Fast trading and the virtue of entropy : evidence from the foreign exchange market

Focusing on the foreign exchange reaction to macroeconomic announcements, we show that fast trading is positively and significantly correlated with the entropy of the distribution of quoted prices in reaction to news: a larger share of fast trading increases the degree of diversity of quotes in the order book, for given liquidity, order book depth and size of order flows. Exploiting the WM Reuters' reform of the fixing methodology in February 2015 as a natural experiment, we provide evidence that fast trading raises entropy, rather than reacting to it. While more entropy in quoted prices means noisier information and arguably complicates price discovery from an individual trader's perspective, we show that, in the aggregate, more entropy actually brings traded prices closer to the random walk hypothesis, and improves indicators of market efficiency and quality of trade execution. We estimate that a 10 percent increase in entropy reduces the negative impact of macro news by over 60% for effective spreads, against over 40% for realized spreads and price impacts. Our findings suggest that the main mechanism by which fast trading may have desirable effects on market performance specifically hinges on enhanced heterogeneity in trading patterns, best captured by entropy.