Debunking Some Myths About The "Greek CDS Contagion" Threat

Tyler Durden's picture

Now that the Greek bailout is topic front and center for the second year
in a row, it means that it is time for the mainstream media to once
again prove to the world that in the past year it has learned precisely didley squat about
how the more complicated securities used in capital markets operate.
Such as CDS. Just like in May 2010, the prevalent trope among the clickbaiters
is that CDS written against Greece will destroy the world, in
superficial attempts to bring about panic induced by the faulty
conventional wisdom that CDS was the cause for the implosion of AIG.
Well, wrong.

First, for those who actually care, CDS is nothing more than a low margin synthetic method to express a bearish position on an underlying credit (such as US government bonds), in the process facilitating market clearance and price discovery. Period. And as long as there is an idiot who is willing to take the other side of a trade which expresses nothing more than appreciation of risk, CDS can be written into existence (which is how Paulson made his Abacus money) or traded on the secondary market (the fact that said idiot is still trading CDS is thanks to the US government policy of making financial risk and failure a thing of the past, but that is the topic for another post). Another key CDS feature that is constantly ignored by nearly everyone is variation margin, or the daily posting of cash collateral based on intraday moves of CDS. This was extensively discussed in our post on the threats posed by the $600 trillion OTC market. A third key fact is that all the data on Greek CDS is publicly available yet nobody bothers to actually check it: after all why let facts stand in the way of a good story. So here are the facts: there is $5 billion in net notional daily margined risk exposure on the Hellenic Republic (also known as Greece). This is a 28% reduction in net notional risk over the past year, the largest drop in risk exposure of any of the top 30 credits tracked by DTCC. The total notional of Greek CDS outstanding is the 21st biggest in the world, behind such names as Italy (at the top), France, Spain, Goldman, JPMorgan, Berkshire and Wells Fargo. Lastly, 5 Year Greek CDS (at last check about 45 points upfront) is currently trading wide of cash bonds, which upon an event of default will likely rise on short covering and higher recovery expectations, which means that sellers of protection will actually receive a cash payment from this point until an actual EOD occurs when the basis between cash and CDS collapses.

The truth is that there is certainly risk from a Greek contagion effect, and far more than a risk -  it is a certainty, and the catalyst will be none other than the world's largest and most undercapitalized hedge fund - the ECB, which holds tens of billions of Greek debt as cash collateral which would have to remarked 50% lower in the process making the key European liquidity backstopper insolvent (in practice if not in theory: after all the ECB will just print more €), forcing a self-fulfilling liquidity run prophecy. But the contagion risk is not in Greek CDS, where risk sellers have not only contracted their exposure by the largest amount of the most active contracts, but where daily steady cash outflows to satisfy variation margin mean that banks have "overreserved" for an event of default, and may in fact be cash inflow beneficiaries.

As the DTCC-sourced chart below shows Greek CDS has seen the largest drop in net notional outstanding among the 30 largest positions. Furthermore, as can be seen it is well down on the table. The names in the top 5 positions should be a far greater source of risk, as that is where the perception of risk, and the market "reality" have not yet converged.

Most important, at least in our opinion, in the table above is not the 21st ranked name, but the 26th one, which has seen the change in its net notional CDS outstanding increase by 136%, or the most of the top 30, over the past year.

It should be rather obvious which "entity" is ranked #26... soon to be #1.

And for those who missed it, such as the entire mainstream media, here is your guide to the uber secret intricacies (yet completely public to those who search) that actually happen in CDS daily margin flows.

CCPs typically rely on four different controls to manage their
counterparty risk: participation constraints, initial margins, variation
margins and non-margin collateral.

A first set of measures are participation constraints, which aim to prevent CCPs from dealing with counterparties that have unacceptably high probabilities of default.

The second line of defense is initial margins in
the form of cash or highly liquid securities collected from
counterparties. These are designed to cover most possible losses in case
of default of a counterparty. More specifically, initial margins are
meant to cover possible losses between the time of default of a
counterparty,8 at which point the CCP would inherit its positions, and
the closeout of these positions through selling or hedging. On this
basis, our hypothetical CCP sets initial margins to cover 99.5% of
expected possible losses that could arise over a five-day period. CCPs
usually accept cash or high-quality liquid securities, such as
government bonds, as initial margin collateral.

As the market values of counterparties’ portfolios fluctuate, CCPs collect variation margins,
the third set of controls. Counterparties whose  portfolios have lost
market value must pay variation margins equal to the size of the loss
since the previous valuation. The CCP typically passes on the variation
margins it collects to the participants whose portfolios gained in
value. Thus, the exchange of variation margins compensates participants
for realised profits/losses associated with past price movements while
initial margins protect the CCP against potential future exposures.
Variation margins, typically paid in cash, are usually collected on a
daily basis, although more than one intraday payment may be requested if
prices are unusually volatile.

Full link for the above.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
scatterbrains's picture

wow fukin scary. I wonder where the U.S. ranks on a "greatest year over year increase" ranking ?

Pure Evil's picture

Beyond fekin' scary.

And, we're supposed to believe that ordinary humans are more than capable of managing a system this complex. And of course all tangibles and random events have been taken into account, much less, someone hacking into the system and causing chaos.

No black swan events here. The smartest men in the room, no scratch that, Top Men, are large and in charge.

You know, everything is hunky dory inside the hydrogen bomb, right up to the point where fission occurs. After that its just a cascading event that can't be stopped, well, not by mere humans anyway, until the nuclear fuel is spent.

Transformer's picture

First word of explanation and I am lost.  What is a CCP?

AbandonShip's picture

CCP = Central Clearing Party

Like the Intercontinental Exchange (ICE) in the U.S. or London Clearing House.  Basically these CCPs bring together the various CDS dealers (banks) and their clients (hedge funds) to try and "clear" the trades so that we can reduce the systemic risk (the CCP would have all the risk theoretically, I know I'm really watering this down so don't bother objecting.)  Think of going from a messy spiders-web design to a hub-and-spoke model like in a bicycle wheel. The second design is "better" because all the "counterparty" risk goes to the middle (the CCP).

ZeroPower's picture

Nice, finally an up-to-date net notional number for GRE CDS.

Tyler, an MD informed me apparently no single CDS has to be reported to the DTCC - hence the wanting of an exchange by stakeholders to finally bring in true verified numbers. Any idea if the DTCC only sources and requires US banks to provide this information? Cant imagine all desks worldwide would want to or care enough to report to the DTCC.

Tyler Durden's picture

Wrong. Since 2009 every CDS transaction has to be reported to the DTCC. Does this mean the publicly reported data is ironclad? Or that has to be means is? Of course not.

unununium's picture

One naturally wonders how much exposure remains from contracts written pre-2009.

Tyler Durden's picture

Very little due to 3 years of rolling

oogs66's picture

yes, the rolling has played a big role, and even the implementation of SNAC had a big part as many existing trades were converted to SNAC trades at the time to ensure liquidity

Arius's picture

interesting stuff...

gwar5's picture

So, if/when Greek defaults or restructures, Greek bonds held by the ECB and all the other central banks will be slashed, shrinking their books, creating a liquidity crisis because they will all have to scramble for cash to replace the bond losses to maintain reserve requirements. I think I got it.

No doubt the other PIIG bonds will be selling at a discount at that point too, making things worse. Dominos.

Well it's certainly good to know that amidst a massive liquidity crisis big enough to threaten the global monetary system, the puny derivative market is safe and sound and fears of a secondary chain reaction are unjustified.


Fascist Dictator's picture

"the puny derivative market is safe and sound and fears of a secondary chain reaction are unjustified."


thanks for the laugh. a quadrillion or two ain't what it used to be.

Daneric's picture

It is of my opinion the banksters are fighting the Greeks so hard about haircuts is because they cannot afford to set a precedent for others to follow.  If the Greeks get off the hook and default on part of the debt, it is clear to me every other country that has debt problems will obviously be looking for the same - defaulting or forcing haircuts on chunks of debt. 

Greece may not blow up the system in and of itself, however if every other nation follows Greece's lead - and why wouldn't they? - well, yeah you got a problem.

EDIT: And to add that is probably why they are trying so hard to figure out how to go about a restructuring without triggering a credit event - not because Greece will blow it up - but because precedent will be set and any other nations that follow that have a higher ranking on this chart WILL blow it up. 

So it all comes down to precedent which is why they are squirming so hard to come up with a scheme that will NOT set precedent necessarily. It also comes down to: If we let them off the hook - or even to be PERCEIVED to be letting them off the hook, we open the Pandora's box to some of these other countries that have even bigger debt problems and it snowballs.

This is all a very pyschological game being played here for world audiences by the banksters.  In the end, it'll fail, because escalating riots will do that. 

treasurefish's picture

Precedent = Sovereign Iceland

bonderøven-farm ass's picture

A drab of state, a cloth-o'-silver slut,

     To have her train bourne up, and her soul trail in the dirt

Daneric's picture

Yes they like to keep that out of the news. 

dcb's picture

I believe this view 100%. in fact if one thinks about future access to capitial markets, in a word flush with capitial, I'll buy the new debt of any country that repudiates the old debt. I have troied to get my handds on icelandic bonds without success (any help here). hedge funds, the brics, etc will all step intot he market. I make aceptions for banana republics like ecuador. the power  that be make it worse by extending loans, instead the total amount lost increases almost certain to become a systemic event. then we can just bail the banks out more. it is so disgusting, that I can't figure out why people are taking pot shots at these people. I am astounded at how corrupt the system is and thet at least in the us we aren't rioting saying these people need to go.


I think there is only one outcome in the end, and that is civl war. In fact I think the administration knows this and has been laying the ground work with increased domestic spy apparatus, etc. if you can't see the set up for the crisis, where we force the populace to be hostage to the bankers you are in fact blind.

sgt_doom's picture

"It is of my opinion the banksters are fighting the Greeks so hard about haircuts is because they cannot afford to set a precedent for others to follow."

Naaah.....they just want to PRIVATIZE EVERYTHING.

First they establish all those "public-private partnerships"....

Phase II:  Next, they privatize everything.

Lord Welligton's picture

"And as long as there is an idiot who is willing to take the other side of a trade which expresses nothing more than appreciation of risk, CDS can be written into existence"

Maybe I don't get it.

But idiots are not allowed.

Participation constraints?

Lord Welligton's picture

"First, for those who actually care, CDS is nothing more than a low margin synthetic method to express a bearish position on an underlying credit (such as US government bonds), in the process facilitating market clearance and price discovery."

Where is the price discovery on banks, including the ECB, that hold Greek debt?

jm's picture

The point is that CDS allow an investor to hedge his exposure.  Without it you have very inadequate hedge instruments, so investors will sell and credit markets lock up.  Financing for all enterprises and govts will be reduced by a lot, as spreads will rocket.

Reese Bobby's picture

Uh, uh.  So an investor who owns Greek bonds can buy protection on a like notional amount to hedge exposure.  And there is a new party that sold the protection long the original notional amount.  What does that accomplish besides levering up the original mess?

XPolemic's picture

The theory is that it is better for a billion people to lose* 1$ than it is for 1 person to lose a billion dollars.

The problem with that theory is that a billion people disagree with it.


* Okay okay: risk a dollar.


jm's picture

That's not the point at all.

IF spread jump and stay there, then a lot of business wil go under becasue they can't roll their debt.

Goverments won't be able to roll their debt.  There will be no umemployment checks.  There will be no food stamps.  There will be lot less employment.

Your house will lose even more value, because nobody will take the credit risk of a borrower.

Go ahead, be bitter.  I am.  But don't be stupid.  Refusing to see the big picture is like cutting your dick off to spite an unfaithful wife.


jm's picture


You know nothing about me, you fucking idiot.  And all you ever do is say stupid stuff that shows you know nothing about the issue.  Just shut up and fuck off.


ZeroPower's picture

Ignore the trolls man.

Lord Welligton's picture


Well bugger me.

I didn't realise that CDS were essential for the efficient allocation of capital.

But then the East India Company never used them.

Who's calling the margin and who has the cash to meet the margin.

jm's picture

You asked a question.  I answered.  Doesn't seem to be much use in trying to explain anything to closed minds.

Lord Welligton's picture

Where is the price discovery on banks, including the ECB, that hold Greek debt?

That was my question.

Thanks for attempting to answer.

oogs66's picture

price discovery on what?   The CDS on the banks is very liquid.  BAC  168/172  6 tighter on Friday, JPM 81/84  4 tighter,  and GS  147/151.  

Those are 5 year prices for the banks.  They equate to values under SNAC.  Very liquid, very transparent and the closing prices should be publicly available on Markit website.

Curves are less liquid.  1 year CDS is 42/62 for BAC.  Although it is 20 bps wide, since it is only 1 year, that equates to about 0.2% bid/offer, similar to what a 5 bp mkt would be on 5 year CDS.


Or are you asking about price discovery on Greek CDS?  There is a market for that as well, though just like the Greek bonds, it is less liquid than it was.


5 year Greece finished Friday 1875/1975.  On a "price" basis, you could sell CDS at 35 points up front + 500 bps running, or you could buy CDS at 36.3 points up front plus 500 bps running.  That is a bid/offer of 1.3% of notional  (as the spread widens, the convexity means each bp of bid/offer is less than for a tighter spread name, so although 100 bps sounds like a lot, it is only 1.3%).  My understanding is that the bid/offer on Greek bonds in the 5 year tenor is at least 2 pts wide, so the CDS is more transparent and more liquid.


If you are saying that we don't know the exact outright exposure of each bank to Greece, that is true, but we don't know their exposure on the bond side either for most banks unless they made a specific disclosure. 

If your main concern is the all the counterparty risk the banks have, no we do not know it, but so far banks have actually demonstrated a pretty good ability to manage that.  AIG seems to have been an exception, but that was not allowed to play out fully. 

Bank financial reporting leaves a lot to be desired in my opinion, but that applies to things as simple as their loan loss reserves - can you tell which loans they reserved it against?  what is loan loss reserves and what is being squirrelled away for fraudclosure settlement? 


Reese Bobby's picture

I know who bails out the TBTF banks.  I know the potential regulation of derivatives was killed so that the TBTF banks could use hyper-leverage to print money until it all came crashing down.


Why should it take years to move CDS to an exchange traded platform where the taxpayer is not on the hook?  The answer is that the Global Bank Cartel that runs our Governments and controls our money supply is in no hurry.


You try to make CDS sound like Lloyd Blankfein's "God's work."  It is really just leverage.

ZeroPower's picture

Poor argument. The taxpayer is not "on the hook" for CDS. The only logical argument is against bailing out TBTFs. Something which unfortunately the US gov was quite keen to do.

sgt_doom's picture

"The point is that CDS allow an investor to hedge his exposure."

Negative, the point is that when an unlimited number of CDSes can be written against a particular category of bonds or a specific financial entity, and an unlimited number of commodity futures contract can be created as well, coupled with an unlimited number of investors allowed per hedge fund (in the USA, the Investment Company Act (amended) of 1996), you have unlimited potential for ultra-leveraged speculation.

Next question?????

Lord Welligton's picture

So. All that is being said here is nothing more or less than that CDS are traded on margin.

And margin may and will be called.

So what.

Does this make Greece more solvent?



The USA?


All margin products are Ponzi products.

scatterbrains's picture

my take from the article is that even though US 10 yr trades at 3 the CDS on the same is sling shotting.

oogs66's picture

so there should be no options, no futures, no shorts, and heck no ability to borrow money to buy stocks?  or homes?

All of those products have features similar to CDS, particularly options.

Reese Bobby's picture

"...faulty conventional wisdom that CDS was the cause for the implosion of AIG."

I'd like to hear the correct non-conventional wisdom on AIG.  Because you're right, I'm still walking around thinking AIG sold protection on sub-prime by the boat-load because they didn't have to mark to market or post collateral given their AAA rating...


And to me, "participation constraints, initial margins, variation margins and non-margin collateral," sound like 1980's portfolio insurance: they work until you get a cascade-like sell-off.

Tyler Durden's picture

AIG was going long the synthetic underlying. Since they were expressing unlimited demand, Goldman or whoever could simply continue issuing increasingly lower yielding cash paper which AIG would buy to collect the same yield that they collected by receiving the protection on sold CDS. The point is the instrument is not the issue, but the underlying decision making stupidity.

And furthermore in a normal world, if the government really wanted to reallocate profits in a somewhat fair way than merely defrauding taxpayers, it would take the profits made from buyers of AIG protection and use it to fund AIG capital deficiency. Of course, that never happened as it would mean Goldman, which in this case was spot in in predicting the housing collapse (ethics aside), would see a massive loss. This all goes to the point of endless moral hazard which as noted above, is the topic of a different post entirely, althugh is currentl being manifested in the cash market: instead of one organization selling protection you are getting every financial company buying up record amounts of debt issued by corporates at record low rates, many of which variable. How is that any different than what happened with AIG?

The point is the demand for risk is there, and it will always be met with one instrument or another.

If one wishes to blame someone for the AIG collapse, blame the Fed.


Reese Bobby's picture

I understand your, "guns don't kill people" viewpoint.


The sad thing is we can question whether the crew who blew-up AIG were even stupid.  Who went to jail?  Who had to disgorge profits?  Cassano looks to be doing quite well.  Sad...


Which speaks to your point on moral hazard.  The only lesson to market participants has been: if there is a risk you may fail, make sure you fail really big.

Thanks, (man I get so nice when I type to a Tyler).

falak pema's picture

I find you "gayness" practically contagious when you speak to TD. No trolling and vociferous  name calling and lolling even want to know "so who went to jail?"...Now that is called "having a social conscience" you are no longer an advocate of Huntsville summary justice in the lone star state?

Wow, its raining unicorns...but there, I am dreaming ...of a white xmas of patriot's act repeal!

XPolemic's picture

I'm not sure why in a normal world you would forceably take money from Goldman and give to AIG.

Yes, AIG mispriced their risk premiums, but Goldman also failed to manage the issuer risk.

I say, let them all go down. How else are would you reset 20 years of capital misallocation and malinvestment?


narnia's picture

Tyler's "justice" + a whole lot more pain would have been served had AIG been rightfully forced into bankruptcy court.  This, of course, assumes AIG really "lost" all these bets definitively.  I'm still perplexed how they did if GSEs covered 100% of par for their "implied" guarantees on the underlying debt instruments.

XPolemic's picture

Correct me if I am wrong, but were the instruments underlying the CDS not MBS or CDO?

Can you tell me which GSEs guaranteed 100% of par on those instruments?

It's been a long time since I looked at this, but my (probably incorrect) understanding of what happened was this:

Goldman (and Bear and Lehman and everyone else) tranched mortgage notes together and issued CDO/MBS.

Goldman foresaw a housing correction, and so purchased CDS from AIG on primarily CDO/MBS.

AIG (rather stupidly) sold these CDS at increasing discounts because they thought the probability of default was 0, and hence it was free money all round.

Interest rates climbed roughly 425bips, bottom fell out of the housing market, subprime default rate exploded, causing credit events in MBS/CDO and subsequent calls on CDS written on said MBS/CDO which AIG couldn't cover.

Could you elaborate on which GSEs were backstopping MBS/CDO paper?


narnia's picture


This is the Fed's picture of real estate financing in the US:


If I am reading this properly, the total of all single family mortgages outstanding in the US in 2010Q1 was $10.7 trillion, of this:

$2.7 trillion was held directly by commercial/savings banks (which should not have been in a syndicated pool)

$0.9 trillion was held by non-financial individual / hard money lenders (which should not been in a syndicated pool)

$4.8 trillion was in MBS directly held by GSEs 

$1 trillion was in MBS still held outside the GSEs but guaranteed by the GSEs 

$1.3 trillion was in MBS held privately & not guaranteed by GSEs (pretty close to the amount purchased at near 100% par by the Fed, probably not by coincidence)

I'm just not seeing much MBS out there that would have been part of a syndicated pool, upon which CDS was issued, that was not guaranteed by the government.  My two cents on all of this from the very beginning was: the rating agencies gave these instruments AAA and AIG priced the CDS without even looking at mortgagee risk because of the government credit enhancement.  Given how it has all played out, you have to wonder whether the mortgagees not paying even qualified as a credit event.

XPolemic's picture

Given how it has all played out, you have to wonder whether the mortgagees not paying even qualified as a credit event.

Superb. I'm going to put that in my keep file.

What I didn't get from the link is what form the guarantee took. Was it similar to deposit insurance? Sort of like goverment mortgage insurance except that no one paid any premiums?


narnia's picture

FNMA, FHLC & GNMA provide fee based sureties, backed by the full faith & credit of the US Treasury, for MBS based on criteria defined by politicians.  These entities racked up pretty signficant fees using the government's credit to amass a brutally underwater multi-trillion $ MBS portfolio. (FDIC is also not free, it's substantially paid for by member banks).

Perhaps Goldman & others saw an opening to make these fees using a CDO structure. Maybe AIG looked at FNMA's past exposure in this game as low, so priced the CDS to allow the structure to work.  Perhaps the ratings agencies also looked at FNMA's past exposure to rate it.  Maybe this model was used for the lion's share of the $1.3 trillion non-GSE guaranteed MBS issued.  Maybe that's why the Fed bought all this stuff (if they, in fact, did) because they were exposed to it anyway with AIG.  

XPolemic's picture

So let me see if I have this right. The US Treasury effectively supplied a form of fee based mortgage insurance, which essentially guaranteed the underlying mortgage note, regardless of the LTV of the mortgage vis a vis the actual property.

Because these mortgages were essentially insured by the UST the following happened.

1) They gave anyone who walked in the door essentially any size mortgage they asked for (and sometimes a bigger mortgage than they asked for).

2) They tranched these notes up, and despite the fact that the borrowers were not creditworthy, the mortgage notes were, essentially because of the sureties provided by FNMA, FHLC & GNMA.

3) SURPRISINGLY, people with no job, no income and no assets experienced severe repayment stress when Greenspan raised interest rates ~425bps to halt the descent of the USD.

4) The subprime default rate exploded and the 30-90 day cashflow dried up.

5) Yada yada yada. US government steps in and makes good on it's sureties, everyone happy. Ok, not everyone.

Sounds to me like the collapse of socialism, not capitalism.


narnia's picture

The easing of credit standards (and, to some degree, the gamesmanship that went with meeting artificial government standards), the expansion of other homeownership subsidies (direct & indirect community development programs, tax incentives, regulatory control) & development oriented tax regimes caused a major earthquake in the demand & supply sides of real estate pricing.  The flooding of cheap liquidity backed by the US taxpayer in this asset class was the tsunami.

When the wave subsided, the US has a supply of ownership quality real estate that maybe 50% of the sustainable portion of our centrally controlled economy can support at historic interest rates.  

Valuations are inversely related to the interest rate.  When the interest rate hikes went into effect, it crumbled refinancing activity (which fueled what keynesians like to call "aggregate demand"), which caused an economic slowdown, which concurrently led to liqudity drying up & defaults, which led to liquidity injections, which led to price confusion, which led to panic, which led to bailouts, which led to more liquidity, which led to prologued price discovery...  which leads us to where we are now, an entire asset class in need of bankruptcy court.

The simple version of cause and effect is taxation (the state spending money, directly or indirectly, through subsidy, regulation, or printing money) leads to inflation which ultimately deflates to the price it would have been without the taxation.