Reaaly? February 13, 2013 at 7:59 am
What’s a credit event? It’s a difficult question. Dealbreaker is exercised on this, or more specifically on the issues with CDS protection holders getting paid on some unusual credit-event-like happenings:
This is a reasonable criticism. Certainly it seems odd that in something like the SNS Reaal bail in, sub CDS have not yet triggered (although they might today). Moreover, even if the CDS do trigger, because in SNS’s case the bonds have been expropriated, physically settled CDS holders may not have anything to deliver. Thus there are really two issues: the problem of what is and is not a credit event (a perennial issue in the CDS market); and the problem that even if there is a credit event, the things that the bond turned into (including, err, thin air) might not be deliverable. As Dealbreaker says
It may be “difficult to formulate contracts in an entirely watertight fashion,” but getting to a seaworthy concept doesn’t seem all that hard: essentially, define “Reference Obligation” to mean “(i) the Reference Obligation or (ii) whatever package of things the Reference Obligation poofed into in a Credit Event.”
I think the recent problems in the CDS market stem from the market morphing from one in which participants used CDS to trade the pure credit risk in a fixed income instrument or at an issuer level, to a broader market for asset insurance.
The defintions have always been quirky, but in that context, it’s perhaps easier to see why an expropriation would (a) not be a credit event, and (b) should lead to buyers getting no payment as the issuer is now more creditworthy.
I agree that the CDS market was set up to trade the pure credit risk in an instrument or issuer rather than as quasi-insurance, but this still needs a clear connection from the floating leg of the CDS to the ultimate outcome of placing credit in the instrument or issuer.
As far as I can see it will otherwise become impossible, not just to construct an arbitrage-based price for a CDS, but to arrive at any sort of rational price. Imagine how few people would gamble on horse-races if they weren’t allowed to bet on who would win, but had to bet on who would do ‘quite well’ according to the post-race decision of a committee of bookmakers.
In this particular case I think it could be argued that expropriation is not a contractual provision of the instruments involved, and therefore if SNS pays the government it will not be making the payments in any method allowed by the terms of the contract. I realise this is not an entirely watertight argument though.
In this particular case I think it could be argued that expropriation is not a contractual provision of the instruments involved, and therefore if SNS pays the government it will not be making the payments in any method allowed by the terms of the contract
This doesn’t sound like an argument that would fly (under English law). The legality of the expropriation may be challengeable under Dutch law, but until that is overturned, if SNS pays the government, there is no failure to pay.
Anyway, the committee has decided it’s a restructuring event.
GY – yeah, I saw the DC decision. It doesn’t feel good – it clearly isn’t a restructuring, but rather something else that, err, they didn’t think of.
JH – Following on from that, it might have been rather better just to wait for a failure to pay event. The problem with that is, though, that some CDS might expire in the meantime.
P – Sorry, I don’t follow. I had a bond worth, say, 80. Now I have nothing, thanks to government action. Why shouldn’t the CDS on that bond pay out?
I wonder if we’ll see a boom in political risk insurance for bank sub debt from EU countries.
The argument would be that the CDS shouldn’t pay out unless there is a credit-related writedown or default. In this case it’s perhaps more arguable given the expropriation was linked to the potential failure of the bank, but as a general proposition should CDS cover non-credit related government actions, or is that, as per Ginger, a risk to be covered by insurance?
I’d argue CDS shouldn’t cover non-credit risks, but in a world where the exact mechanisms for writing down hybrid capital are in a state of flux, arguably it should cover any effective writedown regardless of the form in which it is implemented. At the end of the day, if CDS can’t (in principle, for contracts written in the future) cover this sort of bail-in, it’s worthless.
The point surely is that resolution risk isn’t in addition to credit risk, it is (mostly) instead of it. Banks will be resolved rather than failing. Therefore, as GY says, if CDS don’t certainly cover this risk, they are of very uncertain utility on financials.