Lots of people have been rightfully concerned
about just how much of the trading in the financial markets is done algorithmically via high frequency trading systems that execute trades faster than any of us can comprehend, based on various algorithms, trying to shave little bits and pieces of profit. The key worry, of course, is that these algorithms can get into something of an infinite loop problem that spins the markets out of control. We've had momentary blips
, at times, that happened so quickly it wasn't even clear why they happened. And now we have a case where overreacting to a fake tweet
may have briefly cost the financial markets $136 billion (yes, with a b).
The story is that the Associated Press's twitter feed was hacked and a bogus tweet was posted, reading: "Breaking: Two Explosions in the White House and Barack Obama is injured."
And off went Wall Street, sending the stock markets down quite rapidly. Of course, after the report was quickly corrected and shown to be false, the market rapidly recovered -- though, in the down and upswings, it's likely some people lost a fair bit of money while others made out quite well. Still, this raises a different set of questions even on top of the worries about high frequency trading, pertaining to the various inputs it receives. Reacting to stories on Twitter is an interesting way to try to beat the news cycle. Since so many stories break on Twitter first, it's no surprise that Wall Street is eagerly scanning the service for market-impacting news. The real question is if anything can or should be done about this. One argument is that we can leave it alone because it's self defeating. If your system jumped the gun and traded down on this report, well, good
, you deserve to lose money for trading on a bogus tweet. Similarly, the fact that the market bounced back showed that it can self-correct fairly quickly, even if there are billions to be made and lost in between.
For those who think this is
a problem, the bigger question might just be: and, so, what do you do about it? I'm just as worried as the next guy about the problems of such inter-connected, algorithmic trading systems spinning out of control, but I can't think of any way to prevent it that doesn't also lead to great collateral damage for more efficient or reasonable markets. That doesn't necessarily mean there isn't an answer, but it doesn't appear that there's an obvious response, seeing as no one has introduced any. In the end, it may be that this is just a fact of life. People are always going to seek out ways to beat the market by being first. And many times that will lead to great profits. But sometimes, when a story like this comes out, you get burned. Maybe that's perfectly reasonable.