Harry Wilson today outs Allen & Overy’s David Benton as the legal mastermind behind the mess that is sovereign CDS documentation. Benton’s certainly coming under a lot of criticism these days, and not just on the ultra-wonky end of the spectrum from people like me. Even Pimco’s Bill Gross seems to have a beef with these rules — and Pimco’s on the Determinations Committee!
“If I were a buyer of protection on Greece and have seen the result this morning in terms of no protection, then I would be upset,” Gross, manager of the world’s largest bond fund, said on CNBC television of the ISDA’s decision.
So when Wilson says that ISDA’s decision not yet to declare default was “controversial”, he’s not wrong. Here, for instance, is Barry Ritholtz:
Here is a question for the crowd: Exactly how brain damaged, foolish and stupid must a trader be to ever buy one of these embarrassingly laughable instruments called derivatives?
The claim that Greece has not defaulted — despite refusing to make good on their obligations in full or on time — is utterly laughable.
And Peter Eavis is back with more CDS criticism, too:
One of the decisions of the swaps association on Thursday underscored how swaps can be disconnected from actions that harm investors’ economic interests. As part of the Greek debt deal, the European Central Bank will be shielded against losses on the Greek bonds it holds, a move that relegates, or subordinates, the claims of private creditors who hold the same bonds.
But the swaps association said the plans to subordinate private creditors do not meet the definition of subordination in the swaps contracts, so they do not have to pay out.
All of which says to me that ISDA and Greece have done an incredibly bad communications job here. Because ISDA’s decision was, clearly, the correct one.
The point here, which is easy to miss, is that credit default swaps only get triggered when there’s a real-world event of default. Yes, the deal with the ECB is indeed going to subordinate private-sector bondholders. And yes, Greece is indeed going to fail to make good on its obligations come March 20. There will be an event of default in Greece. But swaps don’t pay out on future events. They pay out on past events. And Greece hasn’t actually defaulted yet: every payment it has promised to make has, to date, been made in full and on time.
Now there are exceptions to this rule. If a government explicitly repudiates its outstanding debt, then that can count as an event of default even before a payment is missed. But Greece hasn’t actually done that. And most of the time, default swaps only pay out when there’s a default. Which is as it should be.
Is that reason for bondholders to be upset, pace Gross? Absolutely not. If you own a credit default swap on Greece, you own a piece of paper worth about 75 cents on the dollar. If you want to realize that 75 cents right now, you can: you can just sell your CDS. If and when the CDS is officially triggered, there will be an auction, and the CDS will be found to be worth roughly 75 cents on the dollar. In that case, you will wind up with 75 cents whether you like it or not.
In other words, when Greece finally defaults, owners of credit protection will be forced to get a payout. Whereas those owners right now have the option: they can take the payout if they want it, or they can hold on to their CDS position if they would rather do that. I don’t see why having that option would make anybody upset.
This is why the CDS market has been so successful: it’s a liquid, market instrument, which prices in expectations of future default. Ritholtz is right that Greece has refused to make good on its future obligations. And as a result, default protection on Greece is extremely valuable. When that future date comes and goes without a bond payment, the CDS will get triggered, and holders of protection will get a lot of money. There’s nothing broken there.
The subordination question is a bit messier, but it’s fundamentally the same idea. Greece has now created two classes of bonds: the ones held by the ECB, and the ones held by private bondholders. There’s nothing in the documentation of those bonds which might indicate the ECB’s bonds are senior to the private sector’s bonds. Right now, they’re all, legally, pari passu.
Again, in future, that’s not going to be the case. Greece is going to privilege the principal and coupon payments to the ECB, while imposing a massive haircut on the payments due private bondholders. That’s both subordination and an event of default. And when it happens, the CDS will get triggered. And that trigger is priced in to the CDS market.
In many ways it’s the genius of the CDS market — at least in theory — that there’s no rush to trigger CDS, because if you know that the instrument is going to get triggered very soon anyway, it’s going to be worth pretty much the same today as it will be when it’s triggered.
That’s my problem with the way ISDA rules cover bonds covered by CACs. Because of technical issues surrounding the availability of deliverables, it’s possible that if you wait for the default to happen, you’ll be too late to get what by rights should be your payout on the bonds. But this is a separate issue from what Gross and Ritholtz and Eavis are worrying about. They seem to think one of two things: either that Greece has already defaulted, and that therefore the CDS should have been triggered by now, or else that a Greek default is so certain at this point that the CDS should have been triggered by now. The first isn’t true. And the second is silly.
Eavis has another point, which is that default swaps are used for a purpose, and that purpose is to hedge against falling bond valuations. (That’s what he means by “investors’ economic interests”.) He is worried that the payout on the bonds might not be entirely in line with the loss of value on the bonds. And that’s a reasonable worry. But it’s also, right now, a pretty theoretical worry. Because in practice, the value of Greek CDS has tracked the value of Greek bonds extremely closely. In other words, even if there are possible problems with them in theory, they seem to have worked OK in practice.
I’ve got a few questions for ISDA about the way that CDS documentation works in the sovereign context in particular, and I’ll be wonking out about this issue further going forward. Because I think that the combination of CACs and CDSs is potentially extremely dangerous. But what I’m emphatically not worried about is ISDA’s decision not to trigger the CDS just yet. That decision was exactly correct. Even Pimco voted for it.