Automated trading has taken over the market. It’s rare now to have a human trading directly on the market. Gone are the trading pits, replaced with racks of servers. Humans now set buy or sell limits and trigger events that are matched automatically in the exchange.
The vast majority of the bid/offers we see on quote systems today come from High Frequency Trading machines. These groups are using computer programs which can see and act on market information millions of times faster than a human can act. So fast that they are making micro second decisions – about 150,000 times faster than you can click a mouse.
These algorithmic computer trading machines rely on high frequency, high turn over. They don’t buy and hold a stock, instead buy a stock and then quickly sell it in a very short time, usually less than a second, later. While they don’t make very much on a trade, sometimes less than 1 cent, they do thousands of these transactions per day.
Just like when people decide a stock is overpriced, the algorithms in use can do the same. But because these algorithms work so fast, they can and have caused major drops in individual stocks, groups of related stocks and even the entire market in the span of minutes rather than hours or days – the so called flash cashes.
For example, on May 6, 2010 an event occurred that’s now called “The flash crash of 2:45”. Starting around 2:30 in the afternoon, the US stock index of Dow Jones Industrial average, S&P 500 crashed by about 9% within 10 minutes. The biggest single day drop, ever. Then 20 minutes later the market had mostly recovered.
According to the experts at ASIC this kind of trading is costing everyone money.