The Continuing Deterioration of the Collateralized Debt Obligation Market

If you really want to worry about the credit markets, it might behoove you to turn your attention from subprimes to the vastly more arcane, opaque, and larger problem of collateralized debt obligations. A structured credit product, they are so heterogeneous in terms of structure and composition that it is difficult to make meaningful generalization about them.

However, quite often they took the difficult-to-sell weaker tranches of seucritizations and resecuritized them, and through the magic of credit enhancement, assignment of priorities in payment, default histories of little relevance, and complicit rating agencies, a good bit of the value of these CDOs were rated AAA. $2.5 trillion of CDOs was issued in 2006 alone.

Easier-to-grasp stories have captured the headlines, but the bad news about CDOs continues. The latest sighting is in the Financial Times’ Lex column:

It is August all over again. Or maybe worse. An index used as an imperfect proxy for collateralized debt obligations is pricing supposedly safe AAA securities dramatically below par. Having knocked the banks on the head, the CDO crisis is reaching into sleepier corners of finance: the bond insurers. MBIA and Ambac, the two largest, saw their shares slump on Thursday. This is scary because, bluntly, what hurts Ambac and MBIA hurts the rest of the credit market. So many transactions, and not just linked to CDOs, depend on the guarantee these two provide. Complex structured finance deals have been made possible, or priced attractively, because the insurers have shouldered some part of the credit risk that others could not. If that guarantee were not worth what it was, say because MBIA or Ambac were downgraded, all those transactions that included the guarantee would get knocked. This panic has “contagion” written all over it.

Why are CDOs, specifically, so troubling? After all, many investors – and the insurers – concentrate their exposure at the tippity-top of the capital structure. In MBIA’s case, the losses on certain pools of securities would have to range between 15 and 60 per cent before affecting MBIA’s insurance. Under a stress scenario run a couple of months ago, Standard & Poor’s projected a deterioration in the collateral of mortgage-backed securities that underlies CDOs. It still felt the insurers’ capital cushion was safe. For instance, it assumed that half of all BBB tranches examined were wiped out – not just in default but valueless.

This would be statistically extraordinary. First, it would imply historically huge losses on the underlying mortgages. Second, it would imply extreme correlations of default between different geographic regions of the US. If you can imagine there are CDOs out there whose collateral is made up entirely of tranches of BBB mortgage securities – so called mezzanine CDOs – and these tranches are precisely the ones that become hypothetically worthless, all at the same time, then one-off hits could be scary.

Print Friendly, PDF & Email

3 comments

  1. bob

    Yves,

    First, congrats on the interview at Blown Mortgage – it was good to hear you.

    As far as the CDOs, I think they are all crap. If the examples Tanta used in http://calculatedrisk.blogspot.com/2007/05/mbs-for-ubernerds-iii-credit-risk.html and http://calculatedrisk.blogspot.com/2007/07/leverage-ratings-and-forced-unwind.html are typical, the BBB tranches of MBS are only thinly protected, and are complete junk to begin with. (This “credit enhancement” stuff is just smoke and mirrors based on wildly low default assumnptions.) In her example, only 7% of the MBS has to go south before the CDO based on it is totally wiped out.

    Given all the news lately, I don’t see any reason to assum that at least most of the 2005 – 2007 stuff was similar garbage. I’m not an investor or mortgage banker, however.

  2. bob

    Oops, that should have been: “I don’t see any reason not to assume that most of the 2005 – 2007 stuff was similar garbage. Not to mention bigs chunks of everything else since the start of the decade.

  3. Anonymous

    As usual banks are so inspired that they all do the same and will end up with a « zero of conduit »
    That means they all have their conduits and they all have questionable balance sheets and income statements (HSBC has 32 Billion USD managed under the same scheme)

Comments are closed.