Thursday, October 23, 2008

There are many Pension Funds about to lose bigtime

CDO Cuts Show $1 Trillion Corporate-Debt Bets Toxic

Investors are taking losses of up to 90 percent in the $1.2 trillion market for collateralized debt obligations tied to corporate credit as the failures of Lehman Brothers Holdings Inc. and Icelandic banks send shockwaves through the global financial system. The losses among banks, insurers and money managers may spark the next round of writedowns on CDOs after $660 billion in subprime-related losses. They may force lenders to post more reserves after governments worldwide announced $3 trillion in financial-industry rescue packages since last month, according to Barclays Capital. ``We'll see the same problems we've seen in subprime,'' said Alistair Milne.

Posted by lvmreader @ 02:17 PM (1182 views)
Please complete the required fields.



14 thoughts on “There are many Pension Funds about to lose bigtime

  • “””
    The banks that structured the securities and investors both failed to do “fundamental credit analysis,” said Janet Tavakoli, president of Tavakoli Structured Finance in Chicago. “They were using correlation models, they were using spread models, but they weren’t doing analysis on the underlying corporations.”
    “””
    – Who the hell would try to use a correlation model for CDOs? Typically there would be CDSs on 80 to 120 issuers providing the premia to the investors, one would need an accurate model with 80 to 120 dimensions of the corerlation of an unobservable phenomena (the default of the company in question is only observable once and that event will be in the future)
    Even if one uses a copula function of the cumulative density functions (as used in the three ‘standard’ models I know of) instead of a correlation matrix, one must assume some pretty hairy stuff (worst that risk of default is independent of interest rates, but also pretty bad that recovery rates are independent of interest rates – both of which run directly against the most basic common sense) and furthermore there has been no legal test of what really matters in the ‘worst case scenario’, although therewas a case in the U.S. of a pool of Mortages and the investors did not get to recover the assets (houses) – so the S.P.V. won the recovery.

    “””
    Some investors are choosing to buy protection and determine their losses now, according to Edmund Parker, head of derivatives at law firm Mayer Brown LLP in London.
    “””
    – I get this if one is forced to, but personally I’d prefer to just write it off completely if I had the capital base since I’d know I was paying way over the odds to protect something I’ve already written down to 10%. Maybe I’d get lucky and end up more than zero! (Oh – and the ex-head of treasury from KBC Financial Products agrees with this approach… maybe why he left).

    One point I think is important to note in this whole mess is that as the rate of defaults increases the recovery realised from defaults goes down – something very few of the models take into account – this is why the senior tranches of CDOs were so mispriced (that and the fact that not only were rating agencies paid by the people selling the products but that their models were open, so one could create a product in such a way as to ensure a good rating!)

    Reply
    Please complete the required fields.



  • @ 51ck-6-51x ,

    So you seem to be implying that we cannot know the probability of default of a company and past events are no reliable guide. Which current experience seems to be brutally confirming.

    Are you a proponent of structural models or intensity models? And do you think that the 40% Recovery Rate so beloved of so many models is a good guide? It seems to me that 5% – 10% is more realistic in the last 2 years.

    Hull-White is one model I studied.

    Reply
    Please complete the required fields.



  • The probability of default is not much of an issue – just look at CDS. The problem is the correlation of a large basket. What is the probability that ACME defaults given that DodgyCorp has defaulted. This really is unobservable until after the event.

    So when trying to use such a correl model it would be hard to get the input correct ANd even harder (mainly computationally expensive) to run the model with so many dimensions.

    The usual fix is to use a copula function, however to do so requires the assumptions I listed, which are not just assumptions they are blatently incorrect assumptions (To assume is to make an ass of U and me ;p)

    I would say that there is no good model for these bad boys (of course there may be – but those profiting from it are going to keep quiet for a while yet). I do think in this case the intensity model would be the better of the two.

    The structures had the best model – callibrate it to the rating agency models to maximise what we can sell it for!

    Reply
    Please complete the required fields.



  • The burden of being a Recession can be a heavy one. In 1996, one of Mr. Recession’s cousins, Amschel, having been asked to fill a leadership position in the family bank in London, hanged himself in a hotel bathroom at the age of 41.

    Four years later Raphael de Recession, also a cousin, died on a sidewalk in Manhattan at the age of 23 from a heroin overdose.

    sorry, I had a bad day yesterday and finished off with ‘not the 9 o’clock news’ and a bottle of wine

    how do they get away with that stuff?

    Reply
    Please complete the required fields.



  • @malct,

    Isn’t Nat Recession currently living it up with Yachts, Hedge Funds and dodgy talk of illegal donations?

    Reply
    Please complete the required fields.



  • ivmreader – I’ve no idea

    Reply
    Please complete the required fields.



  • @ 51ck-6-51x ,

    Thus how about a scenario matrix where we do not believe in the probabilities of defaults between connected companies, but instead we borrow from Structural Mechanics and Finite Element Analysis to create a nodal structure connected by rods of varying stiffness, whose stiffness we do not know but can begin to guess as events progress?

    What do you think of CreditGrades & KMV-Moody models?

    Reply
    Please complete the required fields.



  • @malct,

    Good answer.

    Reply
    Please complete the required fields.



  • 51ck-6-51x and Ivmreader, for the benefit of the humans listening in can you explain in simple terms wtf you are talking about?

    Please not both at the same time, I’ve only got two years.

    Reply
    Please complete the required fields.



  • Not on this topic!!!! Not on this forum.

    Reply
    Please complete the required fields.



  • OK Ivm I can see you are a gluten for punishment

    And like a true Recession he has a taste for the good life: as an avid skier, his principal residence is in Klosters, Switzerland, and he uses his Gulfstream jet to shuttle among his other homes in Paris, Moscow, London, New York and Greece.

    But he is also a man of contradictions: he dates supermodels and actresses, sits on an advisory board of the Brookings Institution, a research organization in Washington, and serves his guests the best wines from the Recession vineyards, which he himself will not drink.

    Reply
    Please complete the required fields.



  • my mum used to have an ACME mangle and I used to turn the handle

    the water fell into a bucket and the clothes came out creased

    Reply
    Please complete the required fields.



Add a comment

  • Your email address is required so we can verify that the comment is genuine. It will not be posted anywhere on the site, will be stored confidentially by us and never given out to any third party.
  • Please note that any viewpoints published here as comments are user´s views and not the views of HousePriceCrash.co.uk.
  • Please adhere to the Guidelines

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>