CDS vs TROR -- confusion about buyer and seller

[sridhar]

David,

Help me with this confusion.

If you treat CDS and TROR as two varieties of swaps -- it is clear to me who the swap buyer and sellers are in a CDS. The swap buyer in a CDS owns the reference asset and transfers the credit risk to the swap seller.

In the case of TROR, it seems to me that the swap seller (aka the TROR payer) has ownership of the reference asset and transfers the credit risk to the swap buyer.

In a CDS, the swap seller ends up with the credit risk.
In a TROR, the swap buyer ends up with the credit risk...

Do I've this right? I remain confused My confusion is I don't quite understand in a TROR, who is buying protection and who ends up with the credit risk -- using the language of "swap buyer" and "swap seller."

--sridhar

 

[David Harper CFA FRM]

Hi sridhar,

First a note which I do not think is central to your point: in either a CDS or TROR, no ownership of the reference asset is necessary. In our readings (de Servigny was the assigned here, now is Hull), TROR implies ownership by the TROR payer but it is not essential. Even further, the CDS does not imply ownership (currently in the market, I believe, there is more non ownership; it is exactly why you will read savvy commentators say the notional growth in CDS is not meaningful - because most/many of the credit sellers do not own the reference). It will help to think of both of these (CDS, TROR) as "merely" bilateral (between two parties) contracts. One party may happen to own the reference asset, maybe not. Put another way, you and I can craft a CDS on IBM; e.g., I pay you credit spread (Hull's payment), you give me contingent payment if IBM defaults (Hull's payoff). We don't need to own IBM bond to make this contract, we only need to "reference" it for the objective trigger

Your language point is interesting to me. But you are basically correct. Partly it is semantics: Culp (who is not assigned for this) does in fact call the TROR parties credit protection buyer and seller. If we used Culp, you would have no confusion!

The others do not, and IMO this is because a CDS is truly a credit derivative (as you say, credit risk is transferred to the "swap" buyer) but many have suggested that swap is a misnomer here - a CDS is less like a swap and more like an option...

...but the TROR is truly a swap. So, the language typically switches to payer/receiver. The TROR is using language similar to the interest rate swap, for good reason. But Meissner's TROR Receiver = Culp's Credit protection seller. My minor point is that viewing the CDS as a true credit derivative, but the TROR as swap (that additionally swaps market risk) arguable justifies buyer/seller versus payer/receiver swap language. hope that helps...

David

append: i meant you are correct with this: "that the swap seller (aka the TROR payer) has ownership of the reference asset and transfers the credit risk to the swap buyer" in the sense that TROR payer is transferring credit risk to the TROR receiver; the receiver is assuming default, credit deterioration & market riks.

 

[sridhar]

Thanks David...As always, you deliver....My motivation in asking the question was this: I saw today somewhere a practice question that started off with:

"An analyst is examining a total rate of return swap....." with some details.....And then goes on to ask which of the following stmts is least accurate. The choices all referred to "swap buyer" and "swap seller" as opposed to payer and receiver. I guess this is why you and others have said that it is truly important to understand what's going on and not just skim stuff -- to have a shot at the FRM.

--sridhar