Short interest represents the total number of a company's shares sold short that have not yet been covered and is commonly expressed as a percentage of shares outstanding. Short interest accordingly can serve as a gauge of sentiment given the metric measures the degree to which investors are betting on a decline in a company's share price. Research has also widely shown that investors engaging in short selling typically focus on valuations and fundamentals to a greater extent than the average market participant. At the same time, these investors can also have superior access to information on a particular company. As a result of these characteristics and when considering additional variables such as the costs associated with short selling, short interest has traditionally carried credence as an informational sentiment factor.
The short interest metric has recently garnered exponentially increasing interest from market participants as companies with high short interest have generated notably strong returns over the past year. In particular, heavily shorted companies have generally outperformed amid the recent strong risk-on market environment, while the high-profile retail trading frenzy that occurred in early 2021 also resulted in several highly shorted companies posting extraordinary returns. While short interest has long been seen as an important barometer of the negative sentiment, or lack thereof, surrounding a company, some have recently begun to actively seek out companies with high short interest levels following the strong returns generated by these companies. Thus, many market participants have started to view high levels of short interest as an indicator that a company is an attractive investment opportunity.
Many market participants have been quick to adopt the view that short interest levels are positively correlated to future returns. As a behavioral manager, we recognize this premise should be effectively tested over a robust long-term dataset that extends far beyond anecdotal observations or a small sample size of recent returns. Accordingly, the table and chart below summarize the backtest results of short interest expressed as a percentage of shares outstanding. The table on the left displays the average twelve-month returns by decile over the test period, with the first decile representing companies with the lowest short interest levels. Conversely, the tenth decile represents the returns of companies with the highest short interest levels. It is important to note that these deciles are sector neutral, meaning that each decile is comprised of the same sector weights as the overall universe. The chart on the right breaks out how the factor performed throughout the test period, with the data specifically representing the quarterly return spread of the bottom three (highest short interest) deciles of the factor versus the factor average. The results reflect a test period of 2000 - 2020 that was run with a quarterly frequency and the Russell 3000 used as the universe.
The data above validates the observation that companies with high short interest levels have recently generated strong returns. As shown in the time-series data, stocks with elevated short interest have widely outperformed the average company in the universe over the past several quarters. However, looking beyond this small handful of recent observations, the broader dataset directly opposes the notion that high short interest is an attractive investment characteristic. This can be seen as the results above exhibit that low short interest companies outperform the average company in the universe while companies with high short interest conversely underperform by a material degree. We would also note that the tenth decile of data representing the companies with the highest short interest levels is the single worst-performing decile over the entire test period. Furthermore, the time-series data also shows that the poor results for companies with high short interest are not concentrated in a specific period as such companies have only outperformed the average decile in an abysmal 30% of periods even when considering the four most recent quarters.
The strikingly consistent poor returns generated by companies with high short interest begs the question of why market participants have so quickly embraced the view that elevated short interest is somehow indicative of an attractive investment opportunity despite the evidence that suggests the contrary. We believe this phenomenon results from market participants falling victim to two behavioral errors, the recency bias, and the availability heuristic. The recency bias refers to the tendency to value recent data with a greater weight than earlier data. The influence of this bias can be seen when considering that the newfound investor infatuation with high short interest companies is largely based on the strength of the recent returns, while the persistent longer-term underperformance of companies with similarly elevated short interest levels has not been considered. Furthermore, the availability heuristic refers to the tendency to judge the likelihood of an event based on how easily an example comes to mind. While this heuristic can overlap with the recency bias, the availability heuristic refers to the broader propensity to overweight examples easily recallable for any reason, usually due to prominence or memorability, when making forecasts. The impact of this bias can be seen when observing the extent to which investors have continued to drawback to the extreme returns of several highly shorted companies amid the retail trading frenzy as validation of the belief that elevated short interest is now indicative of future outperformance.
Beyond roundly disproving the emerging notion that investors should pursue companies with high short interest, this dataset also provides insight into the specific level of short interest that should be categorized as “elevated” and thus avoided by the prudent investor. The chart below displays the average short interest of the companies that comprise several key decile groups in the backtest data at the end of every year over the entire 2000 - 2020 test period. The top three deciles of data, comprised of companies with the lowest short interest, have typically been characterized by short interest levels in the low single-digit range. The bottom three deciles of the data, representing companies with elevated short interest, have typically been characterized by short interest levels greater than 10%. Finally, the tenth decile that specifically represents the most heavily shorted companies has typically seen short interest levels near or greater than 20%. While the exact level of short interest seen among the deciles has exhibited a degree of variation across different market environments over time, we believe the data provides worthwhile insight into the general amount of short interest that should cause investors to become increasingly cautious.
Looking forward, instances in which short squeezes trigger a large increase in a company’s share price will certainly continue to occasionally occur due to underlying market mechanics. However, this data overall details the extent to which such instances are outliers, given that companies with elevated short interest have consistently proved underperformers over long periods. This research further underscores the value in resisting the natural cognitive tendency to blindly follow the latest market trends simply because the most recent or easily recallable examples of a trend have generated strong returns.
Any forecasts, figures, opinions, or investment techniques and strategies explained are Hillcrest Asset Management, LLC's as of the date of publication. They are considered to be accurate at the time of writing, but no warranty of accuracy is given, and no liability in respect to error or omission is accepted. They are subject to change without reference or notification. The views contained herein are not to be taken as advice or a recommendation to buy or sell any investment. The material should not be relied upon as containing sufficient information to support any investment decision.
Data is provided by various sources and prepared by Hillcrest Asset Management, LLC and has not been verified or audited by an independent accountant. Test results are not indicative of future results and should not be relied upon. While the information provided above is not based on the performance of any individual security or group of securities, the methodology used to provide the information can be obtained by contacting Hillcrest Asset Management, LLC.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.