Lack of Anonymity and the Inference from Order Flow

Abstract

This paper investigates the information content of signals about the identity of investors and whether they affect price formation. We use a dataset from Finland that combines information about the identity of investors with complete order flow records. While we document that investors use multiple brokers, our study demonstrates that broker identity can nonetheless be used as a powerful signal about the identity of investors who initiate trades. This finding testifies to the existence of frictions in the economic environment that prevent investors from completely eliminating the information content of broker ID using mixed strategies. We show that the broker ID signal is important enough to affect prices: The permanent price impact of orders coming from different brokers fits the information profile of the investors associated with these brokers. Our results suggest that the market correctly processes the signal embedded in broker identity, and liquidity improvements documented in the literature when exchanges adopt a more anonymous market structure could arise because prices adjust less efficiently to order flow information when the degree of anonymity increases.