Comments on: Floyd Norris: Off the Mark on Subprimes

By: Anonymous

Anonymous — Fri, 20 Jul 2007 23:25:00 +0000

All very interesting, thanks.
I was talking to a banker the other day — a fellow I did CDOs with for five years. His group seems all smiles — because they are short subprime MBS to the gills, in their proprietary accounts. THis suggests to me that despite the lack of a track record on mortgage ABS, people focused on the matter were able to judge, going back 18 or 24 months, what was coming down the pipe, and went out and bought a ton of paper as “protection buyer.” This makes me a little more sympathetic to Floyd’s silk purse/sow’s ear comment. At the same time, I think you’re observation that if the enhancement had been layered a little less … greedily by the spreadsheeters (aka “bankers”) selling it, things might have worked out for, eg, the investors in the BSC High Grade funds. My last observation: anyone who thinks the so-called bankers at the major broker-dealers care about ANYTHING but getting rich quick is deceived. Many were children during the 80s, when the corporations took over the mainstream media and began feeding them “Greed is good.” The effects are telling today.

By: david

david — Fri, 20 Jul 2007 21:56:00 +0000

Thanks for the explanation anonymous.
It was clear and succint.

By: Lipstick on pig

Lipstick on pig — Fri, 20 Jul 2007 20:27:00 +0000

Re: 4:18 p.m. comment. Actually a mortgage can outperform: as to term (if the borrower stays in the loan longer than expected.) But still the whole subprime mess is a souffle now deservedly collapsing.

And one way corporate debt and residential mortgage backed debt are comparable is that the same agencies are rating them: Moody’s and S&P. If the market starts doubting their ratings because they f***ed up on subprime, that doubt will spread to their ratings of corporate debt. Just one way among many others that the subprime problems can affect the broader economy.

By: Anonymous

Anonymous — Fri, 20 Jul 2007 20:18:00 +0000

Norris’s first comment makes me curious as to how much research he has done regarding ratings across asset classes. At this point I’m shocked that anyone following structured finance would still compare a rmbs rating to corporate.

Let me elaborate on what Yves was discussing comparing static pools to dynamic pools – as I’m sure many of you have questions (that haven’t read the paper). Dynamic pools (corporate debt) and static pools (RMBS) cannot both have a rating system that is static. Lets start with corporate, a corporation picks two projects, both expecting 8% yields, one fails and looses 2%, the other outperforms and returns 18% the following year, so the company recovers from a loss of 2%. Lets take two mortgages, both yield 8%, one fails, you loose 2%, can the other one outperform??? no. So as you can see the models were flawed from the beginning. Hence the reason rating agencies no longer claim ratings as being applicable across asset classes. This is one of many great points in the paper ‘where did the risk go’. If you haven’t read it I highly suggest it before monday morning.

By: Anonymous

Anonymous — Fri, 20 Jul 2007 16:19:00 +0000

You don’t need to be an economist to figure this one out. Even a lay person would know that no matter how you repackage a loan, if the borrower is borrowing above his means and the value of the collateral is tied to the amount of loans being written, rating the security AAA is putting lipstick on a pig.

By: david

david — Fri, 20 Jul 2007 11:53:00 +0000

Well, there was overaggressiveness both in lending and in structuring.
It looks to me that this only has exacerbated the problem and might possibly lead to worse consequences. The root of the problem it appears to me is the fact that no one had an iota of an idea how the subprime would perform under stress and as you mention the situation the were even a lot more clueless regarding the crap from late 2005 and 2006.

It looks to me that your models can be very sophisticated and elaborate, but it is almost impossible to get your assumptions to reflect reality unless they are tested and shaped by reality, by experimental tests, that is market peformance under different conditions.

In the absence of lots of data points, one should try to understand, model and predict the behavior of the securities and their underlying assets. On the other hand, one should be very very dubious and sceptical if the
models have not been tested by market as market behaviour is subtle and hard to grasp model and predict.