A Study On The Phenomenon Of My Country's Stock Market Plummeting

This paper studies the short-run crash phenomenon in stock market, that is, the stock index drops dramatically in a single trading day. The drop basically is an outlier, out of the regular stock movement such as the one on Feb.27th, May 30th,2007, and August 31th,2009. The most extant studies focus on those stock drops lasting for a while, such as the global stock crash in 1987.Though there are a lot of research papers on stock crash, there is no consensus on the definition, either theoretically or empirically. This paper defines the stock crash as the outlier of regular movement. Using Sornette’s method to identify the outliers that assumes the stock appreciation follows a normal distribution and based on the stock market data spanning from 1996 to 2010, we compute the distributions of the SHSE composite index, SZSE component index, and SZSE composite index respectively. Then we define the crash phenomenon as the one probabilistically happening in more than 15 years. According to this definition, there are 24 crashes across 1996 to 2010.After identifying the bull market and bear market according to Pagan and Sossounov (2003), we find that there are 11 crashes in bull market,13 ones in bear market, and 5 ones in the transition month. In addition, we find that there were no significant negative affairs happening in several days before and after the 15 crashes out of the 24 ones.This paper mainly focuses on the empirical studies on the stock crash phenomenon and includes three parts, the stock indexes, stocks, and the behavior of investors during crashes. In the first part, we examine the excess returns and excess trading volumes before and after the crashes. In the second part, we build the empirical model to examine the factors impacting on the volumes and returns during the periods. We also run some robust tests using all period data, data filtered by in bull or in bear markets, and data filtered by with significant affairs or without significant affairs during the period. In the third part, we design questionnaires to survey the views of investors about the reasons of stock crashes, and the behaviors of investors during stock crashes. We get some useful results and conclusions about stock crashes from these three empirical studies. The following important empirical results and conclusions are helpful to explore the reasons of stock crashes and to explain the stock crash phenomenon in China:First, from the perspective of stock index, positive cumulative abnormal return will occur after market crash, meaning that rebound will occur after market crash (market crash due to significant affairs excluded). This indicates that irrational sellout happens during market crash, which is just a transitory phenomenon; when the panic disperses, the stock price will rebound.Second, from the perspective of individual stock performance, riskier stocks are more easily to hit the Limit Down, along with smaller trading volume and larger decline; stocks with relatively better performance before the crash exhibit a larger trading volume and smaller decline; stocks held by a larger proportion of institutional investors and big investors exhibit a smaller trading volume and smaller decline; stocks held by a larger proportion of retail investors exhibit faster rebound after market crash. These results indicate that different types of investors act differently during market crash.Third, information or affairs contribute significantly to both index and individual stock performance before as well as after the crash. In terms of market crash brought about by significant affairs, stock index exhibit larger trading volume before the crash, and exhibit negative cumulative abnormal return approaching the crash; however trading volume shrinks after the crash, and rebound takes place slowly. When there is no significant affair, cumulative abnormal trading volume before the crash is relatively smaller, without significant negative cumulative abnormal return, and rebound occurs after the crash. When it comes to individual stocks, after market crash brought about by significant affairs, riskier stocks exhibit smaller rise, and stocks held by a larger proportion of institutional investors and big investors exhibit a bigger rise; when there’s no significant affair, riskier stocks exhibit bigger rise, and stocks held by a larger proportion of retail investors exhibit a bigger rise. The above findings suggest that information asymmetry exist in China’s stock market, resulting in information disparity among different investors. Information has a significant impact on investor decision. Different types of investors with different level of information make different decisions, influencing the equilibrium price of the stock market in different ways.Forth, according to the questionnaires, most investors view market crash and normal market volatility differently; colonial irrational behavior by investors is the main cause of market crash; when market crash happens without clear and definite reasons, less-informed investors would think that well-informed investors had received significant bad news. Such bad news is the major cause for market crash, even if the news may do not exist at all; government policy is another important cause for market crash.From the investigation of investor action before and after the crash, we find that:Fifth, market crash day along with the following days witness highly anxious investors, and a relatively higher proportion of them sell out stocks on the event day, indicating panic stock dump during market crash.Sixth, herd behavior exists during market crash, with evident imitation and contagion herd behavior and information flow herd behavior by investors; meanwhile, investor disposition effect becomes more evident during market crash, with a higher proportion of investors prone to sell out profit-making stocks, which further contributes to the crash; policy effect, though not the decisive factor, also impacts the crash.To sum up, we can achieve the following conclusions:first, information asymmetry exist in China’s stock market, with different types of investors boasting different levels of information; second, information (affair) has an important impact on investor decision-making; third, different types of investors act differently; fifth, imitation and contagion herd behavior and information flow herd behavior by less-informed investors is the major cause of market crash, while disposition effect and policy effect further contribute to the crash, i.e., when market crash happens without clear and definite reasons, less-informed investors would think that well-informed investors had received significant bad news, and they will follow and sell out the stocks, when the stock market decline to some extent, market crash happens.According to the results and conclusions from the empirical studies, and assuming there exist information asymmetry in stock trading, we try to explain the stock crash phenomenon in China theoretically by analyzing the different demand functions and curves of traders received different quantity information and exploring how to reach equilibrium prices in stock markets. (This theory is just discussed but not completed. Theoretical researches are not the point in this paper)How to theoretically interpret the crash phenomenon is a tough question and there is popularly accepted theoretical model, in despite of many relevant literature.Assuming there exists information asymmetry in stock trading, this paper stratify the investors based on the quantity of received information. Barlevy and Veronesi (2003) (hereafter B-V model) break down the investors into informed trades, uninformed traders and noise traders and build a model to explain the crashes according the demands respectively. It is critical to have a upwards demand for the uninformed investors. The crash defined by B-V model is the discrete change of the stock equilibrium price and the crash defined in this paper is a special B-V crash. Therefore, this paper uses B-V model to explain the stock crash phenomenon in China. Meanwhile, we extend Shiller (1984,1989)’s behavioral finance model under the condition with stratified investors and information gathering costs, and model the demand of less informed investors, and get the locally upwards demand function. This complements the B-V model and contributes to explain the crash phenomenon in China stock market. In this paper, we explain the stock crashes as the panic of less informed or lower level stratification investors.