They all happened so fast: the 2010 flash crash that sent the Dow Jones Industrial Average plunging almost 1,000 points, Facebook (FB) and BATS' botched IPOs and most recently, Knight Capital’s (KCG) $440 million trading loss that nearly destroyed the firm.
These market-rattling events cost millions of dollars and shook investor confidence. But now some experts argue that all of them could have been prevented by a "kill switch" that would immediately shut down any abnormal trading activity.
“Knight was lucky to survive,” said securities lawyer Gregg Berman, a partner in Fulbright & Jaworski in New York City. “The algorithms that are out in the market now are so complicated that when it comes to trading, there can be glitches that aren’t discovered until too late. When things get too complicated without some sort of breaker or kill switch, we are going to continue to see these types of events.”
The Securities and Exchange Commission held a roundtable Tuesday to discuss potential market regulations and technology review standards. The meeting included representatives from stock exchanges, brokerage firms, financial institutions and the Financial Industry Regulatory Authority. The group submitted a letter to the SEC at the end of September outlining potential market regulations — including a kill switch to prevent massive money-losing technology glitches that have plagued markets recently.
A kill switch or trigger mechanism would establish a formula that tracks the “peak net notional exposure” of a trading or brokerage firm to identify any abnormal activity.
“They would establish a formula based on each individuals firm’s trading activity,” explained Edward Kim, a consultant with Grant Thornton and former senior vice president at the NASDAQ. “Every firm will have a different level based on their average amount of activity…and if you trigger some level, maybe five or 10 times that level, a shut down might be set off.” He added that the exchanges and trading outlets will have to come to agreements on what is defined as “excessive market activity.”
The goal is to limit a firm’s exposure in the market and the trigger formula would take into account long and short positions.
“It’s really the combined amount of financial exposure a firm has taken throughout the course of a day," Kim said. "If they were long $5 million and short $3 million that firm would have peak net notional exposure of $8 million. The actual math involved is going to be simple. It’s the question of where to set the trigger if averaging $8 million and one day you are doing $40 million or $80 million -- that should set something off.”
While the experts agreed that a kill switch is a smart idea to keep up with technological advances, it shouldn’t be the only line of defense.
“Kill switches shouldn’t replace internal risk controls, they should be a last resort for the firms,” said Jim Toes, president and CEO Security Traders Association. “They should be looking to internal risk controls as their first line of defense.”
Technology has changed the trading world, and according to Kim, 50-70% of trading on exchanges is of the high frequency variety.
“That means the majority of trades in the market are not done by investors seeking to invest long term in a company, these are short-term, hyper-active traders who have zero interest in the fundamentals of the company, they are purely interested in taking advantage of the minute movements of a stock price.”
While the formula behind a trigger mechanism is pretty basic, implementing it and coordinating among financial firms and exchanges might prove to be a little more difficult. Toes pointed out that financial firms will have a hard time figuring out how to shut down just one unit of their business without it affecting other parts.
“You still need reports to come back once that trigger is pulled, you still have orders residing in the marketplace. They will need to figure out how to kill the outbound message without compromising the inbound messages.”
Containing a software glitch or erroneous trade from spreading is top priority for regulators.
“The goal of the switch is to prevent contagion from occurring; you have an event and you need to fence it in and avoid it from leaking and spreading to other parts of the firm or within an exchange,” said Toes.
While bad trades can cause a massive ripple effect throughout an exchange, they also weigh on investor confidence and cause some traditional buy and hold investors to leave the market.
“It’s clear individuals have lost confidence in this market,” said Kim. “The market performance has been strong recently and yet investors are fleeing the market -- that tells you that investors don’t know what makes the stocks move, and that’s a scary thought.”
But whatever regulators decide to do, one thing is for sure: Technology will continue to advance, making trading algorithms that much more complex.
“High-frequency trading is not going away, there’s no reason for it to go away," said Berman. "Computers will become more powerful, allowing for more entries into the marketplace and the trades will become bigger, faster, stronger. It’s what we all expect."
Follow Kathryn on Twitter @kathrynvasel