The anatomy of short sales and price adjustment: evidence from the Hong Kong stock market

Purpose – The purpose of this paper is to study the relationship between excess returns and short sale activities under unexpected informed and uninformed short sales. It is found that in general, the unexpected informed short sale is a bearish signal and has a negative effect on excess returns, while the unexpected uninformed short sale is a bullish signal and has a positive effect on excess returns. It is also found that short sales of large‐cap and frequently traded stocks are more likely to be uninformed, whereas short sales of small‐cap and infrequently traded stocks are more likely to be informed. Design/methodology/approach – Following Conrad's method, the paper differentiates between two short‐sale scenarios: the expected and the unexpected short sales. For the unexpected short sales, the author further differentiates between informed and uninformed short sales. The paper then anatomizes the process into three periods to study the whole process. Findings – The results indicate that in general, the unexpected informed short sale is a bearish signal and has a negative effect on excess returns, while the unexpected uninformed short sale is a bullish signal and has a positive effect on excess returns. The paper also examines the stock characteristics under each scenario and finds that short sales of large‐cap stocks and frequently traded stocks are more likely to be uninformed, while short sales of small‐cap and low frequency stocks are more likely to be informed. Originality/value – In contrast to Conrad, the paper uses daily short interest and trading data instead of bimonthly short interest data. This is the first contribution of this study. As Diether et al. have stated, it is important to study short‐selling activity at a higher frequency. Although Conrad focused on the price reaction prior to the short interest announcement day, this paper investigates the price reaction both prior to and following the short interest announcement day. This is the second contribution of this study. The third contribution of the paper involves anatomizing the short‐sale process into three periods and investigating the price movement.