Those who believe that HFT trading firms are making easy money through arbitrage strategies in the equity markets have missed a tremendous evolution in arbitrage over the last 10 years.
When I graduated from high school in Holland in the 1980s, I learned about the business of arbitrage. This is a type of trading that is defined as realizing a price difference in the same underlying asset between different markets. I also learned that there were real-world examples of this type of trading all around me. For example, on the stock exchange in Amsterdam, a small number of traders were receiving prices from the local trading floor and simultaneously comparing them to prices for similar assets in overseas markets. The traders would note any price differences, taking conversion rates and costs into account. These arbitrageurs had to make their calculations rapidly without any help from a computer, but the pricing differences often were quite large and they made a good living from this arbitrage business.
In the 90s data vendors started providing market information electronically, which meant participants could process price updates faster and with more accuracy than before. Market makers and automated trading firms were among the first to integrate these electronic data sets into automated trading models so they could react in real time to changes in the market. As a result, markets became more connected, and prices started to align increasingly across markets in Europe and the US. The inefficiencies of the old price differences were gone, so assets were priced fairly across markets.
Today automated trading firms are able to provide prices in the same stock on various platforms at the same time. So for instance the same firm (i.e. the same trader and his identical algorithm) provide a bid and an ask quote on LSE and BATS CHI-X in a cross-listed security (asset listed on multiple exchanges) at the same time. The trader sends his quote and size out to each platform simultaneously. He does not have a bias where to trade: he simply waits to see where his order gets filled first. Once he has bought his stock on the LSE (for example), he needs to sell it, again without any bias as to which platform the trade occurs on.
Of course, if one order is filled, the market price for that security will have moved, and the automated trader accordingly adjusts the prices on all the other platforms on which he’s trading. This adjustment takes the form of a cancellation of the old order plus reinsertion of a new order, or a modification, and explains why there is such a high ratio these days of cancelled and modified orders versus actual trades. This activity actually improves transparency and reduces volatility by eradicating pricing discrepancies faster than we have ever seen before. The faster price updates occur, the safer the business model (from an internal and a market perspective). So automated trading firms create efficiencies for all participants by offering to trade on multiple platforms.
Arbitrage still exists in equity markets, and that's good for everyone, because without arbitrage our markets would be much less efficient. But sending the same prices to all platforms at the same time is not the same business as arbitrage. Arbitrage is a strategy involving simultaneous trading on disconnected markets – where the price movements could be independent. Providing prices across multiple exchanges within connected markets simply creates pricing efficiency across platforms, from which all market participants benefit by having fair prices when they want to trade, with decreased risk and less volatile markets.
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.
