After three years of severe market volatility, and a jaw-dropping flash crash, the financial rock star phenomenon known as high frequency trading (HFT) looks a little burned out. Like many rock stars, HFT seems to be suffering from too many late night parties and some incredibly unflattering press. And its business managers -- regulators and exchanges -- are watching closely for signs of abuse, making it more difficult for HFTs to get away with anything.
HFT has had a good run. Relatively cheap stock prices made it easy to trade billions of shares with limited capital. High volatility gave HFT algorithms a chance to dive in and out of the market constantly, chipping off fractions of pennies until they added up to substantial profits. But they may have partied just a little too hard, causing more damage than just the May 6 flash crash.
According to a new study by physicist Neil Johnson and his colleagues at the University of Miami, "Financial black swans driven by ultrafast machine ecology", more than 18,000 instances of ultrafast mini-crashes have occurred over the past five years -- almost ten per trading day on average. These mini-crashes took place in under 1.5 seconds, with many happening in less than one-tenth of a second, and moved the stock price by more than 0.8 percent. There is some evidence that the faster the trades, the higher the likelihood of an incident.
These mini-crashes, added to some of the more obvious glitches such as fat fingers and rogue algorithms, are adding to the tarnish on HFT. Investors have noticed and voted with their feet; trading volumes in the U.S. stock market have plummeted to the lowest levels of the year despite rallying prices. According to Bloomberg, current levels are the lowest (excluding holiday weeks) since Bloomberg began tracking the data in 2008.
This is partly because global economic woes are causing market uncertainty, but it also reflects fears over competing with 'robot' traders. The faster these trading systems get, the more danger lies in market abuse, errors and mini- or full-blown flash crashes occurring. There is also the probability that increased interdependence of asset classes will lead to cross asset flash crashes -- a domino effect in which the crashes 'splash' across equities, commodities, foreign exchange, and derivatives.
As Johnson said in his report: "The downside of society's continuing drive toward larger, faster, and more interconnected socio-technical systems such as global financial markets is that future catastrophes may be less easy to forsee and manage."
This is true. Because the mechanics of anticipating and responding to market abuse or flash crashes are complex, everyone -- not just regulators and exchanges but brokers and traders -- has to be proactive in using the correct tools to monitor algorithmic trading. The detection of abusive patterns must happen in real-time, before any suspicious behavior has a chance to move the market. Sensing and responding to market patterns before aberrations or errors have a chance to move prices is the right thing to do -- in all asset classes.
The rock star-like HFT firms, especially, have to be more proactive to ensure that market abuse doesn't happen and that errors and are caught before the market is affected. Otherwise their business managers might shut them down and they will be partying like it's 1999 -- not 2012.