You can't argue too much with the SEC's gentle suggestion that maybe banks should tell people, in a consistent format, what's up with their European debt exposure. It seems to be a thing that is on investors' minds, so why not have the SEC try to put their minds at ease:
"Our staff has been working with banks to improve their disclosure about sovereign-debt exposure for several months," SEC Chairman Mary Schapiro said in a written statement released Monday. "Even so, I understand this is an area of focus and uncertainty that could really benefit from further transparency and consistency, particularly as we head into annual reporting season. I think the staff's guidance should help achieve that goal."
Yep. The release is here and contains a good list of things you might want to know, including things like "The effects of credit default protection purchased separately by counterparty and country," "The fair value and notional value of the purchased credit protection," and "The types of counterparties that the credit protection was purchased from and an indication of the counterparty’s credit quality." It's not exactly a standardized form for disclosure that will allow everyone to do detailed comparison among the banks and/or sleep well at night, but it should at least shame people into giving reasonably detailed substantive information so that when your bank blows up you at least won't be surprised at which European country did it. That seems good. It even seems like what the SEC is supposed to do.
The Journal, ever fair, finds a token objector, sort of:
Bert Ely, a bank consultant based in Alexandria, Va., called it "really significant," even though the guidance is nonbinding. ... Mr. Ely cautioned that it may be "tricky" for large financial institutions to comply, because it often is difficult for them to pull financial information in a consistent fashion from all of the different countries in which they operate.
Well let's really hope that isn't true, but on the off chance it is, then that makes the SEC's gentle suggestion an even better idea! If it's "tricky" for banks to disclose what their European exposure is because they don't know it themselves, then this is an excellent New Year's motivator to get right on that.
In other shadow-of-credit-derivatives news from semi-comical financial regulators*, here is a paper from the European Central Bank, who ought to know, about how moral hazard in CDS protection selling can increase risk. The basic intuition goes like this:
1. Bank A buys CDS from Bank B
2. Underlier deteriorates
3. Bank B is like "oh no, now all my assets will go to go first to Bank A to pay off my CDS obligations, so I have less incentive to preserve assets and more incentive to gamble them all because I have limited liability"
4. Bank B gambles
5. Bank B goes bankrupt with some probability
6. Bank A now is unprotected
7. Madness, death, etc.
Now this is both obviously right economics 101 and also sort of a deeply weird way to think about the issue. (Because why CDS? That list of events is like the list of everything that happens to a bank. Banks are just piles of liabilities with a thin candy shell of equity. If you really think that every time a bank's assets lose value everyone there is like "shit, now all our earnings go to creditors, let's just take the rest of this stuff to Vegas," then how do you sleep at night? There are things you can and can't think here about incentives and whatnot.) But anyway the authors try to solve the problem with margining and transferability rules, and fine, whatever.
A throwaway bit caught my eye though:
[First] we consider the case in which the protection buyer can observe the seller's risk-prevention effort so that there is no moral hazard and the first-best is achieved. While implausible, this case offers a benchmark against which we will identify the inefficiencies generated by moral hazard. In the first-best, efficiency requires that the protection seller exerts effort and that margins are not used. ... [math math mathety math math math] ... When effort is observable, the optimal contract entails [proper risk management] effort [by the protection seller], provides full insurance and is actuarially fair.
So "while implausible," monitoring of counterparty behavior by protection buyers is the first-best case. "Implausible" seems about right from a mathematical-model perspective - you can't really monitor in real-time whether your CDS counterparties are turning into reckless jerks - but obviously there's some scope for monitoring. And not much is done. It's kind of weird. Lots of CDS and related contracts require counterparties to maintain decent ratings (or post collateral if they get downgraded), and not default on their other debt's covenants, which are about the bluntest monitoring tools available - but the idea of bank-written credit insurance contracts containing covenants, or requiring more granular risk disclosure to counterparties than shows up in SEC filings, would probably strike most people as absurd.
But covenants are one thing; market discipline - protection buyers abandoning dicey-looking sellers - is another. In a world where the norm is for banks to keep their hot-button risk exposures - this week, European debt - a closely guarded secret, counterparties have to do their monitoring based on market rumor and Egan-Jonesery. In a world where the norm is for the SEC to push reasonably thorough, reasonably comparable disclosure of those risk exposures as they become hot buttons, actual efficiency-enhancing monitoring of counterparty risk becomes a possibility.
* Whatever, the ECB isn't really a regulator, fine.