管理评论 ›› 2026, Vol. 38 ›› Issue (4): 206-217.

• 会计与财务管理 • 上一篇    

创业板改革并试点注册制会降低权益资本成本吗?——基于双重机器学习的实证研究

胡磊1, 何重阳1, 祝平衡2   

  1. 1. 湖南工商大学财政金融学院, 长沙 410205;
    2. 中南大学商学院, 长沙 410083
  • 收稿日期:2024-04-08 发布日期:2026-05-14
  • 作者简介:胡磊,湖南工商大学财政金融学院副教授,硕士生导师,博士;何重阳,湖南工商大学财政金融学院硕士研究生;祝平衡(通讯作者),中南大学商学院副教授,硕士生导师,博士。
  • 基金资助:
    国家社会科学基金青年项目(24CJY123)。

Does the Pilot of Registration System on the ChiNext Reduce the Cost of Equity Capital?: A Double Machine Learning Model

Hu Lei1, He Chongyang1, Zhu Pingheng2   

  1. 1. School of Finance, Hunan University of Technology and Business, Changsha 410205;
    2. School of Business, Central South University, Changsha 410083
  • Received:2024-04-08 Published:2026-05-14

摘要: 研究注册制改革的溢出效应,有利于优化资本市场资源配置功能。本文基于2016—2022年创业板和主板上市公司的数据,采用双重机器学习模型探究创业板改革并试点注册制产生的溢出效应。研究发现,改革后存量上市企业的权益资本成本得到显著降低,且这一溢出效应主要通过提高股票流动性和信息披露质量实现。此外,在分析师关注度低、审计质量低和融资约束高的企业中,创业板改革并试点注册制对降低权益资本成本的效果更加显著。本文丰富了注册制溢出效应的相关研究,对全面深化资本市场改革具有重要的实践启示。

关键词: 注册制改革, 创业板, 双重机器学习, 权益资本成本, 溢出效应

Abstract: Studying the spillover effects of the registration-based IPO system helps optimize the capital market’s resource allocation function. Based on data from ChiNext and main board listed companies from 2016 to 2022, this study uses a double machine learning model to explore the spillover effects of the registration-based IPO reform on ChiNext. The findings reveal that the reform has significantly reduced the cost of equity capital for ChiNext-listed companies, with the spillover effect mainly driven by improvements in stock liquidity and information disclosure quality. Moreover, the impact of the registration-based IPO reform on reducing the cost of equity capital is more pronounced for companies with low analyst coverage, low audit quality and high financing constraints. This paper enriches the relevant research on the spillover effects of the registration-based IPO system and has important practical implications for comprehensively deepening the capital market reform.

Key words: the registration-based IPO system, ChiNext, double machine learning, the cost of equity capital, spillover effect