Are bankruptcy futures a good idea?

Alea, who broke the news that the CFTC's proposed CDS futures were being tranformed into bankruptcy futures, has a provocative post up saying that this is actually a good thing.

The CFTC was right to restrict the credit event to bankruptcy only.
Otherwise sellers of protection would been in the same situation that an insurance company having to pay up in full for a new house if there was a fire in the attic.Not an attractive proposition.
Main benefit of the CME contract:
given that the recovery rate is fixed ,it will allow for discovery of the implied hazard rate.This a Good Thing.

Sellers of protection are always in the position of that insurance company, though – which is why they insist on getting a bond back in return for paying out. It's a bit like the insurance company saying that they'll pay out in full on the fire in the attic, but only if they get the house in return.

Now I'm unclear on how the new CFTC contract will work, but I do like the idea that it will help create a very clear hazard rate, which is similar to the default probability. I know that Jochen Andritzky at the IMF has a complicated formula for deducing the default probability from the combination of CDS spreads and bond spreads, but that doesn't work when the CDS market simply dries up, as it has done in the case of Ecuador.

My worry is that bankruptcy futures will make workouts much messier, but Alea, by email, reckons otherwise:

I don't think it will complicate workouts because the event "bankruptcy" itself terminates the contract and if one party has an incentive to game the event, the other party, the seller of protection, has an incentive to counter the gaming.

I'm looking forward to learning more about how these contracts work – and seeing whether they take off. As Alea points out, it's been tried before, unsuccessfully.

Posted by Felix at 16:51 EST

Comments

Felix
I believe from reading the Enron emails that the CFTC actually was approached in 2000/2001 regarding the so-called bankruptcy swaps and they have had plenty of time to reflect on the issue of moral hazard that you rightly raise.

Posted by: jck at 17:59 EST, February 06, 2007

There are zero-recovery CDSs (though generally of low liquidity), and of course the spreads on them are wider; there's no reason not to do a binary bet on whichever set of criteria you wish - without delivery of a reference obligation but the price is going to change.

The average insurance company will insist that you actually own the house you're insuring, and that you not overinsure ("I'll pay twice the premium if you'll pay double my losses"), largely, as I understand it, for moral hazard (and fraud) reasons; CDS prices are typically quoted on the assumption that neither party has substantial ability to actually cause the default to take place. You can buy protection on a bond you don't own, and you can buy protection that will pay the full par value of a bond even if there is a nonzero recovery; there's no moral hazard reason not to.

Posted by: dWj at 23:38 EST, February 06, 2007