Regulators globally are concerned that dark trading harms price discovery. We show that dark trades are less informed than lit trades. High levels of dark trading increase adverse selection risk on the lit exchange by increasing the concentration of informed traders. Using both high- and low-frequency measures of informational efficiency we find that low levels of non-block dark trading are benign or even beneficial for informational efficiency, but high levels are harmful. In contrast, we find no evidence that block trades in the dark impede price discovery.

Stock exchanges compete for order flow through their fee models. A traditional model pays rebates to liquidity suppliers, and an inverted model pays rebates to liquidity demanders. Using a regulatory intervention to examine the interaction between tick size, restrictions on dark trading, and exchange fees, we show that traders use inverted venues to adjust for suboptimal tick sizes. Increased inverted venue activity improves pricing efficiency and liquidity, especially when the tick size is binding. We show that the sub-tick price improvement offered by inverted venues enhances competition for liquidity provision and increases information impounded into prices through nonmarketable limit orders.

Dark pools can restrict access for specific trader types. We compare execution outcomes between dark pools that restrict high frequency trader access and those that do not. We find that trades executed in dark pools with more access restrictions have less order flow information leakage, adverse selection risk and post-trade order imbalances than trades in less restricted pools. Evidence from exogenous dark pool closures demonstrates that these differences are causal. The ability to segment order flow can benefit investors because it allows them to make trade-offs between execution risk and information leakage across different dark venues.

We examine the impact of a rule in the Canadian equities market that requires dark orders to offer price improvement over displayed orders. We show that this rule eliminated intermediation of retail orders in the dark and shifted retail orders onto the lit market with the lowest trading fee. Intermediaries shifted liquidity supply to this market leading to increased displayed liquidity. We conclude that reducing retail order segmentation enhances lit liquidity. Despite the improvement in liquidity, retail traders receive less price improvement. Retail brokers pay higher trading fees to exchanges, and high-frequency traders earn higher revenues from trading fees.

We examine returns, order flow, and market conditions in the minutes before, during, and after NYSE and Nasdaq short sales. We find two distinct types of short sales: those that provide liquidity, and those that demand it. Liquidity-supplying shorts are strongly contrarian at intraday horizons. They trade when spreads are unusually wide, facing greater adverse selection. Liquidity-demanding shorts trade when spreads are narrow and tend to follow short-term price declines. These results support a competitive rational expectations model where both market-makers and informed traders short, indicating that these two shorting types are integral to both price discovery and liquidity provision.