Size Effect In The Chinese Stock Market

In numerous empirical asset pricing models,firm size is often utilized and considered a determinant of expected stock returns.However,there is little consensus on the extent and stability of size premiums among existing research.Moreover,the manifestation of the size effect varies across different levels of the stock market.In previous studies,scholars have employed two primary methods to estimate the size effect: one approach uses the raw firm size or the logarithm of the size as the explanatory variable,while the other defines the size factor(SMB)as the difference between the monthly average returns of the smallest and largest stock groups,incorporating it into the study of stock return variations.This paper applies a metaanalysis method to studies employing raw size values,examining publication biases in the published literature and estimating the true "size effect" within the Chinese market.Simultaneously,the Fama-French five-factor model,with SMB as a proxy for firm size,is used to examine the presence of the "size effect" in different sectors of the Chinese stock market.In evidence-based econometric research,this paper collects 312 slope coefficient estimates reflecting the relationship between company size and stock returns from 35 articles based on the Chinese stock market,conducting a meta-analysis on the size premium of the Chinese stock market for the first time.The study finds that:(1)empirical research on the size premium in the Chinese stock market generally exhibits publication bias,as scholars tend to publish statistically significant negative slope coefficients.The degree of publication bias is related to the publication date,with more recent articles reporting smaller results.(2)After isolating the influence of publication bias,the true effect value is-0.082,supporting the existence of the size effect in the Chinese market.(3)By collecting factors that may influence the slope coefficient in published research and using Bayesian model averaging to explore the causes of heterogeneity,it is found that,apart from financial leverage,other factors have no significant impact on the slope coefficient size.This supports,to some extent,the theoretical hypothesis that company size is a pricing factor for stock returns.This paper collects A-share data from May 2000 to April 2021,as well as data from the main board,SME board,and GEM board from May 2010 to April 2021.The Fama-French five-factor model is employed to examine the presence of the size effect in each market and compare the significance and magnitude differences of the size effect across markets.Additionally,the study analyzes the performance of other market anomalies and ultimately uses the Fama-French three-factor model to test the robustness of the conclusions.The research finds that:(1)the five-factor model is applicable in the A-share market,and company size significantly negatively affects Ashare market stock returns,with this influence being more pronounced for small companies.(2)A significant size effect exists in each board market,with varying significance and strength of the size factor.(3)The performance of the five-factor model,as well as of the remaining factors,varies across board markets,with the main board’s investment factor,GEM board’s value factor,and SME board’s orthogonalized profitability factor displaying greater significance,distinct from the other two markets.Moreover,the significance of the investment factor is weak in both the SME board and GEM board markets,indicating that the investment factor may be unable to explain the risk associated with investment behavior in these two board markets.Finally,based on the research conclusions,this paper offers recommendations to researchers,investors,and regulatory authorities.