CDS is a derivative. That is to say, it is an insurance contract on "something"; this "something" being collateral debt obligations (CDO), the sliced-and-diced securities that take in a bunch of questionable loans and create AAA bonds out of them.
Now if you're a financial institution, then you are concerned about the risk of these CDOs, and the risk of these CDOs will determine their values. Very risky? then the price of these CDOs will be low, and vice versa [1].
Well, banks had no way of valuing this risk, because the underlying loans for these CDOs can number in the thousands- to model all that would have been too intractable. If you can't value them, then you can't buy them.
That's when Wall Street hit on this idea: why not use the value of the options made on the CDOs to value the CDOs themselves? In other words:
- Efficient market says that price encapsulates all the information available
- CDSs are insurance contracts made on CDOs
- These CDSs are traded are certain prices
- therefore, these prices tell you how to value these CDOs
One slight problem. The assumption that these CDS derivatives are made in "efficient market" turned out to be horribly wrong. To see how badly you can abuse the market, consider this example:
- Bank A bought XYZ CDO at $100M
- Bank A creates two fake brokerage firms: Stool Pigeon LLC and Suckers Inc.
- Stool Pigeon LLC and Suckers Inc buy and sell CDS on XYZ CDO (derivatives) to and from each other at completely made-up prices. Since one subsidiary's loss and another subsidiary's gain, these sham transactions cost Bank A zero dollars.
- By manipulating CDS prices, Bank A just manipulated XYZ CDO valuation
- Bank A then sells XYZ CDO to Iceland at $110M. Profit!
Something was wrong.
[1] The risk here is not just default risk, but its correlation risk. I'm going to gloss over that for the sake of expositional clarity.
No comments:
Post a Comment