Latency Arbitrage Reduces Liquidity


“We find that market fragmentation and the presence of a latency arbitrageur reduces total surplus and negatively impacts liquidity.” 

The above statement is taken from a new paper that has just come out of the University of Michigan.  The paper titled “Latency Arbitrage, Market Fragmentation, and Efficiency: A Two-Market Model”  was written by Elaine Wah and Michael Wellman.  One interesting note is that these two researchers are not from the business school but rather they hail from Computer Science and Engineering department.  Also interesting is that this paper was supported from grants from the National Science Foundation.  Based on those facts alone, this paper earns immediate respect from us as opposed to the pro-HFT puff piece that recently came out of Columbia Business School that was sponsored by Citadel.

The paper studies the effects of what is known as latency arbitrage.  Long time readers of Themis Thoughts of course are very familiar with this term and probably recall a white paper that we wrote back in December 2009 titled  Latency Arbitrage: The Real Power Behind Predatory High Frequency Trading .

The authors of the University of Michigan paper define latency arbitrage as:

“Given order information from exchanges, the SIP takes some finite time, say δ milliseconds, to compute and disseminate the NBBO. A computationally advantaged trader who can process the order stream in less than δ milliseconds can simply out-compute the SIP to derive NBBO*, a projection of the future NBBO that will be seen by the public. By anticipating future NBBO, an HFT algorithm can capitalize on cross-market disparities before they are reflected in the public price quote, in effect jumping ahead of incoming orders to pocket a small but sure profit. Naturally this precipitates an arms race, as an even faster trader can calculate an NBBO** to see the future of NBBO*, and so on.”

Allow us to digress for a moment since it’s important to note this one critical statement from the paper:

Lacking suitable data to study these questions empirically, we pursue a simulation approach.  Order activity at the temporal granularity of interest here is generally unavailable for public research.”

One of the favorite defenses of the pro-HFT crowd, which includes some regulators, is that they have not seen any data that proves HFT is harmful.   Of course they haven’t, because they make it very difficult and expensive for researchers to access the raw data.

The authors conclude:

“Our results demonstrate that market efficiency is negatively affected by the actions of a latency arbitrageur, with no countervailing benefit in liquidity or any other measured market performance characteristic.”

They go on to recommend the elimination of continuous trading and replacing it with call markets:

“Virtually all modern financial markets employ continuous trading, which enables speed- advantaged traders to make risk-free profits over fragmented markets and which degrades overall efficiency. Our proposed alternative is a discrete-time call market, which eliminates latency arbitrage opportunities and improves efficiency. A call market prevents high- frequency traders from gaining a latency advantage, thereby increasing surplus for back- ground traders. Aggregating orders over small, regular time intervals provides additional efficiency gains, and in fact these benefits appear to overshadow the gains attributable specifically to neutralizing latency arbitrage.”

This University of Michigan paper is similar to the recent Chicago Fed paper because they both offer new innovative solutions to the current market structure problems.  We encourage our regulators to stop trying to defend the Rube Goldberg machine stock market that they helped to create and start trying to experiment with innovative solutions like this paper has proposed.