Whenever markets go haywire as they did during the ‘flash crash’ of May 6th, there are inevitable calls for changes to ensure it doesn’t happen again. This time around the pundits and politicians are calling for circuit breakers to slow down trading when individual stocks or the overall markets goes into freefall. CNBC did a good job of summarizing what’s on the table in the way of circuit breakers in this recent article.
Assuming we get circuit breakers, there is no doubt that they will have a dramatic impact on high-frequency and algorithmic trading. How can you enter into a trade or a hedge in a fast market if you are at risk of one half of the trade being busted after the fact? If you had a buy and a sell that were 30 seconds apart to capture a ¼ point move leveraged 20 times, you have a 5% profit. If you later learn that your sell has been busted and that instrument is down 10% then your position is down 200%. Same thing if one half of a hedge is canceled and you have to assume the risk of an unhedged position with a much longer time window that intended when you enter the trade.
As with many regulations, the unintended consequences are often as bad or worse than what you are aiming to regulate.
I’m not arguing there is no need for circuit breakers, there is no question that unchecked electronic trading can spiral out of control. I’m just wondering how it will all play out and who will get a voice.
By now we should all have learned to never underestimate the resourcefulness of the quants and their ability to succeed given a defined set of rules. Whatever circuit breakers end up looking like, I am confident the algorithm designers will adapt, but if the result is that program trading disappears whenever volatility occurs we may see more flash crashes, not fewer. Electronic trading represents more than half of all trading, and if that much liquidity gets turned on or off like a faucet it will continue to lead to erratic markets.