High frequency trading, noise herding and market quality

We develop a discrete-time infinite-horizon model to investigate the role of a high fre-quency trader’s speed advantage when he interacts with a low frequency trader and with noise traders who herd on informed trading. We find that speed in trading rather an in-formational advantage can make the high frequency trader a temporary monopolist in the market, and thus lead to a higher trading volume and trading profits for the high frequency trader. Furthermore, noise traders herding behavior combined with high frequency trading leads to slower price discovery, i.e., the market is not semi-strong as in Chau and Vayanos (2008). In addition, when the high frequency trader can adopt an order splitting strategy, he can generate more profit, and the market quality, for instance, in terms of price discovery and market depth is further improved. JEL classification: G11; G12; G1