Tuesday’s ‘flash crash’ precipitated by a fake tweet attributed to a major news organization is old news by now. No long term investors got hurt in the debacle and by the end of the week it was business as usual in the nation’s exchanges. That doesn’t mean it won’t happen again and again.
In just four minutes $136 billion was erased from the S&P 500 Index, the main benchmark average of the stock market. The Dow lost 145 points at the same time. Both averages regained those losses over the next two minutes causing some traders to check whether the elevator ride was real or a computer glitch. It was neither.
Having a front-row seat to the now-notorious tweet from the Associated Press claiming that the White House had been bombed and the president injured, my first reaction was to check its authenticity. I guess human nature is a lot more cynical than computers because in the time required to determine the tweet had been a hoax, billions in securities had changed hands.
Although no one is completely sure exactly what happened, we do know that the brunt of the decline was caused once again by algorithmic trading programs, which are these proprietary software trading systems that are lightning fast and take advantage of minute disparities among stock prices. The newest wrinkle in these high frequency platforms involves computer scanning for key words among various news agencies.
Computers are now programmed to search out words that may impact stock price movements. In this case, words such as White House, explosions, bombs and injuries could and did trigger a sell program among Wall Street’s R2D2s and C3POs this week. Once the ball started rolling downhill, armies of computers began spewing out sell orders instantaneously. Once the truth was discovered, these same computers stampeded in the opposite direction. This is what most observers believed happened this week.
It wasn’t the first time. In May of 2010 the now famous “flash crash” occurred. It erased $862 billion from the market’s value in less than 20 minutes. Last year the same kind of computer glitch wiped $440 million off the books of one trading firm and sent it to the brink of bankruptcy.
Regulators have been working to safe guard the public from these on-going speed traps. Under what is called the “limit/up-limit/down system,” trades in stock prices will gradually be limited to a certain percentage range above and below the average price of a stock over the last five minutes. It will act as a speed bump. It won’t be perfect but could at least slow down the declines we have experienced from seconds and minutes to hours or more.
That may just be wishful thinking on my part. More and more of the trading in the stock market is done by these “algos,” as they are now called. It has changed the nature of the stock market and has precipitated the demise of many day traders that are simply outgunned by these R2D2s. What worries me most is that some day in the future these systems may accidentally short circuit in a big way and suddenly we are all looking down the barrel of a financial death star.