The myths around latency arbitrage

28 November 2013


One of the criticisms of High Frequency Trading is that they gain an unfair advantage by receiving data faster than others. This enables them to react to market events before anybody else and effectively ‘front-run’ the rest of the market.

One strategy that is meant to be a popular one is called latency arbitrage. A recent paper by the University of Michigan calculates that this ‘tactic’ generates a whopping $21 billion by taking advantage of the time it takes for trade price information from the various markets to reach a central repository that publishes a public quote, known as the National Best Bid and Offer (NBBO). According to the researchers, HFT firms can make risk free profits by receiving market data directly from the various exchanges and compute their own NBBO. The research reads: ‘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 sure but small profit’.

Wouldn’t it be great if life was that simple? The research makes a number of basic mistakes, not least in the fact that most of the institutions (‘the rest of the market’) trade through algorithms provided by brokers that use exactly the same direct data feeds that these HFT firms use. In fact, it would be pretty much impossible nowadays for any broker to execute a client order by merely relying on the SIP, which is the slower, consolidated data feed.

A more glaring mistake in their assertion is that by receiving direct data feeds from the exchanges, the HFT firms can somehow predict the future. It is based on a false premise that the slow, consolidated data feed is the past and the direct exchange data feeds are the future. Imagine a race-track where the spectators are the first to know which horse has won the race. The reason they are the first to know is that they are physically located next to the track. The people watching the race on their television sets will see the result of the race with a slight delay, the time it takes for the TV equipment, cables, antenna, etc. to process and disseminate the picture. The fact that the people watching the race from their TV set are seeing the result later than the people physically located at the track does not mean that the latter are able to profit from this information advantage. Why not? Because they are both observing an event that already has taken place, i.e. history. The one that is physically located at the track is merely observing it sooner after the event has taken place than the one sitting at home. They cannot miraculously make a bet on the winning horse in between the time that they have observed the result and the time that the people at home have observed it. Yet for some reason this is what the researchers at the University of Michigan, and a number of HFT critics, think happens in the stock market.

To drive the analogy a bit further, professional traders and brokers need to receive market data information as quickly as possible in order to update their prices. In the same vein, bookmakers are physically located at the race-track because they are making prices to customers and they need to be aware of all available information as quickly as possible. This lowers the risk of their prices being stale and them losing money. It doesn’t give them the ability to predict who will win the race any more than people sitting at home.

The views expressed in this blog post are the personal opinions of the author and do not necessarily reflect the official policies or positions of the FIA European Principal Traders Association or the Futures Industry Association.

  • Blog