OTC Derivatives Reform (Maybe I’ve spent too many months of reading regulatory proposals and legislation, or maybe I’ve just grown pessimistic, but it seems a credit default swap loophole exists in the otherwise mundane definition of “swap” released by the Commodity Futures Trading Commission on April 27.
The CFTC fact sheet on the proposal tells us that consumer and commercial transactions, loan participations, some forwards and insurance contracts are not swaps. That makes sense. But reading into the definition of insurance made me stop and think – if you take out point four (below), doesn’t this sound very much like a covered credit default swap (CDS)? My comments in parentheses:
- (Must own the bond.) The beneficiary must have an insurable interest that is the subject of the contract and thereby bear the risk of loss with respect to that interest continuously throughout the duration of the contract;
-
(The underlying has a credit event.) The loss must occur and be proved;
-
(the notional of the bonds held) any payment or indemnification for loss must be limited to the value of the insurable interest;
- (I’ll get back to this one.) The contract must not be traded, separately from the insured interest, on an organized market or over-the-counter; and
- (Contract terms are up to the dealer.) With respect to financial guaranty insurance only, in the event of a payment default or insolvency of the obligor, any acceleration of payments under the policy must be at the sole discretion of the insurer.
Comments | Post a Comment
8 Comments to "A new CDS Loophole":
Anonymous
02 May 2011
This seems to state that an insurance company, say AIG, can sell protection up to the notional amount of the underlying credit they hold? Would this not be a transaction that increases leverage and thus exposure to the underlying credit? Isn't this in fact how AIG got in trouble in the first place?
gbollenbacher
02 May 2011
Great observation, Kevin. An additional wrinkle - the asset manager owns the bonds when they buy the protection. It's insurance, not a tradeable swap. The credit risk goes down, the bonds go up in value, the asset manager sells the bonds, asks the insurer to terminate the contract. The insurer's price to do that is too high for the asset manager, so he does a back-to-back, which has now become a tradeable swap and must be done on a SEF. Right?
Comments (48)
Anonymous
02 May 2011
Isn't the point and the purpose for the new regs that AIG got in trouble by selling protection (thus payment to GS from TARP funds), not by buying it?
kmcpartland
02 May 2011
gollenbacher - interesting point. But if the back to back trade was considered a hedge than maybe it too would escape regulation as an "end user"? Not sure on that one. Anonymous - AIG sold insurance that it could never repay, and the same could happen here. Hence my point about the loophole. But, they couldn't sell more protection than bonds actually existed - taking out a lot of the leverage. When they sold CDS on CDOs, the amount of CDOs in the market was unlimited and most of the people by CDS on CDO didn't actually hold the underlying. That couldn't be true in this case - they'd have to hold the underlying hence limiting the amount of credit protection they could sell. All that's left then is counterparty risk - not dissimilar to my assumption that my home owners insurance company would pay if my house burned down.
Comments (80)
Anonymous
02 May 2011
Selling (CDS) protection = long Buying (CDS) protection = short If they are long the credit and sell protection, they are long x 2. Isn't the point of the regulation that they can only buy protection as a hedge and only for the stated notional? Notwithstanding the fact that the legislators do not understand this market...
kmcpartland
02 May 2011
I think we've gotten off a bit on a tangent - what I'm saying in the piece is that it might be possible to disguise a CDS as insurance, and hence evade CFTC/SEC regulation. Therefore what you're suggesting - assuming all of the above caveats - could in fact happen which is as you say contrary to the point of the regulation.
Comments (80)
gbollenbacher
02 May 2011
Anonymous has an interesting point. The perceived risk in the CDS market relates to insurance sellers who can't fulfill their obligations. There may be a risk that some buyers (premium payers) might default if the risk premium drops sharply early in the contract, but the regulation doesn't say much about that. On the other hand, the CFTC definition places most of the criteria for classification as insurance on the buyer, not the seller. If a firm approaches AIG, or some other insurance carrier, and asks to buy protection on a single bond (a single-name CDS) is it the responsibility of the seller (the carrier) to determine if the buyer owns the single name asset? Have any of the sellers agreed to do that? As to my previous point, what happens if the buyer sells the underlying; does the insurance contract now become a tradeable swap? Further, any future AIG FP, wanting to take a flyer, would sell some of the CDSs as insurance, to holders of the assets, and the rest as swaps. If the swaps transactions ere big enough, they would be exempted from SEF, and presumably from CCP, reporting as block trades. Are we better off then?
Comments (48)
Anonymous
02 May 2011
The irony in this debate is that in the past, investment banks would often structure credit derivatives transactions to avoid being considered insurance. The regulatory burdens were too cumbersome, and in many cases, the institution was not allowed to sell insurance.