Are Frequent Batch Auctions a Solution to HFT Latency Arbitrage?

As the frequency of my blog posts on these issues attest, I’ve been closely monitoring HFTs, dark pools, and market fragmentation. So when I attended the European Capital Markets Institute’s Annual Conference in Brussels on 29 October, of particular interest to me was a presentation on frequent batch auctions. Eric Budish, associate professor of economics at the University of Chicago Booth School of Business, presented this potential solution to HFT latency arbitrage. Previously in this blog, I have highlighted two other possible solutions — the IEX speed bump and information transmission zoning.

The main thrust of Budish’s argument is that continuous time auctions will, by design, always afford the opportunity for latency arbitrage since someone must always be first to the top of the order book under continuous price-time priority. The phenomenon of HFT in this context is therefore not in any sense new or special or concerning; rather it is simply a rational response to existing market design.

The solution proposed is to eliminate this race entirely and have batch auctions with some high frequency, say once every 100 milliseconds. In this auction, the market would establish an equilibrium price that matches supply and demand, minimising the proportion of unfilled buy-and-sell orders. It would now no longer matter if an order reached the auction first or last, as long as it reached it during the 100-millisecond window — an eternity for computer networks.

It certainly sounded like a silver bullet to the problem of latency arbitrage, but Mark Hemsley, CEO of BATS Chi-X Europe, then took the opportunity to highlight some problems with implementing these frequent batch auctions. First, it is unclear how a discrete batch auction exchange would function alongside continuous time exchanges that would continue to co-exist, at least for some time. A second issue he raised is that it is not clear how a frequent batch auction equity exchange would interact with a derivatives exchange. In both cases, the problem for the batch auction model is that pre-trade transparent continuous-auction exchanges would be constantly updating prices in real time, providing true price discovery, which could complicate the discrete-time batch auction process.

To get around the asynchronicity of price discovery between exchanges, all exchanges would have to become frequent batch auctions and have the auctions perfectly synchronised for the discrete-time auction model to work, in its strictest interpretation. This would clearly be a regulatory headache, if technically relatively straightforward.

Budish’s rebuttal was that any frequent batch auction exchange would eliminate latency arbitrage on its exchange, while agreeing that latency arbitrage could continue on parallel continuous market exchanges. In this case, the proposal is similar to IEX’s speed bump, which is designed to protect traders only on the IEX platform.

Interestingly, there is a dark pool that operates as a frequent batch auction. The London Stock Exchange’s Turquoise Midpoint Dark Pool has a service known as Uncross, which matches orders at the midpoint of the bid and offer with auctions held randomly every 5 to 10 seconds. This frequency is far lower than that suggested as being most likely by Budish during his keynote and is therefore designed mostly for long-term investors trading in large sizes.

Clearly, there is a lot going on in capital markets design and regulation, driven by these issues of fragmentation, HFT, and dark pools. However, a key point made by panelist Rhodri Preece, CFA, head of capital markets policy for the Europe, Middle East, and Africa (EMEA) region at CFA Institute, is that the profitability of HFTs has declined rapidly in recent years, most likely as a result of the physical limits of reducing latency being reached. The question was then raised — if these rents from latency arbitrage have been competed away by HFTs then why bother intervening just as the party is finishing?

CFA Institute was an institutional sponsor of the conference.

If you liked this post, consider subscribing to Market Integrity Insights.

Photo credit: iStockphoto.com/eternalcreative