Information Processing Skills of Short Sellers: Empirical Evidence from the COVID-19 Pandemic

Abstract

We aim to answer if superior performance by short sellers’ is generated by processing public information rather than by exploiting private information. To achieve this, we analyze if short sellers with healthcare expertise outperform in short selling of non-healthcare stocks compared to those with no healthcare expertise. Since we expect that any short sellers’ private information about healthcare stocks is unlikely to be material for non-healthcare stocks, we conclude that any observed outperformance in non-healthcare stocks is more likely caused by processing public information. As an identification strategy, we interpret the outbreak of the COVID-19 pandemic as a treatment to short sellers with healthcare expertise. Our measures of healthcare expertise are based on pre-COVID-19 performance related to either holding or covering a short position in healthcare stocks. Using a unique German sample of daily short selling data, we find that treated short positions identified by general shorting (covering) outperformance are associated with lower 10-day CARs for non-healthcare stocks by an economically significant magnitude of 4.3 percent (7.2 percent). Robustness test rule out that our results are also driven by the use of private information or non information-based trading advantages such as better funding or borrowing ability of observed short sellers.

Keywords: short selling, healthcare, COVID-19, information processing skills, private information

JEL Classification: G14, G23