“Don’t tell me you never heard of Goldman Sachs?!” Eddie Cantor

Here is a thought but more of a question…does Wall Street manage risk as well as a casino? I realize this is a preposterous question because of differences in scale, in operations and in the “things” bet upon. This said, it is possible to say that casinos seem better able to account for the occurrence of devastating fat tails. Remember fat tails? They are things like the market crashes of 1987 and 2000, Long-Term Capital Management, the collapse of Bear Stearns, the Savings and Loan Crisis, the crash of 1929, the collapse of Northern Rock, the Russian Debt crisis, the 1997 Asian financial crisis, the 1990 Japanese asset bubble crisis, the 1973 oil crisis and 1978 energy crisis, the Credit Crisis, etc. Anyway, casinos, unlike investment banks seem at times to display a better operational grasp of risk. With fewer investment banks and universal banks like Chase and Bank of America left after the Credit Crisis, the possibility of prospective catastrophic failure may still be greater than before.

Two years ago, things were going along well on Wall Street before the current Credit Crisis. So well, that we stopped talking about Barings, National Australia Bank, Kidder Peabody, Enron and other things like that. Talking about the possibility of operational failure at global investment banks was until last September 2008, would have been odd. Somewhat akin to warning a teenager about old age-a futile and strange exercise.

Consider that in 2006, Goldman Sachs announced, before it was on the cover of the Economist in 2007, that it had turned out $2.6 billion in profits in just three months-nearly half of what it earned in the entire 2005 year. Perhaps not coincidentally, Goldman also put a record amount of the firm’s capital at risk of evaporating on any given trading day. Goldman’s value at risk jumped to $92 million, up 135% from $39 million in 2001.

When asked in November 2006 about the jump in Goldman’s value at risk and the possibility of a market catastrophe, Goldman’s then CEO, Hank Paulson stated that Goldman mainly used value at risk models to measure risk. He went on the say that VaR “always assumes that the future is going to be like the past.” While acknowledging a weakness in Wall Street’s most prevalently used risk model, Paulson went on to explain that, “the one thing we do know, is [that] if and when there is another shock, things you hope wouldn’t correlate [or trade in tandem] are going to correlate.” What this means is that seemingly uncorrelated things like gold and commercial mortgages could go into free fall…

Lehman Brothers and Morgan Stanley were determined to keep pace with Goldman. They had all hired scores of the greatest mathematical minds from New Delhi to M.I.T. to create increasingly complex risk models-all essentially around VaR and variations of VaR risk models. They believed that the gigantic investment houses were invested in so many markets at any one time that their diversification made them less risky. Their thinking was that the chances of all of various global markets going haywire simultaneously was statistically low, therefore an investment bank with hyper-diversification of assets would face less risk.

During all this, record profits were being made by all the investment banks. The scale of trading from the size of the bets placed at investment banking trading desks to the volume of trades, at the all the major investment banks was never greater.

Fast forward to the present. For a variety of reasons, from the lack of transparency of many of the bets being made by hedge funds and investment banks to the fact that the value of the transactions and the underlying assets behind many of the mortgage backed securities traded were not marked to market or known, and the fact that Wall Street’s most prevalently used risk models made no account for the occurrence of fat tails, things do not look as good for the investment banks as they did in 2006 and 2007.

Casinos understand the law of large numbers. In a casino, the average bet size at any given moment for all bets outstanding is very small. Assume for the sake of argument that the average bet size is $10. There may be $30 million in outstanding bets at any given time, but the number of bets is so great and the bet size so small, that it is not probable that any one player will break the house. Also, each individual better is restricted to a relatively small loss level. Casinos seem to operate using the law of large numbers.

The law of large numbers states that as the sample size increases or the number of players in the casino example, the observed average or variations in a casino’s returns will be extremely small. The greater the sample size, the distribution of the results becomes narrower and narrower. The more people there are that are betting at any given time, the smaller the dispersion of results will be for the casino. In theory, the greater and greater the number of players, the more the casino’s results will converge to occur almost along a vertical line.

The law of large numbers does not apply to the world’s remaining mammoth investment banks and universal banks. There are fewer betters than ever before. However, the bet sizes have continued to increase. By varying estimates there are between 500 trillion to one quadrillion dollars worth of outstanding securities. If this is true, then it is possible that a substantial percentage of the hundreds of trillions of dollars of securities in existence not only have no market value, they are not subject to any regulatory oversight.

In 2009, Wall Street is almost the opposite of the casino model in terms of the number of players and their average bet size. We may already have hundreds (if we take into account all global counter-parties including hedge funds) or fewer players with loss limits greater than the value of the figurative casino, or the value of the global banking system. What if many or all of the hundreds of trillions of dollars of derivatives have notional values that are less than their market value? We do not presently know because these transactions are not transparent. What if some of these derivatives have no market value and are being carried on the books of investment banks and other counterparties with only notational values?

The globalization of derivatives transactions and the value of the outstanding bet sizes may set the stage for a prospective and potentially catastrophic crisis for the world’s banking system. Shouldn’t we revisit the way we look at risk?

R. Tamara de Silva