Did S&P Really Stress Test the Monolines?
By: HousingWire staff
February 27, 2008
On the heels of an S&P report affirming MBIA’s AAA insurer rating, and giving a similar stamp of approval to Ambac, HW’s received more than a few emails questioning the stress test used to give both monolines a top grade.
One reader noted in an email to us:
Net net, the model doesn’t seem to indicate any increase in correlations across portfolios which was the CDO and CDO2 structure’s weakness. I think they are fudging by releasing more data, but not really acknowledging the structural changes that have gone on in these markets. It feels like a CYA move to disclose a little more but use the same approach with a little more stress on the model.
Most of the AAA stuff was indicated by MBIA as having 10-15% OC. The latest document showed them having 15% most likely through CDS. The current market could blow through that in sub-prime and everything I have heard about ALT-A is that it is following close behind. Think of subprime blow ups at 30% with 60% recovery put you at a 12% loss unstressed. It isn’t too tough to imagine a little extra hitting subprime and ALT-A. The OC really isn’t going to be enough.
That opinion contrasts with the trader that runs the Accrued Interest blog, who said he thought the stress test was just that; a highly-stressful scenario.
Get your HW Fix
Join more than 3,000 bold subscribers who already get HW's daily email delivered to their inbox -- it's free, and a great way to ensure you don't miss something.
Events
2008 Oct 19 -- 2008 Oct 22
MBA Annual Convention & Expo
The annual conference for MBA members and affiliates, and the largest industry event each year.
2008 Nov 05
Mortgage Fraud Blog Conference
Miami, FL Stay one step ahead of the fraudsters, and survive the current marketplace by attending this event.
Print This Article






Got something to say?
Remarks by first-time commenters will not appear immediately, and must be approved before they will appear. All comments are subject to HousingWire's usage policy.