Empirical Evidence Of Herding Behavior In The Chinese Stock Market

This thesis provides a comprehensive study of herding behavior in the Chinese Stock Market using the cross-sectional absolute deviation of returns method（CSAD）proposed by（Chang et al.,2000）,which captures the non-linearity relationship between the dispersion of individual returns and market return.According to（Christie ＆ Huang,1995）and（Chang et al.,2000）,in a stock market,herding behavior occurs when individual returns begin to converge towards the consensus of the market,leading to a decrease in the dispersion of stock return from the market return.First,this study investigates the scope to which herding behavior towards the market consensus for the Chinese Stock Market is driven by fundamental and non-fundamental factors.The empirical evidence documents a market-wide herding with reference and without any reference to fundamental factors within all Chinese Stock Markets during the sample period.Nevertheless,in most of the cases,fundamental herding is greater in magnitude than non-fundamental herding.The findings also suggest that the magnitude of herding intensity varies respectively within the markets and that the reasons for investors’ herding behavior vary under specific market conditions such as market phases,extreme market conditions,liquidity effect,financial turmoil,macroeconomic announcements,and contagion effect.Furthermore,the findings interestingly suggest that the A-shares markets are informationally richer environment and that investors in the A-shares market now rely more on fundamental information when trading.This inference is consistent with some studies that document enhanced informational efficiency in the Chinese Stock Market.In contrast,in B-shares markets,herding seems to be more intentional among B-shares market traders.This implication is consistent with some studies that suggest that since Chinese authorities have lifted trade restrictions to give access to foreign institutional investors,herding formation has become more noticeable without reference to fundamentals,suggesting that foreign institutional investors are rational but less informed relative to Chinese local investors.Hence,it is reasonable to conclude that the herding formation should prosper in a market where investors are informationally underprivileged.Second,this study examines how limits of arbitrage can affect herding behavior.Using both unique trading restrictions in the Chinese Stock Market and other commonly used limits of arbitrage metrics,a comprehensive limits of arbitrage index is constructed.The findings support that limits of arbitrage are essential for herding as it may trigger different herding patterns among investors with regard to different market conditions such as market trends,financial crisis,and under extreme market conditions.The findings imply that herding in most cases is more pronounced in falling markets in both high and low limits of arbitrage during the full sample period,FC period,and AFC period which supports the theory of homogenous trading due to flight to safety.Nevertheless,during the BFC period,investors tend to herd more during rising markets with a greater magnitude in periods of high limits of arbitrage.Besides,the findings also suggest that the pattern of herding propensity is variant over the years and is more noticeable in stock portfolios with high limits of arbitrage especially during turmoil periods coupled with high uncertainty such as the great FC period and the Chinese Stock Market 2015 crash.Finally,the results suggest that the trading constraints established in the name of protecting individual investors may in fact be detrimental to them since these additional limits of arbitrage will enhance herding behavior and will increase the inefficiency of the security market.Third,this study analyzes the effect of investors’ fear on their investment behavior in the Chinese Stock Market.This study uses the “volatility index（VIX）” as a gauge of investors’ fear.The results suggest that herding behavior in the Chinese Stock Market is stimulated by an increase in investors’ fear.The results are consistent with（Huang ＆ Wang,2017）and（Bekiros et al.,2017）who advocate that implied volatility（VIX）plays a crucial role in fueling this irrational behavior.Besides,the results indicate that even with small changes in VIX values,herding is more pronounced during the FC period,which further supports the evidence of herding behavior existence when investors’ fear emerges.Furthermore,the findings interestingly reveal that the intensity of herding in large changes in the VIX is higher than that in small changes in the VIX for most of the markets and periods,implying that compared with large changes in the VIX,investors in the Chinese Stock Market are more sensitive to large changes in the VIX and this consequently has a greater impact on their investment strategies.Fourth,this study inspects the impact of idiosyncratic volatility on the investors’ herd behavior in the Chinese Stock Market.The study employs the single factor model by（Bali ＆ Cakici,2009）to estimate the idiosyncratic volatility.The empirical results show that herding behavior exhibits diverse patterns under different equity portfolios according to the levels of idiosyncratic volatility as well as the market trend.Moreover,the FC period increases herding,especially within stock portfolios with higher idiosyncratic volatility.Finally,using monthly data,this study analyses the effect of investors’ sentiment on herding behavior.This study uses C.I.C.S.I index and I.S.I index as a proxy of investors’ sentiment in the Chinese Stock Market.The finding provides no empirical evidence of the existence of herding propensity in the Chinese Stock Market covering the whole period from 2003 to 2018 and overall sub-periods including BFC,during FC,and AFC,which is consistent with the idea that herding behavior is a short-lived phenomenon（Christie ＆ Huang,1995）.While,when conditioning herding on investors sentiment,herding intensity is more evidenced within the Chinese A-shares market,where individual investors are the major traders,implying that herding within the Chinese A-shares market is a long-lived phenomenon.Finally,this study documents that herding exists in the Chinese Stock Market and varies with levels of investors’ sentiment and market trends.Moreover,this finding provides supplementary evidence that the level of investors’ sentiment is a crucial element to identify different herding patterns among individual investors.