Registration-Based IPO Reform and the Validity of Multifactor Models: An Empirical Test Based on Chinese A-Shares

Abstract

The registration-based IPO reform signifies a substantial institutional transformation within China's capital market. This study utilizes a dataset comprising 389,140 observations from Chinese A-share listed companies between 2016 and 2024, employing the launch of the STAR Market in 2019 as a quasi-natural experiment. This study employs a multifaceted methodological approach, encompassing the difference-in-differences (DiD) method, mediation model, and panel error correction model (ECM), to examine the impact of this reform on the validity of the Fama-French five-factor model (FF5). The findings of the study indicate that following the implementation of the reform, there was a substantial increase in the model's R² by 2.1%, as well as a significant rise in the model's |α| ( where |α| denotes the absolute value of the intercept) by 5.97%. The analysis indicates that the mediating effect of disclosure quality accounts for 89.30% of the total effect, while the half-life experienced an extension from 2.42 months to 3.51 months, marking a 45% increase. This study makes a significant contribution to the asset pricing research field by expanding its scope to include institutional dependence. By examining this phenomenon, it becomes possible to understand the trade-off between disclosure quality and pricing efficiency resulting from the reform.