Economic theory suggests that all else equal, higher levels of short interest imply a built in demand for an asset. This is due to the mechanics of short selling. In order to sell short, one should first borrow the security (usually from a broker-dealer or an institutional investor) and then sell it. Eventually, the short seller must buy back the shares and return them to the lender.
However, since most short sellers do not cover their short positions at the same point of time, this potential demand is difficult to estimate empirically. Therefore, in order to observe the potential effect of short covering on security prices, one must create an environment in which short sellers are encouraged to cover their short positions around a short period of time. One potential environment is to observe the market reactions to extreme surprises around quarterly earnings announcements. This is a natural setting to examine the price pressure effect because we can compare the market reactions of firms with high levels of short interest with those of firms with low levels of short interest while controlling for the level of the surprises.
The focus of this paper is to examine the effect of the potential demand created by short selling on equity price by estimating the price reaction to earnings announcements based on the levels of short interest. If short interest creates a demand one should observe a positive relation between the level of short interest and the price reaction to an extreme earnings surprise. Our results show that, in fact, there is a significantly positive association between the level of short interest and the price reaction to the most positive and negative earnings surprises.
This finding suggests that the potential demand created by a relatively large amount of short interest is realized when extremely good news or bad news is released. Specifically, shorts appear to cover when very good news is released and tend to take profits when particularly bad news is released. Our results confirm the inference from previous studies that the demand curves are not perfectly elastic (e.g., Harris and Gruel (1986), Shleifer (1986), Loderer, Cooney, and Van Drunen (1991), Kaul, Mehrotra and Morck (2000), and Wurgler and Zhuravskaya (2002)).
A recent paper by Fishman, Hong and Kubik (2007), examines what they describe as the arbitrageurs’ amplification mechanism in the context of short selling in equity market. They argue that the magnitude of arbitrage traders is positively correlated to the price reaction from news releases. In an experiment similar to ours, they find that the prices of highly shorted stocks are in fact more sensitive to good earnings news than those of stocks with little short interest but not for bad news releases. We conclude that their results like ours are simply due to a rightward shift in the demand curve upon the release extreme information.
Download
The Effect of Short Selling on Market Reactions to Earnings Announcements
