John Carney has a really smart post over at Clusterstock regarding the pricing of credit default swaps for Campbell Soup, JPMorgan and the US government. In case you haven’t been following all of this, the cost to insure credit risk on Campbell Soup and this country are essentially the same while the cost to insure JPMorgan risk is approximately three times as much.
Bloomberg makes an argument that this makes no sense since Morgan is essentially a GSE. In the event of failure, strike that word failure, in the event of a problem with Morgan we can all count on the government coming once more to the barricades. True, but as Carney points out this misses some important elements of risk:
Let’s run through some risks that Reilly seems to be overlooking when it comes to JP Morgan’s debt.
We know that the government isn’t going to let JP Morgan go bankrupt. But we definitely do not know what form future bailouts will take. Maybe creditors will be protected in a future bailout. Maybe they won’t. Keep in mind that in the recent bailouts of the auto sector forced bondholders to take deep haircuts.
We also know that the failure of JP Morgan would almost certainly mean that the financial system was in great distress. In that case, anyone who sold insurance on JP Morgan would likewise probably be distressed, making paying off the insurance more costly. It seems what’s happening here is that sellers of swaps are smartly taking into account this risk.
That kind of risk doesn’t apply to isolated failures due to corruption or embezzling. If Campbell’s went down, the credit markets wouldn’t suddenly freeze up. Those who sold the credit default swaps wouldn’t necessarily have trouble getting the liquidity they need to fund the obligations.
Overlooking these kinds of risks is a problem for Reilly’s argument. It’s hard enough to establish you know how to price risk better than the market. And it’s pretty much impossible when you over look important risks.
I particularly like the way John gets to the real point about letting the markets sort the risks out, but I still don’t quite understand the concept of writing CDS on US government debt.
It seems to me this is a pointless exercise. A default by the US would be such an outlier that the ultimate risk and fallout would seem to be non-quantifiable. Therefore, either buying or selling such insurance would seem to warrant the gambling pejorative so often hurled at the swaps market. Now there may indeed be some perfectly valid reasons for hedging against this risk that involve other parts of the credit markets, and if that is the case I would love to hear from anyone out there who can shed light on the topic.
I suppose I can understand the motivation to purchase protection against a US default, I just cannot imagine why any purchaser assumes they would ever collect on such a policy. In the event of a true default, it’s likely that any writer of CDS would be rendered insolvent due to other exposures. The probability of collection would most likely be rather small if it existed at all. If that’s indeed the case then those who are writing these policies are clever indeed since they probably recognize the Armageddon consequences if they are called upon to pay. Basically, they’re collecting premiums against risk they could never cover. On the other hand, buyers would seem to be spendthrifts with nary a clue.