Transparency: Sometimes Even the I-Banks Like It

Gari has a question about the new disclosure rules surrounding credit default swaps. Basically, up until now they've been treated, for disclosure purposes, like derivatives, under FAS 133; as of November 15, they'll be treated like guarantees, which come with a lot more detail, under FIN 45. And Gretchen Morgenson, for one, welcomes the increased transparency.

But Gari notes that the monolines, who have always treated their CDSs like guarantees, are very glad that they've been able to do that, rather than having to mark them to market as they would do if they were treated like derivatives. So what does this change mean for the banks? Might they actually like being able "to make predictions of how the instruments will perform to term," he asks, as well as (or alongside) any marking-to-market that they have to do for accounting reasons?

Gari finishes with this:

I'm sure Jack Ciesielski, who's quoted by Morgenson and weighed in on Schwarzman, would be able to clear it up. But I'm too poor and shallow to pay for his insight.

Well, all you have to do is ask. Which is what I did, and this is what Ciesielski replied:

FIN 45 was "tweaked" only to the extent that it will now require disclosure of the current status of a guarantee's payment/performance risk. That disclosure could be made by disclosing say, the current external credit rating of the underlying, if it's available. Or it could be a disclosure of the firm's own internal assessment/rating of the underlying.

That disclosure hadn't been made for credit-indexed derivatives that were accounted for under Statement 133. They didn't have any of the narrower guarantee disclosures in FIN 45 now being extended to them. They were accounted for under 133 @ FV; the FIN 45 disclosures for guarantees, however, were more specific for the kinds of risks entailed in those arrangements. A credit-indexed derivative carries the same kind of risks, but not those specific disclosures because they weren't required under 133.

So this amendment gets instruments that are doing the same thing to make the same level of disclosures. It seems like a good idea, no?

It does seem like a good idea. Up until now, CDS-as-derivatives have just had mark-to-market disclosures; from November on, they will have CDS-as-guarantees disclosures, which are much fuller, and which can involve things like "the firm's own internal assessment/rating of the underlying" credit being guaranteed.

But here's the question: Morgenson quotes Ciesielski as saying that the rule change could cause some volatility in the market, since banks won't want to reveal all this extra information:

Fearful of how investors will react to the extent of their swap holdings, companies may move to unwind them or offset them when they can.

"This is something that will change behavior," Mr. Ciesielski said. "If you don't want to look so bad, you're going to have to be busy in the next few months to work these down, wriggle out of them or offset them."

On the other hand, Gari suspects that investment banks might embrace the new rules. Yes, officially, they will still have to mark their CDS to market, just as they have done until now. On the other hand, at the same time they can point to credit ratings and internal assessments saying that the mark-to-market values are massively out of whack with fundamentals, and that they don't in reality have anything like the kind of liabilities that the market is saying they have.

I suspect the truth will be somewhere in the middle. Just like fund managers like to buy outperforming stocks at the end of the quarter in order to improve their optics (if not their returns), investment banks will want to show that they've written protection on reasonably safe-sounding real-world credits, rather than impossible-to-fathom CDOs and the like. So there might be some mildly panicked selling of CDS on CDOs, and other such exotica. But on the other hand the banks might quite like being able to show that the stuff they're taking writedowns on does not in reality look nearly as dangerous as the mark-to-market losses might suggest.