We study the association between daily changes in short selling activity and financial stock prices during extreme events using TailCoR, a measure of tail correlation. For the largest European and US banks, as well as European insurers, we uncover a strong relation during exceptional (extreme) days and a weak relation during normal (average) days. Examining days with large increases in short positions and large downfalls in stock prices, we find evidence of both momentum and contrarian short selling taking place. For North American bank stocks, contrarian short selling appears more practiced than for European bank and insurance stocks. We find that the uncovered relationship decreases with firm size and increases during ban periods, which is in line with short selling becoming more informative when constrained.

We study the impact of the arrival of macroeconomic news on the informational and noise-driven components in high-frequency quote processes and their conditional variances. We decompose bid and ask returns into a common (“efficient return”) factor and two market-side-specific components capturing market microstructure effects. The corresponding variance components reflect information-driven and noise-induced volatilities. We find that all volatility components reveal distinct dynamics and are positively influenced by news. The proportion of noise-induced variances is highest before announcements and significantly declines thereafter. Moreover, news-affected responses in all volatility components are influenced by order flow imbalances.