Friday, August 6, 2010

May 6th Flash Crash is still not well understood

Today is the 3 month anniversary of the "Flash Crash", the peculiar trading behavior that occurred during the afternoon of May 6th, wildly disrupting the markets and resulting in thousands of cancelled trades.

Today, the Wall Street Journal has a fascinating, long, and detailed article about what is known, and what still isn't known, about the Flash Crash. If you aren't a subscriber, you may be able to get to the article by searching for it via Google, since something about the redirect-via-Google-search triggers the WSJ to allow you to read the full article rather than stopping at the paywall.

Some of the details in the article were quite new to me, for example:

At one point, Apple traded for nearly $100,000 a share on Arca, according to NYSE officials, after a buy order for 5,000 shares entered the market and only 4,105 shares were available. When Arca's computers saw that no more shares were available to sell, the system automatically assigned a default price of $99,999 to the remaining 895 shares. Those trades later were cancelled.

A default price of one hundred thousand dollars per share? That's ridiculous, and is just as ridiculous as the one penny per share "stub quotes" that also occurred during the crash.

According to the WSJ, the investigation continues, and "the SEC and Commodities Futures Trading Commission expect to issue a final report on the flash crash within a few months." So hopefully we will soon know more about the details and can study them and discuss what they mean.

In the meantime, the WSJ article has some fascinating tidbits about the interactions between humans and their computers during the crash:

Rumors swirled about of an erroneous "fat-finger" order by a trader at Citigroup Inc.—that the trader mistakenly entered extra zeros, turning millions into billions. Citigroup and regulators later said such an errant trade did not appear to have taken place.

But the rumor helped stabilize the market. If the massive decline was the result of a mistake and not some terrible news, that meant there were bargains to be had.


I'm intrigued by the notion of the rumor stabilizing the market. It is very interesting to reflect on what that tells us about how we reason about our computers, and our algorithms, and our observations of systematic behavior.

The WSJ also confirms what others said about the disturbing behavior of so-called "stop-loss" orders during the Flash Crash:

Some of the biggest ETF traders are firms that try to profit from discrepancies between prices of ETFs and the stocks that they track. But as questions mounted about pricing of individual stocks, these firms pulled back from trading. This hurt small investors who had placed "stop-loss orders," aimed at protecting against big losses by automatically selling once prices fell below a certain level. Those orders hit the market when there were virtually no buyers to be found.

Unfortunately, as the article notes,

Exchanges are unlikely to be able to prevent high-frequency trading firms or statistical-arbitrage firms from bailing out of the market en masse.


So it's still quite confusing about exactly what went wrong, and what should be done to improve the market behaviors, other than continuing to educate the public about how modern markets operate, and requiring consumer-facing financial firms to provide appropriate tools, and appropriate training, for the individual consumers who are trying to participate in the market themselves (e.g., me, with my Rollover IRA).


Also, if you're still listening, the gang over at Nanex are continuing to explore visualizations of some of the algorithmic trading behaviors that they observed during the Flash Crash. I'm not quite sure what to make of this, but the displays are quite intriguing to scroll through. They also maintain a short links page with some further reading.

No comments:

Post a Comment