ 1 θ ( t ) Y ( t ) α 1 + σ 2 1 - θ ( t ) ) Y ( t ) α 2 ( 0 < α 1 , α 2 < 1 ) （9）
式中：α为对应产业产出和研发投入水平对该产业能源消耗量的共同影响弹性;σ为各产业影响系数;根据前期假设,各产业能源消耗量随对应产业产出的增长符合稻田条件,则有0<α1, α2<1。对上式两侧取对数,可得：
ln E 1 ( t ) = ln σ 1 + α 1 lnθ ( t ) + lnY ( t ) ln E 2 ( t ) = ln σ 2 + α 2 ln ( 1 - θ ( t ) ) + lnY ( t ) （10）
对上式两侧对时间取倒数,各产业能源消耗量变化率则为：
E ˙ 1 ( t ) E 1 ( t ) = α 1 θ ˙ ( t ) θ ( t ) + Y ˙ ( t ) Y ( t ) E ˙ 2 ( t ) E 2 ( t ) = α 2 - θ ˙ ( t ) 1 - θ ( t ) + Y ˙ ( t ) Y ( t ) （11）
那么,区域能源消耗的变化量则为：
E ˙ ( t ) = α 1 E 1 θ ( t ) - α 2 E 2 1 - θ ( t ) θ ˙ ( t ) + α 1 E 1 + α 2 E 2 Y ˙ ( t ) Y ( t ) （12）
由式（12）可知,区域能源消费量的增量 E ˙ ( t ) 与产业结构的特征 θ ( t ) 及其变化方向 θ ˙ ( t ) 具有显著联系。其中, θ ˙ ( t ) 的系数是分析这一影响过程特征的关键。令式（12）的斜率为u,并将式（9）代入,则有下式：
u = α 1 σ 1 θ ( t ) α 1 - 1 Y ( t ) α 1 - α 2 σ 2 1 - θ ( t ) α 2 - 1 Y ( t ) α 2 （13）
假设以上两个产业部门用于研发的投入占各自产出的系数相同,即有 σ 1 = σ 2 。那么,式（13）的正负性将取决于两个产业部门产业产出和研发投入对该产业能源消耗量共同影响弹性α1和α2。假设有0<α1<α2<1和0<α2<α1<1两个基本情景,而 θ ˙ ( t ) 为产业部门1的份额变化量,则有：
第一,当0<α1<α2<1时,则显然有 α 1 Y ( t ) α 1 α 2 Y ( t ) α 2 < 1 。那么,当 θ ( t ) α 1 - 1 1 - θ ( t ) α 2 - 1 < 1 时,则 θ ( t ) > 0.5 。此时u显然为负值,即当产业部门1占经济产出份额超过一半时,其份额的继续增长,将引起区域能源总消耗的负向变化。而当 θ ( t ) α 1 - 1 1 - θ ( t ) α 2 - 1 > 1 时,u的正负性无法确定。
第二,当0<α2<α1<1时,则显然有 α 1 Y ( t ) α 1 α 2 Y ( t ) α 2 > 1 。那么,当 θ ( t ) α 1 - 1 1 - θ ( t ) α 2 - 1 > 1 时,则 θ ( t ) ＜ 0.5 。此时u显然为正值,即当产业部门1占经济产出份额未达到一半时,其份额的继续增长,将引起区域能源总消耗的正向变化。而当 θ ( t ) α 1 - 1 1 - θ ( t ) α 2 - 1 < 1 时,u的正负性无法确定。
简而言之,上述有关两个产业部门的四个情景分析显示,产业份额变迁对区域能源消费的影响亦取决于产业属性和已有份额等要素的共同限制。结合现实情况中的产业产出对能源影响的弹性,可假定情景a中产业部门1和部门2分别为服务业和制造业,隐含着部门1的产出对自身能源消费的影响弹性,亦即能源消耗强度较低。此时,当服务业的份额低于50%时,其份额的增长对区域能源总消耗变化量的影响方向无法确定;但其超过50%时,则显然将对能源总消耗产生负向效应。类似的,可假定情景b中产业1和产业2分别为制造业和服务业,部门1的产出对其能源消费的影响弹性,亦即能源消耗强度较高。此时,当制造业的份额低于50%时,其份额的增长将显然对区域能源总消耗量产生正向影响;而当其份额高于50%时,其份额变迁对能源总消耗的影响方向则无法确定。
以上通过数理推倒的方式建立了产业结构在指征区域经济发展阶段的基础上对区域碳排放的影响模型。模型的动态均衡分析显示,经济发展初期的碳排放由能耗强度高的产业份额增长产生的正影响所控制;而随着经济发展阶段的不断推进,能耗强度低的产业份额增长所产生的负影响将逐步主导碳排放变动。因此,碳排放在经济发展过程中逐步上升后达到峰值,最终出现降低趋势是人类经济行为引起的排放变动的基本客观规律。

3 数据来源

全球各国的燃料燃烧产生碳排放数据来源于国际能源署（IEA）的《燃料燃烧产生碳排放年鉴（CO2 Emissions from Fuel Combustion）》系列,基本研究时段为1998-2011年。全球各国分产业增加值数据、人口规模和单位GDP能耗（Energy use (kg oil equivalent) per $1,000 GDP (Constant 2005 PPP $)）由联合国官方数据库①（①United Nations Statistics Division(UNSD)中的Global Indicator Database数据库（http://data.un.org/DataMartInfo.aspx）,指标名称为Gross Value Added by Kind of Economic Activity at constant 2005 prices。）获取。其中分产业增加值数据中包含的产业类型有：农林牧渔业（A-B）,第二产业（含：采矿业和制造业（C-E）、制造业（D）和建筑业（F）等）,服务业（含：批发零售及餐饮住宿业（G-H）、交通运输和仓储业（I）和其他产业（J-P）等）。基于对数据稳健性的考虑,将剔除掉经济发展水平极低或处在政治局势不稳定的国家和地区,仅考虑人类发展水平②（② 目前在全球范围内受到广泛认可的用于划分国家发展水平的指标为来源于联合国发展计划署编撰的《人类发展报告》的人类发展指数（Human Development Index,HDI）。本文使用2010年该报告最新修正后的划分结果对各国进行讨论。）为“极高”、“高等”及“中等”的国家。故样本数量为95个;其中,极高、高等和中等发展水平国家数量分别为38个、33个和24个。
产业结构特征虽然能够反映区域经济发展方式和阶段,但因其以份额形式表达导致仍会遗漏诸多区域的相关属性信息,例如区域规模、技术水平等。因此,在现有数据条件下,将不同产业的份额作为核心解释变量,并选取人口数量和单位GDP碳排放量作为控制变量（表1）。
Tab. 1 The definitions of the variables in the model

表1 模型变量定义及说明

变量类别 变量名称 变量含义 对应指标
被解释变量 LCAR 区域碳排放指数 由区域燃料燃烧产生碳排放的对数表示
解释变量 工业化水平 IND 第二产业份额 由采矿业和制造业、建筑业总产业份额值表示
服务化水平 SER 服务业份额 由批发零售及餐饮住宿业、交通运输和仓储业和其他产业等总产业份额值表示
控制变量 LPOP 区域规模指数 由区域人口数量的对数表示
LENE 区域技术水平指数 由区域单位GDP能源使用量的对数表示
实证模型主要关注在三次产业分类下所能够反映出的区域工业化水平和服务化水平分别对区域碳排放的影响作用强度。影响机制模型式（12）的动态分析显示,任意产业份额对区域碳排放的增速均存在潜在影响,那么产业份额的二次项则有可能对碳排放产生影响。因此,实证模型具体形式如下：
LCA R it = β 1 IN D it 2 + β 2 IN D it + β 3 SE R it 2 + β 4 SE R it + β 5 LPO P it + β 6 LEN E it + u i + ε it i = 1 , 2 , ⋯ , N ; t = 1 , 2 , ⋯ , T (14)
式中：i为研究样本中各个国家或地区;N为本文总样本数量;t代表年份;T为研究年份; u i 和 ε it 共同代表符合扰动项,分别表示个体效应变量和随机干扰项。

4 结果分析

4.1 全球各国尺度下的回归结果与讨论

为了确定回归方法,使用加入了个体虚拟变量的LSDV法估计双向固定效应模型。结果显示,多数国家或地区的虚拟变量很显著,即存在固定效应。同时,亦对固定效应和随机效应两模型进行Hausman检验,结果强烈拒绝“H0: ui与xit, zi不相关”的假设,故应使用固定效应模型进行估计。固定效应模型的回归结果显示第二产业份额的二次项显著性极低,故可剔除。固定效应回归结果的异方差情况进行沃尔德检验,结果强烈拒绝“组间同方差”的原假设。在考虑模型一阶差分情况下进行Wald检验,结果强烈拒绝“不存在一阶组内自相关”的原假设。同时,组间截面相关的LM检验结果显示残差相关系数矩阵不可逆,无法执行命令,且该方法更适用于长面板。故选用Pesaran、Friedman和Frees的检验方法,三个方法的检验p值均低于0.01,即强烈拒绝“不存在组间截面相关”的原假设。为了解决原始数据存在的组内和组间自相关的问题,且本文面板数据时间维度较短、样本数量较多,故采用适合于大N小T的D-K标准误回归模型（Regression with Driscoll-Kraay standard errors）进行校正（表2）。结果显示,经过校正的模型方差显著提升。
Tab. 2 The.results of the regression with Driscoll-Kraay standard errors

表2 D-K标准误回归模型估计结果

变量名称 系数 回归结果
IND β2 5.482***
(16.57)
SER2 Β3 -2.583**
(-4.08)
SER β4 7.674***
(9.13)
LPOP β5 1.224***
(17.69)
LENE β6 0.218***
(8.09)
_cons -22.58***
(-21.56)
N 1326
R2 0.427

注：括号中为t值;*、**、***分别表示在10%、5%、1%的水平。数据来源：由STATA软件分析结果整理得到。

上述回归结果显示了以下信息：
（1）第二产业份额的系数显著为正值,说明在全球各国视角下,当第二产业份额增长一个百分点,各区域碳排放随之增长5.48个百分点。第二产业多为能源开采及加工等能源工业部门以及制造业部门,总体上对能源的消耗量大,且引起的负面环境影响显著,而这些特征是第二产业自身的属性所造成。在第二产业份额上升过程中,碳排放作为其中一项伴生结果,必然也会随之上升。
第二产业的发展是区域经济中早期崛起的引擎和保证,同时,由于本文的分析限定在全球极高、高等和中等发展水平的背景下,这些国家多数已经经过或正在稳定朝向工业化的顶峰迈进,即处于正常的经济发展过程中,其经济结构性特征也随之发生渐进式的演化。这些国家的第二产业份额部分逐年上升、部分相对稳定,亦有渐进下降,这说明研究时段已经能够较为完整地体现伴随着工业化过程的区域经济发展规律。因此,现有的研究时段内各样本已经能够共同体现出第二产业份额的变化规律,故该份额对区域碳排放的影响及强度特征的可信度也较高。那么,对于正处于工业化过程中的区域,第二产业份额的抬升是实现经济发展的基本保证,而其所引起的碳排放增长也是必然结果。在这个意义上,实现碳减排的一个可能方式就是限制粗放型第二产业的发展以降低其份额或不断实现第二产业内部的结构升级以降低其对区域碳排放的影响强度。
（2）服务业份额体现为SER2和SER两个解释变量,其回归系数分别为-2.583和7.674,说明服务业份额每上升一个百分点所对应的区域碳排放上升幅度呈现逐步降低的趋势,即服务业份额增长所对应的区域碳排放增量随经济发展水平的提高而呈现降低态势。服务业份额回归系数显示服务业份额与碳排放两者间倒U型结构的拐点位置。这一开口向下的二次函数的对称轴位于SER=148.5%,远大于1,而SER ⊂ [0,1],这说明在服务业份额的有效区间内,该产业份额对区域碳排放的影响为正向作用。同时,定义该影响模型的一阶导数为服务业份额对区域碳排放的影响强度③（③ 表示为 ∂ LCAR ∂ SER = - 5.166 SER + 7.674 。）,不难发现这一影响强度随服务业份额上升在逐步降低,且当服务业份额高于42.43%时,第二产业份额对碳排放的影响强度高于服务业份额。第二产业和服务业两者份额对于区域碳排放影响强度的差异将是区域碳排放增长速度发生改变的重要动因。
（3）人口规模作为控制变量之一,对区域碳排放的影响系数为显著正值。结果显示当区域人口规模上升一个百分点时,将带来区域碳排放1.224个百分点的增长。在本文的研究时段内,全球平均人口自然增长率由1.28%下降至1.10%附近,故在上述回归结果基础上,在全球整体视角下,每年大约1.4%的碳排放增量来源于人口规模的扩大。因此,对区域人口规模和自然增长率的有效控制也是可行的控制碳排放方法。但是,由于人口规模控制的限制性较强,并且全球大多数国家在短期内无法实现人口负增长,故仅从控制该要素增长的角度进行减排的作用并不明显。
（4）以单位GDP能耗作为指标的区域技术水平与碳排放呈现显著正相关。结果显示当单位GDP能耗每降低一个百分点,区域碳排放下降0.218个百分点。1998-2011年全球单位GDP能耗的年均增量为-1.38%,技术水平提高促使碳排放在整体上每年降低大约0.3个百分点。但是,技术水平的变化波动性较强,并不具有明显的趋势性特征。这可能与技术水平高度依赖区域创新能力,这导致技术突破具有较高的偶然性。因此,通过技术水平的提高进行碳减排是可行的路径,但该指标的短期稳定性较低,从长期角度来看则仍是较为可靠的减排手段。

4.2 不同发展水平国家集团背景下的回归结果与讨论

对不同发展水平国家集团分类进行计量回归,首先对固定效应和随机效应模型使用Hausman检验,结果显示均应使用固定效应模型。但是,三个发展水平国家集团的固定效应回归均显示存在显著异方差和序列相关,因此对三个类别均使用D-K标准误回归模型进行校正,结果如表3所示。
Tab. 3 The results of the regression with Driscoll-Kraay standard errors in different groups of countries

表3 不同发展水平国家集团的D-K标准误回归模型估计结果

变量名称 系数 极高发展水平 高等发展水平 中等发展水平
模型一 模型一 模型二 模型一
IND β 2 3.367** 5.870*** 5.830*** 5.270***
(-4.16) (-7.81) (-7.73) (-9.73)
SER2 β 3 -3.723*** -0.259 -4.671**
(-5.07) (-0.21) (-3.50)
SER β 4 8.172*** 5.415** 5.130*** 9.365***
(-6.00) (-3.37) (-5.19) (-6.31)
LPOP β 5 0.605*** 1.644*** 1.646*** 2.140***
(-28.67) (-27.7) (-27.78) (-35.6)
LENE β 6 -0.0591 0.333*** 0.336*** 0.434***
(-1.81) (-5.00) (-5.14) (-9.51)
_cons -9.842*** -29.81*** -29.76*** -40.31***
(-10.15) (-23.25) (-23.30) (-57.04)
N 532 458 458 336
R2 0.488 0.373 0.373 0.660

注：括号中为t值;*、**、***分别表示在10%、5%、1%的水平。数据来源：由STATA软件分析结果整理得到。

上述回归结果显示：
（1）极高发展水平国家的回归结果显示第二产业份额每增加一个百分点,区域碳排放则随之增长3.367个百分点,低于全球平均水平。这很有可能是由于该类发展水平国家集团第二产业份额已经普遍较低。1998-2011年,极高发展水平国家的第二产业份额平均值由30.40%下降至28.36%,仅就2011年而言,第二产业最高的份额为63.60%,最低仅为7.23%。可见,该类国家的第二产业份额变动总体上对区域碳排放具有负向贡献。极高发展水平国家的服务业份额与全球整体情况相似,均表现为与区域碳排放的倒U型关系④（④ 强度为 ∂ LCAR ∂ SER = - 7.446 SER + 8.172 。）。随着区域服务业份额的上升,极高发展水平国家的碳排放随之上升,但增速却随份额的升高而降低。结合第二产业份额对区域碳排放的影响强度特征可以发现,对于此类国家集团来说,当服务业份额高于64.53%时,服务业份额的增长所带来的碳排放增速将低于第二产业份额降低所带来的碳排放的降速,这意味着此时产业结构的升级将带来区域碳排放的降低。
（2）高等发展水平国家的回归结果显示,服务业份额的二次项对区域碳排放的影响并不显著,故选用无服务业二次项的模型进行回归分析。模型二的结果显示,高等发展水平国家集团第二产业份额的变化对区域碳排放呈现显著正向影响,且当第二产业份额每上升一个百分点将引起区域碳排放增长5.83个百分点。研究时段内,高等发展水平国家集团第二产业份额平均值由36.08%略下降至35.35%。可见,第二产业份额的变化引起高等发展水平国家集团碳排放下降了约4.26个百分点。高等发展水平国家集团服务业份额变动对区域碳排放的影响呈现显著的线性特征,当服务业份额每上升一个百分点会引起区域碳排放增长5.13个百分点。研究时段内,高等发展水平国家集团服务业份额平均水平由54.10%明显上升至57.07%。这一定程度上说明该类型国家集团的服务业份额变化大约引起了15.03个百分点的区域碳排放增长,这一增幅显著高于第二产业份额变动所引起的降低幅度。以上结果说明,高等发展水平国家的第二产业和服务业对碳排放均具有正向影响强度,但服务业的强度较低。因此,从长期来看,第二产业份额向服务业流动的过程会带来碳排放的降低。
（3）中等发展水平国家回归结果显示第二产业份额对区域碳排放呈现显著正向效应,当该份额每上升一个百分点时,将会带来区域碳排放5.27个百分点的抬升。研究时段内,中等发展水平国家第二产业份额平均值由35.04%逐步下降至33.62%,从整体上为该类型国家的区域碳排放带来了7.48个百分点的下降。中等发展水平国家服务业份额对区域碳排放的影响呈现显著倒U型结构⑤（⑤ 强度为 ∂ LCAR ∂ SER = - 9.342 SER + 9.365 。）。研究时段内,该类型国家服务业份额的平均值由48.52%明显上升至53.14%,增幅为4.62个百分点。相应的,当中等发展水平国家服务业份额在50%附近时,该份额每提升一个百分点,大约带来区域碳排放4.69个百分点的增长,这意味着该类型国家在研究时段内由服务业份额增长引起了约有20%的排放量抬升。这一增幅远高于第二产业份额下降带来的碳排放降低。

4.3 不同发展水平国家集团的回归结果比较分析

为了便于分析,假设第一产业份额稳定在一个较低的水平,这意味着第二产业和服务业两者的份额处于此消彼长的状态。那么,定义产业结构调整过程所引起的区域碳排放变动强度为Q,具体如下：
Q = - ∂ LCAR ∂ IND + ∂ LCAR ∂ SER （15）
因此,极高、高等和中等发展水平国家集团在第二产业份额降低、服务业份额上升相应幅度的经济发展阶段推进过程中对区域碳排放的影响强度分别如下：
Q Vhigh = - 7.446 SER + 4.805 Q High = - 0.7 Q Medium = - 9.342 SER + 4.095 （16）
根据式（16）得到图1,发现极高和中等发展水平国家集团随经济发展阶段推进过程中碳排放的总影响强度逐步降低。两者分别在服务业份额达到43.83%和64.53%之后,随着经济发展阶段的推进,区域碳排放将出现负增长。高等发展水平国家由于服务业份额二次项不显著,使产业结构调整对区域碳排放的影响强度为负常数,这说明对于这类发展水平国家集团来说,经济发展阶段下降将始终为其区域碳排放带来相同的负贡献。
Fig. 1 The schematic diagram of the influencing intensities of carbon emissions during the evolution in industrial structure in different groups of countries.

图1 第二产业降低、服务业份额上升过程中各类型国家集团区域碳排放变化强度示意图

从横向对比的角度出发,截至2011年,中等发展水平国家集团服务业份额的平均值已经超过51.33%,而极高发展水平国家集团服务业份额的平均值还介于64.53%和73.93%之间。因此,仅从平均水平来看,目前各类型国家产业结构调整对区域碳排放的影响强度均为负值;从强度绝对值来看,由大至小分别为中等发展水平、高等发展水平和极高发展水平国家集团。这说明服务业和第二产业相同幅度的此消彼长,在中等发展水平国家和地区所带来的区域碳排放降低量最大,并且碳排放峰值到来的更早。进一步说,中等发展水平国家当前的整体技术水平已经相对较高,促使碳排放随经济发展虽然仍表现为先增后降的基本规律,但峰值较极高和高等发展水平国家的历史水平有降低趋势,且出现时点明显提前。
造成上述强度差异的原因可能有：第一,中等发展水平国家第二产业份额普遍低于其他类型国家,产业结构调整的潜力更大;第二,中等发展水平国家目前的发展尚对第二产业有较强依赖,其中不乏生产方式粗放的产业,率先退出此类型产业将为区域碳排放降低做出贡献。而对于极高和高等发展水平国家集团来说,社会经济发展已相对成熟,产业结构调整相对缓慢,且第二产业整体能源效率高、集约性强,促使其份额降低对碳排放的负贡献低于中等发展水平国家。

5 结论

经济发展对区域碳排放的影响研究已经相对丰富,但仍有必要进一步挖掘可能的影响因素以寻求更多的减排路径。作为区域经济发展阶段的替代性指标,产业结构变迁与区域碳排放变动具有显著相关关系。在全球各国的空间尺度下,在现有大量相关研究基础上着重分析产业结构对碳排放的影响,有以下初步发现：
（1）产业结构对区域碳排放理论模型的动态学分析显示,经济发展初期的碳排放变动由能耗强度更高的第二产业所控制,其较高的影响强度将促使排放抬升;随着经济发展阶段的推进,能耗强度较低的服务业逐步对碳排放产生主导作用,其份额增长带来的较低影响强度将替换第二产业的影响,促使碳排放增速放缓乃至逐步降低。
（2）全球尺度下产业结构对区域碳排放影响的实证分析结果显示,制造业与服务业份额增长均对碳排放具有正向效应;其中,制造业的影响强度为恒正值,服务业的影响强度随其份额增长而逐步降低。因此,区域处于中早期经济发展阶段时,制造业份额的快速增长将带来碳排放的上升;当区域经济发展到一定水平时,服务业份额逐步抬升,制造业份额相应下降,二者对碳排放的总影响度逐步降低,产生抑制碳排放的作用。
（3）不同发展水平国家集团背景下的实证分析结果显示,碳排放在经济发展过程中均存在倒U型演化趋势,其峰值对应特定产业结构特征。其中,极高和中等发展水平国家碳排放峰值分别对应服务业份额达到64.53%和43.83%时,这为判断各国合理达峰时点提供了一定依据。同时,产业结构升级对于中等和高等发展水平国家集团的碳减排效率明显高于极高发展水平国家集团,是较有效率的减排方法。同时,极高发展水平国家集团通过主导第二产业影响强度的制造业和服务业内部的结构调整进行碳减排亦是较具潜力并具有针对性的路径。

The authors have declared that no competing interests exist.

参考文献

原文顺序 | 文献年度倒序 | 文中引用次数倒序
[1]
Nordhaus W D.Economic growth and climate: the carbon dioxide problem. The American Economic Review, 1977, 67(1): 341-346.

[2]
胡鞍钢, 管清友. 中国应对全球气候变化的四大可行性. 清华大学学报: 哲学社会科学版, 2008, 23(6): 120-132.胡 鞍钢、管清友在《清华大学学报》(哲学社会科学版)2008年第6期撰文认为,应对全球气候变化已经成为世界各国政治领导人的共识,但是减缓和适应气候变 化趋势在发达国家、发展中国家和最不发达国家之间却面临巨大的不平等。应对全球气候变化问题是一项全球性公共产品,没有任何

[Hu Angang, Guan Qingyou.Four feasibilities for china to fight global climate changes. Journal of Tsinghua University: Philosophy and Social Sciences Edition, 2008, 23(6): 120-132.]

[3]
胡鞍钢, 管清友. 应对全球气候变化: 中国的贡献: 兼评托尼·布莱尔《打破气候变化僵局:低碳未来的全球协议》报告. 当代亚太, 2008, (4): 7-25.全球气候变化是人类面临的共同挑战，中国也是全球气候变化的最大受害者之一。由于成本和收益分担不均等问题，应对气候变化的全球治理面临很大困难。总体而言，减排的成本损失有限，但收益显而易见，而不减排的成本却相当高昂。发达国家和发展中国家都应调整其经济结构承担减排义务。中国有能力也有必要积极参与减排，应对气候变化，为世界做出绿色贡献。

DOI

[Hu Angang, Guan Qingyou.To cope with the global climate change: Chinese contribution. Contemporary Asia Pacific, 2008, (4): 7-25.]

[4]
IPCC. Climate Change 2013: The Physical Science Basis. Cambridge, UK: Cambridge University Press, 2013.light vehicles-hybrids, electric and fuel cell vehicles intelligent transportation systems infrastructureaviation encouragment of substitues for SF 6 supermarket refrigeration:HFC emissions advancedagricultural systems options for Economically efficient supply and use of energy

DOI

[5]
李俊峰, 马玲娟. 低碳经济是规制世界发展格局的新规则. 世界环境, 2008, (2): 17-20.低碳经济是当今世界热炒的话题,它似乎是一个技术问题,或者足经济发展模式问题.然而笔者认为,低碳经济的背景和意义远不止如此,也许,它的背后和联合国宪章、关贸总协定一样,是规制世界发展新格局的又一个新的联合国宪章.

[Li Junfeng, Ma Lingjuan.Low carbon economy is the new rules restricting the world's development structure. Environment of World, 2008, (2): 17-20.]

[6]
Li G, Yuan Y.Impact of regional development on carbon emission: Empirical evidence across countries. Chinese Geographical Science, 2014, 24(5): 499-510.Global warming is recently an urgent issue worldwide. The increase of carbon emissions induced by human economic activities has become a major driving force behind global climate change. Thus, as a matter of social responsibility, reasonable carbon constraints should be implemented to ensure environmental security and sustainable development for every country. Based on a summary of studies that examined the relationship between carbon emissions and regional development, this paper shows that human activity-led carbon emission is caused by the combination of several influencing factors, including population size, income level, and technical progress. Thus, a quantitative model derived from IPAT-ImPACT-Kaya series and STIRPAT models was established. Empirical analysis using multivariate nonlinear regression demonstrated that the origins of growing global carbon emission included the increasing influencing elasticity of the population size and the declining negative effect of technical progress. Meanwhile, in context of classification of country groups at different income levels, according to the comparison of fluctuating patterns of the influencing elasticity, technical progress was found as the main factor influencing carbon emission levels in high-income countries, and population size might be the controlling factor in middle-income countries. However, for low-income countries, the nonlinear relationship between carbon emission and its influencing factors was not significant, whereas population growth was identified as an important potential driving force in future carbon emissions. This study can therefore provide a reference for the formulation of policies on carbon constraints, especially to develop more efficient carbon mitigating policies for countries at different income levels.

DOI

[7]
Houghton J.Climate change 1995: The Science of Climate Change: Contribution of Working Group I to the Second Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press, 1996.Reducing greenhouse gas emissions is a defining challenge for China in two senses. First, it is required to avert dangerous and potentially catastrophic climate change. And, second, an effective response to climate change requires fundamental economic reforms and poses

DOI

[8]
Sommer U.World energy outlook 1996. Agrarwirtschaft: International Energy Agency, 1997, 46(3): 158.The global energy landscape is evolving at a rapid pace, reshaping long-held expectations for our energy future. The 2014 edition of the World Energy Outlook (WEO) will incorporate all the latest data and developments to produce a comprehensive and authoritative analysis of medium- and longer-term energy trends. It will complement a full set of energy projections – which extend from today through, for the first time, the year 2040 – with strategic insights into their meaning for energy security, the economy and the environment. Oil, natural gas, coal, renewables and energy efficiency will be covered, along with updates on trends in energy-related CO2 emissions, fossil-fuel and renewable energy subsidies, and universal access to modern energy services.

DOI

[9]
Ehrlich P, Holdren J.Population and panaceas a technological perspective. BioScience, 1969, 19(12): 1065-1071.The article outlines the problems encountered by mankind and their effort in addressing them. It reveals that the biggest problem of mankind ever encountered is population explosion, which grows at a rate of 2% per year. Such problem has a corresponding chain reaction including malnutrition and transportation. Concerned critics believed that technology holds the key in addressing the situation but experts believed that technology remains inadequate if the problem on population is not effectively addressed by the governments in various countries. Scientists said that without such effort in reducing world population, their moves in revolutionizing tropical agriculture, increasing yields of fisheries, desalting water for irrigation or exploiting new power resources are not worthwhile.

DOI

[10]
Shafik N, Bandyopadhyay S.Economic Growth and Environmental Quality: Time-series and Cross-country Evidence. Washington DC: World Bank Publications, 1992.ABSTRACT The authors explore the relationship between economic growth and environmental quality by analyzing patterns of environmental transformation for countries at different income levels. They look at how eight indicators of environmental quality evolve in response to economic growth and policies across a large number of countries and across time. Several conclusions are drawn; (1) income has the most consistently significant effect on all indicators of environmental quality; (2) many indicators tend to improve as countries approach middle-income levels; (3) technology seems to work in favor of improved environmental quality; (4) the econometric evidence suggests that trade, debt, and other macroeconomic policy variables seem to have little effect on the environment, although some policies can be linked to specific environmental problems; (5) the evidence shows that it is possible to"grow out of"some environmental problems, but there is nothing automatic about doing so - policies and investments to reduce degradation are necessary; and (6) action tends to be taken where there are generalized local costs and substantial private and social benefits.

[11]
Grossman G, Krueger A.Environmental impacts of a North American free trade agreement. Cambridge, USA: MIT Press, 1993.A reduction in trade barriers generally will affect the environment by expanding the scale of economic activity, by altering the composition of economic activit

DOI

[12]
Selden T, Song D.Environmental quality and development: Is there a Kuznets curve for air pollution emissions?. Journal of Environmental Economics and Management, 1994, 27(2): 147-162.Several recent studies have identified inverted-U relationships between pollution and economic development. We investigate this question using a cross-national panel of data on emissions of four important air pollutants: suspended particulate matter, sulfur dioxide, oxides of nitrogen, and carbon monoxide. We find that per capita emissions of all four pollutants exhibit inverted-U relationships with per capita GDP. While this suggests that emissions will decrease in the very long run, we forecast continued rapid growth in global emissions over the next several decades.

DOI

[13]
邹秀萍, 陈劭锋, 宁淼, 等. 中国省级区域碳排放影响因素的实证分析. 生态经济, 2009, (3): 34-37.文章在能源消费导致的碳排放核算的基础上,利用1995～2005年中国30个省市区的面板数据定量分析了各地区碳排放与经济发展水平、产业结构、能源效率之间的关系.分析显示各地区碳排放与经济发展水平存在倒U型曲线趋势,而与能源消耗强度呈U型曲线关系,与第二产业产值的比重呈N型曲线关系.从各地区碳排放量对平均碳排放量的偏离量的大小,可以看出中国碳排放存在显著的东南部低中北部高西北部低的空间分布格局.

[Zou Xiuping, Chen Xunfeng, Ning Miao, et al.An empirical research on the influence factor of carbon emission in Chinese provincial regions. Ecological Economics, 2009, (3): 34-37.]

[14]
Tucker M.Carbon dioxide emissions and global GDP. Ecological Economics, 1995, 15(3): 215-223.A positive relationship between carbon dioxide emissions, the most important greenhouse gas (GHG) implicated in global warming, and GDP is shown in this paper, examining per capita income and CO 2 emissions of 137 countries across 21 years. It also appears that as per capita incomes accelerate across countries emissions increases, for the most part, tend to decelerate. It could be that higher income levels lead to increased demand for environmental protection. Only emissions reduction proposals that assure incomes will not be adversely affected, particularly those of less developed countries (LDCs), will have any possibility of successful implementation.

DOI

[15]
Moomaw W, Unruh G.Are environmental Kuznets curves misleading us? The case of CO2 emissions. Environment and Development Economics, 1997, (2): 451-463.

[16]
Arrow K, Bolin B, Costanza R, et al.Economic growth, carrying capacity, and the environment. Ecological Economics, 1995, 15(2): 91-95.income goes upthere isincreasing environ- mental degradation uptoa point, after whichenvironmental quality improves. (Therelation hasan"inverted-U" shape.) Oneexplanation ofthis finding isthat people inpoorcountries cannot afford to emphasize amenities overmaterial well-be- ing. Consequently, intheearlier stages of economic development, increased pollution isregarded asanacceptable side effect of economic growth. However, whenacountry hasattained asufficiently highstandard of living, people give greater attention toenvi- ronmental amenities. Thisleads toenviron- mental legislation, newinstitutions forthe protection oftheenvironment, andsoforth. Theabove argument does not, however, pertain totheenvironmental resource basis ofmaterial well-being, amatter weshall

DOI PMID

[17]
Duro J, Alcántara V, Padilla E.International inequality in energy intensity levels and the role of production composition and energy efficiency: An analysis of OECD countries. Ecological Economics, 2010, 69(12): 2468-2474.This paper analyses the inequality of energy intensity levels between OECD countries, its causes and evolution. The paper develops a methodology which allows the inequality in energy consumption per capita to be decomposed into explanatory factors. It also analyses the contribution of different groups of countries to this inequality. The results show that although differences in affluence are the most significant factor in explaining inequality in energy consumption per capita, the inequality in energy intensity levels plays a prominent role in reducing the inequality in energy consumption per capita over the analysed period. The paper also develops a methodology which determines the importance of different production structures and energy efficiency of productive sectors in the differences in energy use per unit of GDP between the countries analysed. The results show that sector specialisation becomes increasingly important in explaining the inequality of energy intensity, while there is a significant trend towards the convergence of energy efficiency between countries sector by sector. This trend would explain the decreasing weight of energy intensity as an explanatory factor of the inequalities in energy consumption per capita.

DOI

[18]
Sun J.The decrease in the difference of energy intensities between OECD countries from 1971 to 1998. Energy Policy, 2002, 30(8): 631-635.<h2 class="secHeading" id="section_abstract">Abstract</h2><p id="">This study uses mean deviation to measure the level of difference in energy intensities within a given group or between groups. The energy intensities of OECD countries from 1971 to 1998 are observed. The results show that the level of difference in energy intensities between OECD countries has decreased in the observed period.</p>

DOI

[19]
Duro J, Padilla E.International inequalities in per capita CO2 emissions: A decomposition methodology by Kaya factors. Energy Economics, 2006, 28(2): 170-187.<h2 class="secHeading" id="section_abstract">Abstract</h2><p id="">In this paper, we provide a methodology for decomposing international inequalities in per capita CO<sub>2</sub> emissions into Kaya (multiplicative) factors and two interaction terms. We use the Theil index of inequality and show that this decomposition methodology can be extended for analyzing between- and within-group inequality components. We can thus analyze the factors behind inequalities in per capita CO<sub>2</sub> emissions across countries, between groups of countries and within groups of countries. The empirical illustration for international data suggests some points. Firstly, international inequality in per capita CO<sub>2</sub> emissions is mainly attributable to inequalities in per capita income levels, which helps to explain its recent reduction, while differences in carbon intensity of energy and energy intensity have made a less significant contribution. This result is strongly influenced by the performance of China and India. Secondly, the between-group inequality component, which is the biggest component, is also largely explained by the income factor. Thirdly, the within-group inequality component increased slightly during the period, something mainly due to the change in the income factor and the interaction terms in a few regions.</p>

DOI

[20]
马颖. 论结构主义经济发展思路的假定前提. 经济评论, 1997, (6): 40-44.论结构主义经济发展思路的假定前提马颖在战后发展经济学演进史上，持结构主义思路的发展经济学家们同新古典主义发展经济学家们展开了长期而激烈的论战。① 在６０年代中期“新古典主义复兴”回潮到来之前，结构主义者曾在发展经济学领域占据了优势地位并在发展中国家产生...

[Ma Ying.The premise of development ideas of structuralism economy. Economic Reviews, 1997, (6): 40-44.]

[21]
库兹涅茨. 各国的经济增长. 常勋译. 北京: 商务印书馆, 2009.

[Kuznets S.Economic Growth of Nations: Total Output and Production Structure. Translated by Chang Xun. Beijing: The Commercial Press, 2009.]

[22]
Chenery H.Patterns of industrial growth. The American Economic Review, 1960, 28(3): 624-654.The kinetics of the slow combustion of the three xylene isomers have been studied manometrically in a quartz vessel, under static conditions at subatmospheric pressures over the temperature range from 410 to 550 degrees, employing 1:1 to 1:20 hydrocarbon: oxygen mixtures. It was shown that W-max = k.P-n. where P is the total initial pressure, W-max the maximum rate developed and n is 2.8, 1.9 and 1.5 for m-, o- and p-xylene, respectively. Although W-max is affected by changes in mixture composition and temperature, the value of n is independent of these parameters. Arrhenius plots were linear between 410 and 550 degrees and the activation energies for the over-all oxidation process were calculated to be 38, 39 and 40 kcal./mole for o-, m- and p-xylene, respectively. The greater ease of oxidation of o-xylene was ascribed to the greater reactivity of the chain branching intermediate. The lifetime of this intermediate was calculated to be 2 min, for o-xylene as contrasted to 20 and 17 min, for m- and p-xylene, respectively. Addition of inert gases (He, A, N-2 and SF6) increased W-max thus suggesting that this rate is governed by a diffusion controlled process at the walla of the reaction vessel. Studies of the competitive oxidation of binary mixtures of the xylenes indicated that the chain propagation reactions proceed at essentially equal rates in all three oxidations and have nearly equal activation energies.

DOI

[23]
Syrquin M, Chenery H, Mundial B.Patterns of Development, 1950 to 1983. Washington, DC: World Bank Publications, 1989.

[24]
Syrquin M, Chenery H.Three decades of industrialization. The World Bank Economic Review, 1989, 3(2): 145-181.Economists have long searched for patterns that relate successful development to structure and policy. This article reviews the experience of growth and industrialization in the postwar period in more than 100 economies, drawing on time-series data over a three-decade period. Economies are classified according to their population size, the share of primary or manufactured goods in their exports, and the weight of exports in gross domestic product (GDP). We examine the composition of demand, trade, output, manufacturing type, and factor use overall and between sectors as they relate to income growth. Higher income growth and more marked transformation are found among the groups with large populations, a predominance of manufactures in exports, and a larger role of exports. We also find that the patterns suggested by cross-country analysis are robust when tested using the time series data now available. Although development experiences may vary over time and across countries, there is sufficient uniformity within them for the main features of structural transformation to emerge as clear and consistent patterns of modern economic growth.

DOI

[25]
钟学义, 王丽. 产业结构变动同经济增长的数量关系探讨. 数量经济技术经济研究, 1997, (5): 22, 29.正 一个时期以来,产业结构作为经济结构中的一个最基本的核心的结构形式被许多经济学家所注视,如美国哈佛的大学教授霍利斯·B·钱纳里、库兹涅茨、弗来明、 克拉克、赛尔奎因、鲁宾逊等等,由于研究经济问题的最终目的之一是如何促进经济增长,因此在产业结构问题的研究中,经济增长同产业结构变动的关系也就成了 经济学家最感兴趣的研究论题之一。为此,他们写了大量的经济学著作,库兹涅茨的《投入产出经济学》、钱纳里的《结

DOI

[Zhong Xueyi, Wang Li. Investigation of the relationship between the economic growth with the industrial structure change. The Journal of Quantitative and Technical Economics, 1997, (5): 22, 29.]

[26]
Combes P.Economic structure and local growth: France, 1984-1993. Journal of urban economics, 2000, 47(3): 329-355.For 52 industry sectors and 42 services sectors, this paper tests how the local economic structure (local sectoral specialization and diversity, competition, average size of plants, and total employment density) affects the 1984-1993 employment growth of 341 local areas. These areas entirely and continuously cover the French territory. The impact of the local economic structure differs in industry and services. In industrial sectors, local total employment density, competition, and plant size always reduce local growth. Sectoral specialization and diversity have a negative impact on growth, but also increase the growth of a few sectors. Service sectors always exhibit negative specialization effects and positive diversity effects. Competition and plant size have a negative impact and density a positive one, but exceptions are observed for some sectors.

DOI

[27]
薛白. 基于产业结构优化的经济增长方式转变: 作用机理及其测度. 管理科学, 2009, 22(5): 112-120.经济增长方式转变是由高级生产要素的生产和流动所引致的国民经济 中一系列生产函数配置方式由低级向高级的动态演变过程,产业结构优化体现在发生配置方式变革的产业内部和产业之间各种生产函数的结构性调整,两者可归结为 要素配置结构变革在不同层面上的体现.以要素配置结构变革为桥梁,从产业结构合理化和高级化角度研究经济增长方式转变的机理,并对大道定理进行拓展,认为 产业结构优化与经济增长方式转变之间相互推动,使经济增长过程表现出阶段性的动态演变路径.基于此,从产业结构的微观要素配置层面和宏观动态演进层面构建 判别经济增长方式转变的衡量体系,并指出经济增长方式转变取决于政府诱导性结构变迁手段和市场内生性结构变迁动力间的兼容程度.

DOI

[Xue Bai.Transformation of economic growth patterns under the view of optimizing industrial structure: Mechanism and measurement. Journal of Management Science, 2009, 22(5): 112-120.]

[28]
Paul S, Bhattacharya R.CO2 emission from energy use in India: a decomposition analysis. Energy Policy, 2004, 32(5): 585-593.<h2 class="secHeading" id="section_abstract">Abstract</h2><p id="">This paper aims at identifying the factors that have influenced the changes in the level of energy-related CO<sub>2</sub> emissions. By means of decomposition method the observed changes are analysed in terms of four factors: pollution coefficient, energy intensity, structural changes and economic activity. The study refers to the major economic sectors of India for the period 1980&ndash;1996. The results show economic growth has the largest positive effect in CO<sub>2</sub> emissions changes in all the major economic sectors. Emissions of CO<sub>2</sub> in industrial and transport sectors show a decreasing trend due to improved energy efficiency and fuel switching. However, the reducing effect of the pollution coefficient and energy intensity on CO<sub>2</sub> emissions in agricultural sector is almost nil. The energy intensity varies over a wider range and has had a greater impact on energy-induced CO<sub>2</sub> emissions than the pollution coefficient.</p>

DOI

[29]
Ma C, Stern D.China's changing energy intensity trend: A decomposition analysis. Energy Economics, 2008, 30(3): 1037-1053.<h2 class="secHeading" id="section_abstract">Abstract</h2><p id="">China experienced a dramatic decline in energy intensity from the onset of economic reform in the late 1970s until 2000, but since then the rate of decline slowed and energy intensity actually increased in 2003. Most previous studies found that most of the decline was due to technological change, but disagreed on the role of structural change. To the best of our knowledge, no decomposition study has investigated the role of inter-fuel substitution in the decline in energy intensity or the causes of the rise in energy intensity since 2000. In this paper, we use logarithmic mean Divisia index (LMDI) techniques to decompose changes in energy intensity in the period 1980&ndash;2003. We find that: (1) technological change is confirmed as the dominant contributor to the decline in energy intensity; (2) structural change at the industry and sector (sub-industry) level actually increased energy intensity over the period of 1980&ndash;2003, although the structural change at the industry level was very different in the 1980s and in the post-1990 period; (3) structural change involving shifts of production between sub-sectors, however, decreased overall energy intensity; (4) the increase in energy intensity since 2000 is explained by negative technological progress; (5) inter-fuel substitution is found to contribute little to the changes in energy intensity.</p>

DOI

[30]
Ang B.Decomposition analysis for policymaking in energy: Which is the preferred method?. Energy Policy, 2004, 32(9): 1131-1139.<h2 class="secHeading" id="section_abstract">Abstract</h2><p id="">Although a large number of energy decomposition analysis studies have been reported in the last 25 years, there is still a lack of consensus among researchers and analysts as to which is the &ldquo;best&rdquo; decomposition method. As the usefulness of decomposition analysis has now been firmly established in energy studies and its scope for policymaking has expanded greatly, there is a need to have a common understanding among practitioners and consistency on the choice of decomposition methods in empirical studies. After an overview of the application and methodology development of decomposition analysis, the paper attempts to address the above-mentioned issues and provide recommendations.</p>

DOI

[31]
Al-Ghandoor A, Al-Hinti I, Mukattash A, et al.Decomposition analysis of electricity use in the Jordanian industrial sector. International Journal of Sustainable Energy, 2010, 29(4): 233-244.Industry is responsible for about 31% of total electricity demand in Jordan. This paper analyses the changes in industrial electricity demand during the years 1998-2005 and identifies the factors affecting this demand. In order to gain greater insight into past electricity use changes, a Laspeyers decomposition approach was used to disaggregate changes in the electricity demand of the Jordanian industrial sector into production, structural and efficiency effects. To accomplish the objectives of this paper, the Jordanian industrial sector was disaggregated into seven sub-sectors: mining of chemical and fertilizer minerals, paper, plastics, petroleum, cement, iron and steel, and other industries. A major finding of this paper is that, although increased industrial production caused electricity demand to increase between 1998 and 2005, significant improvements in energy efficiency and structural shift have contributed to reducing the rate of this increase.

DOI

[32]
王铮, 朱永彬. 我国各省区碳排放量状况及减排对策研究. 中国科学院院刊, 2008, 23(2): 109-115.温室气体导致全球气候变暖已为 世界所公认,其中又以CO2气体的排放为主。本文针对主要排放源——能源消费导致的碳排放进行核算,并在省级尺度上对中国1995—2006年的碳排放进 行对比发现,碳排放较高的省份集中在消费结构以煤为主的地区,如山西;以及第二产业比重较大的地区,如山东、河北、江苏等省。而一些经济发达、科技领先的 省市,如北京和上海碳排放有明显下降趋势。

DOI

[Wang Zheng, Zhu Yongbin.Study on the status of carbon emission in provincial scale of china and countermeasur es for reducing its emission. Bulletin of Chinese Academy of Sciences, 2008, 23(2): 109-115.]

[33]
郑长德, 刘帅. 产业结构与碳排放: 基于中国省际面板数据的实证分析. 开发研究, 2011, (2): 26-33.本文采用我国30个省份2000-2008年的相关数据,使用面板数据的分析方法对我国各省份的产业结构与碳排放的关系进行了实证分析,分析结果表明:总体上,经济增长是导致我国碳排放增加的主要因素,具体到各产业而言,第二产业对碳排放的影响最大,第一、三产业的影响较小;此外,第二产业的发展结构也不合理,"三高"和对能源依赖较高的企业居多,应大力发展高新技术产业和低碳产业;各省份三次产业的发展结构也不尽合理,应鼓励发展第三产业,保持第一产业,改造第二产业.

[Zheng Changde, Liu Shuai.Industrial structure and carbon emissions: An empirical analysis based on provincial panel data Chinese. Research and Development, 2011, (2): 26-33.]

[34]
刘再起, 陈春. 低碳经济与产业结构调整研究. 国外社会科学, 2010, (3): 21-27.在共同应对全球气候变暖的大背 景下,低碳经济无疑成为世界各国寻求经济增长与环境保护的平衡路径和寻求长期的可持续发展模式的解决之道。本文选择全球具有代表性的7个国家(美、日、 德、法、英、俄、中)的面板数据,运用类似似乎不相关回归方法(Cross-section SUR)对各国产业结构调整对二氧化碳排放量影响的变系数不变截距模型进行实证分析,结果表明:各国产业结构的变化对碳排放量的影响程度不一,而且几乎所 有产业的发展均会增加二氧化碳排放量,但第一、二、三产业的影响度逐次递减。因此要发展低碳经济,必须视国情合理选择主导产业,加快产业结构调整。

[Liu Zaiqi, Chen Chun.Research on low carbon economy and industrial structure adjustment. Abroad Social Science, 2010, (3): 21-27.]

[35]
刘卫东, 张雷, 王礼茂, 等. 我国低碳经济发展框架初步研究. 地理研究, 2010, 29(5): 778-787.<p>在哥本哈根世界气候变化大会上,我国向世界承诺2020年单位国内生产总值的二氧化碳排放量比2005年下降40％~45％。本文在对已有研究成果进行系统梳理的基础上,分析了影响我国碳排放的主要因素,核算了主要减排途径的碳减排潜力,提出了至2020年我国发展低碳经济的基本框架。研究发现,碳排放强度与产业结构演化之间存在倒U字形曲线关系,发展模式转变和产业结构调整取得实质性成效是实现2020年减排目标的前提。此外,工业技术节能、建筑节能和道路交通节能也还都有一定的潜力。在不同情景下,工业技术节能对实现2020年减排目标的贡献程度在12%~14%之间,建筑节能和增加非化石能源规模分别可以起到10%左右的贡献,道路交通节能的贡献率在2%~3%之间。</p>

[Liu Weidong, Zhang Lei, Wang Limao, et al.A sketch map of low carbon economic development in China. Geographical Research, 2010, 29(5): 778-787.]

[36]
谭飞燕, 张雯. 中国产业结构变动的碳排放效应分析: 基于省际数据的实证研究. 经济问题, 2011, (9): 32-35.在碳排放模型框架内，利用测算的全国省际二氧化碳排放数据，通过设定多种模型形式考察了各种 因素特别是产业结构变动对二氧化碳排放产生的影响。分析结果表明，产业结构的工业化进程直接加剧了二氧化碳的排放，产业结构变化是中国碳排放增长的重要驱 动因素之一，其影响程度较大；同时，FDI环境效应的合力是负面的，贸易并非国际碳污染转移的主要渠道。

[Tan Feiyan, Zhang Wen.Industrial structure and carbon emissions in China: Evidence from province level data. Economic Issues, 2011, (9): 32-35.]

[37]
Inada K.A note on the simple majority decision rule. Econometrica, 1964, 32(4): 525-531.Data on the postembryonic development, longevity, ageing and rejuvenation of planarians are reviewed. From the available information, including the results of their own experiments, the authors conclude that planarians display a pattern of ageing intermediate between that of the "non-ageing" (fully-renewable) invertebrates and typical senescence as seen in mammals. The reversibility of planarian age changes and the measurement of "rejuvenation" in planarians are discussed.

DOI

Options
文章导航

/

〈 〉