ESG信息摩擦与收益可预测性:ESG评级分歧的新证据  

ESG information frictions and return predictability:new evidence on ESG rating divergence

作　　者：张伟伟 张景静 赵宇 ZHANG Weiwei;ZHANG Jingjing;ZHAO Yu(Northeast Asia Research Center,Jilin University,Changchun Jilin 130012,China;School of Economics,Shandong University,Jinan Shandong 250100,China;School of Finance,Changchun University of Finance and Economics,Changchun Jilin 130117,China)

机构地区：[1]吉林大学东北亚研究中心,吉林长春130012 [2]山东大学经济学院,山东济南250100 [3]长春财经学院金融学院,吉林长春130117

出　　处：《中国人口·资源与环境》2024年第12期98-107,共10页China Population,Resources and Environment

基　　金：国家社会科学基金重点项目“金融发展对我国绿色经济发展的影响机理研究”(批准号:21AJY014)。

摘　　要：随着ESG投资理念日趋主流化,越来越多的投资者开始依赖评级机构的ESG评级评估企业ESG表现,但由于缺乏共识性的ESG评估标准,各评级机构关于同一家企业的ESG评级结果存在差异,这一现状可能对资本市场的估值逻辑产生影响。鉴于此,该研究基于2009—2022年10家评级机构关于中国A股企业的ESG评级数据,采用多种统计方法验证了中国资本市场ESG评级分歧的存在性。在此基础上,采用各评级机构对同一家企业ESG评分的标准差测度ESG评级分歧,考察以ESG评级分歧为代表的信息摩擦与股票市场表现的关系,从而探索成熟资本市场估值理论在中国的适用场景。(1)研究发现,ESG评级分歧显著降低了未来股票收益,即验证了ESG评级分歧对未来股票收益的预测能力,肯定了ESG评级分歧的定价效应。该结论在经过内生性处理、更换核心变量测度等系列稳健性检验后依旧成立。(2)作用机制检验显示,ESG评级分歧加剧了未来股价的下降幅度,降低了投资者对股票的市场认可度以及分析师对企业的关注度和荐股评级情况,从而降低了未来股票收益表现。(3)拓展性讨论部分,该研究基于Fama-French3因子及5因子模型估计的收益残差标准差作为特质波动率的测度变量,使用截面Fama-MacBeth回归方法考察控制了ESG评级分歧后特质波动率对股票收益率预测能力的变化,发现ESG评级分歧与未来股票收益负相关的原因有一部分来自“特质波动率之谜”的定价异象。As ESG investment principles increasingly gain traction,a growing number of investors are turning to ratings from agencies to assess corporate ESG performance.However,the lack of consensus on ESG evaluation standards leads to discrepancies in the ratings assigned to the same firm by different agencies,which may impact the valuation logic within capital markets.In this context,this study utilized ESG rating data from ten agencies regarding Chinese A-share companies from 2009 to 2022 and employed various statistical methods to validate the existence of ESG rating divergence in the Chinese capital market.Building on this foundation,the study mea-sured ESG rating divergence through the standard deviation of ratings assigned to the same firm,examining the relationship between this divergence—representing information frictions—and stock market performance,thus exploring the applicability of established capi-tal market valuation theories in the Chinese context.The findings indicated that.①ESG rating divergence significantly reduced future stock returns,thereby confirming its predictive ability concerning future stock performance and affirming its pricing effect.This conclu-sion remained robust following a series of robustness checks,including addressing endogeneity and changing the measurement of core variables.②Mechanism analyses revealed that ESG rating divergence exacerbated future declines in stock prices,diminished inves-tors'market recognition of the stocks,and reduced analysts'attention and recommendations regarding these firms,consequently leading to lower future stock returns.③In the broader discussion,the study employed the standard deviation of residuals estimated from the Fa-ma-French three-factor and five-factor models as a measure of idiosyncratic volatility,using cross-sectional Fama-MacBeth regression methods to investigate changes in the predictive power of idiosyncratic volatility over stock returns when controlling for ESG rating di-vergence.It found that part of the negative correlation bet

关 键 词：ESG评级分歧 异质信念 信息噪声 特质波动率异象 资产定价 

分 类 号：F270[经济管理—企业管理]