I love Matt Levine's articles. He's good at simply explaining the arcana of Wall Street.
But when I read this article, I can only say one thing: Why?
Why is Wall Street spending all this time and energy in creating all these different securities? I suspect that any excess "profits" or "returns" to be had from them accrue 90% to the bankers and 10% to the muppets funding them. So why do the muppets keep playing this game?
Muppets are usually pension funds, where the people doing the investing want to do it as easy as possible and with high rewards so they get the bonus for doing nothing.
In this case, the sucker would actually have been DB's investors if the worst case ever did happen. (They would have gone under like many other banks - the bailout saved them.)
In this case the investors they sold protection to (hedge funds) got better prices than they ordinarily would have, because of the risk hidden in the conduit.
To expand on the other replies, and based on experience:
1) The way you get to sell products to investors is by promising some sort of alpha: "My fund is less risky, my derivative pays more for the same risk, etc."
2) But magic does not happen in the financial markets (and when it does is through the old stories.. diversification, transferring the risk to those of broader shoulders, etc.)
3) So what you do is HIDE THE RISK. There are many ways to do that, and they usually involve COMPLEXITY. Complexity and opacity are your friends, as they confuse investors who don't realize there is a hidden risk somewhere there, or a scenario you haven't told them (but put on page 350) where if A B and C happen, then he will lose a lot of money.
I'm not joking here.. sometimes complexity is good as it allows you to e.g. reduce mismatches in your risks (your income is in USD and expenses in EUR, your debts are long term and assets short term, etc.). However, the ultimate purpose of most complex products is that they allow you to sell stuff that appears better than it really is. And IT WORKS. Everytime. Against millionaires and against folks buying silly combos of insurances plus deposits (forgot their names).
I've been rereading my copy of Taleb's Antifragile. He keeps hammering similar points over and over: complexity, risk, opacity, susceptibility to Black Swans, etc.
I think I'd be depressed if I worked on Wall Street and the better part of my compensation resulted from peddling all these confusing products.
Still, as you point out, sometimes complexity is good. I do understand that.
And I would almost be ok with this if the complexity-generators understood all their complexity and were using it as a great trick to steal money from idiots. What really scares me is that a) even the complexity-generators can easily miss potential issues, and b) when you throw all this complexity together, nobody has any clue what's going on.
And then we wrap it up all with a bow by allowing finance people to alternately claim that a) they are so smart they deserve to personally make millions to billions, and b) they couldn't possibly be held accountable for systemic fuckups, because nobody could possibly have known. I'll accept one or the other, but the combination is incredibly dangerous.
>Why is Wall Street spending all this time and energy in creating all these different securities?
The primary purpose of these trades is to offload super senior risk from DB's balance sheet.
Super senior risk is the risk that DB retained when it (very profitably) sold various other tranches on a group of assets (typically CDS on a bunch of corporate and sovereign credits).
This type of risk is very difficult to sell as the return is very low and most entities they might buy unfunded protection from are likely to not be able to pay in the event that the swap pays out.
Offloading that risk through a swap to a conduit like this allows DB to release (expensive) capital/reserves that would otherwise need to hold.
It's a regulatory arbitrage. Everyone knows that the risk on those super-senior tranches is basically zero (it would take something like 60% of the world's investment-grade companies going bankrupt for them to actually pay off), but the regulations say that Deutsche would have to hold assets to cover it. So they find some investors who declare that they'll cover it instead, in exchange for a few percentage points, and everybody wins: the investors get (a small amount of) free money, Deutsche can hold fewer assets and use its balance sheet for more profitable things. Even the regulators probably don't mind so much - the regulations are overly simplistic (dare I say populist?) in circumstances like this, and everyone knows that.
I think anyone who has programmed for a medium- or large-sized organization will be familiar with the experience of getting a stupid technical directive from above (e.g. "this part must be XML") and treating it as an obstruction to be worked around (while technically abiding by the letter of it). It's not an undefined state, it's just a case where the rule is unhelpful.
The "muppets" are multi-billionaire asset holders. That money has to go somewhere and they want a return on it, or at least the promise of a return.
Quite a lot of financial strangeness can be answered with "where else are you going to put your money?" Western treasury bonds are the safe option, but they're close to negative yields. Spending $1.01 at auction to buy a bond which nominally costs 95c and pays back $1 in five years doesn't look great. So there is a huge demand for "riskless" positive-yielding investments.
That's why London (and elsewhere) property continues to shoot upwards in value way beyond the ability of the inhabitants to afford it. That's why Switzerland has negative interest rates. And so on.
Every time I think I know a little bit about economics, I read articles like this and realise I'm playing with wooden blocks while the big boys are playing with space shuttles. ...except they made the space shuttle out of wooden blocks, and they're just pretending really hard that it's going to work.
If you have a small group of brilliant computer scientists, they can create world changing software. How many folks were required to get Snapchat or Youtube off the ground?
What happens when this level of genius in small groups is cut loose on financial markets where "Make money is the only goal"? You get these space shuttles. They are so complex because that's all these people ever think about.
The silicon valley is hardly a counter example of "make money is the only goal". In fact for those who question the compensation of bank executives, the compensation of silicon valley executives should raise a few eyebrows.
> If you have a small group of brilliant computer scientists, they can create world changing software. How many folks were required to get Snapchat or Youtube off the ground?
Snapchat is world changing and the work of brilliant computer scientists? Did I miss something? I can only hope that your post was intended to be sarcasm.
I think the meaning is "It only takes a small high powered team to 'change the world'/add tremendous value if they have skills and a good idea." Snapchat is one example of many.
Youtube owes just as much of its success to an army of lawyers achieving two things: filling up with copyright content without getting sued and/or jailed, and banging codec licensor and standards body heads together until video worked everywhere without wildly unreasonable cost or transcoding for every device.
Technological advancements are made. UDP Broadcast was completely broken in the 90s in Linux and was fixed by people in the finance industry. High speed network cards are constantly improving thanks to those same people.
I'm playing "with the big boys" i.e. 500 million market cap enterprise financial. I'm not sure why I do this other than it's easy to disappear into the noise occasionally and put my feet up. In fact if we disappeared tomorrow there would be a chorus of "meh".
The space shuttle we have is, to paraphrase [1], made from piss-soaked snowflakes and we are crossing fingers that it isn't all going to melt in our hands overnight.
Sometimes it does dissolve in our hands and at that point, that's when the extensive hordes of ancillary staff roll out to lick butt.
So when you think that you're playing with wooden blocks, you're right, but you're probably doing it a bit better than piling them up randomly for the highest margin to support a massive hierarchy of unnecessary staff. And you're right, it is pretending. But it makes money, and that's all they care about at the end of the day.
Good luck for you still have hope of doing an honest job.
This is not economics. Its "finance", well really just a bunch of very long contracts that people pretend to "value", by making up numbers as you see from this article, in order to make money from people.
You're correct, although I'm unable to edit the original post now. Thanks for the correction too - it made me look up the definitions of both. Always good to learn something!
The major difference being that traditional insurance is heavily regulated, limiting the shenanigans. Finance, on the other hand, is more an anything-goes-until-you-catch-us model.
Having worked in both, I'm not sure I agree. I will agree that the incentives to skirt regulation seem higher in finance than in insurance, but my personal opinion (and I could be very wrong) is that is a result of the regulation not due to its lack.
Long time back i was studying Asset Backed Commercial Papers (ABCP) as a special purpose vehicle conduit and its effects on market runs during the financial crisis.
For those who want to understand it , these might possibly help(These are only personal notes.):
Excellent explanation. CDSs are insurance. So why can't you buy the equivalent product from an insurer, labelled and regulated as insurance? Because the capital requirements under insurance regulation would have been prohibitive.
So why weren't the banks charged with illegally selling insurance products and/or violating insurance regulations? Because fuck you, peon.
Imagine an empty company loans 100 companies $1 each.
To get the $100 to loan, that company sells bonds to 2 investors for $50 each.
One investor is senior and one is junior.
What this means is that if any of those 100 companies default and don't pay back their loan, then the junior investor loses money.
The junior investor loses money until their entire $50 is lost, and only then does the senior investor have to suffer loses.
The senior investor therefore only has losses if more than $50 of the total $100 loans is lost.
This is basic credit tranching in funded format.
Unfortunately things get a lot more complicated when instead of $1 loans, we have $1 credit default swaps, which instead of loaning someone $1, you promise to pay them if a 3rd party company defaults.
This trade is DB buying protection from a company on a very senior tranche of this type of trade, but they are buying it leveraged, so for example they buy $100 of protection from a company that has $10 of cash inside (provided by the third-party investors), with a promise to pay more if the tranche deteriorates.
As far as rationale, the primary purpose of these trades is to offload super senior risk from DB's balance sheet, which they accrued through doing a different type of trade.
Deutsche Bank is hardly a bank in the traditional sense considering their balance sheet. With $72.8 trillion euro in derivative exposure (the largest in the world, 21x the GDP of Germany) this will have to blow up in a spectacular fashion once one of these counterparties fail:
The notional amount doesn't tell you anything meaningful in term of risk. Say I am Deutsche, you are Citi. I trade a €1bn 10 year interest rate swap with you and then another €1bn 10 year interest rate swap with you in the opposite direction. The risk is exactly zero (the two transactions will be netted to zero in case of an insolvency) and the cash flows will perfectly offset each others. We each have €2bn of notional of derivatives outstanding.
Banks like Deutsche aren't in the business of taking large directional derivative bets. They are market makers so the quasi totality of these 72 trillions are offsetting positions, which are mostly collateralised, i.e. each counterparty has to post assets to match the evolution of the value of the derivative in order to keep the credit exposure minimal.
With such a massive book, there will be residual risks all over the place but 72 trillion is a meaningless number for assessing them.
Absolutely correct. Notional is an absolutely meaningless figure. If anything it is comparable to the total notional of derivatives ever traded: contracts are rarely torn up (1), more usually offsetting derivatives are entered into in order to cut the exposure.
And anyway notional tells you nothing about the risk or exposure. 1bn of 30y is a lot riskier than a 1bn of 1y etc.
1 - End customer trades do sometimes get torn up. Also trade compression services between street counterparts also result in bulk tear-ups from time to time. But still the vast majority of trades are legacy and largely offset each other.
Well, in theory, in the example you site, "the risk is exactly zero". However, it's not always this way, because there is also gambling, err speculation, err market making, err ...
Not everything is always perfectly hedged. And when it's not, there are mark to market issues. And, the market isn't always liquid. Levine is in fact writing about a market that "froze up".
One of my favorite anecdotes (pg 212) from The Big Short involves Deutsche. It was perhaps embellished a bit for the book. It involved credit default swaps:
In early July, Morgan Stanley received its first
wake-up call. It came from ... Deutsche Bank ...
the CDS (sold to) Deutsche Bank had moved in
Deutsche Bank's favor
Deutsche: Triple-A-rated subprime CDOs ... were
now worth ... 70 cents on the dollar
Morgan: Our model says they're worth ninety-five
Deutsche: our model say's they're worth seventy
Morgan: Well, our model says they are worth
ninety-five, repeated the Morgan Stanley person
Deutsche: Dude, fuck your model. I'll make you a
market. They are seventy---seventy-seven. You have
three choices. You can sell them back to me at
seventy. You can buy some more at seventy-seven.
Or you can give me my fuckimg one point two
billion dollars.
That's exactly what happens at just the wrong point in time when there is stress in the markets. This isn't something from ancient history, this all happened less than 10 years ago.
So, while 72 trillion is indeed a "meaningless" number, quite often the "residual risks" you mention will greatly exceed the equity in the big investment banks. Notably, Deutsche has in the past been considered to be very over-leveraged.
The banks smartly foisted off some of their "residual risks" onto muppets and companies like AIG, who stupidly accepted the business. But AIG wasn't liquid enough to pay off. If the Fed hadn't bailed out AIG, who then bailed out their counter-parties, the fallout would have utterly destroyed the world's financial system.
As it was, we came pretty close to everything crashing down. IIRC at one point the Fed had to provide liquidity to e.g. Verizon, because the banks weren't willing to. Sheesh.
Yes, gross derivative exposure is irrelevant... until there is a breach in the counterparty chain and suddenly all net becomes gross (as in the case of the Lehman bankruptcy), such as during a financial crisis.
Perhaps rather than total notational outstanding, a more meaningful number is the amount of real assets this is all backed by. According to the IMF we have $600 trillion in gross notional derivatives backed by a $600 billion in real assets, i.e. 1000x systemic leverage:
Well the point still stands... in case of an "event" that blows up collateral chains, Deutsche will be in big trouble.
Also, interestingly, from the piece I quoted, a lot of these derivatives get to be created without any collateral being posted at all:
"Another important metric is the under-collateralization of the OTC market. The Bank for International Settlements estimates that the volume of collateral supporting the OTC market is about $1.8 trillion, thus roughly only half of exposures. Assuming a collateral reuse rate between 2.5-3.0, the dedicated collateral is some $600 - $700 billion. Some counterparties (e.g., sovereigns, quasi-sovereigns, large pension funds and insurers, and AAA corporations) are often not required to post collateral. The remaining exposures will have to be collateralized when moved to CCP to avoid creating puts to the safety net. As such, there is likely to an increased demand for collateral worldwide."
> Well the point still stands... in case of an "event" that blows up collateral chains, Deutsche will be in big trouble.
Not for bilateral derivatives that can always be netted, even in bankruptcy, which make up the bulk of the notional outstanding. If you read the ISDA Master Agreement[1], derivatives can always be netted as there is a provision that exempts derivatives from the usual bankruptcy automatic stays to stop literally this exact threat you posited [2]. Even beyond the bilateral case, the multilateral cases are increasingly cleared through central clearinghouses[4]. But even when they aren't, multilateral nets can still be pushed through, in one of two ways: 1) even when Lehman failed, there were some multilateral nets that greatly reduced the exposure, under mutual agreement by all creditors, and 2)The Dodd-Frank Act, since 2009, has also greatly helped this, giving the FDIC and Fed enhanced powers under the Orderly Liquidation Authority[3], which allow them to multilaterally net to reduce contagion risks even if there is no such master agreement that a priori would have let them do so[5].
[2] A Dialogue on the Costs and Benefits of Automatic Stays for Derivatives and Repurchase Agreements, by Duffie and Skeel, alternatively see [3], or [4]
[3]The New Financial Deal: Understanding the Dodd-Frank Act and Its (Unintended) Consequences by David Skeel.
[4] The Risk Controllers by Peter Norman.
[5] Even pre Dodd-Frank, when there wasn't a law that let this happen, usually the creditors come together and fix things, a la LTCM "bailout"[6], which was funded entirely by the creditors and unwound in an orderly fashion through multilateral netting. Bear Stearns and Lehman Brothers were the only creditors that stiffed everyone else, by refusing to participate, and that was partly why they got fucked in 2009 since the other bankers remembered their unwillingness to help out in 1998, and reciprocated.
[6] I hate that this is called a Bailout. No taxpayer funds were spent. The creditors (i.e. big banks) bailed out one of their own by pooling together money and fixing their shit. The only governmental involvement was that the Fed president called the CEOs of the big banks and said "yo fix your shit before it spreads", and they did. That's not a bailout in the sense of "taxpayers bailed them out". They bailed themselves out.
The money moves around and around and around (notionally) a dizzyingly large number of times because the more opaque the process and end product, the less immediate introspection it receives, the less well the average Joe can understand it, and the firm can continue 'alchemy'.
Kudos to the regulators for the arduous task of unravelling the large ball of string and following it to the conclusion!
It reminds me a lot of large code bases. At some point people face a choice between a) forcing sufficient simplicity and clarity that one has a chance of understanding what's going on, and b) saying "fuck it" and letting complexity metastasize to the point nobody really understands what's going on.
The difference, of course, is that with software the opaqueness is pure cost. Whereas in the financial markets the opaqueness is often a profitable resource from somebody.
The insurance in question would only pay off under some very improbable and dire circumstances. And under these circumstances the chances of getting a payoff are rather slim because, you know, the world is crashing and burning. Some time during the crisis the risk became very real which was reflected in the market prices of credit protections. But DB and "investors" continued to act as if the risk was small.
What I don't get is who paid for that risk in the end. I suspect it was the government which bailed everyone out with public money so that the doomsday scenario was never realized.
Off topic but this reminded me of a movie i saw last year, "Blackbox BRD" (http://en.m.wikipedia.org/wiki/Black_Box_BRD) which relates to Deutsche Bank past and has an interesting interview with a high level staff (Bank president?) and found it curious to glimpse on how they think (or thought).
52 comments
[ 4.0 ms ] story [ 114 ms ] threadBut when I read this article, I can only say one thing: Why?
Why is Wall Street spending all this time and energy in creating all these different securities? I suspect that any excess "profits" or "returns" to be had from them accrue 90% to the bankers and 10% to the muppets funding them. So why do the muppets keep playing this game?
In this case, the sucker would actually have been DB's investors if the worst case ever did happen. (They would have gone under like many other banks - the bailout saved them.)
In this case the investors they sold protection to (hedge funds) got better prices than they ordinarily would have, because of the risk hidden in the conduit.
1) The way you get to sell products to investors is by promising some sort of alpha: "My fund is less risky, my derivative pays more for the same risk, etc."
2) But magic does not happen in the financial markets (and when it does is through the old stories.. diversification, transferring the risk to those of broader shoulders, etc.)
3) So what you do is HIDE THE RISK. There are many ways to do that, and they usually involve COMPLEXITY. Complexity and opacity are your friends, as they confuse investors who don't realize there is a hidden risk somewhere there, or a scenario you haven't told them (but put on page 350) where if A B and C happen, then he will lose a lot of money.
I'm not joking here.. sometimes complexity is good as it allows you to e.g. reduce mismatches in your risks (your income is in USD and expenses in EUR, your debts are long term and assets short term, etc.). However, the ultimate purpose of most complex products is that they allow you to sell stuff that appears better than it really is. And IT WORKS. Everytime. Against millionaires and against folks buying silly combos of insurances plus deposits (forgot their names).
I've been rereading my copy of Taleb's Antifragile. He keeps hammering similar points over and over: complexity, risk, opacity, susceptibility to Black Swans, etc.
I think I'd be depressed if I worked on Wall Street and the better part of my compensation resulted from peddling all these confusing products.
Still, as you point out, sometimes complexity is good. I do understand that.
And then we wrap it up all with a bow by allowing finance people to alternately claim that a) they are so smart they deserve to personally make millions to billions, and b) they couldn't possibly be held accountable for systemic fuckups, because nobody could possibly have known. I'll accept one or the other, but the combination is incredibly dangerous.
The primary purpose of these trades is to offload super senior risk from DB's balance sheet.
Super senior risk is the risk that DB retained when it (very profitably) sold various other tranches on a group of assets (typically CDS on a bunch of corporate and sovereign credits).
This type of risk is very difficult to sell as the return is very low and most entities they might buy unfunded protection from are likely to not be able to pay in the event that the swap pays out.
Offloading that risk through a swap to a conduit like this allows DB to release (expensive) capital/reserves that would otherwise need to hold.
Fail early, fail hard, then go fix it. Don't go off operating in an undefined state.
Quite a lot of financial strangeness can be answered with "where else are you going to put your money?" Western treasury bonds are the safe option, but they're close to negative yields. Spending $1.01 at auction to buy a bond which nominally costs 95c and pays back $1 in five years doesn't look great. So there is a huge demand for "riskless" positive-yielding investments.
That's why London (and elsewhere) property continues to shoot upwards in value way beyond the ability of the inhabitants to afford it. That's why Switzerland has negative interest rates. And so on.
What happens when this level of genius in small groups is cut loose on financial markets where "Make money is the only goal"? You get these space shuttles. They are so complex because that's all these people ever think about.
Snapchat is world changing and the work of brilliant computer scientists? Did I miss something? I can only hope that your post was intended to be sarcasm.
The space shuttle we have is, to paraphrase [1], made from piss-soaked snowflakes and we are crossing fingers that it isn't all going to melt in our hands overnight.
Sometimes it does dissolve in our hands and at that point, that's when the extensive hordes of ancillary staff roll out to lick butt.
So when you think that you're playing with wooden blocks, you're right, but you're probably doing it a bit better than piling them up randomly for the highest margin to support a massive hierarchy of unnecessary staff. And you're right, it is pretending. But it makes money, and that's all they care about at the end of the day.
Good luck for you still have hope of doing an honest job.
[1] http://www.stilldrinking.org/programming-sucks
In practice AIG managed to get itself into a situation where the Fed needed to give it $182.3 Billion dollars to keep it afloat.
That's $182.3 Billion, not Million. https://en.wikipedia.org/wiki/Aig#Liquidity_crisis_and_gover...
For those who want to understand it , these might possibly help(These are only personal notes.):
[part 1]https://medium.com/finance-and-economics-studies/read-note-o...
[part 2]https://medium.com/finance-and-economics-studies/asset-backe...
I got lost somewhere around the first appearance of the word tranch,
Deutsche bank sold lot's of this type of insurance.
If the markets had turned and some large companies had gone bankrupt they would not have had the means to pay the people who bought this insurance.
Basically they took very low quality insurance and re-packaged it as high quality insuarance.
So why weren't the banks charged with illegally selling insurance products and/or violating insurance regulations? Because fuck you, peon.
Imagine an empty company loans 100 companies $1 each.
To get the $100 to loan, that company sells bonds to 2 investors for $50 each.
One investor is senior and one is junior.
What this means is that if any of those 100 companies default and don't pay back their loan, then the junior investor loses money.
The junior investor loses money until their entire $50 is lost, and only then does the senior investor have to suffer loses.
The senior investor therefore only has losses if more than $50 of the total $100 loans is lost.
This is basic credit tranching in funded format.
Unfortunately things get a lot more complicated when instead of $1 loans, we have $1 credit default swaps, which instead of loaning someone $1, you promise to pay them if a 3rd party company defaults.
This trade is DB buying protection from a company on a very senior tranche of this type of trade, but they are buying it leveraged, so for example they buy $100 of protection from a company that has $10 of cash inside (provided by the third-party investors), with a promise to pay more if the tranche deteriorates.
As far as rationale, the primary purpose of these trades is to offload super senior risk from DB's balance sheet, which they accrued through doing a different type of trade.
http://imgur.com/MNECKC4.jpg
A former Kansas Fed president called them "horribly undercapitalized", with a leverage ratio of 1.63 percent:
http://www.reuters.com/article/2013/06/14/us-financial-regul...
Banks like Deutsche aren't in the business of taking large directional derivative bets. They are market makers so the quasi totality of these 72 trillions are offsetting positions, which are mostly collateralised, i.e. each counterparty has to post assets to match the evolution of the value of the derivative in order to keep the credit exposure minimal.
With such a massive book, there will be residual risks all over the place but 72 trillion is a meaningless number for assessing them.
And anyway notional tells you nothing about the risk or exposure. 1bn of 30y is a lot riskier than a 1bn of 1y etc.
1 - End customer trades do sometimes get torn up. Also trade compression services between street counterparts also result in bulk tear-ups from time to time. But still the vast majority of trades are legacy and largely offset each other.
Not everything is always perfectly hedged. And when it's not, there are mark to market issues. And, the market isn't always liquid. Levine is in fact writing about a market that "froze up".
One of my favorite anecdotes (pg 212) from The Big Short involves Deutsche. It was perhaps embellished a bit for the book. It involved credit default swaps:
That's exactly what happens at just the wrong point in time when there is stress in the markets. This isn't something from ancient history, this all happened less than 10 years ago.So, while 72 trillion is indeed a "meaningless" number, quite often the "residual risks" you mention will greatly exceed the equity in the big investment banks. Notably, Deutsche has in the past been considered to be very over-leveraged.
The banks smartly foisted off some of their "residual risks" onto muppets and companies like AIG, who stupidly accepted the business. But AIG wasn't liquid enough to pay off. If the Fed hadn't bailed out AIG, who then bailed out their counter-parties, the fallout would have utterly destroyed the world's financial system.
As it was, we came pretty close to everything crashing down. IIRC at one point the Fed had to provide liquidity to e.g. Verizon, because the banks weren't willing to. Sheesh.
Perhaps rather than total notational outstanding, a more meaningful number is the amount of real assets this is all backed by. According to the IMF we have $600 trillion in gross notional derivatives backed by a $600 billion in real assets, i.e. 1000x systemic leverage:
http://www.imf.org/external/pubs/ft/sdn/2012/sdn1212.pdf
One can imagine this to become just a tiny bit problematic once collateral chains start breaking, such as when "AAA-rated" AIG failed :)
Also, interestingly, from the piece I quoted, a lot of these derivatives get to be created without any collateral being posted at all:
"Another important metric is the under-collateralization of the OTC market. The Bank for International Settlements estimates that the volume of collateral supporting the OTC market is about $1.8 trillion, thus roughly only half of exposures. Assuming a collateral reuse rate between 2.5-3.0, the dedicated collateral is some $600 - $700 billion. Some counterparties (e.g., sovereigns, quasi-sovereigns, large pension funds and insurers, and AAA corporations) are often not required to post collateral. The remaining exposures will have to be collateralized when moved to CCP to avoid creating puts to the safety net. As such, there is likely to an increased demand for collateral worldwide."
Not for bilateral derivatives that can always be netted, even in bankruptcy, which make up the bulk of the notional outstanding. If you read the ISDA Master Agreement[1], derivatives can always be netted as there is a provision that exempts derivatives from the usual bankruptcy automatic stays to stop literally this exact threat you posited [2]. Even beyond the bilateral case, the multilateral cases are increasingly cleared through central clearinghouses[4]. But even when they aren't, multilateral nets can still be pushed through, in one of two ways: 1) even when Lehman failed, there were some multilateral nets that greatly reduced the exposure, under mutual agreement by all creditors, and 2)The Dodd-Frank Act, since 2009, has also greatly helped this, giving the FDIC and Fed enhanced powers under the Orderly Liquidation Authority[3], which allow them to multilaterally net to reduce contagion risks even if there is no such master agreement that a priori would have let them do so[5].
[1] http://en.wikipedia.org/wiki/ISDA_Master_Agreement
[2] A Dialogue on the Costs and Benefits of Automatic Stays for Derivatives and Repurchase Agreements, by Duffie and Skeel, alternatively see [3], or [4]
[3]The New Financial Deal: Understanding the Dodd-Frank Act and Its (Unintended) Consequences by David Skeel.
[4] The Risk Controllers by Peter Norman.
[5] Even pre Dodd-Frank, when there wasn't a law that let this happen, usually the creditors come together and fix things, a la LTCM "bailout"[6], which was funded entirely by the creditors and unwound in an orderly fashion through multilateral netting. Bear Stearns and Lehman Brothers were the only creditors that stiffed everyone else, by refusing to participate, and that was partly why they got fucked in 2009 since the other bankers remembered their unwillingness to help out in 1998, and reciprocated.
[6] I hate that this is called a Bailout. No taxpayer funds were spent. The creditors (i.e. big banks) bailed out one of their own by pooling together money and fixing their shit. The only governmental involvement was that the Fed president called the CEOs of the big banks and said "yo fix your shit before it spreads", and they did. That's not a bailout in the sense of "taxpayers bailed them out". They bailed themselves out.
The difference, of course, is that with software the opaqueness is pure cost. Whereas in the financial markets the opaqueness is often a profitable resource from somebody.
What I don't get is who paid for that risk in the end. I suspect it was the government which bailed everyone out with public money so that the doomsday scenario was never realized.