A Principal-Agent Model of Trading Under Market Impact -Crossing networks interacting with dealer markets-

Abstract: We use a principal-agent model to analyze the structure of a book-driven dealer market when the dealer faces competition from a crossing network or dark pool. The agents are privately informed about their types (e.g. their portfolios), which is something that the dealer must take into account when engaging his counterparties. Instead of trading with the dealer, the agents may chose to trade in a crossing network. We show that the presence of such a network results in more types being serviced by the dealer and that, under certain conditions and due to reduced adverse selection effects, the book's spread shrinks. We allow for the pricing on the dealer market to determine the structure of the crossing network and show that the same conditions that lead to a reduction of the spread imply the existence of an equilibrium book/crossing network pair.