Dark pools—that is, equity trading systems that do not publicly show orders—have raised concerns with such groups as the European Commission and the US SEC because they may diminish price discovery. But the author finds that including a dark pool alongside an exchange enhances price discovery under certain conditions.
What’s Inside?
The author analyzes the impact of dark pools on price discovery. Although some regulators and market practitioners have argued that dark pools reduce price discovery, he finds that including a dark pool alongside an exchange improves price discovery under natural conditions but also reduces exchange liquidity. The author uses a two-period model of strategic venue selection by liquidity traders and informed traders.
How Is This Research Useful to Practitioners?
Dark pools have increasingly become a critical component of US equity markets. The market share of dark pools has increased from approximately 6.5% in 2008 to approximately 12.0% in 2011. One of the reasons for the existence of dark pools is that institutional investors need nondisplayed venues to trade large blocks of stock without signaling to the broader equity market. The author challenges the conventional wisdom that dark trading diminishes price discovery. He uses a model of dark pool trading and highlights the effects on liquidity and price discovery. Under natural conditions, the inclusion of a dark pool concentrates informed traders on the exchange while uninformed traders are pushed into the dark pool. This self-selection mechanism lowers the noisiness of supply and demand on the exchange and improves price discovery.
Because of the growing importance of dark pools to equity market structure, the author concludes that they actually enhance market efficiency.
How Did the Author Conduct This Research?
The author models the exchange and dark pool with a two-period model of strategic trading venue selection by informed traders and liquidity traders. Informed traders seek to profit from proprietary information; liquidity traders need to meet their specific liquidity needs. Both optimally choose between a dark pool and the exchange as part of a self-selection mechanism. The exchange displays bid and ask prices and executes all submitted orders at the bid or ask. In contrast, the passively modeled dark pool matches orders at the midpoint of the exchange bid and ask. Unlike the exchange, the dark pool cannot guarantee execution; thus, submitting an order to the dark pool involves an inherent trade-off between possible price improvement and no execution. The author’s model results are driven by execution risk in the dark pool.
He provides four predictions that are derived from the model where the degree of volatility, the addition of a dark pool, or the horizon of information serves as the exogenous variable and thus causes the prediction of the effect.
Abstractor’s Viewpoint
The author successfully demonstrates that under natural conditions, the inclusion of a dark pool alongside an exchange improves price discovery, which is in contrast to conventional wisdom. This conclusion needs to be balanced with some of the author’s observations, including that improved price discovery on the exchange coincides with reduced exchange liquidity, that delay costs can limit the self-selection mechanism, and that dark pools can reduce price discovery in some rare instances of uninformed trader order imbalance. Recent empirical data seem to support the relevance of the self-selection mechanism described by the author, which is a key component of his conclusions.