Algorithmic trading and market efficiency around the introduction of the NYSE Hybrid Market

I study the effect of algorithmic trading on market efficiency, taking into account past market and limit order flows alike. I find that an exogenous increase in algorithmic trading around the introduction of the NYSE Hybrid Market leads to a significant decrease in the predictive power of surprises in market order imbalance and limit order book imbalances, especially at the outer levels of the limit order book. However, the predictive power of past returns remains largely unchanged. This suggests that algorithmic trading improves market efficiency by facilitating the incorporation of information embedded in both market and limit order flows.