is back to his CDS obsession.
Imagine I’m a banker and I have this problem that they tell me I have to mark assets to market. I don’t like that because it makes my stated profits vary and my bonus depends on accounting profits and I don’t like the risk. How can I get around this ?
Well I can set up a special purpose entity and have it write me a CDS on some instrument I hold (they certainly have done this).
Now this doesn’t protect me from the risk that my instrument will default, because my special purpose entity won’t be able to pay up in case of default. However, it does remove my irritating mark to market problem.
How ? Now I have more assets to mark to market – the original instrument and the CDS that my SPE wrote me. Well fortunately AA rated AIG has also written CDSs on the same instrument and people are buying and selling AIG-CDSs on the instrument.
So I book my SPE-CDS as equal in value to the AIG-CDS. Huh ? That’s crazy. They are totally different instruments with totally different counterparty risk. In my example, the SPE has almost no assets, so its CDS is really worthless. However, the most similar asset which is bought and sold on a market is the AIG-CDS so the mark to market standards tell me to mark the value of my worthless CDS to the value of the AIG-CDS.
I am Enroning. Enron’s main scam was to set up special purpose entities and to book Enron’s claims on the SPEs as assets even though the SPEs were never ever going to pay up.
Now the scam as I’ve described it wouldn’t work — certainly not after the Enron bankruptcy. The transaction with my SPE is anything but arms length. I can’t pretend the price I chose as both parties in the CDS is a market price.
However, I think a similar scam can work if I have a friend at another bank who owns the same instruments. We each set up an SPE to write CDSs for each other. This transaction is still not arms length (I mean this is a scam) but it looks OK.
I think a lot of this was done, that is, a lot of CDSs were totally fake CDSs which everyone knew wouldn’t pay as promised if the insured instruments defaulted. I strongly suspect that a lot of CDSs were really MRSs, that is, a swap of mark to market risk, where one agent buys insurance against changes in the perceived probability of default on an instrument which hasn’t defaulted yet, but not against actual default.
It seems to me that such CDSs are not socially useful. They are useful as an accounting trick so that bankers, traders and hedge fund managers can hide their losses from investing in instruments whose value has declined by fake profits from instruments whose value is falsely marked to the value of a quite different instrument.
Does anyone doubt that people were doing this ? The question is whether a CDS was ever booked as equal to value to another CDS written by a different entity with different counterparty risk. Does anyone doubt that the answer is yes ?
Welcome to the modern world of finance. Aloha.