For example: in the 00’s, Wall Street began using then-new Credit Default Swaps to assess and predict the risk of bond and loan defaults through the so-called Gaussian Copula Function.Simply put, you expected the prices of CDS protection to reflect the underlying risk; the higher the price, the bigger the risk.

There were two problems with this:

1)Credit Default Swaps were fairly new, so you couldn’t tell from such a small historical sample whether they were liable to non-standard deviations (see:AIG).

2)If it made sense for the banks to assess risk by looking at CDS prices, it also had to make sense for CDS investors to assess their future returns by looking at bond prices. You could invert the Copula! The result, in theory at least, can be a perfect feedback loop: investors on the long side have confidence because of lack of confidence on the short side; investors on the short side have lack of confidence because the long side is confident etc.

Of course, in the long run, fundamentals have to correct market irrationality. Unfortunately, what the promoters of Portfolio theory, Gaussian Copula, “Wisdom of the crowds” etc. don’t understand is that widespread adaptation of market-following techniques can reinforce existing errors and irrationalities.