Banks Sell Products That Make Them Money. So Do Ratings Agencies

Publish date:

The role of the hero who has been in the belly of the beast and emerged to slay it seems to be psychologically rewarding,* because people keep trying to claim it for themselves. Like this Geoffrey Tomes gentleman, who bared his soul to tell DealBook that he "was selling JPMorgan funds that often had weak performance records, and I was doing it for no other reason than to enrich the firm." You imagine him racing to tell DealBook about this revelation, imagining his welcome into the Greg Smith Hall of Heroes, and being a little disappointed that everyonewas all "wait, what, did anyone on earth not know that was exactly what you were doing? Why did they think you were pushing JPMorgan products? That's what banks do."

Similarly there is a certain amount of "duh, obvs" to the recently unsealed documents in the lawsuit over the Cheyne SIV. But they're still funny. Cheyne was a conduit stuffed with mortgages and home equity loans that exploded in the face of some people, who are now suing Morgan Stanley, who built it, and the rating agencies, who did sort of a not-so-great job of kicking its tires. From their filing unsealed yesterday:

In relation to CDO criteria changes, Perry Inglis, the head of the S&P group that rated the Cheyne SIV notes, wrote: "[I]t would be good to get an idea of how far these would have to change for us to be 'competitive' on these types of deals. I'm a bit unclear if it is a big change or a 'wee itty bitty no-one's going to notice' change!" On the same topic, Gilkes wrote to Inglis: "As usual, things reach crisis level due to competitivepressures, rather than being properly thoughtout and debated over time." Two months before the Cheyne SIV launched, Gilkes lamented, "[r]emember the dream of being able to defend the model with sound empirical research? The sort of activity a true quant CoE should be doing perhaps? If we are just going to make it up in order to rate deals, then quants are of precious little value." Frank Parisi, S&P's Chief Credit Officer for structured finance, testified that the RMBS model in effect in 2005 and 2006 was only marginally more accurate than "if youjust simply flipped a coin."

Not quite as evocative as the cows but not bad. Gilkes, by the way, is this guy, who seems more directly connected to the Cheyne ratings than I'd thought, but who now makes a nice living shitting on his former firm. Also feasting off the carcass of the rating agencies' credibility is the plaintiff's expert in the case, who was dredged up to say "these ratings were obviously wrong even at the time," but with math.**

It's hard to count any of this as much new news - if there's one thing we knew, it's that ratings of structured products in 2006 sucked, and if there's another thing we basically knew, it's that they sucked because rating structured products was too lucrative a business for the agencies to risk pissing off the banks by rating conservatively. These new documents mostly just provide more embarrassing evidence of that.

But I suppose that Abu Dhabi Commercial Bank - the named plaintiff in the Cheyne case, who got SIV all over them when Cheyne blew up - didn't know that in 2006, and I suppose that whatever widows and orphans Geoffrey Tomes was selling underperforming JPMorgan funds to using imaginary performance numbers didn't know that they could get other, non-JPMorgan funds with better imaginary performance numbers. So I guess these revelations are useful for the next wave of suckers - though they might have been more useful before, y'know, the last wave of suckers lost all their money.

Morgan Stanley Got S&P To Inflate Ratings, Investors Say [Bloomberg]

* Also sometimes financially rewarding.

** Are you persuaded? Here, here, here, here. He relies a lot on ratings transition correlation matrices rather than spreads or default probabilities, which strikes me as fake-ish, but I have no expertise.