Please Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.

A Strange CDO Rating

I spent all day at an "economic crisis" conference, where one of the topics was mortgage backed securities and CDOs. Of course, ratings was one of the key issues.

The fact of the matter is, there isn't that much AAA product out there. But of course there's plenty of demand for it. So the brokerage houses tried to manufacture some.

One way to create mortgage pools is to "stratify" them. If FICO scores are the operative measure, you might have an AAA pool of 750 and above, a AA pool of 720-749, an A pool of 690-719, etc. Then you'd have pools of mortgages with similar pedigree characteristics. The "disadvantage" is that you would have relatively small pools of high quality product.

The more conventional way was to mix large numbers of mortgages of differing quality, and then rely on a statistical property called the law of large numbers to bail you out. Even if the average FICo score was say, 620, the reasoning was that no more than, say,  20% would default in the worst case scenario, less than that under most others.

In that case, rating was by seniority. In the above example, the most senior  80% of the paper would be rated AAA (since there would "never" be losses above the bottom 20%), with lower seniority paper with lower ratings to "mimic" the loss experience of bonds.

All in all, the whole business struck me as "trying to make a silk purse out of a sow's ear." I don't buy that. At the risk of being politically incorrect, my take on these financial instruments is, "what's descended from junk" is junk. Some things are basically not upgradable.