摘要
本文采用我国深沪两市A股上市公司面板数据,利用同伴企业股票收益波动作为工具变量,首次证实我国上市公司资本结构、负债期限结构决策受到同伴决策的显著影响。进一步检验发现,同伴效应在方向、范围上表现出不对称性;资本结构指标表现出"乘数效应";采用基于同伴企业集团内非同一行业企业构建的工具变量去除关联效应后,企业间资本结构决策显著相关性仍然存在。为使实证结果更加可靠,本文分别从统计模型设定、工具变量和同伴企业衡量指标三个方面进行稳健性检验。本文的研究对我国学术界开展企业财务决策同伴效应及其属性的理论与实证研究具有开拓性价值。
Peer effects can be characterized as a commonly seen phenomenon among firms, not only in operational decisions, but also, as many scholars contend, in financial decisions, such as capital structure. Although there are several theories explaining peer effects in firms＇ decisions, it is tricky to empirically identify the existence of peer effects due to potential endogeneity problems. Using unbalanced panel data of all non-financial Chinese A-share listed firms （listed on Shanghai and Shenzhen market） and within-industry peer firms＇ compounded yearly stock return shock as instrument, our study, for the first time, empirically presents evidence that peer effects exist in capital structure decision, as well as in the determination of debt maturity. Further investigations show us that firstly, peer effects show asymmetry with respect to peer firms＇ moves in different directions and scales. More specifically, focal firms would negatively adjust capital structure when peers drop their debt ratio, but would sit back when peers tend to increase borrowing. Also, focal firms response only to peers＇ minor adjustments and choose to ignore sharp changes; secondly, there are social multiplier effects in the leverage ratio. In other words, small changes at firm level will transfer to peer firms, which eventually end up in fluctuations of multiple scales at industry level; thirdly, the statistically significant correlation in capital structure decision among firms is still robust even if we use another instrument （based on peer firms＇ business group member firms who operate in different industries） to account for the correlated effects. All these results point to the fact that peer effects, instead of correlat- ed effects, explain the strongly correlated capital structure among firms in each industry. In order to guarantee that our previous find- ings are convincing, we perform robustness tests along three important dimensions, basic empirical setting, construction of instrumental variable and measurement method
出处
《南开管理评论》
CSSCI
北大核心
2017年第2期58-70,共13页
Nankai Business Review
基金
国家自然科学基金面上项目(71150002)资助
关键词
中国上市公司
资本结构
同伴效应
工具变量
Chinese Listed Firms
Capital Structure
Peer Effects
Instrumental Variable
