Give me a large enough lever and a place to stand, said Archimedes, and I can move the world. Well, our lever may finally be big enough. Apparently, the International Swaps and Derivatives Association is announcing that the market in credit default swaps is $62.2 trillion, just about double its size last year.
That's a big number: $62,200,000,000,000. But the mind can't grasp all those zeros--it just sort of slides past them, like a car skidding on too much ice. But here's something that might make it a little easier: $62.2 trillion is just about equal to the entire output of the world last year. Every single thing made or grown, bought or sold--at least as best we can count.
Kind of makes you wonder what a run on the market might look like, hmmm?
But what fraction of this represents zero-sum debts that banks owe to themselves (indirectly)? I can imagine different divisions of individual banks owing and lending money to to each other in different deals. Or three-way deals such as bank A lending to B which lends to C, which in turns lends to A, etc.
Kind of makes you wonder what a run on the market might look like, hmmm?
I don't know what the run will look like, but the legal fees from untangling the mess will run approximately $62,200,000,000,000.
derivatives involve double and triple counting the same exposures
In fact,derivatives can be written on more than the face value of ll the actual bonds that exist (ever heard of the ABX index(s)?). If I buy a derivative for $1 that says Megan pays me $5 dollars if the value of a specified $1,000,000 General Motors bond declines to say, $950,000, then the ISDA statistics will say we have a $1 million derivative, even though in reality it is a $5 "bet"
it is quite easy to demagog these numbers, which is why people do so
Tell me, what bank or hedge fund has failed due to over-indulging in credit default swaps?
It's also important to remember that the funding costs of these are relatively low for banks. There's not as much balance sheet pressure to get rid of them at once. With the LBO junk and MBS debt it gets expensive to keep it on long.
Jozef -
True on both counts (the inflated nature of notional value and the lack of failures tied specifically to CDS and other large swap transactions).
However, I'm pretty sure that the notional value of Bear's counterparty obligations on this basis, which are rumoured to have measured in the trillions, played a big part in the Fed's rescue of the firm. And if you follow the link, Yves indicates some of the problems of unwinding these things. You can either:
1) create an offsetting transaction with the same or other firm
2) ask the counterparty if you can assign your interest in the contract to another counterparty. You need their permission because firms manage these things.
So the problem is - how do you unwind these things in a world where financial institutions are desparately trying to wind down their counterparty exposures? Who, for instance, would have accepted on March 12 an assignment of a large CDS contract written with Bear Stearns?
Notional value isn't interesting most of the time, but it is when your in extreme conditions..as we are.
I recomment you read Yves post and the linked analysis. (I believe Megan picked it up off my Google share, but I could be wrong)
Nice metaphor.
The face values of these transactions are a large multiple of realistic exposure of the two parties to the transaction. The school my daughter attends has an interest rate swap under which the school pays a fixed rate and the counterparty pays a rate that floats with the floating rate on the school's bonds. This protects the school in the event interest rates spike. The amount of the bonds (and the "size" of the swap transaction) is $7,000,000. The difference in the fixed rate vs. the floating rate has never exceeded $100,000 in any of the last seven years. If that same ratio (1:70) applied to your $62 trillion, the actual exposure people are facing out there is "only" in the billions.
johnF,
The face value is cited because that is "firm".
The key problem with the contract you described is that it is open ended --- the school gets a fixed rate, but the counter-party has an open ended liability regardless of what interest rates spike up to.
While the historical on the rate may not have varied by more than 100k in the last 7 years, what if history is not a guide?
What if rates spike to 15% above current rates?
At some level, the counter-party will crack and fold, and with it, the contract blows up.
That is the systemic problem --- no one can plausibly claim they really know what is at stake.
Your guide of 1:70 cannot be plausibly applied without being able to mathematically lay out every individual contract, and for each contract, every independent variable (and both expected and maximum / minimum ranges) and then add the mess back up under different assumptions as to what is the "possible" range.
Look back at Long term credit --- they had a great formula until history proved to be a poor guide.
When a market is at a disjuncture, those sharp changes are never modeled properly --- or there would be no huge fortunes made by people who bet right against them.
Nobody is clicking through! These are *Credit Default Swaps*. If you wrote one on Linens & Things, you are now insuring a bond at 100 (face or notional) that is trading at 40.
CDS are many times the value of the outstanding debt covered. But I'll give you an example from another market that will make you shake your head.
The oil platform Piper Alpha was insured (and reinsured many times) across the Lloyd's market. It was destroyed in a fire, giving rise to an insurance claim of about $250 million, if I remember correctly.
Oddly enough, given the size of the original loss, the losses related to Piper Alpha in Lloyd's, thanks to the so-called were over a billion dollars. Sometimes the very act of dragging something that has been written and re-written several times through the market can take out small players - Snakes and Ladders.
One way this can cause more damage is when people start litigating the terms of CDS. There is a lot of room to argue the reference instruments and substitutable instruments in the terms.
johnF is describing an interest rate swap
MindlesDreck and johnF make reasonable points, but I will say that
no, the liabilities are not open ended - the reason tehre are so many swaps is because people cap their exposure, or buy/write a swap going the other direction
Mondles' point about the collateral risk is well taken (in the sense of posting cash collateral, not neo-military use of the term) Even if you haven't lost money on the thing, you might have to cough up more margin, etc. Still - from my knowledge, which is not slight - not a significant contributor to Bear's demise.
johnF is describing an interest rate swap
MindlesDreck and johnF make reasonable points, but I will say that
no, the liabilities are not open ended - the reason tehre are so many swaps is because people cap their exposure, or buy/write a swap going the other direction
Mondles' point about the collateral risk is well taken (in the sense of posting cash collateral, not neo-military use of the term) Even if you haven't lost money on the thing, you might have to cough up more margin, etc. Still - from my knowledge, which is not slight - not a significant contributor to Bear's demise.
Arrghh, considerable knowledge apparently doesn't prevent a misreading, or failure to read.
I didn't say CDS contributed to BSC's demise, I said they were a factor in the need to rescue BSC. The Fed was clearly concerned about the effects on the financial system of unwinding or rendering worthless BSC's trillions of CDS and other derivatives.
Anybody who's been on the zillions of street conference calls that Tim Geithner et al have given to the street since Mid-March knows this.