Hi David, I've got a short question on the above mentioned. Am I correct to say that the CDS is an unfunded protection due to only contingent payoffs? And TRS would be partially funded and lastly CLN is funded.
There are two good points of difference here: funding and risk transfer.
With respect to funded/unfunded, the FRM has consistently viewed the CDS and TRS (aka, TRORS) generically as unfunded. I guess partially funded could apply depending on the definition, but by "funded" we aren't referring to collateral or margin (of course collateral is counterparty risk mitigation), but with respect to "funded" we ask: have the investors, who have assumed the risk, prepaid with their cash such that there is not material counterparty risk. Fully funded would refer to cash "prepaid" to cover the full exposure (in a CDS the exposure may include the full notional; whereas M2M collateral may fall way short of that).
In the case of a TRS, we say "unfunded" because (eg) the reference deteriorates in value, the protection buyer counts on the protector seller (counterparty risk) to pay instead of withholding prepaid funds (or getting from trustee). In this way, the way i look at funded/unfunded is simply: where is the cash that will be used to cover default/deterioration? If it was prepaid by investors and available to protection buyer, it is "funded;" if protection buyer assumes counterparty risk to recover, unfunded.
Funded = investors prepay cash; unfunded = counterparties merely have contractually promised to cover
CDS: unfunded, hedges credit default only (+ credit deterioration if M2M)
TRS/TRORS = CDS + market risk: i.e., unfunded, but hedges credit + market risk (most comprehensive hedge among credit derivatives)
CLN: funded equivalent of CDS (why funded? b/c unlike the CDS protection seller, the CLN seller prepaid!), hedges credit but not market risk
Have a quick question. Is it safe to say that the CLN investors are short the CDS and the CLN issuer is long the CDS. In case of a default by the reference, the CLN investors only get the recovery rate. And as per my understanding, CLN investors get the CDS premium through a higher yield on the note...?
I 100% agree with your characterization (as you know "short the CDS" is to sell credit protection, which is synthetically long the reference) and it, IMO, perfectly comports with Culp: "A CLN is thus economically equivalent from the issuer's perspective to issuing a normal note plus buying credit protection [i.e., "long the CDS"] from the bond investor through a CDS ... the above-market interest earned by investors is equal to the normal interest on the bond plus the CDS spread ... in the event reference asset default does occur, the par value of the reference asset minus the expected recovery rate is diverted from CLN holders to the issuer."(Culp, Structured Finance, p 264)
"The investor selling the CDS is viewed as being “long” on the CDS and the credit, as if the investor owned the bond. In contrast, the investor who bought protection is “short” on the CDS and the underlying credit"
At least what you don't disagree on is that buying protection is short the reference and selling protection is long the reference.
Wow! I typically consider wikipedia trustworthy, given it is crowdsourced, but I disagree totally wrt the CDS instrument: I have *never* seen long/short CDS employed that way. I have only seen long CDS = protection buyer and short CDS = protection seller. Off the top of my head, at a minimum, I know that Gunter Meissner (previous FRM authority) and Jorion use it thusly, and actually highly authoritative, imo, is Moorad Choudhry who's written a whole book on CDS and I know he uses long CDS = protection buyer. I don't have a source that employs it otherwise ....
but as you suggest, the semantic of long/short CDS are more debatable than buying protection is unequivocally to be (synthetically) short the reference: in both cases, your position gains on MTM deterioration. Thanks for the link, it's not every day we enjoy the honor of disagreeing with wikipedia
Hah - no problem. I've heard both used commonly...there are some people who swear by calling everything in bond terms so they would agree with wiki. I'll do a strawpoll on Monday when I'm back in the office...
I just find it easier to talk in terms of protection....oh, and, you could always edit wikipedia when you get the chance?!
could you please do the strawpoll (I don't live in the real world, seriously)? makes total sense re: consistency with bond terms.
re: wikipedia: no thanks, i try to limit my free work to 4 or 5 hours per day, i'm just not that altruistic that i feel obliged to change wikipedia
Just to jump in on the topic because this has caught me out a few times in practice q's! At my institution (a global IB) we refer to it the same way as Mark W mentioned - in bond terms, i.e. 'long' CS01 is either long the bond or selling protection in CDS.
I've had to make a mental note to flip the direction on any long/short CDS type Q's.
Well, I counted at least 3 questions in the P2 on Saturday that referred to either 'long' or 'short' CDS. I chose to follow David's view and take long CDS to mean buy protection, i.e. short the credit. I'm very glad I asked this questionn now and got clarification prior to the exam otherwise my inclination would have been the other way around (for bond consistency).
I do think it's a point GARP should be clear on...after all, it's purely semantic (though they may profess consistency with the source text...?).
Hi Mark - really? P2 questions that refer to "long [| short] CDS" without further specification?
What worries me is that, in this thread alone, we've heard two reality-based (actual practice) citations, plus wikipedia, that cite a definition contrary to the readings' authors ... I am going to query GARP to confirm that their usage was (is) the same as Jorion (and previously assigned Meissner). It seems risky to use naked "long CDS" without clarification, given the exam is supposed to be practical and practitioners appear to a definition which is the opposite of the readings. Thanks!
Yes, there was one direct question along the following lines where it mattered explicitly:
Position; Rating Before; Rating After
Long CDS; XXX; XXX
Short CDS; XXX; XXX
Long Bond; XXX; XXX
Q: Which scenario of ratings causes biggest decrease in credit quality?
And two separate ones where you needed to be clear on which way around you're. I was surprised, if I'm honest, as I thought GARP would be clear in the exam.
Anyway, straw poll at work underway. Though one comment from a former banker/hedgie guy was "the bank always sells the CDS (i.e. long credit) - they are the ones who know more than you and are trying to rip you off!"
Hi Mark, Thanks, that is really interesting! I just forwarded a link to your post to GARP's Research Director (it's nice that the majority agree with the text, but clearly there is ambiguity in practice). Thanks again for taking the time to post your findings, I think it will be helpful,
Just to reiterate can we say that short CDS is a kind of option to the buyer where he has to pay premium to avail that and vice versa for seller. Can we look in terms of hedging their reference asset. But a lot of people just go on long on the CDS ( many hedge funds who made money in the crisis they just paid premium even if they are not holding toxic asset on their portfolio yes but they are paying huge premiums to stay i the hunt ) Is that naked short? I am not sure if they are having any referenced asset in the naked short like option. Is my understanding correct?
Buying protection through a CDS is not buying an option...it's 'like' buying insurance - but they are not the same as you don't necessarily need to have an insurable interest. In other words, you can buy protection on a company/sovereign whose bonds you don't own and so profit if they suffer a credit event (let's say default to keep it simple). Note that neither party has an option over whether to pay/exercise or not - they are legally bound to do so.
Quoting from the FT, who are in turn quoting SIFMA:
"In this respect, CDS are like all other derivatives (listed and unlisted) that do not require the incurrence of a loss as a condition to payment under the derivative instrument…the absence of any requirement that a CDS holder incur a loss as a condition to payment is critical to the efficiency of, and the benefits afforded by, the CDS market. Because CDS do not require an insurable interest, or the incurrence of a loss as a condition to payment, if treated as insurance, they would be prohibited as a result of the very feature that has contributed to the success of this product."
As to whether you should have to have an insurable interest when buying CDS protection, that's a whole different debate.
Hope that helps. Thanks - Mark
P.S. Reuters article regarding EU banning 'naked' CDs, i.e. buying protection without owning the bonds. Quote:
"But in a surprisingly blunt paper, the International Monetary Fund weighed in strongly against the EU ban last week - saying it found little evidence that SCDS (Sovereign CDS) overall had been out of line with bond spreads and that for the most part premiums reflected the underlying country's fundamentals - even if they reflected them quicker than the cash market"
A CDS only outsources Credit Risk while a TRS (Total Return Swap) outsources Credit + Market Risk.
In a Total Return Swap the seller of credit risk (the bank or called TRS payer) pays 1. the coupon and 2. the price appreciation to the investors (TRS receiver or the buyer of credit risk) who pay 1. LIBOR + some given basispoints and 2. the price depreciation to the seller of the credit risk (the bank or TRS payer).
A TRS is more of a hedge than a CDS from the seller perspective of credit risk (or also called the buyer of credit protection or TRS payer). The TRS payer is hedged against market & credit risk & credit deterioration (price depreciation of the underlying credit/asset).
Here are some good references about TRS by Choudhry:
TRS = unfunded
CDS = unfunded CLN = funded (funded equivalent of a CDS)
If a credit event happens, the CLN investor (protection seller OR called buyer of credit risk) must forego some coupon payments or principal. CLN are on-balance sheet debt instruments and are linked to the performance of the referecne asset/credit
Bear in mind: Crouhy & Choudhry are NOT consitent with terminology:
1. Seller of credit risk (the bank) = buyer of credit protection 2. Buyer of credit risk (investors) = seller of credit protection
Unfunded: no upfront payment is made by the total return receiver at inception of the transaction. The TRS receiver pays no upfront amount in return for the total return of the reference asset/credit. Under an unfunded credit derivative, the seller will make a payment only if the conditions to settlement are met.
In an unfunded credit derivative, typified by a credit default swap, the protection seller does not make an upfront payment to the protection buyer. Thus the main difference between funded and unfunded contracts is that in a funded contract, the insurance protection payment is made to the protection buyer at the start of the transaction; if there is no credit event, the payment is returned to the protection seller. In an unfunded contract, the protection payment is made on termination of the contract on occurrence of a triggering credit event. Otherwise it is not made at all. When entering into a funded contract transaction, therefore, the protection seller must find the funds at the start of the trade.Compared to cash market bonds and loans, unfunded credit derivatives isolate and transfer credit risk. In other words, their value reflects (in theory) only the credit quality of the reference entity.
The CAIA writes about CLNs:
CLN are bonds issued with an embedded credit option. Typically, these notes can be issued with reference to a single corporation or to a basket of credit risks. The holder of the CLN is paid a coupon and the par value of the note at maturity if there is no default on the underlying referenced corporation or basket of credits. However, if there is some default, downgrade, or other adverse credit event, the holder of the CLN will receive either a lower coupon payment or only a partial redemption of the CLN principal value.
Why would an investor purchase a CLN? The reason is simple. By agreeing to bear some of the credit risk associated with a corporation or basket of other credits, the holder of the CLN will receive a higher yield on the CLN. In effect the holder of the CLN has sold some credit insurance to the issuer of the note. If a credit event occurs, the CLN holder must forgo some of his coupon or principal value to make the seller of the note whole. If there is no credit event, the holder of the CLN collects an insurance premium in the form of a higher yield. The investor in the CLN is, in fact, selling credit protection in return for a higher yield on the CLN. CLNs appeal to investors who wish to take on more credit risk but are either wary of stand-alone credit derivatives such as swaps and options or limited in their ability to access credit derivatives directly. In contrast, a CLN is just that, a coupon-paying note. They are on-balance-sheet debt instruments that any investor can purchase. Furthermore, they can be tailored to achieve the specific credit risk profile that the CLN holder wishes to target.