December 6, 2013
Bastian von Beschwitz and Massimo Massa, INSEAD, and Donald Keim, University of Pennsylvania
von Beschwitz, Keim, and Massa investigate whether providers of high frequency media analytics affect the stock market. This question is difficult to answer because the response to news analytics usually cannot be distinguished from the reaction to the news itself. The authors exploit a unique experiment based on differences in news event classifications between different product releases of a major provider of news analytics for algorithmic traders. Comparing the market reaction to similar news items depending on whether the news has been released to customers, the authors are able to determine the causal effect of news analytics on stock prices, irrespective of the informational content of the news. They show that coverage in news analytics speeds up the market reaction by both increasing the stock price update and the trading volume in the first few seconds after the news event. Such coverage also increases prices if the content of the news is positive. Placebo tests and econometric robustness checks, either based on difference-in-difference specifications or different samples, confirm the results. The fact that a provider of media analytics impacts the market in a separate and distinct way from the underlying informational content of the news has important normative implications for the regulatory debate.
Songzi Du, Simon Fraser University, and Haoxiang Zhu, MIT
Du and Zhu characterize a dynamic ex post equilibrium in a sequence of uniform-price double auctions. Bidders start with private inventories, receive over time a sequence of private signals, have interdependent and linearly decreasing marginal values, and trade with demand schedules. In the authors' ex post equilibrium, each bidder's strategy remains optimal even if he observes the concurrent and historical private information of other bidders; therefore, the ex post equilibrium is robust to distributions of signals and inventories. The equilibrium prices aggregate dispersed private information, and the equilibrium allocations converge to the efficient allocation exponentially over time. The socially optimal trading frequency is low for scheduled arrivals of information but high for stochastic arrivals of information.
Bart Zhou Yueshen, Tinbergen Institute
In a high-speed trading environment, traders simultaneously react to public information not knowing the sequence in which their orders arrive at the exchange. Yueshen develops a theoretical model to capture such queuing uncertainty. Market makers strategically choose the size of their limit orders to compete for a common profit opportunity in liquidity provision. In equilibrium, liquidity overshoots: orders at the end of the queue make expected losses. Once realized, liquidity provision in the bottom of the queue is withdrawn, resulting in "flickering orders". A boost in the trading speed amplifies the overshoot but the effect on order book dynamics (strategic order submission and cancellation) depends on the source of the speed increase. These predictions echo empirical evidence on and policy concerns over "quote stuffing," order-to-trade ratios, and minimum quote life. The model points to an optimal level of queuing uncertainty to which the exchange can steer by carefully randomizing the limit order queues.
Chen Yao and Mao Ye, University of Illinois
When price competition over nominal bid-ask spreads in a stock market is constrained by tick size, liquidity providers either compete on trading speed attributable to the time priority rule or they are willing to pay a fee to make the market. Yao and Ye argue that price constraints created by regulation (SEC rule 612, or the Minimum Pricing Increment Rule) are among the factors driving high-frequency trading and the proliferation of taker/maker-fee markets. They find greater high-frequency liquidity provision for lower-priced stocks with higher market caps, whereby the one-cent tick size exerts a higher constraint on price competition. The same stocks also achieve higher market share in markets with inverted fees, whereby liquidity providers must pay a fee instead of receiving a rebate. A reduction in the nominal share price resulting from a stock split increases the trading speed of a stock and leads to a migration of trading volume to the market with inverted fees. Price constraints also prevent speed competition from improving the price of liquidity, although time priority can determine which traders are able to provide liquidity at the constrained price. The authors find that exogenous technology shocks that improve speed at the millisecond, microsecond, or nanosecond levels do not improve quoted spreads, effective spreads, or depth. Their results suggest that deregulation of tick size can reduce speed competition and increase price competition.
Jonathan Brogaard, University of Washington; Bjorn Hagstromer and Lars Norden, Stockholm University School of Business; and Ryan Riordan, University of Ontario Institute of Technology
Using user-level data from NASDAQ OMX Stockholm, Brogaard, Hagstromer, Norden, and Riordan investigate how different network connectivity speeds influence market participant dynamics. They find that colocated traders have an informational advantage relative to non-colocated participants. The authors use an exchange system upgrade that allows colocated traders to upgrade to an even faster connection to identify a shock to the speed hierarchy. Participants who upgrade reduce their adverse selection costs and improve their inventory management ability, allowing them to increase their market share in liquidity provision. Non-colocated traders incur higher adverse selection costs after the event. Overall, however, the introduction of speed differentiation improves both bid-ask spreads and market depth. The authors' results suggest that the liquidity improvements are related to the fastest traders' increased market share and their enhanced inventory management abilities.
Sabrina Buti, University of Toronto; Francesco Consonni and Barbara Rindi, Bocconi University; and Ingrid Werner, Ohio State University
Sub-Penny Trading (SPT) is a form of dark trading that allows traders to undercut displayed liquidity. Buti, Consonni, Rindi, and Werner distinguish between SPT that is queue-jumping (QJ) and mid-crossing (MID) and find that QJ is higher for NASDAQ than for NYSE stocks. Consistently with Buti, Rindi, Wen, and Werner (2013), QJ is positively related to depth and negatively related to stock price. The authors also find that QJ is associated with improved lit market quality, especially for large capitalization stocks. Sub-penny quotes are allowed for stocks priced below $1, and the authors use this fact to show that when the price of a stock crosses from above to below $1, QJ increases, and the spread improves, but depth deteriorates.