| The China Securities Regulatory Commission initiated a pilot scheme on March 31,2010,allowing margin trading and short selling of stocks on a designated list.As of December 31,2017,the average of monthly margin debt is 180 times of the average of monthly shorting debt.Moreover,the margin debt surges to above 2,000 RMB billion during the dramatic increases and decreases in 2015 Chinese stock market.Thus,the impact of margin trading on stock market should be more pronounced than that of short selling.However,prior literature mainly focuses on the overall impact of this margintrading and short-selling program,or the impact of short-selling activities only.As margin trading is much more active than short selling in China and that margin trading has been blamed for the abnormal fluctuation of the Chinese stock market in 2015,this dissertation concentrates on characteristics of margin-trading underlying stocks,trading strategies possibly used by margin traders,and the effect of margin-trading activities on information efficiency and price crashes risk.Specifically,our main findings are as follows:Firstly,being aware of that individual investors,who are widely known as speculative investors,may be the primary group engaging in margin trading,we study the speculative characteristic of margin-trading underlying stocks and its impact on margin-trading activities.Using various measures of speculative stock,e.g.,small size,high P/E ratio,high turnover,lottery stock,and number of individual stock holders,we employ one-way ANOVA and panel regression methods to study the reaction of margin trading to speculative underlying stocks.We find that more speculative stocks are associated with increased margin-trading activities,indicating that margin traders are dominated by individual investors and they prefer speculative underlying stocks.Secondly,we investigate trading strategies possibly used by margin traders.By studying the return predictability of margin-trading activities,we find that stocks with heavily margin-buying activities do not outperform lightly margin-bought underlying stocks,indicating that margin traders do not follow informational trading strategies when opening a margin-trading position.In addition,we find that margin-covering activities are positively related to subsequent returns,indicating that margin traders possess no new negative firm-specific information when covering their positions.Thus,margin traders in China may be uninformed;i.e.,they possess no new firm-specific information about future returns when conducting margin trading.Moreover,we examine the relationship between margin-trading activities and previous returns,and find that margin-trading activities are positively correlated with past stock returns,especially at weekly or monthly frequency.This suggests that margin traders may follow positive-feedback strategies;that is,they are likely to trade on past returns instead of new firm-specific information.Furthermore,by analyzing the relationship between margin-trading activities and the trading signals of moving-average trading rules,we find that margin trading is more active when the short-run moving average of an individual stock price is above the long-run moving average,especially at daily or weekly frequency,indicating that margin traders may follow moving-average trading rules.Thus,these results provide further evidence that margin traders in the Chinese stock market are noise traders rather than informed traders because they mainly follow non-informational trading strategies used by noise traders,e.g.,positive-feedback strategies and moving-average rules.Thirdly,we examine the effect of margin trading on information efficiency and reconcile the debate regarding the effect that margin trading has on price efficiency.We separate information efficiency into information content and price-adjustment speed,and use various measures to reflect these two aspects of information efficiency,respectively.With this specification,we use a difference-in-difference(DID)approach to verify whether implementing the margin-trading program affects price efficiency.We find that when compared with non-pilot stocks,pilot stocks' information content drops substantially,which decreases price efficiency,but price-adjustment speed increases rapidly,which increases price efficiency.These results indicate that the implementation of margin trading has opposite effects on information content and price-adjustment speed.Moreover,as there is much variation in margin-trading activities among eligible stocks,we cross-sectionally investigate the impact of margin-trading activity on two aspects of price efficiency.We find that more margin-trading activities are associated with increased pricing errors but reduced idiosyncratic volatility,implying that more margin-trading activities are associated with lower information content.In contrast,margin-trading activities are negatively associated with price-delay measures and with the cross-correlation between stock return and lagged market return,indicating that more margin-trading activities are associated with higher price-adjustment speed.Overall,margin trading lowers information content but speeds up price adjustment.This is consistent with results of DID analyses,which show that margin trading has opposite effects on information content and price-adjustment speed.Our results are robust after controlling for endogeneity,the effect of short selling,and the interference of outliers.Finally,we investigate the impact of margin trading on price crashes risk of the underlying stock.We use frequency of price's extremely downwards,negative conditional skewness(NCSKEW),and down-to-up volatility(DUVOL)as measures of crashes risk,and examine the effect of margin trading on future crashes risk.We find that intensed margin-trading activities are associated with higher NCSKEW and DUVOL,indicating that margin trading is not beneficial to price's crashes risk of underlying stocks.Moreover,this harmful impact of margin trading is more significant for stocks with lower negative firm-specific information,e.g.,speculative stocks,less short-selling stocks,and stocks with higher asymmetric information.In addition,we find that margin-trading activities has no significant effect on manager behavior of hiding firm's negative information.These results suggest that the non-beneficial effect of margin trading on future crashes risk may due to that margin trading decrease information content of pilot stocks.In addition,we examine the relationship between margin-trading volatility and current price crashes,and find that greater margin-trading volatility is associated with higher current price crashes.This implies that margin traders do not increase margin-buying activities during a price decline,thus do not help to decrease current price crashes. |
PDF Full Text Request