Honestly, we don’t have an answer to the question of whether high-frequency trading (HFT) is good or bad. But before you either pat us on the back for that statement or call us “crooks” let us explain why the problem is a lot more nuanced than the hyperbolic headlines and sound-bites would have you believe.
The arguments about HFTs have bloomed again following the publication of Michael Lewis’ new book Flash Boys, but what exactly is HFT anyway? If we define it as fast, automatic (executed by computers) trading then a lot of traders would fall into that category – even many individual traders work like that. Some of them have been playing an important role in the markets for many years before the flash crash of 2010.
Market makers would be considered HFTs and even a lot of fund managers employ automatic or algorithmic trading as they accumulate or distribute positions. Those are probably valuable functions for the market and result in benefits to all traders.
The Arguments for HFT
The primary argument HFTs make about value is that it improves liquidity and transparency in market prices. Depending on how you measure volume, HFTs account for 50-70% of daily trading volume on the exchanges. Theoretically, this improves liquidity and tightens the bid-ask spread.
Many of us remember very well the days when options had a minimum spread of 5 cents per share and stocks were priced in 1/16ths of a dollar. In those days, the normal bid-ask spread on a stock was 6.25 cents per share. For active stocks the spread can be a penny or less these days. That may not sound like a big difference but keep in mind that the spread for a $10 stock would have cost you 0.625% right off the top. That kind of expense can add up quick.
HFTs will also argue that they improve transparency because they are active on the short side of the market as well. That means bad news is priced into share prices quickly and longer term investors can benefit from that information. We would agree that short-sellers are an important component of transparency, but this is also where the arguments for benefits start to break down.
What’s the problem with HFTs?
In our opinion, the first real issue with HFTs is that they can create feedback loops of selling. Its also true that they can do the opposite for buying, but most market participants don’t mind if the market ‘crashes up’. These selling frenzies are rare but they happen from time to time and have increased one of the most important costs in the market – systemic risk. Systemic risk is also known as ‘undiversifiable risk’ which means there is basically nothing you can do about it.
Traders typically refer to the Flash Crash of 2010 as the best example of the risks of HFT feedback loops. On May 6th 2010 the Dow Jones Industrial Average dropped 1,000 points (9%) only to recover nearly 600 of those points by the close. This is a classic example but its also a little deceiving. Flash crashes happen much more frequently than that but they are not always as systemic as the one in 2010.
For example, last week on April 4th, the Nasdaq’s stock (NDAQ) dropped $.98 in a little more than two seconds because of HFT activity. The stock recovered (mostly) almost immediately but that more or less triggered the subsequent selling of so-called “momentum stocks” over the next week. Nothing fundamental had changed but trader expectations for systemic risk went through the roof.
The other issue with HFTs that they definitely don’t want to talk about is the fact that they get information before the rest of the market. Is it illegal for them to get that information? Well, not exactly but it could become illegal if the investing public is able to really push the issue towards a regulatory ‘solution’.
Without getting into the technical-weeds, HFT traders have access to information faster than other traders. They locate their trading servers inside or virtually next door to the exchanges themselves and they are very good at detecting when orders are hitting the market so they can buy first and then sell at a slightly higher price. Routing orders between exchanges where very subtle differences in prices emerge also allows HFTs to act on information before anyone else is even aware something has happened.