Allison Pyburn has some numbers on just how much the notorious "super-senior" tranches of subprime-backed CDOs are actually worth in the market. It's probably no surprise that the junior tranches are changing hands in "the high single digits," but more to the point the super-senior tranches, she says, are fetching no more than 60 cents on the dollar, which of course corresponds to a 40% write-down. By contrast, she says, Merrill's latest write-down was just 19% on its super-senior CDO holdings.
These numbers are scary enough that there's a serious chance of a domino effect. Banks for instance were able to hedge their CDO holdings by insuring them against loss – but now the insurers are at risk of going bust. The titles of two blog entries on this subject say it all: "So Much For Being Hedged" and "Another word for hedged… leveraged".
We've already seen something along these lines at Morgan Stanley, where the "hedge" on its bearish mortgage-bond bet seems to have been much riskier than the original bet itself. But I have to say that the Morgan Stanley losses don't make a lot of sense to me. For all the talk of "negative convexity", the trade looks as though it was after all a simple long-senior, short-junior play. If you really wanted to make a bearish bet on mortgage bonds, why on earth would you fund it with a long position in mortgage bonds, of all things? Why not use emerging-market debt, or Treasury bonds, or something – anything – which wouldn't go down when mortgage debt fell in value?