How Does One Fake Tweet Cause a Stock Market Crash?

  • Share
  • Read Later

At 1:07 p.m. on Tuesday, the Twitter feed of the Associated Press told us that Barack Obama had been injured in an explosion at the White House. The tweet was fake — the product of a hack — but given the events in Boston last week, the news spread like wildfire, garnering more that 4,000 retweets.

The AP quickly addressed the situation, suspending its Twitter account, and alerting readers through associated accounts that the tweet describing an explosion at the White House was the result of a hack.  No harm, no foul, right?

Well, not exactly. According to the Financial Times, that one tweet sent shock waves through the stock market — causing the S&P 500 to decline 0.9% — enough to wipe out $130 billion in stock value in a matter of seconds. The market quickly recovered that value, but the breakneck pace at which the stock market tumbled reminded many people of the infamous 2010 “flash crash,” or last year’s crisis at Knight Capital Management, in which a computer glitch cost the firm $440 million and nearly sent it into bankruptcy.

(MORE: High Frequency Trading: Wall Street’s Doomsday Machine?)

Both of these events were caused by the proliferation of high-frequency trading, or the practice of Wall Street firms using high-powered computers to execute thousands or millions of trades per second, making miniscule profits — that add up in a big way — on each trade.

Though nobody knows for sure what exactly precipitated Tuesday’s volatility, many market watchers blamed high-frequency traders, and more specifically the variety that use algorithms to comb through the internet at lightning-quick speeds, actually “reading” news items and tweets, and making trades based off of that information.

How do these computer programs do this? According to Irene Aldrige, managing partner and quantitative portfolio manager at ABLE Alpha Trading, and author of a new book on high frequency trading, the “method is actually quite simplistic.” High frequency traders compile a list of news sources like SEC filings, business publications, and, yes, Twitter, and tell their computer programs to comb through those sources looking for specific words or phrases like “bankruptcy” or “merger” that signal something about the broader market or specific companies. Obviously, not every instance of a negative word in a story is going to mean that a specific company, or the broader market, is going to lose value, but these programs are able to filter through so much information that in the aggregate, these methods are often money-makers for HFT firms.

(MORE: Are Average Investors Getting Bilked by Wall Street Supercomputers?)

As for Tuesday’s incident, it’s possible that many firms had the words “White House,” “explosion,” or “Barack Obama” in their databases as key words that could trigger selling given the right circumstances. According to Aldridge, “If a trusted news source with a lot of followers like the Associated Press sends out those words close together that may have triggered some selling.” Aldridge says the fact that so many people re-tweeted the message — many of whom were trusted journalists themselves — would likely make the news appear even more trustworthy to these bots.

In the grand scheme of things, Tuesday’s mini-crash was not a big deal. “No long-term investors lost any money,” says Aldridge. The market recovered almost instantaneously, and an optimist may look at the event as an example of high-frequency algorithms behaving in a more orderly manner than they have in the past.

For others however, the event proves that high-frequency traders are nothing more than a fair-weather friend. Proponents of high-frequency trading argue that the preponderance of activity they bring to the marketplace adds liquidity and brings down trading costs. There is probably some truth to this argument, as trading costs have never been lower. At the same time, Tuesday shows that when the going gets tough, these computers tend to sell quickly and run for the hills, actually reducing liquidity when the market needs it most.

In the end however, high-frequency trading isn’t going anywhere, and neither is Twitter. Federal regulators are in the process of developing new rules to reign in volatility caused by trading algorithms, and that will be an ongoing process that will probably take years. But Tuesday’s mini-crash also teaches us that Twitter and news organizations themselves have to be even more vigilant about their own security. After all, it’s not just humans who are reading, and reacting, to information anymore.

(MORE: 10 Best and Worst Sports-Team Relocations in History)