A Model of Maker-Taker Fees and Quasi-Natural Experimental Evidence

Abstract

Contrary to conventional belief, who pays the tax in overlapping markets, or the exchange access fee in the case of multiple exchange venues, always matters. As the make rebate increases, pricing efficiency improves in the maker-taker venue at the expense of lowering investor surplus. But summed over both venue types, total investor surplus is increasing in the absolute magnitude of the make rebate in the maker-taker venue and the make fee in the inverted venue. Once arbitrage and endogenous information acquisition are introduced into a model of the limit order book, traders are driven to optimally adjust the likelihood of an informed trade to match fee structures. We test this implication using NASDAQ’s ‘quasi-natural’ 1.9 trillion-dollar experiment with a unilateral maker-taker fee/rebate reduction. We show that the cum take-fee spread narrows due to a deterioration in informational content resulting in investor surplus gain but worsening platform price efficiency.

Keywords: Maker-Taker Fee, Exchange Competition, Market Efficiency, RegNMS, Fee Pilot

JEL Classification: G12, G14