| 11 May 2010
Participating in Folly
(The full letter, parts 1-6 in pdf form can be found here)
We do not mean in any way to tell you that investing is folly, or a casino or any of the usual terms thrown around when people bemoan Wall Street or the stock market or any of those things. The preservation and growth of your capital is a serious business, and we take it very seriously. However, we do believe that the financial markets feature silliness, myths and pure speculative folly. After the last decade that is certainly an argument that resonates with many. Serious people need to account for it, understand it, and recognize whether they like it or not, that they have to decide when they will or will not participate in it, and how.
So serious we are, but we do think a sense of humor is important, if only to keep one’s mind fresh and sharp. Serious matters can dull the senses without a bit of perspective about the often absurd nature of what we are dealing with.
Which brings us to last Thursday, when the markets showed us why any simple thoughts about what will happen, or can happen, in the short term are ridiculous. In a few short minutes, we saw Procter and Gamble collapse by about 30% and then rise almost all the way back. The entire market melted down and wiped out the entire years gain, and then proceeded to climb back to only a loss of over 3% (only!) Speculation abounded about the cause, but few of them really made sense. We may find out differently, but I suspect nothing went wrong. At least there were no errors in order entry. Alan Abelson points out:
Proponents of high-frequency trading, which these days accounts for between 50% and 70% of trading, never tire of citing as its supposedly incomparable virtue that it supplies gobs of liquidity to the market. It sorrows us to report that the bare bones of what happened on Thursday is that when the going got rough, the high-frequency crowd stampeded for the exits and their vaunted pools of liquidity vanished with them.
As the Themis pair describe the debacle, "once the market sensed stress, the bids were canceled and market sell orders chased prices down to the lowest possible point." Investors who had sought to protect themselves with the prudent use of stop orders got creamed because those or-ders were often filled at prices that were temporarily and wildly depressed.
Many of you may not realize that well over half the trading out there is by firms that buy, and then sell, securities in under a second, also known as high-frequency trading. It is a strange world when you are in fact a long-term investor if your holding period is longer than 45 seconds. Many of the millisecond traders (forget “day traders”) who throughout the day supply both buys and sells for each other (and for us) through computerized programs just stopped being involved once the market began a strong downward move. Therefore when you wanted to sell in those few terrifying moments nobody was there. With no bid the price was, well zero, for a few moments. Then traders came back in and the ascent of the market back to a mere horrible day was over in minutes.
Two of my favorite quotes come to mind once again, as we see the chaos that quantitative models from finance and economics wreak havoc in both financial markets and the real economy, from Thursday’s meltdown to the flawed models underlying the mortgage market. All are based on the hubris that we can adequately model mathematically complex events:
"The real trouble with this world of ours is not that it is an unreasonable world, nor even that it is a reasonable one. The commonest kind of trouble is that it is nearly reasonable, but not quite. Life is not an illogicality; yet it is a trap for logicians. It looks just a little more mathematical and regular than it is; its exactitude is obvious, but its inexactitude is hidden; its wildness lies in wait."
-G.K. Chesterton
“Even though economics is a very old subject, it has not truly come to grips with the main difficulty, which is the inordinate practical importance of a few extreme events.”
-Benoit Mandelbrot
We do not wish to imply that people should change anything. We do feel that the idea that a bit of algebra or calculus can solve the investment problem is wrong. The confidence with which much of the “smart money” proceed however is dangerous, because excess confidence means that generally, risk is too cheap, and stocks (or bonds for that matter) are overpriced.
We also do not wish to blame the trading programs. History shows us that this condition is not unique to today. In moments like Thursday, all kinds of wackiness can ensue and has as long as markets have traded:
"Of all the mysteries of the stock exchange there is none so impenetrable as why there should be a buyer for everyone who seeks to sell. October 24, 1929 showed that what is mysterious is not inevitable. Often there were no buyers, and only after wide vertical declines could anyone be in-duced to bid ... Repeatedly and in many issues there was a plethora of selling orders and no buyers at all. The stock of White Sewing Machine Company, which had reached a high of 48 in the months preceding, had closed at 11 on the night before. During the day someone had the happy idea of entering a bid for a block of stock at a dollar a share. In the absence of any other bid he got it."
-John Kenneth Galbraith, 1955, The Great Crash
"I started accumulating stocks in December of '74 and January of '75. One stock that I wanted to buy was General Cinema, which was selling at a low of 10. On a whim I told my broker to put in an order for 500 GCN at 5. My broker said, 'Look, Dick, the price is 10, you're putting in a crazy bid.' I said 'Try it.' Evidently, some frightened investor put in an order to 'sell GCN at the market' and my bid was the only bid. I got the stock at 5."
-Richard Russell, 1999, Dow Theory Letters
The issue for us is how much should we participate in the folly of an overpriced market, and how much to devote to lower risk assets or strategies that are not dependent on rising markets. Because, despite last week, as soon as issues are pushed off or bailed out for the short-term, markets could very well renew their upward march.
To the extent that we accept market risk, we are knowingly participating in folly. To the extent we do not, we are knowingly subjecting you to the risk of underperformance, an underperformance that might persist for some time. Wile E. Coyote of Roadrunner fame would often chase his prey off a cliff and seemingly defy gravity, at least until he looked down. Did we just see the markets peek down to see what was under their feet? Or, is today’s recovery a sign that it didn’t look too closely.
Comments
On Mandelbrot, I agree with you on the value of his theoretical work. And, I will say I cannot disagree or agree with you on the timescale. You may be right, I just don't know. However, the way many of the HFT guys work is based on the idea that the patterns Mandelbrot describes already existed on that timescale. That is what they are counting on. Once again, it isn't a disagreement, it is ignorance. I just haven't been able to pin down how they add to volatility or extreme outcomes in ways that don't already exist.
By the way, I am about to put up a post about Taleb, who happens to be connected to the "Flash Crash."
Thinking about the tax point: isn't it odd that you have to pay roughly 1000 basis points to buy a book in La, but zero to buy a stock?
Great stuff, Lance. All your posts this week have been top-notch.
Soros' views are interesting on this score, since I think he would view it similar to me, as a general issue with markets rather than a special case. I wish I could ask him why he feels that eliminating HFT would make such a thing less likely. Frankly, I am not sure it would. That isn't a disagreement, just an honest admission of not knowing.
Interestingly Tim, I back in the Fall of 2005 at Jeremy Grantham's annual conference they unveiled their own HFT program using mean reversion algorithms. I am curious what his view would be as well.
Mandelbrot is also interesting in this context. Since he views markets as showing a fractal pattern and power laws at all scales, wouldn't he also feel that what happened needs no special explanation?
Once again, I don't know, and I would like to ask them, even if they supported curbing HFT, why in light of their beliefs about markets why HFT is a special case to be curbed. I think the discussion would be fascinating.
Thus I am unsure what kind of regulation is needed, if any, to keep these kinds of things from happening. My guess is that like most things with markets, we have to accept that people react and interact in complex ways that cannot be modeled or curbed to any great degree. We encourage markets because they work over time, not in any particular moment.
It might, but I am always wary of unintended consequences, though unlike many, I realize the unintended consequences of private actors are an issue as well.
I would probably make it smaller, these guys trade so much, and scalp such small amounts of profit per trade it doesn't need to be very large, but I will admit to not having studied what it would take to be effective.
Also, I am not sure that it solves the issue. Traders, whatever their style, can walk away when fear hits. We'll never solve the argument I suspect since many different lines of argument that are hard to prove can be put forward.
For example, markets have certainly experienced panic before, as the above examples suggest. High frequency traders don't seem to have caused the decline, and that makes sense if you think about what they do. They just disappeared when their supposed virtue was most needed. Though we may find that judgment is incorrect later.
RSS feed for comments to this post.