加载中...
Green Credit Policy and the Agglomeration of Emission-intensive Industrial Chains
产业链上各环节的集聚不仅有助于我国从全产业链角度实现延链、补链和强链的目标,对于高排放产业而言还能实现循环减排,推动产业链绿色化转型。但对于如何推动高排放产业链集聚,现有文献鲜有分析。本文在划分我国制造业产业链关联产业集合的基础上,考察了绿色信贷政策对高排放产业链集聚的影响效应。研究发现,绿色信贷政策有助于推动高排放产业从单一产业集聚转向产业链集聚。产生这一作用的原因是,绿色信贷政策在减少高排放企业的银行信贷后,迫使相关企业加强了商业信用的使用,进而寻求与产业链关联产业共同集聚。进一步分析发现,不同于单一产业集聚通过规模效应实现减排,产业链集聚能够产生循环减排效应,尤其促进水资源重复使用并减少污水排放。本文不仅从产业链集聚的崭新视角补充了绿色信贷政策的效应研究,还为优化我国产业链空间布局、推动高排放产业链绿色化转型提供了有益的政策启示。
Abstract
Against the backdrop of rising global uncertainty, local governments in China have increasingly shifted their development strategies from traditional industry-oriented investment promotion toward an industrial chain-based approach. This transition aims to encourage the spatial co-agglomeration of upstream, midstream, and downstream activities within the same industrial chain. For emission-intensive industries, such co-agglomeration is particularly important. Beyond reducing production costs and increasing efficiency through labor sharing and knowledge spillovers, closer spatial integration along the chain facilitates circular production processes and supports the green transformation of the entire industrial chain. Despite its growing policy relevance, however, little is known about how industrial chain agglomeration can be effectively promoted in emission-intensive industries. This paper argues that whether emission-intensive industries can move from single-industry agglomeration to industrial chain agglomeration crucially depends on the financing environment they face. Emission-intensive industries are typically capital-intensive and face substantial financing needs, making their location choices highly sensitive to external financing conditions. In China 's bank-dominated financial system, changes in the availability of bank credit to emission-intensive industries can therefore serve as a key lever in reshaping the spatial patterns of their industrial chains. To operationalize this idea, we construct an industry relatedness matrix for China 's manufacturing sectors based on multiple dimensions, including labor-pooling linkages, technological spillovers, input-output relationships, and shared markets or resources. This matrix allows us to identify sets of industries that are connected along the same industrial chain. We then measure the agglomeration level of each industry and its chain-related industries at the prefecture-city level using location quotients. Industrial chain agglomeration is captured by the correlation between the agglomeration of a focal industry and that of its related industries within the same chain. Building on this framework, we exploit the green credit policy as an exogenous shock to bank credit availability faced by emission-intensive industries. Using firm-level data from China's Industrial Enterprise Database (2004-2015), we examine how this financing shock reshapes the spatial co-agglomeration of emission-intensive industrial chains. Our empirical results show that the green credit policy significantly promotes the spatial co-agglomeration of different production stages within emission-intensive industrial chains at the city level. This finding remains robust after excluding alternative explanations, mitigating endogeneity concerns, and conducting pre-trend tests as well as a series of additional robustness checks. Mechanism analysis indicates that, following the policy-induced green credit shock, emission-intensive industries relocate toward clusters of upstream and downstream industries in order to obtain greater access to trade credit, which constitutes the primary channel through which industrial chain agglomeration is strengthened. Moreover, we find that, unlike single-industry agglomeration that mainly reduces emissions through scale effects, industrial chain agglomeration generates a circular emission-reduction mechanism by enhancing water recycling and reducing external wastewater discharges. This paper contributes to the existing literature in three main ways. First, while prior studies on green credit policy focus primarily on firm-level outcomes such as emission reduction and green innovation, this paper extends the analysis to the spatial patterns of emission-intensive industries. Second, whereas most studies on industrial spatial patterns in China emphasize single-industry agglomeration, this paper shifts attention to industrial chain agglomeration and identifies a policy channel through which such agglomeration can be fostered in emission-intensive industries. Third, in contrast to existing studies that highlight scale-based emission reductions within single industries, this paper demonstrates that industrial chain agglomeration can generate additional environmental benefits through enhanced resource recycling and inter-industry coordination.
绿色信贷政策显著促进高排放产业链集聚。
基于2004-2015年中国工业企业数据,利用2007年《关于落实环保政策法规防范信贷风险的意见》作为准自然实验,采用三重差分模型。结果显示,本产业集聚水平×高排放产业×2007年以后的三重交互项系数在1%水平上显著为正(系数0.0251),表明相比未受冲击的产业,绿色信贷政策显著增强了高排放产业与其产业链关联产业集聚水平的正向关联性,推动高排放产业从单一产业集聚转向产业链集聚。
绿色信贷政策显著降低高排放企业银行信贷。
机制分析显示,高排放产业×2007年以后的双重交互项对银行信贷的影响系数在1%水平上显著为负(系数-0.0079),证实绿色信贷政策平均而言显著降低了高排放企业获得的银行信贷。同时,该政策对高排放企业获得或提供的商业信用并无显著平均影响,说明商业信用不会自然增加以弥补信贷缺口。
高排放企业为获取商业信用而向产业链关联产业集聚地靠拢。
企业层面机制检验发现,在绿色信贷政策冲击下,当高排放企业所在城市的产业链关联产业集聚程度更高时,其获得的商业信用显著更多(关联产业集聚水平×高排放产业×2007年后的系数为0.0043,显著为正)。而本产业集聚水平高反而会减少商业信用(系数-0.0029),说明高排放企业唯有向产业链关联产业集聚地靠拢才能增加商业信用,单纯向本产业集聚靠拢反而加剧融资约束。
产业链集聚产生循环减排效应,促进水资源重复利用并减少污水排放。
进一步分析产业链集聚的减排效应,本产业集聚水平×关联产业集聚水平的交互项对废水排放强度的影响系数显著为负(-0.9787),但对烟粉尘和二氧化硫排放强度无显著影响。产业链集聚显著减少了单位产值的废水产生量(系数-2.2746),且显著增加了水重复使用强度(系数0.3788),表明其减排机制是通过促进水资源在产业链上下游之间循环利用实现,区别于单一产业集聚的规模效应。
政策效应在银行垄断城市、环境违规较多城市更为显著。
排除替代性解释的异质性检验显示,在银行市场相对垄断的城市,绿色信贷政策显著促进了高排放产业链集聚(系数0.0324),而在银行竞争城市无显著效应;在环境违规企业较多的城市效应显著(系数0.0348),而在环境违规少的城市不显著。这证实确实为绿色信贷政策而非其他干扰性政策驱动的结果。
核心解释变量
核心解释变量为绿色信贷政策冲击(高排放产业×2007年以后虚拟变量的交互项),以及本产业集聚水平(区位熵对数)与其交互构成的三重差分项。高排放产业根据产业层面废水、二氧化硫及烟粉尘排放强度中位数划分。
被解释变量
被解释变量为产业链集聚程度,以焦点产业及其产业链关联产业集聚水平(区位熵)之间的相关性刻画。机制检验的被解释变量包括银行信贷(剔除应付账款后流动负债/总资产)、获得商业信用(应付账款/总资产)、提供商业信用(应收账款/总资产)及净商业信用。减排效应检验采用烟粉尘、二氧化硫及废水排放强度,以及废水处理量、废水产生量和水重复使用强度(均除以工业产值)。
样本与数据
主要样本为2004-2015年中国工业企业数据库(剔除2009、2010及2014年),覆盖162个三位数制造业产业。产业链划分数据还包括2007年投入产出表、2005年中国人口抽样调查数据。稳健性检验使用2008-2020年全国税收调查数据。污染排放强度数据来自中国工业企业污染数据库。
识别方法 / 模型设定
以2007年《关于落实环保政策法规防范信贷风险的意见》出台作为绿色信贷政策的准自然实验,构建三重差分模型(本产业集聚水平×高排放产业×政策时间)识别政策效应。采用产业—年份、城市—年份及城市—产业联合固定效应,标准误聚类至产业—年份层面。稳健性检验包括事前趋势检验、安慰剂检验、Bartik工具变量、调整空间范围和产业链划分方式等。
内生性及稳健性检验
稳健性检验包括:(1)调整空间范围为城市群和县级层面;(2)调整产业链关联产业集合的划分方式(次优分组、单一投入产出关联矩阵);(3)构建Bartik工具变量处理内生性;(4)使用2008-2020年全国税收调查数据并以2012年《绿色信贷指引》为冲击;(5)事前趋势检验与三种安慰剂检验;(6)调整集聚水平计算方法(就业人数替代企业数)及更换关联产业集聚水平的计算方式;(7)调整政策冲击时点为2012年;(8)保留区位熵小于1的样本;(9)修改标准误聚类层次。排除替代性解释的检验基于银行市场结构与环境违规企业占比的异质性分析。
中国工业企业数据
工业企业可支持的研究问题:本文核心数据为中国工业企业数据库,该数据可用于分析绿色信贷政策冲击下高排放企业的空间选址、产业链上下游匹配及集聚水平变化,也可用于测度城市-产业层面的集聚程度。
中国工业企业基本信息扩展数据
工业企业可支持的研究问题:基于绿色信贷政策实施前后,如何利用企业基本信息的精确与模糊匹配结果,识别排放密集型产业链上下游企业的空间区位变化。
更多相关数据正在补充