It’s Clear to the WSJ, Anyways
The WSJ published an editorial expressing deep skepticism of regulators’ rush to force the creation of a CDS clearinghouse. The editorial’s argument incorporates many elements I’ve raised repeatedly on SWP (and since I know several folks at the Journal read SWP, perhaps the blog contributed to their thinking.) Most importantly, the piece emphasizes the potential for balance sheet risk and moral hazard inherent in the central clearing structure:
[T]he Fed-imposed architecture should still cause taxpayer concern. That’s because it takes the widely dispersed risk in the CDS marketplace and attempts to centralize it in one institution. If not structured correctly, it may reward the participating firms with the weakest balance sheets. For this reason, some of the dealers who have resisted a central counterparty because it threatens their profits may now embrace it as a way to socialize their risks.
. . . .
Here’s how the New York Fed’s central counterparty would change the market: Right now, CDS trades are conducted over-the-counter as private contracts between two parties. They are reported to the Trade Information Warehouse, so the market has some transparency, but nobody is on the hook besides the two parties to the agreement. This provides an incentive for each party to make an informed judgment on whether the counterparty can be relied upon to pay debts. The buyer of credit protection — who is paying annual premiums for the right to be compensated if a company defaults on its bonds — has every reason to study the balance sheet of the seller of a CDS contract. [The superior ability and incentive of bilateral counterparties to evaluate creditworthiness is the essence of my argument about the virtues of bilateral markets.]
. . . .
[T]his system also introduces new risks, because all participants become liable for the potential failure of the weakest members. How does one appropriately judge the credit risk of a participant? ICE Trust and the Fed haven’t released details. Sources tell us that participants will need to have a net worth of at least $1 billion, and, more ominously, that the Fed wants a high rating from a major credit-ratings agency as a crucial test of financial health.
$1 billion net worth? That’s a real screamer. Just what was Lehman’s net worth before it went bust? AIG’s? So . . . $1 billion is supposed to provide some sort of comfort? And, given that balance sheets are notoriously opaque (especially now, when it is impossible to determine market values for large swaths of the assets of big banks and investment banks who would be members of the clearinghouse) relying merely on balance sheet information to assess creditworthiness is extremely dangerous. Extremely. Just because there’s $1 billion on the books doesn’t mean you can take that billion to the bank.
The editorial’s main original contribution is to highlight the FRBNY’s anointing of the big three rating agencies (S&P, Moody’s, & Fitch) to provide validation of the creditworthiness of clearinghouse participants. (It isn’t clear whether this means that CCP members have to achieve a certain credit rating, or the customers they clear for.)
As the Journal says, these guys are always the last to know. All too often they serve the same function as the crime scene unit guy who draws the chalk line around the corpse on the sidewalk. Or, to put it differently, they’re coroners rather than diagnostic physicians. When it comes to evaluating counterparty risk, however, autopsies aren’t all that useful.
Moreover, the rating agencies have no real skin in the game. Sure, they might take a reputational hit if they miss the impending implosion of a credit, but that hardly compares to suffering a major loss when a counterparty defaults. And hell, if the FRBNY is forcing the rating agencies on the ICE clearinghouse even in the aftermath of their abject failure over the past couple of years, even the reputational penalty doesn’t seem that great. If anything, it seems that incompetence is being rewarded, rather than punished. A party to an OTC trade has a much stronger incentive–and almost certainly, better information–to evaluate counterparty risk than the three ratings stooges.
The editorial also raises the excellent point that it is almost inevitable that the Fed would inevitably bail out the clearinghouse if it failed. Indeed, that implication inheres in the Fed’s rationale for the clearinghouse, namely, to reduce systemic risk. The Fed and other regulators argue that OTC market participants are so interconnected that a major failure poses a risk of contagion. But, wait a minute–a central counterparty connects these very same market participants, so it could also be the source of contagion, rather than a force preventing it. It is by no means clear that a central counterparty structure is less susceptible to contagion. Indeed, for reasons I’ve set out in a (tediously long) academic paper, it could actually be more susceptible. That, plus the inevitable too big to fail aspect, means that a CCP could create additional moral hazard. Great.
It is somewhat encouraging to see that a major publication has not imbibed the CDS clearinghouse Kool-Aid. It is also encouraging that the WSJ recognizes that a clearinghouse just slices and dices and redistributes default losses, rather than eliminate them (“all participants become liable for the potential failure of the weakest members”). Too much sloppy reporting, editorializing, and politicking traffics the lie that a CCP somehow makes this risk disappear, through some mysterious financial alchemy. Wrong again. Clearing is a risk sharing mechanism, not a risk eliminating mechanism. Big difference.
Hopefully, now that some of the hyperventilating over credit derivatives has abated a bit, cooler heads will prevail, and will examine the real (rather than imagined) costs and benefits of clearing before forcing a flawed solution on the market. Perhaps the WSJ editorial is the harbinger of such a welcome development. Here’s hoping.
regarding the opacity of balance sheets, does the SWP have any opinions about the upcoming XBRL (an xml schema for accounting) standards for submission of quarterly reports? I believe that it should be possible to automate wide swaths of the normal “auditing” process with the help of XBRL.
Comment by Surya — January 7, 2009 @ 12:12 am
Every little bit helps, but this isn’t the main issue. The main issue is the quality of the data in that goes into the reports, and the fact that accounting reports seldom provide the relevant information, especially in the current environment where price information is very hard to come by for many instruments.