| Description |
I study the effect of dark trading on the incorporation of firm-specific fundamentals-quarterly earnings information-into stock prices. Theoretically, if dark pools attract more uninformed than informed orders, then the signal-to-noise ratio of the flow of orders submitted to exchanges will increase, improving price discovery on exchanges. Using a comprehensive sample of dark trading activity, I find that a higher level of dark trading is associated with: (1) greater preemption of upcoming earnings news, as evidenced by larger associations between preannouncement abnormal returns and upcoming earnings surprises, and smaller price reactions to earnings surprises surrounding earnings announcements, (2) faster price formation after earnings announcements, as evidenced by larger intraperiod timeliness metrics, and (3) improvement in long-horizon informational efficiency of stock price, as evidenced by larger future earnings response coefficients. Inferences are robust to an instrumental variable analysis that uses dark pool launches and closures as a source of plausible exogenous variation in firm-level dark trading. Overall, my evidence is consistent with dark trading improving the incorporation of firm-specific fundamentals into stock prices. |
