摘要
从市场竞争结构角度出发,采用2000~2020年中国沪深A股上市公司数据,通过传统计量方法和基于连续处理变量的非参数推断双重机器学习方法,考察不同垄断势力企业在杠杆操纵行为上的差异。研究表明:相较于高垄断势力企业,融资难度更大的低垄断势力企业存在更大的杠杆操纵动机;宽松货币政策会缓解企业财务杠杆操纵问题,但增加的资金供给更多流入高垄断势力企业,导致低垄断势力企业杠杆操纵程度下降幅度低于高垄断势力企业。
From the perspective of market competition structure,this study uses the data of Chinese A-share listed companies from 2000 to 2020 to study the differences in leverage manipulation behavior of companies with different monopoly power through the traditional method and DML(Double Debiased Machine Learning)nonparametric inference with continuous treatments.The results show that compared with high-monopoly-power companies,companies with low monopoly power have greater incentives to manipulate financial leverage;Easing monetary policy will alleviate the problem of financial leverage manipulation of companies,but the increased capital flows more into high-monopoly-power companies,leading to a more significant decrease in theirs leverage manipulation degree than high-monopoly-power companies.
作者
彭方平
廖敬贤
何锦安
PENG Fangping;LIAO Jingxian;HE Jin’an(Sun Yat-sen University,Guangzhou,China)
出处
《管理学报》
CSSCI
北大核心
2023年第2期297-307,共11页
Chinese Journal of Management
基金
教育部人文社会科学研究规划基金资助项目(21YJA790044)。
关键词
杠杆操纵
垄断势力
成本加成
双重/无偏机器学习
leverage manipulation
monopoly power
markup
double/debiased machine learning model