Blogging Risks in the CDO Market

After having my coffee with Armond Budish this morning, I found myself in the neighborhood of my old offices at Roubini Global Economics, so I popped in to say hi. Nouriel was there, extremely alert and rested for someone who'd made two trips to Singapore in the space of one week, and so was Brad Setser, who, being Brad, wanted to talk about the potential systemic consequences of a rebalancing of Arab states' foreign exchange reserves. Apparently the blank stare he got from me is pretty much identical to the blank stare he gets from Wall Streeters when he brings up the same issue, which probably means that he's on to something very important. As a general rule, systemic shocks come from somewhere unexpected, which means that something like the currency composition of Middle Eastern oil monies could actually be much more dangerous than the subprime mortgage mess over which so much ink has already been spilled.

It didn't take Brad too long to cut his losses and move on to something I have a bit more interest in: CDOs, especially those made up of subprime mortgages. Brad and I are both fans of the blog Naked Capitalism, and Brad brought up an entry from Friday in which Yves Smith talks about exactly how such things are constructed. Smith quotes Eugene Linden:

In recent years, the money funding these mezzanine tranches has come from a subset of another securitization called collateralized debt obligation or CDO. To form a CDO that invests in subprime mortgages, a securitizer will buy up mezzanine tranches from perhaps 100 different mortgage-backed securities, and then package them in different tranches similar to the way a mortgage backed security was packaged in the first place. Thus, some CDO's can consist entirely of BBB- tranches of subprime mortgage MBS, but still have 95% of their value rated investment grade.

Now Linden, if we can believe his blog entry, is chief investment strategist for a hedge fund that specializes in distressed and bankrupt situations. Which would mean he's not a typical innumerate journalist, prone to getting things very wrong. But Smith is loathe to take Linden at face value. Here's what he writes:

From what I have read, it seems more typical for CDOs to have a mix of paper (unless they are CDO-squared or cubed, which hold only other CDO paper, I can't generalize about them), which can include commercial mortgages and LBO debt along with residential mortgages. One of the reasons to buy a CDO rather than an MBS is to get diversification.

Brad, too, was confused. If a CDO is comprised of the mezzanine tranches of lots of bonds all of which are backed by subprime mortgages, how can most of that CDO be rated investment grade? An investment-grade credit rating on such an animal would have to come from the supposed diversification benefits of owning lots of different subprime-backed tranches. And maybe that's what happened: the people putting the CDOs together persuaded the credit agencies that because the subprime-backed bonds came from all over the country, say, they wouldn't all go sour at once. And maybe the ratings agencies, who were making shedloads of money rating these CDOs, didn't ask too many questions. So maybe it's true that people who think they're holding AA-rated CDO paper are in fact holding something which should be trading much more like junk.

But maybe it's not true. After all, ratings agencies might be stupid, but they're not normally that stupid. And it's not even clear that there was remotely enough mezzanine-rated subprime-backed paper to go around in the first place, which is why a lot of CDOs ended up making leveraged bets on AA-rated paper (just like the younger Bear Stearns fund) rather than investing directly in hard-to-find BBB-rated paper. And it would certainly make a lot of sense, if only to help boost its credit ratings, for a CDO to diversify its holdings out of residential mortgages and into, at the very least, commercial mortgages.

In other words, when you start hearing horror stories about CDOs and other magical pieces of financial engineering, approach with extreme skepticism. Smith finds quite a few other holes in Linden's story, as well, which only serve to make the whole thing more suspect. He's part of a wonderful vanguard of bloggers, including Tanta at Calculated Risk, who are very good at seeing through the hyperbole of media reports to the often much less sensational realities underneath. Both Smith and Tanta are very worried about the effects of the housing market on the market in mortgage-backed securities and thence to the credit markets in general. But they don't exaggerate for effect, and they call out people who do. And that's something all of us, trying to pull a signal from all of the noise, should be very grateful for.