Knight Capital Type ‘Glitch’ Will Happen Again: Author
After Knight Capital Group’s torrent of faulty trades on Wednesday that lost the firm $440 million, an expert warns the same type of glitch will happen again and could potentially afflict any trading firm.
“It will happen again, there is no question. My guess is that the mistake will come from a trading firm that does market-making, a place similar to Knight but probably not as large and dominant,” Christopher Steiner, author of the book “Automate This: How Algorithms Came to Rule Our World” told CNBC on Monday.
“They certainly aren’t specific to Knight Capital and I think what's most concerning is that if this can happen to Knight, it can happen to anybody,” Steiner said on CNBC’s “Worldwide Exchange.”
On Monday, Knight Capital announced a $400 million deal to sell convertible preferred shares to a group of investors, a deal that secures its survival. Some customers have also backed the firm, saying they won’t be dropping their orders that are routed through Knight.
Steiner is incredulous how this could have happened in the first place.
“Knight again is the most practiced hand in this business, so you just wouldn’t expect to see that,” he said. “They handle probably half the retail orders in the United States and to say that if Knight can let something like this go wild for 45 minutes I don’t know what we can expect from everybody else.”
Steiner pointed out that such computer malfunctions have happened in the past, such as the glitch that affected Chicago firm Infinium Capital back in February 2010.
“There are several, including one at Infinium Capital in Chicago that lost more than $1 million in three seconds when a new algorithm began wildly trading crude oil futures. Infinium, however, was able to shut the program down quickly.”
Steiner believes that the industry should expect regulation to curb this from happening again. He doesn’t blame the errors on the influx of automated trading but rather the constant need for an increase in speeds for high frequency trading.
“You really have to question what the value is of getting faster and faster by microseconds. What’s the value to the normal investor? What’s the value to the companies that are trying to raise capital on the markets, which is the whole point?”