معاملات نهادی روزانه و نوسانات قیمت سهام در یک بازار نوظهور تحت سلطه سرمایه گذار خرده فروش

We examine the short-run dynamic relation between daily institutional trading and stock price volatility in a retail investor-dominated emerging market. We find a significantly negative relation between volatility and institutional net trading that is mainly due to the unexpected institutional trading. The price volatility–institutional trade relation differs for institutional buys and institutional sells, and for small and large stocks. Institutional investors herd-trade in large stocks, but do not systematically engage in positive-feedback trading. We argue that the net impact of informational and noninformational institutional trades determines the relation between volatility and institutional trading, and that the relation is negative when informational trading by institutions prevails.

The volatility of securities is a central and fundamental issue to finance, and how trading in general and institutional trading in particular impact the stock price volatility has been the focus of many studies. In this study, we focus on the short-run dynamic relation between daily institutional trading and stock price volatility in a retail investor dominated emerging market. We conjecture that the net effect of informational and noninformational trading by institutions determines the relation between volatility and institutional trading, in that the relation tends to be negative if informational trading by institutions prevails, and non-negative otherwise. This study examines the theoretical prediction on the different impacts of informational and noninformational trading on price volatility (e.g., Hellwig, 1980, Wang, 1993 and Wang, 1994). In a market with asymmetrically informed investors, trades are signals of private information (Hasbrouck, 1991a). Informed traders trade based on private information, whereas liquidity traders trade for reasons not directly related to the future payoffs of assets (Admati and Pfleiderer, 1988). Informed trading, especially private information-driven trading, generally reduces volatility, whereas non-informed or liquidity-driven trading increases volatility (French and Roll, 1986). Relative to retail investors, institutions are generally viewed as better-informed, rational, and prudential investors who have better access to market information and a better ability to time stock price volatility (Aggarwal and Rao, 1990 and Busse, 1999). Institutional trading reflects information and speeds up the daily stock price adjustments (Sias and Starks, 1997), especially in the short run (Yan and Zhang, 2009). Some researchers view individual investors as less-informed noise traders who have a short-term speculative investment perspective and are more vulnerable to the influence of psychological biases (Chopra et al., 1992, Brennan, 1995 and Kaniel et al., 2008). In a market with heterogeneously informed investors, trades of better-informed investors partially reveal their private information on the market and move the price towards its fundamentals. Thus, an increase in informed investors can reduce the price variability (De Long et al., 1990 and Wang, 1993). Institutional investors are also constrained by other considerations, such as liquidity (Hodrick and Moulton, 2007 and Coval and Stafford, 2007). As the percentage of institutional investors in a market reaches a certain level, the effect of noninformational institutional trading may prevail and prices may become more volatile. Over a specific period, or for a particular market, this volatility–institutional trading relation can be either negative or non-negative, depending on the net effect of the informational and noninformational trading of institutions. Based on these arguments, we hypothesize that institutional trading is negatively associated with stock price volatility in a retail investor dominated market when the informational effect of institutional trading prevails. We also examine whether the volatility–institutional trading relation is different for institutional buys and sells. A number of previous studies find that the buy orders of informed investors convey more information than the sell orders, and that institutional buys and sells have asymmetric price impacts (Scholes, 1972 and Kraus and Stoll, 1972). Although an institution’s decision to buy a particular stock from a group of unlimited choices can convey favorable firm-specific information, its decision to sell a particular stock from the limited number in its portfolio does not necessarily convey negative information. Rather, it may be due to liquidity needs (Chan and Lakonishok, 1993 and Keim and Madhavan, 1996). If liquidity or other noninformational considerations prevail, the price volatility—institutional selling relation can be non-negative. We postulate that the relation between price volatility and institutional trade differs for institutional buys and sells. We further examine whether the volatility–institutional trading relation is different for small and large stocks. Information asymmetry is more significant for smaller stocks (Hasbrouck, 1991b), because small firms face more uncertainty, attract less attention from analysts, and their values are more difficult to determine (Wermers, 1999). Institutions trade fewer small stocks, and such trades are less constrained by liquidity and other considerations. They trade more large stocks, and such trades are more likely constrained by noninformational considerations. When the percentage of institutional trading of large stocks researches a certain level, the effect of noninformational trading may prevail. We thus postulate that the relation between price volatility and institutional trading relation differs for small and large stocks. Consistent with our hypotheses, after controlling for trading size, lagged volatilities, and the impact of asymmetric volatility, we find that the trades of better-informed institutional investors tend to be negatively associated with stock price volatility in the retail investor dominated emerging Chinese market. The volatility–institutional trading relation is significantly negative for institutional buy–sell imbalances. The volatility–institutional trading relation is also significantly negative for institutional and buys, except for the largest stocks, but significantly positive for institutional sells. And the negative relation between volatility and institutional imbalances (buys) is mainly due to the unexpected component of institutional imbalances (buys). Institutional net buys stabilize the market during lows but do not destabilize it during highs. Institutional buys also help to reduce volatility more on down return days, whereas institutional sells do not help to stabilize the market on either up or down days. Furthermore, we find that institutional investors herd in their trading, but do not systematically engage in positive-feedback trading, and that their herding behavior is largely limited to large stocks. This “intraday positive-feedback” institutional trading of the largest stocks but not other stocks provides an explanation for the different relations between volatility and institutional trading for the largest stock portfolio and other stock portfolios. Some studies show that the price–trading volume relation can be quite different in different markets depending on maturity and investor structure (Karpoff, 1987, Tauchen and Pitts, 1983 and Daigler and Wiley, 1999). Our study provides an explanation for the mixed empirical findings of these studies. Maturity stage and investor structure can change significantly over different sample periods in the same market, and can also differ dramatically across markets. Emerging markets generally consist of a large number of uninformed retail investors. When noise traders dominate the market, stock prices become more volatile as investors become less informed. Thus, an increase in informed institutional trading is generally associated with a decrease in stock price volatility. Developed markets, in contrast, consist primarily of institutions. When the percentage of institutional trading reaches a certain level, the effect of noninformational trading increases, and this can distort stock prices. Overall, for emerging and developed markets, the price volatility–institutional trading relation depends on the net impact of informational and noninformational institutional trading. As one of the top ten largest stock markets in the world, the Chinese stock market has been dominated by small retail investors and is more volatile than many developed markets. The Chinese securities authorities have been trying to stabilize the stock market by enlarging the base of institutional investors, including Qualified Foreign Institutional Investors (QFIIs). Our empirical findings for the volatility–institutional trading relation will have important policy implications as well. The remainder of this paper is organized as follows. Section 1 provides a brief literature review. Section 2 introduces the institutional background, the data, and the sample. Section 3 examines the impacts of institutional trading on volatility. Section 4 discusses whether the volatility–institutional trading relation is different on up and down days. Section 5 provides robustness tests, and Section 6 concludes.

In this study, we examine the short-run dynamic relation between daily institutional trading and stock price volatility in the retail investor-dominated emerging Chinese equity market. Our study is based on the theoretical prediction that informational and noninformational trading have different impacts on price volatility. We argue that the net impact of the informational and noninformational trades of institutions determines the price volatility–institutional trading relation. Because institutions are better-informed rational investors, we hypothesize that institutional trading tends to be negatively associated with stock price volatility in the retail investor dominated emerging Chinese stock market. Consistent with this hypothesis, we find that price volatility is significantly and negatively associated with institutional imbalance at the market level and for all portfolio sizes, after controlling for average trading size, volatility autocorrelation, and asymmetric volatility. This negative price volatility–institutional trade relation is mainly due to the unexpected component. Second, we find that the volatility–institutional trading relation is different for institutional buys and sells, and for small and large stocks. Price volatility is significantly and negatively associated with institutional buys for all of the stocks in our portfolios except for the largest stock portfolio. Previous studies suggest that the buy orders of informed investors convey more information than the sell orders, and institutional buys and sells have asymmetric price impacts. We find that price volatility is negatively associated with institutional buys but positively associated with institutional sells. Moreover, unexpected buys help to reduce price volatility more than expected buys, long-run and expected institutional sells help to stabilize stock prices, and unexpected sells destabilize stock prices. Relative to institutional buying decisions, institutional selling decisions are driven more by liquidity and other noninformational considerations, and thus tend not to be made optimally. As a result, when such noninformational considerations prevail, especially, when the underlying market is bearish, institutional sells may destabilize rather than stabilize prices. Third, we find that institutional net buys stabilize the market during lows but do not destabilize the market during highs, again except for the largest stock portfolio. Institutional buys help to reduce volatility more on return down days, but institutional sells do not help to stabilize the market on either up or down days. When noninformational considerations prevail, institutional sells may destabilize, rather than stabilize, prices on both up and down days. Finally, we find that Chinese institutional investors herd in their contemporaneous trading but do not systematically engage in positive-feedback trading, and that the herding behavior of institutions is largely limited to large stocks. This “intraday positive-feedback” trading by institutions of the largest stocks but not other stocks provides an explanation for the different volatility–institutional trade relations between the largest stock portfolio and other stock portfolios. Institutional investors also tend to time price volatility and decrease their market exposure when volatility is high. Overall, our findings are consistent with our main hypothesis that the net impact of informational and noninformational institutional trades determines the relation between volatility and institutional trading, and that the relation is negative when informational trading by institutions prevails. We note that, however, there are many different types of institutional investors, such as mutual funds and pension funds, and domestic and foreign institutional investors. Different types of institutional investors may have different preferences, trading strategies, and information. Examining the relation between price volatility and different types of institutional investors in the Chinese stock market will be an interesting issue for future research.