What happens when you sell 1000 puts for $110K/year that give the buyer, Citibank, the right to claim $10M upon when you only have $100 million of cash and assets in hand, and a default actually occurs? You owe almost $10B, and only have $100M to cover what you owe. Thats the risk improperly collateralized credit default swaps pose. Fortunately for you, Citibank isn't permitted to legally hire shady guys to break your legs.
Credit default swaps can be simplified as puts for bonds, but generally without the conventional volatility and time decay exposure that typical options possess. The buyer pays the risk premium over treasuries per year as an insurance policy on his bond holdings, or an outright speculation that the value of the bond will go down. These are done with over the counter markets, and counterparties (buyer and seller without intermediary exchange organization) assume the risk of dealing directly with each other with varying, often none or assumed, collateral requirements. In the popular business media, its becoming known that the amount of total bond insurance they represent has come to 45 trillion dollars, versus only billions literally ten years ago.
The credit default swap seller is obligated to pay the buyer if the bond goes default. Additionally, credit default swaps change value (just as puts do) to match the movement of its risk premium over treasuries. Thus credit default swap value is often equal to risk premium, where a corporate bond total value is risk premium (credit default swap is equal) subtracted from the value of a treasury of comparable maturity.
Total market liabilities, and to the extent their exposure is collateralized (thus ability to pay sour bets), are really unknown. I can just imagine that the superhero of trading and financial markets, Goldman Sachs, must be harnessing the power of these swaps to create extra income on seemingly risk free debt with minimal capital requirements. Regardless if its Goldman, Citi, or even Bank of America who is selling these exposures, capital requirements are a question, and the event of a high quality debt default would probably send some likely overlevered players into collapse and immediate bankruptcy.
This is the underlying reason why I consider corporate bonds, even at current risk spreads, to be relatively unattractive unless you are willing to speculate fear is not properly assessing default risk. A high profile default event, which is likely in a recession, will probably send risk spreads screaming higher across the board.
As evidenced here, its already happening, from the likes of players like MBIA and Ambac. Here is a chart from Markit.com showing this recent move in high quality risk spreads. Remember, as this moves up, short credit default swap players (most likely financials) are losing money. Their risk exposures are likely higher in the aggregate compared to outright underlying bond holders of the same security.
What happens when players start to fail that many were not betting against already, however? I'm sure the biggest uncollateralized leverers (sellers of CDS), unable to pay upon bet failure, have a majority of their income harvesting short CDS positions in 'risk free' areas, companies ideally unexposed to recession and financial market turmoil risk - players like GE. A place you'd never expect a failure.
Perhaps an epic default would wipe the slate clean of anyone overleveraged in any holdings, especially the risk-seeking credit default swap sellers, resulting in a cleaner system for future generations. According to PIMCO's Bill Gross, the ramifications of even a typical default rate amid recessions is $250B of losses on the seller side, equal to worst-case estimates of subprime asset backed loans and their related derivatives. What if the default rate is double Gross' number of 1.25%, instead at 2.5%? $500B of losses. Obviously the answer is clear: many will be insolvent and even CDS insurance buyers will likely not get paid in full, or paid at all.
The High Yield risk spread tells the story of recession fear more obviously, containing a larger portion of risk.
This CDX chart from Markit.com is useful. Keep the link of http://www.markit.com/information/products/cdx.html on hand. If there is another shoe to drop to send us reeling into depression, you'll likely see it in the HY (high yield) and especially IG (investment grade) charts. Any market watcher should have his eyes glued to this along with the S&P and treasuries as a potential indicator of future/existing turmoil.
Monday, January 21, 2008
Credit Default Swaps - Keep Your Eye on the Ball
Subscribe to:
Post Comments (Atom)
NICE Blog :)
ReplyDeleteBad Credit Personal Payday Loan can be understood as fiscal sum supplied to applicant by a lender on his requirement, these loans fund for extreme situations of expenses. Name itself suggest that this loan specifically targets bad creditors records which is a cumulative result of breaching any financial contract i.e. to say having CCJ's, IVA's, defaults in repayments, arrears etc.To find cash advance loans, payday loans, cash advance payday loans, cash till payday visit http://www.cashadvancepaydayloans.org.uk
ReplyDelete(Professor Krause,
ReplyDeleteLet me try and redeem myself. Shorting EUR/GBP here at .7433, profit taking at .7250 and a stop at .7495 [which is roughly a 3:1 reward/risk ratio]. I'm trying to catch this on the way down and then all the way back up, regardless if markets blow up).
go for it... good plan.. just keep re-entering and don't double down.
ReplyDeleteunderstood. eur/usd may look like a buy one it partially retraces its 100 pip bounce off 1.4400. Eur/gbp is going as planned.
ReplyDeleteps. Euro just rocketed almost 100 pips to 1.46 against USD. No word on why yet.
ReplyDelete(Haha, shows you I'm not exactly on a trading desk. 75bp emergency cut. This is nuts.)
ReplyDeleteBe Alerted to the Hottest Stock Picks!!
ReplyDeleteJoin Emerging Growth Alert Newsletter !
By joining the team at Emerging Growth Alert you will be in position to receive stock alerts profiling stocks about to move or already in motion. Our alerts are sent in time for you to research, investigate and make a decision about whether this opportunity is right for you. You will not be bombarded with junk mail. There is ABSOLUTELY NO OBLIGATION, and the service is entirely FREE.
http://stocksrus.wordpress.com
http://financialstocks.webs.com
http://cashstocks.angelfire.com
P.S.
You may think this is spam but this is only an invitation for those interested to receive Stocks alert – ONLY stock alerts... Thanks to those who subscribed :)
I have a question about how CDSs work. I read that if there is a default on the bond, the protection buyer is obligated to buy the defaulted bonds and hand them over to the protection seller, before the protection seller is obligated to pay the stipulated insurance. Now, if neither the protection buyer, nor the protection seller, actually holds the underlying bonds, how can they be sure (1) that the protection buyer will be able to buy the bonds upon default within the specified time-frame and (2) that there are not other Seller/Buyer relationships out-there concerning the same underlying bonds, by which to total dollar value of insurance would be greater than the total dollar value of the underlying bonds?
ReplyDeleteAs per question #1: I refer you to Deutsche Bank v. Ambac Credit Products No. 04 Civ. 5594 (DLC) (SDNY 6 Jul 2006):
http://www.credit-deriv.com/crecases.htm
In this case, Deutsche Bank was not able to secure the underlying bonds within the contractually stipulated time frame, which, thereby, freed Ambac from any obligation to pay. Not only may the bond holder at the time of default not be willing to sell the bond to the protection buyer, who needs to transfer the bond in turn to the protection seller to receive their insurance payment, but, the time frame may not be sufficient when several CDSs are involved. This is especially the case since there is heterogeneity of CDS participants, each with their own customs regarding the transfer of the underlying bonds: DEALERS, such as banks and brokerage firms, usually engage in a high volume of CDS transactions, and may act as a buyer and seller of protection with respect to the same security. NONDEALERS, however, typically engage in a much lower volume of transactions and do not act both as protection buyer and protection seller with respect to the same security. It seems that the timeframe of CDSs involving dealers is not as strictly enforced as it is for CDSs involving non-dealers. As the judge wrote in his opinion,
"Imposing a definite and tight timeframe for delivery has particular benefits for the seller of protection that do not pertain to the regular bond market. If the bond is one of those that is covered by many CDS transactions, as it appears may have been the case with the Solutia Bonds, then more bonds may be sought than are available in the market. When that happens, the buyer may be unable to obtain the bonds by the delivery date, relieving the seller entirely of its obligation to pay, or if the bonds are timely delivered, the intense bidding for those bonds may create a relatively inflated market price, allowing the credit seller to recover some of its loss by reselling the bonds quickly before the market price collapses."
In the aforementioned case, in addition to buying protection from Ambac, Deutsche Bank was also a protection seller to JPMorgan who held the Solutia Inc., bonds; I think. Now imagine something: If the requirement of physical settlement (handing over the bonds) is limited to 30 days in both the Deutsche Bank-JPMorgan and Deutsche Bank-Ambac contracts, then it is possible for JPMorgan to transfer on the last day, without Deutsche Bank then being able to transfer them quickly enough to Ambac. Maybe they could make the quick transfer. But, if Deutsche Bank and JPMorgan abide by the norms of dealers, then JPMorgan might be able to delay its transfer without forfeiting their insurance from Deutsche, while this would lead Deutsche to forfeit its insurance from Ambac. Also, imagine there is another CDS somewhere between Ambac and the bond holder at the time of default, where, every contract requires transfer of the bonds within various time frames in order for insurance payment to proceed? Let's say JPMorgan not only bought protection from Deutsche Bank, but also sold protection to Goldman Sachs who in fact possessed the underlying bonds. Let's say Goldman waits until the last day in the time frame. You would, then, need Goldman to transfer the bonds to JPMorgan, who would have to transfer the bonds to Deutsche, who would have to transfer them finally to Ambac, all on the same day.
Given this chain, it would seem that the risk of not making the transfer to receive in turn the insurance increases as you take each derivative step away from the underlying holder of the bond. JPMorgan is dependent on Goldman and Deutsche is dependent on both. Moreover, given that Deutsche bought protection on X number of bonds, while selling protection to JPMorgan on 50% of the bonds and to Credit Suisse First Boston on the other 50% of the bonds, Deutsche has the additional risk needing both Credit Suisse and JPMorgan to transfer quickly. So, not only does risk to the final protection buyer increase with the number of CDSs between the final protection seller and the bondholder, but it increases when greater percentages of the bonds are insured at the top, i.e., at the Deutsche level, than are held by any one party directly below, i.e., at the Credit Suisse, JPMorgan level. That is to say, if JPMorgan sold protection to Goldman for the same percentage of bonds as it bought protection for from Deutsche, while there is the risk of Goldman not coming through and delivering the bonds, there is no risk of having to pay insurance without collecting any insurance. This is not the case, however, with Deutsche. If Credit Suisse delivered the bonds on time, but JP Morgan didn't (due to Goldman), Deutsche would collect nothing from Ambac, but would have to pay insurance to Credit Suisse.
As per # 2: How do protection buyers know that there will not be other protection buyers out there, from other CDSs, scrambling to get the same bonds in the event of a default? Now, let’s say Deutsche bank was a simple protection buyer without being a protection seller on 50% of the Solutia Inc., bonds, just as Ambac is a simple protection seller on 50% of the bonds. Let’s say they wrote this contract on January 1st. Since neither parties hold the bonds, when Solutia defaults on say May 1st of the same year, Deutsche will need to track down the bond holders and buy 50% of the bonds and transfer them to Ambac within 1 month if they want to get their insurance (we are assuming that in this contract, as it was for their real contract, that a 100% transfer of the number stipulated in the contract (here 50% of total Solutia bonds) is required before any payment; if Deutsche only transferred 50% of the contract (i.e., 25% of total Solutia bonds outstanding) within a timely period, Ambac would not have to pay anything even on that 50%). But, how can Deutsche be sure that between January 1st and May 1st that Bank of America did not buy insurance on 65% of the bonds of Solutia Inc., from MBIA? When Solutia defaults, Bank of America and Deutsche bank would suddenly find themselves in competition for the defaulted bonds. If BAC were faster, it might get its 65%, which would leave Deutsche without the 50% it needs to get its money. If Deutsche were faster and got its 50%, then BAC would be without its requisite 65% and so not collect any insurance. In essence, given that the underlying securities are not held by either side of a CDS, why isn’t it possible for there to be insurance policies out there written for more than 100% of the bonds in existence?
ReplyDeleteFor example, what prevents there being $1 Billion in CDS insurance for $100 Million in bonds? Such a situation obviously would be very ridiculous, as it would lead to the consequence that the bonds would be worth more if they defaulted, than if they paid themselves off in total. What prevents there being 10 different independent CDSs of insurance on the same $100 Million of bonds? In such a situation, of the 10 protection buyers scrambling to collect 100% of the defaulted bonds, only one protection buyer could collect, since the success of one protection buyer would imply the failure of the other 9 protection buyers (zero-sum game). More likely, however, is that none of the 10 protection buyers will be able to get all 100% within 30 days, which means nobody would collect their insurance. Let’s imagine, however, that protection buyer A (PBA) is able get all 100% and get his insurance from protection seller A (PSA) within 15 days of the default. PSA could then turn around and sell the bonds to PBB who would then collect from PSB, who could do the same…until the 30 days dried up. PSA, then, would more or less break even. While the bonds would be worthless to PSA, they are worth a lot to PBB, PBC, PBD, PBE…Only one PS would have to pay up, i.e., which PS is transferred the bonds without sufficient time to sell them in turn to an interested PB, while the others will have gotten away with collecting a premium on something they would never actually have to insure.
Not only is it the case that only one insurer will actually have to pay, but insurers could even make money in the right circumstances when total insurance > underlying security. Let’s say there are 3 CDSs worth $101 Million on $100 Million in bonds. Let’s further say, that CDS #1 is for 40% of the 100 $1 Million bonds (40 bonds worth $40 Million), CDS #2 is for 60% (60 bonds worth $60 Million), and CDS #3 is for 1% (1 bond worth $1 Million). If upon default PBC of CDS #3 is quickest and transfers his %1 to the protection seller, he would collect his $1 Million. Let’s then say that the PBA of CDS #1 is at 39% of the bonds outstanding and PBB of CDS#2 is at 59% of the bonds. It is likely that we will see a bidding war between PBA and PBB for that 1 defaulted bond (1% of the outstanding defaulted bonds). Since the protection buyers need that one percent in order to collect on their insurance, and only one of the two protection buyers will be able to collect, we would probably see that the 1 defaulted bond sell to PBB for millions of dollars. PBA is faced with the prospect of paying anything below $40 million, while PBB will likely offer $40 million to the possessor of that 1 defaulted bond. The net outcome would be the following:
PBA: $0 – X (protection premiums paid to date)
PBB: $20 million – Y (protection premiums paid to date)
PBC: $1 million – Z (protection premiums paid to date)
PSA: $X
PSB: (-)$60 million – Y
PSC: $40 million – $1 million + Z or $40 million – PBC’s net
This primer from the same web site you cite is very informative:
ReplyDeletehttp://www.credit-deriv.com/introduction%20to%20credit%20derivatives%20article%20by%20Vinod%20Kothari.pdf
Your questions are good, and settlement of defaults are dependent on specifications of the contracts. I need to research more. I'll add to your question and ask how (or if) a short CDS seller is hit as credit spreads change. ie imagine being short CDS at 100bp on countrywide, then watching that spread blow up to 5000bp. How is collateral affected on the short seller? Does the short seller potentially owe anything to cover the difference (when the CDS buyer can go on the open market and sell that swap for 5000 bp, realizing almost a 5M profit without default occurring).
So I imagine default isn't really necessary to cause unwinding and defaults amongst counterparties.
Take a look here.
ReplyDeletehttp://accruedint.blogspot.com/2007/04/how-does-credit-default-swap-cds-work.html
The accruedint blog writer trades these professional -- he should be able to resolve the questions.
Thank you for your thoughts. Thank you also for the interesting blog posts.
ReplyDelete-David
Be alerted to the Hottest Stock Picks!
ReplyDeleteJoin Emerging Growth Alert Newsletter
By joining the team at Emerging Growth Alert you will be in position to receive stock alerts profiling stocks about to move or already in motion. Our alerts are sent in time for you to research, investigate and make a decision about whether this opportunity is right for you.You will not be bombarded with junk mail. There is ABSOLUTELY NO OBLIGATION, and the service is entirely FREE.
http://stocks.freetzi.com
http://www.zendurl.com/cashing