This week’s stock market insanity was astounding. Gibbs worries that the next glitch might be even worse.
If you must play, decide upon three things at the start: the rules of the game, the stakes, and the quitting time. – Chinese Proverb
Well, it’s been quite a week. On Thursday we saw the Dow jump off a cliff reputedly because some numbnut at a brokerage house went to place a sell order for $16 million in futures tied to stock indexes. It has been claimed that he actually did was somehow place a sell order for $16 billion!
Not surprisingly, nothing good could come of a faux pas of that magnitude: It is theorized that computerized trading systems saw the huge transaction and blindly followed their programs resulting in a sell-off that, along with the Greek financial crisis, pushed the Dow down more than 1,000 points!
According to the Wall Street Journal, other forces were also in play: “Accelerating the declines, high-frequency hedge funds, which use computers to trade at super high speed, appeared to pull back from the market as prices collapsed. These hedge funds have grown to account for a significant amount of trading volume, and their absence likely created a void into which prices fell.”
It was amazing to see a security like Accenture, which, at 2:46 p.m., had been trading at $39.38, plummet to just 1 cent at 2:49 p.m., and then rebound to $39.51 at 2:50 p.m.
Several things about this amaze me. First, if the trader did, indeed, place this gonzo sell erroneous order how could it have been three orders of magnitude larger than they intended? I could understand one order, maybe even two, but three? To do that you have to add extra digits, three of them to be precise … perhaps the trader was a very bad typist with very large fingers and very bad eyesight.
Second, how could the order entry system not have some kind of validation routine? Surely billion dollar orders are somewhat rarer than million dollar orders. You’d think that the software would respond to such an order with something like “Are you really sure you want to do this because it’s kind of large?” or, better yet, “Dude! Are you kidding?!” But nope, obviously nothing of the kind happened.
Third, why do we have such apparently out of control automated trading? It seems kind of obvious that automating something like this comes with some serious dangers. I know to the greedy #^&*%%*& on Wall Street these risks seem irrelevant when enormous profits are possible through computerization, but the real people taking the risks aren’t them, it’s you and me. If the stock market implodes we’d see our investments tank (again) to be followed by a credit squeeze that would bring our economy to its knees (unless you feel that’s where we are at present, in which case, I guess it would be more of a face plant).
I am not against automation. I’m not against profits. What I am against is allowing greed to play fast and loose as it pleases and use computerization in a way that every IT person knows is extremely risky.
If a simple data entry error can send the market into free-fall so easily you can only imagine what total chaos might be possible from more subtle, complex and harder to predict snafus. What if the next cock-up sends the market down by 2,000 points? Sure, in a situation that catastrophic the markets would be automatically shut down (at least, that’s what we hope would happen), but with the economy in such a bad way at home and abroad we can’t afford this kind of instability.
I doubt very much whether anything will change in the near future when it comes to computerized trading, and perhaps we’ll be lucky and the financial apocalypse will never happen. Then again, it just as easily might. Do you really want to bet on it?
Gibbs doesn’t gamble in Ventura, Calif. Your wager to backspin@gibbs.com.
