Slashdot Mirror
Flash Crash Analysis of May 6 Stock Market Plunge
Jamie found an interesting site that has many charts and graphs about the strange May 6 stock market plunge and rebound. There's a lot of information to consume over there, but it does a pretty good job of showing high-frequency trading is getting to be a real problem.
Wow, I know this is asking a lot, especially given the length and depth of the article, but seriously, go read it. They've clearly put a lot of effort into analyzing the situation immediately before and during the crash and that is not what the evidence says. For one thing, only a single affected stock was in 'Slow trade mode' at the beginning of the crash, and only 3 were by the time the crash was at its worst. Furthermore, the stocks in slow trade mode trailed behind the stocks that actually caused the problem.
Basically, the first thing that went wrong is that the NYSE received too many quotes too fast, faster than they could process them. So their systems put them into a queue and processed them as quickly as possible. The next step where things went wrong was that these quotes were timestamped when they left the queue, instead of when they entered. This means that the apparent price on the NYSE was lagging a little bit behind reality. Problem number three occurred when the high frequency trading systems detected this apparent price difference and attempted to capitalize on it, driving the cost for the affected stocks even low and generating more quotes on those stocks as well, causing a feedback loop that bottomed out the market.
Now the question is, why were there so many quotes for these stocks, up to 5000 a second from a single source in some cases. I'm hardly an expert, so I'll just quote the conclusion the report comes to:
What benefit could there be to whomever is generating these extremely high quote rates? After thoughtful analysis, we can only think of one. Competition between HFT systems today has reached the point where microseconds matter. Any edge one has to process information faster than a competitor makes all the difference in this game. If you could generate a large number of quotes that your competitors have to process, but you can ignore since you generated them, you gain valuable processing time. This is an extremely disturbing development, because as more HFT systems start doing this, it is only a matter of time before quote-stuffing shuts down the entire market from congestion. We think it played an active role in the final drop on 5/6/2010, and urge everyone involved to take a look at what is going on. Our recommendation for a simple 50ms quote expiration rule would eliminate quote-stuffing and level the playing field without impacting legitimate trading.
HFT 99% of the time is actually a healthy process - it allows relative mispricing to be quickly corrected and gives investors a chance to trade at prices which are fair whenever they come in to buy or sell.
HFT firms don't care if the P&G share is fundamentally worth 60, 5 or 100$: they are only concerned about the relative value of the stock versus other financial instruments. If for example you know that product X has a robust 1:1 correlation with product Y (for example cash vs. future), and one goes up while the other goes down for no apparent reason, the HFT guys step in and immediately buy X and sell Y until they are realigned. This will usually be for small volumes because such discrepancies don't last for long and are due to people who just aggressively buy or sell one of the products.
Simply put, HF trading firms inject liquidity into the market, they allow thousands of investors to be able to buy and sell literally any financial instrument with a relatively predictable cost (the bid/ask spread).
The real danger IMHO are the "high volume" traders, whether they are hedge funds who take directional bets or some large bank with dubious moral values. These guys will look for markets where they can push the prices up or down using sheer volume. I see this quite often on stocks which are prone to takeover rumours, "large buyers" or "large sellers" suddenly step in and start buying or selling anything they can get their hands on - all market makers panic and think that there is either something they don't know or someone who has insider knowledge. Once things settle, they calmly sell away their new position to other investors (the trend following sheep) who step in to trade on the unknown rumour.