Goldman Sachs Fund Loses 30%, Wall Street Math Fails to Predict Future

Each era has its own delusions...its own magicians and its own mountebanks. If you had told an investor in the 19th century that he could make a lot of money by applying higher mathematics to market situations, he would have regarded you as a fool. The mathematics of investing was very simple back then. A company either made a profit or it didn’t. You didn’t need fancy arithmetic to figure that out. Besides, investors knew that the secret to making money was not mathematics; it was metallurgy. The whole world was getting hammered out in steel and iron - ships, engines, bridges,
railroads...even the wires that brought the latest in telegraphic communications were made of metal. 

Up until the 1980s...there was no particular use for mathematicians on Wall Street. The maths was basic. Besides, stock prices were quoted in fractions that still had an archaic, pre-digital air about them. 

But then came computers...and computer models...and decimals. And then came the mathematicians who knew how to use them. And then came an explosion of new math-based products - whose actual value had to be tracked by complicated computer formulae. Billions of dollars in speculative positions were no longer “marked to market”. They were now “marked to model”.

“What are those things really worth?” asked the old coots who still ran Wall Street firms.

“What is our real risk?” they wanted to know.

“Don’t worry JB,” came the answers. “The geeks who run the models tell me that it would take a 25-sigma event to cause any real problem.”

“What the f...”

“You know, they talk this lingo...25-sigma means 25 standard deviations from the norm...which, as near as I can figure is about as likely as hell freezing over.”

“Oh...”

Well, dear reader. Last week, hell froze. Clients in Goldman Sachs’ (NYSE:GS ) global equity fund must be asking themselves what is going on, after hearing that the fund lost 30% of its value in the last week. It must be what shareholders of the building stocks are wondering too...and what a lot of homeowners want to know.

A few weeks ago, they had scarcely ever met a question mark. Now they’re finding it essential...taking it wherever they go...and keeping it next to the bedside in case they need it in the middle of the night.  Goldman dropped the fees on its fund to 10% of profits. Still, an investor might want to ask questions: What if there aren’t any profits? What if there are more losses?

Investors in the giant private equity firm Blackstone (NYSE:BX ) are probably asking themselves when the stock, now trading at US$26, will get back to where it was when they bought it - US$31. And the fellows who run rival takeover firm, KKR (AMS:KPE ), are probably asking themselves if they want to bother to go public at all. They already had to amend their offer documents, letting investors know that since the low-hanging fruit had been picked, they would have to skinny up a few trees to get the money they were looking for...and that might mean lower operating margins.

The Financial Times reports:

“In a rare unplanned investor call, the bank revealed that a flagship global equity fund had lost over 30 per cent of its value in a week because of problems with its trading strategies created by computer models. In particular, the computers had failed to foresee recent market movements to such a degree that they labelled them a ‘25-standard deviation event’ - something that only happens once every 100,000 years or more. 

“‘We are seeing things that were 25-standard deviation events, several days in a row,’ said David Viniar, Goldman’s chief financial officer.”

Losses in the Goldman fund could go over US$1.5 billion. But heck, everyone makes mistakes. And even a great mathematician such as James Simons, founder of Renaissance Technologies, takes a loss from time to time. Simons used to do maths for the Pentagon. Then, he discovered that he could make billions running a maths-based hedge fund. But last week, Simons was forced to write a letter to his investors. His fund lost about 9% in the first few days of August...and now Simons says, “we cannot predict the duration of the current environment”.

Ah...question marks. Even the maths whizzes find they cannot live without them when the markets turn down. And no matter what kind of math you do, sometimes things take you by surprise. This is the premise of our friend Nassim Nicholas Taleb’s latest best seller, The Black Swan: The Impact of the Highly Improbable. As he told the attendees at the AF Investment Symposium in Vancouver last month, a ‘black swan’ event is an unexpected event that has grave consequences.

Hmmm...sound familiar? The title of Taleb’s speech at the Symposium was “The Scandal of Prediction: How We Can’t Predict and Why We Don’t Know It.”

“We don’t listen to negative advice – what doesn’t work – we listen to what works – but this means we ignore then the ‘silent evidence’ of what doesn’t work – which is often more instructive,” he told the crowd. 

“We are unable to look at data without trying to make a causative link – it’s human nature – but it doesn’t serve us...”

“Models (ours including) are behaving in the opposite way we would predict and have seen and tested for over very long time periods,” said Lehman Brothers (NYSE:LEH) last week. 

As we explained here in The Daily Reckoning, data cannot replace wisdom, and maths is a poor substitute for good judgment. The mathematicians claim to be able to calculate Value at Risk, or VAR. But they can’t really calculate risk at all. Instead, they use historical volatility as a proxy. But just because a stock only traded as low as US$10 once in the last 10 years does not mean that the real odds that it will trade at US$10 again are only 10 x 365. Just let the private equity boys get ahold of it...strip out the best parts of the business...and load the rest up with debt! 

Bill Bonner
The Daily Reckoning Australia

P.S. to get The Daily Reckoning direct to your inbox sign up to our free e-mail newsletter or if you prefer to use RSS, subscribe to the Daily Reckoning RSS feed.

Related Articles:

• None Found