《保险研究》20260405-《算法交易、流动性消耗与股价崩盘风险》（黄婷婷、李路）

［摘   要］随着算法交易在中国资本市场的快速普及,其对市场稳定性的潜在影响日益引发监管层与学界关注。本文基于中国A股市场高频交易数据,构建算法交易活跃度指标,系统考察其对上市公司股价崩盘风险的影响及作用机制。研究发现,算法交易活跃度与股价崩盘风险显著正相关。机制检验表明,该效应并非主要源于管理层坏消息囤积行为,而是通过改变市场微观结构发挥作用:算法交易消耗市场流动性、加剧订单流毒性,并抑制分析师等信息中介的跟踪与信息挖掘,削弱市场在负面冲击下持续承接卖压的能力,从而放大股价非线性下跌风险。异质性分析进一步显示,上述关系在机构持股比例低、异质波动率高、市值小及上市年限短的公司中更为显著。本文的研究揭示了特定类型算法交易在提升交易效率之外可能带来的潜在负外部性,为完善程序化交易监管制度、防范流动性冲击下的尾部风险提供了微观经验证据。

［关键词］算法交易；股价崩盘风险；流动性；基本面信息挖掘

［基金项目］本研究得到教育部人文社会科学研究规划基金项目“私募基金绿色治理对企业低碳创新的影响机制研究”（23YJA790041）和上海高校教师培养资助计划资助。

［作者简介］黄婷婷,上海立信会计金融学院会计学院讲师、博士,研究方向:资本市场信息中介、高频交易；李路(通讯作者),上海外国语大学国际金融贸易学院教授,研究方向:对冲基金、高频交易。

Algorithmic Trading,Liquidity Consumption,and Stock Price Crash Risk

HUANG Ting-ting,LI Lu

Abstract：With the rapid development of algorithmic trading in China’s capital market,its impacts on market stability has attracted increasing attention from regulators and academics.Using high-frequency trading data from the Chinese A-share market,this study constructs an algorithmic trading activity indicator that examines its impacts on firm-level stock price crash risk and the underlying mechanisms.The results show that algorithmic trading activity is significantly positively associated with stock price crash risk at the firm level.Mechanism analyses indicate that this impact is not primarily driven by managerial bad-news hoarding;rather,it operates mainly through changes in market microstructure.Specifically,algorithmic trading increases liquidity consumption and order flow toxicity,and suppresses the tracking and information production activities of fundamental information intermediaries such as financial analysts,thereby weakening the market’s capacity to continuously absorb sell pressures under negative shocks,which in turn amplifies nonlinear stock price declines.Further heterogeneity analyses show that this positive relation is more pronounced among firms with lower institutional ownership,higher idiosyncratic volatility,smaller market capitalization,and shorter listing histories.Overall,this study highlights the potential negative externalities of a specific type of algorithmic trading beyond its efficiency-enhancing role,and provides micro-level empirical evidences for improving the regulation of algorithmic trading and mitigating tail risks arising from liquidity shocks.

Key words：algorithmic trading;stock price crash risk;liquidity;fundamental information acquisition