Synthetic CDO's: tsunami event when major bankruptcies reaches 9 (currently 6)

December 1, 2008 7:54 PM

Synthetic CDO's are complex little known financial instruments (insurance contracts) that are on the brink of triggering "the most colossal rights issue in the history of the world, all at once .. mandatory." If, out of a list of several hundred major companies, any nine go bankrupt, the CDO's are in default, which would mean a mass transfer of cash (real money) from unsuspecting investors around the world goes into the banking system. How much? Nobody knows, but it’s many trillions. Banks will be flush with cash, perhaps ending the credit crisis, while many investors (individuals, charities, municipalities) will be wiped out. Alternatively, the triggering of default on the trillions of synthetic CDOs could be a disaster that tips the world from recession into depression. Nobody knows, but it won’t be a small event. Thus far the count is six: three Icelandic banks, Countrywide, Lehman and Bear Stearns.
posted by stbalbach (49 comments total)

Previously: Graphical explanation of CDOs (not quite the same as a synthetic CDO).
posted by stbalbach at 7:58 PM on December 1, 2008

To be honest, the complexity of these financial instruments is such that I will be very greatly relieved when the whole house of cards implodes and I am forced to spend my short remaining life as a Road Warrior.
posted by Ryvar at 8:06 PM on December 1, 2008 [17 favorites has favorites]

another CDO explainer...

Understanding Synthetics: I'm going to try to explain a passage in Michael Lewis' cover story for Portfolio in which he talks about synthetic CDOs without actually using the term.
posted by kliuless at 8:08 PM on December 1, 2008 [1 favorite has favorites]

This is the equivalent of a financial single link op-ed. The author of the piece has no clue who the investors are, has no clue how much is outstanding, and has no clue how many of the contracts cancel each other out.

Here's a real story about CDS payouts. Maybe instead of engaging in rampant speculation and fear-mongering, we should just stick to the facts as can be verified by a legitimate source.
posted by SeizeTheDay at 8:08 PM on December 1, 2008 [3 favorites has favorites]

Nice to see high finance operating on the same principles as Japanese console RPGs. Break the nine mystic seals and the financial system will be saved forever!
posted by shadow vector at 8:10 PM on December 1, 2008 [16 favorites has favorites]

Although the implications of the article may have warranted you to post it here as a socially conscious thing to do, this is pure balderdash, based upon a sloppy understanding of a very complex derivatives market structure. Ask any CLO trader about dual-party haircuts that are correspondent to market value, not outstanding principal...when these bonds get downgraded to BBB or worse yet CCC the winners will lose inevitably.

Kohler might want to read Mann's Handbook on Fixed Income Securities to learn a few things about OC Tests...
posted by stratastar at 8:10 PM on December 1, 2008 [2 favorites has favorites]

I am forced to spend my short remaining life as a Road Warrior.


I've been saying this since Y2K.

My bullets are all old cold war Chinese crap, and so is my assault rifle.
But I already have the mesh ready to weld to my windshield, and the tire spikes.
Oh, and the Geo mounted P.A. system. I'm practically a warlord.

(deep forboding voice) "Just Walk Away!"
posted by Balisong at 8:11 PM on December 1, 2008 [1 favorite has favorites]

One last thing:

the OPs usage of the term "unsuspecting investors" is somewhat dubious. We will begin to hear stories of municipalities and pension plans that may get wiped out by contractual default, but it was in the legal writing that they signed.

If you cannot understand the thousand page contractual obligation you are getting into, while being whisked away with guarantees of free basis points on leveraged investments, you are being negligent with your vested interests.
posted by stratastar at 8:17 PM on December 1, 2008 [4 favorites has favorites]

once again i am glad im poor and have been investing in cigarettes and booze while diversifying in friends that sleep with their handgun and head out to the country to shoot the olde AK-47 every couple days.

gotta use the cigarettes and booze on someone.
posted by Glibpaxman at 8:24 PM on December 1, 2008 [4 favorites has favorites]

The recklessness with which these people throw money around is scary, and the way they mislead leaves me in awe. This is what happens when droves of Ivy Leaguers are seduced by the prospect of measuring their lives based on how many millions and billions they're worth.
posted by spiderskull at 8:30 PM on December 1, 2008 [1 favorite has favorites]

How A School Board In Wisconsin (And The NYC Subway System) Became Accidental Hedge Funds
posted by Rafaelloello at 8:35 PM on December 1, 2008 [1 favorite has favorites]

SeizeTheDay: "..rampant speculation and fear-mongering...Here's a real story about CDS payouts."

The "real story" you linked to is about normal CDO's which reference a single company - the "synthetic" CDOs reference 100s of companies, and have a limit of 9 bankruptcies before defaulting. Obviously we don't know how big it is because 9 has not been reached yet, but the author of the article says it's in the range of "trillions" of dollars. It's not fear mongering to say this will be a non-trivial event when it happens.
posted by stbalbach at 8:41 PM on December 1, 2008

Is it actually true that the occurrence of nine defaults is of general significance to synthetic CDOs? I had never heard of that, and the only references I can find are to Alan Kohler articles on Business Spectator.

Also, the instrument he described doesn't sound very much like a synthetic CDO at all. The synthetic CDOs I've worked on all issued debt, which kind of makes sense, since you're synthesizing a debt instrument. There was never any mechanism by which investors would be required to pony up more cash, and losing everything if there were nine credit events out of 100 credits in the portfolio just sounds bizarre and incredibly (and pointlessly) risky.
posted by Mr. President Dr. Steve Elvis America at 8:45 PM on December 1, 2008

ANALYSIS-Synthetic CDO default losses likely to multiply

A CDO hit by all seven failures so far; Lehman Brothers, Fannie Mae, Freddie Mac, Washington Mutual, Landsbanki, Glitnir and Kaupthing, in an equally-weighted portfolio of 100 names and with recoveries set by auction, would have lost up to 4.38 percent of notional value.

Understanding Synthetics

Relates back to the Michael Lewis article previously posted.
posted by dhartung at 8:50 PM on December 1, 2008 [1 favorite has favorites]

stratastar: "the OPs usage of the term "unsuspecting investors" is somewhat dubious."

"Unsuspecting investor" is a direct quote from the article. He goes on to say

investors who bother to read the fine print will see that they will lose some or all of their money if seven, eight or nine of a long list of apparently strong global corporations go broke. In 2004-2006 it seemed money for jam. The companies listed would never go broke – it was unthinkable.

In a way it really is an unsuspecting investor. Would you or anyone think 9 of these companies would ever be bankrupt? Hardly dubious. The only people who really knew where the banks themselves, since they were the ones lending money to these companies and understood how over-leveraged they were - which is why the banks did the CDO insurance in the first place!
posted by stbalbach at 8:50 PM on December 1, 2008

To clarify why this sounds so bizarre: The normal reason to buy a large, diverse portfolio is to manage risk--while some of the securities in the portfolio might be subject to default, it's incredibly unlikely that they all will be. Consequently, it's both more likely that you'll bear some defaults than if you just bought one security, but it's less likely that you'll be wiped out.

The normal reason to buy synthetic diverse portfolios is that buying real ones is a pain in the ass, just from a mechanical standpoint.

An instrument that loses all of its value if 9 credits out of 100 default serves a very different function, though, and I'm not sure what it is.
posted by Mr. President Dr. Steve Elvis America at 8:52 PM on December 1, 2008

From the article (emph. mine):

They [synthetic CDOs] have a variety of twists and turns, but it usually goes something like this: if seven of the 100 reference entities default, the SPV has to pay the bank a third of the money; if eight default, it’s two-thirds; and if nine default, the whole amount is repayable.

It sounds like the number of defaults (7, 8, and 9) that the author gives are just examples. But, further down the article, he goes on to say "If the list of defaults – full and partial – gets to nine, then a mass transfer of money will take place...". Is he talking about one specific synthetic CDO or does every synthetic CDO have the same terms? Maybe I'm just misinterpreting him here.

Also from the article (emph. mine):

Here are some of the companies that are on all of the synthetic CDO reference lists: ... [long list of troubled or bankrupt companies]

What does he mean by "all of the synthetic CDO reference lists"?
posted by mhum at 8:54 PM on December 1, 2008

The "real story" you linked to is about normal CDO's which reference a single company - the "synthetic" CDOs reference

stbalbach, do you know the difference between a CDO and a CDS?
posted by SeizeTheDay at 8:55 PM on December 1, 2008

Isn't there a confusion of CDO and CDS here somewhere?

read read read

ah, synthetic CDOs are made of CDSs, kinda like Soylent Green is made of people!

This part:

If the list of defaults – full and partial – gets to nine, then a mass transfer of money will take place from unsuspecting investors around the world into the banking system. How much? Nobody knows, but it’s many trillions.

seems a bit iffy to me; the top-tier US banks presently have a total market cap of $300B, and of course "many trillions" is much more than the "several" trillion of US external government debt, even accounting for the 0.65 AUD/USD that I assume the article is using for dollars.
posted by troy at 8:56 PM on December 1, 2008

You know, I'm not a communist, but sometimes—when I see people throwing around money on a scale that dwarfs anything my humble life will ever see without regard for how it will affect the billions of people like me or poorer—I can just about understand how you could be a commie. Of course, the same sociopaths take over that system, too.
posted by sonic meat machine at 8:57 PM on December 1, 2008 [5 favorites has favorites]

I can just about understand how you could be a commie

here ya go, how things were in a the lost world.
posted by troy at 9:03 PM on December 1, 2008 [1 favorite has favorites]

I've got an alternative forecast. Everything will be fine by February, the instability will all be blamed on presently invisible factors and by this time next year we'll be arguing about something we are currently ignoring that has, by that time, gotten out of hand.
posted by TwelveTwo at 9:06 PM on December 1, 2008 [5 favorites has favorites]

(I'm not a structured credit trader, so if anyone around here is, feel free to correct any of this.)

Alan Kohler is a good journalist, but this is a bad article.

Firstly, his list of defaults is wrong... last I checked, Countrywide and Bear weren't credit events for the purposes of CDS settlement (and he's left out Fannie, Freddie, and WaMu, which were).

So if anything, the count is seven - or eight if you include collapsed Canadian paper producer Tembec. But every CDO references a different set of underlying entities. They could reference none, or some, or all of the eight defaulted companies, which will change the amount of losses in the CDO portfolio.

More to the point, I don't know where he got this idea of "if nine reference entities default, the whole structure blows up". That sounds like he's latched on to one particular tranche of a CDO.

How these things work is that the lowest-rated tranche takes the first losses (for example the first 3% of the losses), then the higher-rated tranches (for example the next 10%), then the highest rated tranches start taking losses.

So, to take a rough example - if you owned the 6-9% tranche of a CDO, then the seventh, eighth and ninth defaults would indeed cause you to lose one-third, then two-thirds, then the whole lot of your capital. (I'm glossing over recovery rates here, which will change the number of defaults needed.) But that's just one part of the capital structure, not the whole shebang.

The article is pretty badly flawed. It does have a point - these lower-rated tranches were embedded in all sorts of funny places, including the Pinnacle notes that were sold to retail investors in Singapore, then blew up and took two billion dollars of investors' money with them - but as far as I can tell, it gets the facts completely wrong.
posted by The Shiny Thing at 9:12 PM on December 1, 2008 [2 favorites has favorites]

troy: "seems a bit iffy to me; the top-tier US banks presently have a total market cap of $300B, and of course "many trillions" is much more than the "several" trillion of US external government debt, even accounting for the 0.65 AUD/USD that I assume the article is using for dollars."

dhartung's "ANALYSIS" link above quotes Citigroup as saying the synthetic CDO market is around 600 billion, but unclear if that's just USA or global.

SeizeTheDay stbalbach, do you know the difference between a CDO and a CDS?

yes, but you wouldn't know it from that sentence. Anyway hopefully the main point is understood, synthetic CDOs are not the same as the CDSs - although S-CDO's are made up of CDSs (like soylent green is made from people, but people are not soylent green).
posted by stbalbach at 9:13 PM on December 1, 2008

Sorry if that was a bit incoherent. XKCD syndrome strikes again.
posted by The Shiny Thing at 9:13 PM on December 1, 2008

Here's one thing I've been wondering about. When people say there are "trillions" of dollars in play, aren't they actually talking about CDSes backed by other CDSes?

Like say there is a three company chain. A buys a CDS on X (where X is someone like Lehman brothers) from AIG for $10 billion. Then X starts to look shaky and B buys a CDs from A for $10 billion at a higher interest rate. Now there are $20 billion in outstanding liabilities, right?

Great for A, they've now "neted" and have guaranteed free money until the end of their contract or until X goes bankrupt, at which point their costs will be covered by AIG.

It would go like AIG -> A -> B.

But now X really does go bankrupt, and what's worse, AIG might also go bankrupt and be unable to pay A, at which point A won't be able to pay B, who might really need the money. So they all fail.

But only $10 billion really disappears, and only $10 billion ever existed in the first place.

--

So my question is how much of this "$trillions" is like that second $20 billion in my example? And why can't the government just come in and figure out who has what portion?
posted by delmoi at 9:41 PM on December 1, 2008 [1 favorite has favorites]

N.B. Not a student of the markets. This is my perspective. Please correct me because I want to learn.

So my question is how much of this "$trillions" is like that second $20 billion in my example?

Nobody really knows. All you can do is let the market delever and see who owns what but that's probably the worst thing that can happen (which is happening right now)

And why can't the government just come in and figure out who has what portion?

a) Because it's just too big an orgy to work out in any conceivable amount of time.

b) To work out who owns what you effectively have to be able to accurately value the assets that these companies have in the first place. Companies are firmly against this because it'll mean that they'll be caught with not only their pants down but left holding their dicks in their hands.

Think about an investment bank holding a mortgage backed security that is backed by "$10,000,000" worth of property when it was issued. That property today is worth maybe $7,500,000 or even $5,000,000 today. For them to have that revalued in the "who owns what" game down to its "true" value would mean they immediately have to write down half their investment and nobody wants to do that unless the absolutely have to.

You can keep using the value of a security so long as 1) you believe it and 2) the sucker who's relying on that value agrees. Until you run out of 2s the companies that hold these assets keep wanting to use the value of the security that they had determined back when the market was extremely high. If you had a choice of being ignorant and pulling $10,000,000 from a sucker who's accepting a CDO comprised of MBSes as collateral and being right and only pulling $5,000,000 from that same sucker wouldn't you rather remain ignorant for as long as you can?

Obviously that doesn't work when the person knows that it's overvalued and slipping. In the case of property you could usually count on the price of property increasing to cover the buyer defaulting and it's why S&P and Moodys were giving so much of this garbage AAA status. They physically didn't have a model for how a CDO would work with house prices declining. All they knew is "well if the buyer doesn't pay just flip the house and give the 60% of people in the AAA tranche their cash". That's why they were letting Wall Street turn dog shit into pure gold.

Do you ever wonder why mortgage brokers were trying to ship so many shit mortgages out the door? Because as soon as the poor sap defaulted on their first ARM reset they could kick them out of the house, flip it for a huge profit in an ever rising market to the next sucker and begin the process anew. To put it simply, they didn't give a strawberry picker $720,000 because they expected him to pay it back.
posted by Talez at 10:48 PM on December 1, 2008 [6 favorites has favorites]

Nobody really knows. All you can do is let the market delever and see who owns what but that's probably the worst thing that can happen (which is happening right now)

Why can't you find out? I mean, pass a law that says everyone who owns a CDS or derivative needs to write down who they bought it from, what loans it references and so on. Any unregistered CDS would become invalid.

a) Because it's just too big an orgy to work out in any conceivable amount of time.

How long would it take? How many CDSes are there? Both of those values things are real numbers which people can actually figure out.
posted by delmoi at 11:04 PM on December 1, 2008

So basically we have the age old problem of Babylon mashing up the scene and blaming the Rastaman going on here.
posted by sgt.serenity at 11:06 PM on December 1, 2008 [5 favorites has favorites]

Why can't you find out? I mean, pass a law that says everyone who owns a CDS or derivative needs to write down who they bought it from, what loans it references and so on. Any unregistered CDS would become invalid.

That's why people are calling for a clearing house for CDSes. It just doesn't help us in the short to medium term.

How long would it take? How many CDSes are there? Both of those values things are real numbers which people can actually figure out.

For the sake of argument if you had infinite manpower there'd be nothing really stopping the government from going to everyone and getting their outstanding CDS contracts and coming up with a number. I'm sure it'll be a mind blowingly huge number to most people.

But you want to know who owns what. And part of knowing who owns what is what assets/securities people have, what is liquid/what they can liquidate which is the real problem. Nobody wants to be caught naked in the public eye holding all these shitty assets.
posted by Talez at 11:27 PM on December 1, 2008

...doom doom doom doom doom-doom doom doom-doom doom doom BLACK PRESIDENT WOOHOO! doom doom doom doom-doom doom doom I LIKE TURKEY doom AND PIE doom doom-doom doom doom doom-doom EAT THE BANKERS doom-doom doom doom...
posted by loquacious at 11:43 PM on December 1, 2008 [16 favorites has favorites]

You know, the amount of money shit out daily as undigested caviar by these thugs would go a long way towards making my life really pleasant. Maybe a few pitchforked heads would help ease the pain. doom doom doom...
posted by maxwelton at 12:39 AM on December 2, 2008 [1 favorite has favorites]

.... nǐ hǎo ma? . . . doom-doom-doom doom . . .
posted by troy at 12:44 AM on December 2, 2008

That article stinks - the 9 number in particular is pulled from somewhere dark and smelly. I know about these products (it's my day (and night) job) and each one is bespoke -- the size of the portfolios and companies are in them, the number of defaults and the recoveries (i.e. what the debt of the defaulted companies is worth) required to trigger losses change from deal to deal. What's more, synthetic CDOs are funded products. Investors paid money on day one to buy the credit-linked notes so whilst they may (a big if, and not linked to "nine" at all) end up getting nothing back they will certainly not have to pay another cent. Investors also have to account for these notes on their balance sheet in some way - as it becomes more likely that defaults will affect the return of principal their value drops. Again, it doesn't change the result that if there are lots of defaults (maybe 9, maybe more or less) investors will lose their investment, but it does mean that there's no tsunami.
posted by patricio at 1:08 AM on December 2, 2008 [2 favorites has favorites]

Mr. President Dr. Steve Elvis America: An instrument that loses all of its value if 9 credits out of 100 default serves a very different function, though, and I'm not sure what it is.

Robbery.
posted by DreamerFi at 1:19 AM on December 2, 2008

Talez writes "For the sake of argument if you had infinite manpower there'd be nothing really stopping the government from going to everyone and getting their outstanding CDS contracts and coming up with a number. I'm sure it'll be a mind blowingly huge number to most people."

Wait, that's the solution to full employment! Instead of a WPA, hire everyone laid off to enumerate the values of CDS contracts!

And to save the government money, the "employees" aren't atually paid, they just get, say 1/100 of 1% of each CDS's value.

And then, seeing as how they're sure they'll make millions, they can borrow against their anticipated future earnings!

To buy foreclosed houses cheaply. And flip them! Which'll drive up housing and create jobs for mortgage lenders!

And the investment bankers, who can package up the loans to the enumerators, in complicated financial vehicles!

We're saved!!!
posted by orthogonality at 1:48 AM on December 2, 2008 [10 favorites has favorites]

Luckily, I live safely nestled between Russia and Germany. I can't imagine any trouble happening here if civilization collapses.

I could paddle north, but there are Vikings that way. Vikings and then starving polar bears.
posted by pracowity at 2:46 AM on December 2, 2008 [6 favorites has favorites]

Gosh I wish I had more time to critique that article (preparing for a lecture this evening), but I agree with patricio in that somethings rotten. A few points off the top of my head:

• The concentration risk the author alludes to as some kind of event creeping up on the markets was news back in 2005. [.pdf]
• In February 2006 FAS released statement 155 [.pdf], which ruled that Synthetic CDOs featured “concentration of credit risk in the form of subordination."; what this meant was these instruments should be marked to market (i.e., NOT model), therefore we must conclude no crisis, these things have already been battered to hell and back. I'm not near my Bloomberg right now otherwise I'd price a few, but I do recall talking to a Structured Product's trader a few weeks ago, and he mentioned that many of the stuff he deals with day to day (and I'm paraphrasing here) had not only priced in a recession, a depression but now were pricing in Armageddon. Massively priced down in other words.
• There are two classes of Synthetic CDOs; managed and static. Managed Synthetics allow the underlying CDSs at the tranche level to be substituted, or pulled out and replaced with another reference name should credit conditions change. As recently as 2003 investor interest in Static Synthetics was waning because of several high profile, high impact defaults (e.g., Parmalat ). One wonders why the author didn't raise this distinction, detailing issue size amoung the two classes.
• Even if there still were large numbers of Static Synthetics being issued (doubtful, why not purchase the more flexible and robust product?) these things don't simply "blow up" as is being implied; workout (what happens after default, the recovery process) is sometimes a protracted process and, in this economic environment, I doubt few folks will be pushing so hard.

This Nomura paper presents a fairly decent overview of Synthetics [.pdf], if anyone is interested in a more detailed perspective of the instruments.

Hope this helps!
posted by Mutant at 2:54 AM on December 2, 2008 [8 favorites has favorites]

To add to Rafaelloello's great link: Wisconsin Schools Shocked by Bad Investment
posted by anniecat at 6:27 AM on December 2, 2008 [1 favorite has favorites]

Ortho that is so complicated sounding that it just might work. Quickly get on that!
posted by Mastercheddaar at 7:53 AM on December 2, 2008

anniecat: Quite how the administrators of the school board decided that BORROWING money to INVEST in something, even safe corporate bonds, was an appropriate activity for a school board is beyond me. The salesman was clearly way out of line, but still, a SCHOOL BOARD doing anything other than straight investing? Interestingly, the school board thought it was buying into a portfolio of "very safe corporate bonds"... at the time Lehman was thought of as a safe name. Ahhh hindsight.
posted by patricio at 8:04 AM on December 2, 2008 [1 favorite has favorites]

That's a fascinating article, anniecat.
posted by delmoi at 8:32 AM on December 2, 2008

changing my name to 'mutant fan club'...
posted by ilovemytoaster at 8:50 AM on December 2, 2008

Anniecat's link has a similar story:

The school boards had been insuring part of a pool of dozens of bonds. If the bonds in the pool were doing OK, then so was the investment. But if just eight companies in that pool defaulted, Whitefish Bay would lose all of its money.
posted by exhilaration at 10:43 AM on December 2, 2008

unsuspecting investors

Uh huh, and the Germans didn't know what was going on in the concentration camps.

There's a difference between not knowing, and not wanting to know.

stratastar has it right, but forget thousand page contracts, how about the prominent disclaimer that I'm sure was right there on the front page of each one: YOU MAY LOSE PART OR ALL OF YOUR PRINCIPAL INVESTMENT.

(If I Godwin the thread, do I get to keep it? Bet you didn't expect a Godwin here, nuh uh...)
posted by vsync at 10:55 AM on December 2, 2008 [1 favorite has favorites]

The ul is a dead giveaway for a mutant comment on scrolling down.
posted by Samuel Farrow at 3:44 PM on December 2, 2008

Not sure I'm following this, but it looks to me like these institutions realised that they were insuring investments (gambles) that could bring down the insurers, and lacking an insurance system big enough to cover what they were playing with, they created a new form of insurance (S-CDOs), and have been paying into that for some years now.

So while this thing could hurt a lot of people, I'm seeing this invention as the finance boys doing the sorts of cover-your-risk things they should have been doing, rather than another example of the cowboy gambling that recently destroyed the global finance system.

Pity the insurance only closes the door after the horse has bolted, yet gave sufficient false sense of security to encourage people to open the door and start chasing the horse.
posted by -harlequin- at 8:07 PM on December 2, 2008 [1 favorite has favorites]

fwiw, salmon also recently sounded off on super-senior tranches, concluding: Ultimately, then, the error was one of management, not of financial technology. The banks’ balance sheets — and those of their off-balance-sheet vehicles — were expanding faster than the banks’ executives and risk managers could really keep a handle on. And rather than call a halt to that which they didn’t fully understand, they handed down edicts instructing the CDO desks to keep on dancing for as long as the music was playing. Most of the executives probably never even heard the term “super-senior” until those tranches started getting written down. It was their own incuriousness, rather than any CDS technology, which was really their undoing.
posted by kliuless at 4:49 AM on December 3, 2008 [1 favorite has favorites]

Recession jokes. People are adding more in the comments.
posted by Tehanu at 9:29 AM on December 3, 2008