The Journal Tells Us Quants Have Suffered

Yves Smith at Naked Capitalism submits:

Readers have probably figured out that I think the reporting in the Wall Street Journal is overrated. We have a prime example on page one today, "How Market Turmoil Waylaid the 'Quants'."

My understanding is that newspapers are in the business of providing news. The story that the quants are in trouble is a month old and has gotten considerable coverage in the business press. By definition it is no longer news. For the Wall Street Journal to justify publishing a story about it, particularly a prominent story, it should offer new information, new analysis, or provide a particularly good synthesis (the New York Times often executes that sort of story well).

This piece does none of the above. Admittedly, it isn't the worst example of a Wall Street Journal article with virtually no redeeming qualities being on page one (my recent top pick is "How Rating Firms' Calls Fueled Subprime Mess"). But this one is close.

In lieu of providing new information, the story is instead unduly focused on the career of one quant, Peter Mueller of Morgan Stanley. Using him as an organizing device allows the story to mention that Morgan has a largely unknown but large statistical trading unit named PDT, for Process Driven Trading, which contributed $540 million, or about 7.2% of the firm's earnings in 2006, and lost $500 million from early July through August 9.

Now this angle could have been compelling if the reporters had gotten close enough to their sources to get either a blow by blow of what that nightmarish month-plus felt like (a Peeping Tom's view of a meltdown is always fascinating) or a sufficiently good understanding of either Morgan's trading approach or the general issues involved in quantitative investing so as to provide new insight.

But because the writers failed to get the goods, the story, which promises to tell us something new about quants and where they went awry, is instead about one quant, Mueller. In a stereotypic rich man's odyssey, Mueller goes through the stations of life of a successful young hedgie: makes a lot of money quickly, becomes disillusioned, loses his significant other, drops out, travels the world, takes up kayaking and yoga and returns to playing jazz (both recording his own albums and performing in subways), and competes successfully in the pro poker circuit. But like all prodigal sons, he eventually comes home, which in this case it was to Morgan Stanley, doing pretty much what he did before.

And Mueller's tale is flabby, and a reference to a short bio of him on a website suggests he didn't cooperate much, if at all with the authors. Similarly, the information on quant strategies is skimpy and superficial. There's a brief description of pair trading, a mention of the fact that these traders tend to crib ideas from the same sources and often wind up with very similar strategies. It also regurgitates the traders' defense for their poor performance: things were anomalous and there was too much capital devoted to their investment style.

John Dizard provided of the Financial Times provided vastly more insight on August 14 in a few incisive paragraphs. Note that the "Gaussian distribution" he mentions is the bell curve; one of the well-known problems with financial models is that many, such as the Black-Scholes options pricing model, assume a normal (bell curve) distribution, which financial markets don't exhibit (they have "fat tails" and also don't have a symmetrical distribution of outcomes):

As is customary, the risk managers were well-prepared for the previous war. For 20 years numerate investors have been complaining about measurements of portfolio risk that use the Gaussian distribution, or bell curve. Every four or five years, they are told, their portfolios suffer from a once-in-50-years event. Something is off here.

Models based on the Gaussian distribution are a pretty good way of managing day-to-day trading positions since, from one day to the next, risks will tend to be normally distributed. Also, they give a simple, one-number measure of risk, which makes it easier for the traders' managers to make decisions.

The "tails risk" ....becomes significant over longer periods of time. Traders who maintain good liquidity and fast reaction times can handle tails risk....Everyone has known, or should have known, this for a long time. There are terabytes of professional journal articles on how to measure and deal with tails risk....

A once-in-10-years-comet- wiping-out-the-dinosaurs disaster is a problem for the investor, not the manager-mammal who collects his compensation annually, in cash, thank you. He has what they call a "resume put", not a term you will find in offering memoranda, and nine years of bonuses....

All this makes life easy for the financial journalist, since once you've been through one cycle, you can just dust off your old commentary.