Enhanced expected impact cost model under abnormally high volatility

Gabriel Tucci, Sameer Jain, Aiying Zhang and Wenting Ge

Abstract

During recent market turbulence, we have observed an anomalously high impact cost for stock market transactions. After conducting empirical analysis on actual execution data from 2020, we find that during such a market regime there appears to be a stronger crowding effect in the stock market, which means market participants have an increased propensity to trade the same stock on the same side at the same time, leading to larger transaction costs in general. Therefore, to address this problem, we extend our impact cost model beyond the typical factors such as order size, price volatility, trade volume and bid–offer spread. These factors alone are insufficient to fully capture the underlying pattern in this market regime as they would in normal times. We propose to add a size adjustment to the actual transaction size to account for this crowding effect and to achieve a more accurate estimate of the expected impact costs during such periods of elevated volatility. The adjustment is constructed using the Chicago Board Options Exchange Volatility Index, a well-known measure for expected volatility and “fear”. With the enhanced model, we are able to provide a more robust estimation for transaction costs that significantly outperforms the original impact cost model in the face of market turmoil.