Can competition between brokers mitigate agency conflicts with their customers?

We study competitive, but strategic, brokers executing trades for an informed trader in multi-period setting. The brokers can choose to (a) execute the order, as agents, first, and trade for themselves as dealers, afterwards; or (b) trade for themselves first and execute the order later. We show that the equilibrium outcome depends on the number of brokers. When the number of brokers exceeds a critical number (greater than one), the informed trader distributes his order (equally) among the available brokers. The brokers, in turn, execute the informed trader's order first and trade personal quantities, as dealers, afterwards. When the number of brokers is below this critical value, the informed trader gives his order to a single broker, who, in turn, trades personal quantities as a dealer first and executes the informed trader's order second. Since the informed trader is hurt in the latter case, he prefers markets with many brokers. Thus, regulators can mitigate trading abuses arising from a conflict of interest between brokers as an alternative to banning such practices. We empirically show that the critical number of brokers for the favorable competitive equilibrium appears to be satisfied for the futures contracts in our sample.