{"id":208438,"date":"2024-06-26T09:44:07","date_gmt":"2024-06-26T13:44:07","guid":{"rendered":"https:\/\/ibkrcampus.com\/?p=208438"},"modified":"2024-08-12T15:29:33","modified_gmt":"2024-08-12T19:29:33","slug":"when-shorts-dont-short","status":"publish","type":"post","link":"https:\/\/www.interactivebrokers.com\/campus\/ibkr-quant-news\/when-shorts-dont-short\/","title":{"rendered":"When Shorts Don’t Short"},"content":{"rendered":"\n

The article “When Shorts Don’t Short” first appeared on Alpha Architect<\/a> blog.<\/em><\/p>\n\n\n\n

The important role played by short sellers, who, through their actions, keep prices efficient by preventing overpricing and the formation of price bubbles in financial markets, has received increasing academic attention in recent years. Research into the information contained in short-selling activity\u2014 for example the 2023 study, \u201cSwim with Sharks: Are Short Sellers More Informed than their Competitors?<\/a>\u201d\u2014has consistently found that short sellers are informed investors who are skilled at processing information (though they tend to be too pessimistic). That is evidenced by the findings that both stocks exhibiting a higher aggregate short interest ratio in the current month, compared to their counterparts, tend to experience relatively lower future returns and stocks with high shorting fees earn abnormally low returns even after accounting for the shorting fees earned from securities lending (loan fees provide information in the cross-section of equity returns).<\/p>\n\n\n\n

Fund families that invest systematically, such as AQR, Avantis, Bridgeway, and Dimensional, have found ways to incorporate the research findings to improve returns over those of a pure index replication strategy. For example, a long-only fund could screen out from their eligible universe stocks with high short interest and\/or shorting fees. This will likely become increasingly important as the markets have become less liquid, increasing the limits to arbitrage and allowing for more overpricing.<\/p>\n\n\n\n

Increasing Limits to Arbitrage and Greater Risk of Shorting<\/strong><\/h2>\n\n\n\n

Xavier Gabaix and Ralph Koijen demonstrated in their 2021 study, \u201cIn Search of the Origins of Financial Fluctuations: The Inelastic Markets Hypothesis<\/a>,\u201d markets have become less liquid and thus more inelastic. Gabaix and Koijen estimated that $1 in cash flows results in an increase of $5 in valuations. One explanation for the reduced liquidity is the increased market share of indexing and passive investing in general. Reduced liquidity increases the risks of shorting.<\/p>\n\n\n\n

Adding further to the risks is the now-demonstrated ability of retail investors to \u201cgang up\u201d against shorts (who might be right in the long term but dead before they reach it). The first GameStop episode<\/a> (January 2021) in which retail investors banded together to engineer a short squeeze<\/a> (which drove the price of the stock up almost 30 times) demonstrated just how risky shorting can be, with the potential for  unlimited losses. A repeat short squeeze<\/a> occurred in May 2024 with short sellers losing more than $2.2 billion as the stock rose 60% on the 13th<\/sup> and another 120% in premarket trading on the 14th. The result is that the limits to arbitrage have now increased, allowing for more overpricing of \u201chigh sentiment\u201d stocks, making the market less efficient.<\/p>\n\n\n\n

An interesting question is: What happens when the informed short sellers don\u2019t short? Yifeng Zhu sought the answer to that question in his January 2024 study \u201cWhen Short Sellers Do Not Short\u2026\u201d<\/a> He introduced a metric for extremely low short volume for individual stocks based on the daily short volume of the month, denoted as SRMIN (the minimum daily short volume ratio) and SRMIN5 (the average of the five minimum daily short volume ratios for each stock in the month). Zhu collected public news data from Ravenpack, aggregated it on a daily level, and employed daily Fama-MacBeth regressions. He utilized short volume ratios as dependent variables and positive and negative news from the current and previous day as independent variables. He controlled for current and previous daily returns, realized firm size, book-to-market ratio, momentum, turnover ratio, illiquidity, and beta. His data sample covered the period January 2010 to October 2022. The dataset included all common stocks listed on the NYSE, AMEX, and NASDAQ, with a prerequisite of non-missing market capitalization. He also limited his analysis to stocks commencing the month with prices of $5 or more. Following is a summary of his key findings:<\/p>\n\n\n\n