Order Exposure in High Frequency Markets

Abstract

We study the order exposure choice of various trader types in high-speed markets. Using message-level data to identify algorithmic (ATs) and non-algorithmic traders (NATs) we examine how technological differences affect order exposure. While both ATs and NATs hide orders, superior technology enables ATs to receive better hidden order execution. We further separate technologically advanced traders into proprietary (HFT) and agency algorithmic (AAT) traders to examine whether differences in motivation (trading strategies) affect order exposure decisions. AATs hide orders to manage their free-option risk and conceal information whereas HFTs’ hide to limit competition in liquidity provision, consistent with the theory of Buti and Rindi (2013). By contrast, NATs hide larger orders to limit their option value, supporting the empirical evidence from pre-HFT era. Our results highlight that transparency is affected not just by technological differences between traders, which is under regulatory and academic focus, but also by diverging motivations to trade.

Keywords: Hidden orders; transparency; high frequency trading; algorithmic trading

JEL Classification: G11, G12, G14, G15, G24