Shopping at Amazon?
Wait! If you click on the Amazon banner link at the top of the page first, HCN will get a percentage of anything you purchase for 24 hours!
Open predictions
Our local book authors
User login
Who's online
There are currently 12 members and 150 guests online.
Posted at 5:17 on Sunday Morning? Not to imply anything but how many glasses of wine have been consumed before this post?
Thank you Tanta!
Mark to Market exists only where you have a liquid market.
Here, there's 2 more problems:
1. Thin trading at the level of a controlled marketplace.
2. Each of these CDO's from each originator seem to have their own level of quality, since no industry standard seems to have been applied (am I right here Tanta?).
Here's a little Accounting Theory; "All Assets should in theory be valued at the present value of future benefits on a cash basis" - lama. The obvious problem here is that we have no idea what cash will flow in the future from these. We could venture a decent guess in a more stable environment, but not here.
Disclaimer: If you ask a sample of CPA's if they agree with my quote, most will not.
Oh damn, I forgot to change the time stamp. I actually started this at 8:17 am ET, but it being Sunday morning and everything, I got distracted and ended up finishing it about 10 minutes ago. Blogger uses the time you opened the compose window unless you change it.
Anyway, all times on this blog are PT, because CR is the blog boss and he is on PT.
All Tanta posts are actually written in ET. I'm not usually still drunk at 8:17 am.
I think one of the indications of a good writer is that even when you know absolutely nothing about what they are writing about, you keep reading anyway because you enjoy their writing style.
So. How do you pronounce Yves anyway? Eve? Eves? Ives? I give up.
Sometimes, it just amazes me the actual depth of foresight that exists, just damn mind boggling, proactive, cutting edge analysis-
"June 24 (Bloomberg) -- International banks, insurers and pension funds may be vulnerable to a slump in property prices and risks built up by the surge in leveraged finance, the Bank for International Settlements said."
Global Financial System Vulnerable to Past `Excesses,' BIS Says - Bloomberg.com
Each of these CDO's from each originator seem to have their own level of quality, since no industry standard seems to have been applied (am I right here Tanta?).
I think the big problem here is that a lot of people still don't get the diff between asset-backed bonds and CDOs.
A CDO is sort of like a managed mutual fund. It is a vehicle that holds derivatives: tranches of mortgage-backed bonds, commercial RE-backed bonds, aircraft leases, God knows what. But it doesn't necessarily actually own those assets outright: it can be a "synthetic" CDO, in which the "fund" sells credit-default swaps on a set of underlying "reference assets" that it doesn't actually own. The "yield" on the CDO is the spread between the premiums paid by the protection buyers and the losses paid out by the CDO.
And they're managed: unlike a mortgage-backed security, which is "brain dead," a CDO can buy and sell paper in and out of the CDO. So first of all, rating one of these things is often more like rating a business than it is rating a simple asset-backed cash flow.
But all those assets underlying those derivatives have their own credit ratings, so the sum total rating of the CDO has to be based on the prior ratings of the assets plus the rating assigned to the CDO manager.
When you start talking marking, you're looking at a situation in which the underlying assets are probably liquid enough to be marked, but that still doesn't make the CDO itself liquid enough to be marked. Of course that's crazy-making: somehow the valuation of this CDO has to be affected by those "underlying marks," even if the CDO itself isn't really marked to market.
Therefore, many gyrations of many weird models come into play. You can get all hostile and call that "marking to model," but what else, realistically, is there to do? There is no market price.
So instead of investors saying "Damn, I bought something that cannot be adquately marked even though it apparently must be marked," they're saying: the rating agencies are cheating!
Eve.
Tanta
I like your tone this morning, it fits the times we live in. Don't ever think of backing off, we need your voice and ideas to be heard.
Moin from Germany,
time to remember two quotes from
Goldman Sachs
"Our Investment Banking backlog increased during the quarter and reached a new record level, surpassing for the first time the prior record set in the second quarter of 2000"
Bear Sterns
Average customer margin debt balances for the quarter ended May 31, 2007 reached a record average of $95.4 billion, up 40% from an average of $68.4 billion in the quarter ended May 31, 2006. Customer short balances averaged $101.9 billion during the second quarter of 2007, up 27% from an average of $80.2 billion in the second quarter of 2006
just the day before the Hedge Funds drama really impacted the market.......
Don't be too hard on Gretchen - she's one of the only people in a major national publication trying to link the housing and credit bubbles in a credible fashion. So she gets it wrong a bit... she's mostly wrong in the details - and, as an ubernerd supplicant, I have to say - it is complicated.
Tanta,
Thanks, I had no idea that these were managed or the potential diversity of the holdings. I guess I should have put a few more question marks in my post.
From Wikipedia: Collateralized debt obligatio - Wikipedia, the free encyclopedia
Mark to model - hark the twaddle
RE your 9:1 leverage at the BS funds. Not that aggressive, most prime brokers will let you play the LCDX market with 2.5% margin. Return should be 1%, so ROE of 40%. Except if one of those pesky cov-lite critters goes belly-up.
I'm not usually still drunk at 8:17 am.
But if you know the party is going to be that good BEFORE hand... please alert us. We're usually available.
FWIW - as for still being drunk at 8AM... my wife was at a business party Friday night, she actually had to 'work it' and left just as the 'official business' shifted into unofficial business (i.e. monkey business) so wasn't a witness but heard the next morning that two middle aged execs from competing client firms got in a fight - literal fist fight - over a younger female employee of one of the firms.
Yes alcohol was involved - enough to float a small yacht.
And all involved (save the young lady who was quite responsible - did nothing other than being present) were still semi-drunk, fully hung over the next morning for marketing seminars & meetings.
Man I wish I'd been there - I'd paid money (as in pay-per-view rates) to see those old fools go at it. Good thing they were out of shape & relatively harmless (unless they get behind the wheel of their of their Beamers that is).
Wow - I didn't know people sill did those kinds of things. Haven't those guys heard of lawyers?
Oh well back to our regular programing...
lama, be careful with wikepedia on some of that stuff.
This is more reliable, if denser:
http://www.mathematik.uni-ulm.de/finmath/ss_05/fe/JPMorganCDOHandbook.pdf
Why is everyone confused? Hell, the WSJ just printed a "definition" of a CDO the other day that made it sound just like a CMO.
That's one reason I keep "picking on" people like poor Gretchen, even though she does have a fan club. You see idiocies like that in "respectable" sources, you start to think you understand the situation. Frankly, us bloggers have to work ten times as hard as some of these paid media writers, because being just bloggers we don't come with the institutional credibility of the Times's business page. For some reason, Morgenson allows herself to get sloppy and carried away by buzz-phrase o' the day, but that's OK because it's the Times, I guess.
puntino, it's not so much the amount of leverage as it is blaming the losses on the mark to market instead of the leverage. My sense of some of the reporting going on was that we weren't taking into account how small that MTM loss could be and still wipe out a 10X leveraged investment.
So if a CDO is like a levered mutual fund of asset-backed bonds, can't somebody at least TRY to value it the same way, "net asset value"?
Closed-end funds can trade below or above NAV, but at least somebody knows what the NAV is supposed to be.
Who would invest in a CDO without knowing what the NAV is? what the leverage is and what the interest on the debt is?
Consider that, in Friday's conference call, Molinaro stated that the composition of the two funds' assets were roughly the same in quantity and quality. Also, consider that most of the 'real junk' from both funds had already been sold to Everquest.
So, the paper Merrill seized was probably of reasonable quality. The press accounts, although sketchy, seem to confirm this. We also saw press accounts that bids on the paper were unexpectedly low.
If it was paper heretofore considered to be of reasonable quality, and it sold at prices lower than expected, that has some interesting implications, doesn't it?
I still think this has the potential to become a Kreditanstalt-type event, and I think Bear Stearns will end up pumping even more money into the sinkhole.
mp-
I think the tell-tale will come around the second week in July when the statements/hedge updates go out, then we may see who had to take mark-downs, National made what was likely a very broad move, others may follow, what bothers me is what specifically triggered the National pricing move?
"two middle aged execs from competing client firms got in a fight - literal fist fight - over a younger female employee of one of the firms"
Hmmm... middle aged... out of shape... either one of them single? doubt it... younger female...
And we wonder why we're losing our competitive edge?
Tanta - I'm really glad someone has brought up 'models'. My skin crawls when I hear people say they've analyzed this & that and therefore the only likely outcome is [_____].
Especially when their analysis & model assumptions aren't 'transparent' nor 'available'.
Its not that I don't see value in models... its just that I only trust them when there are adequate REAL WORLD controls monitoring & managing the process being modeled... so that when (not if) there are 'variances' the system can still be held together.
I learned this distrust of models as a chemical engineer where we rigorously modeled chemical plants - fairly easy to do compared to complex unquantifiable open ended business models.
And I and my fellow workers were still able on occasion to overflow the tanks & send thousands of gallons into the streets & toward the Cedar River.
Another thing I learned - models aren't much good at placing sand bags. I don't know if the Times article mentions that as I haven't paid to look beyond the paywall either. Someone should call Gretchen & ask her.
risk capital, your point is well-taken, but I think we'll know even before that. If it does occur, there will be blood in the street.
Frankly, us bloggers have to work ten times as hard as some of these paid media writers, because being just bloggers we don't come with the institutional credibility of the Times's business page.
Besides - in the blogs you get ripped when you get sloppy (or in my case 'sloppier than usual'). Gretchen can hide behind the paywall & delete the emails she doesn't like - comment on those she does.
Life is good.
So, the paper Merrill seized was probably of reasonable quality. The press accounts, although sketchy, seem to confirm this. We also saw press accounts that bids on the paper were unexpectedly low.
If it was paper heretofore considered to be of reasonable quality, and it sold at prices lower than expected, that has some interesting implications, doesn't it?
This reminds me of the World Savings thing. World spent decades originating high-quality neg-am ARMs that were highly illiquid. Nobody else understood them; nobody else knew how to service them; nobody had a bid out for them. Yet those loans went along performing very well for years.
Then, all of a sudden, everybody and his pet kitty got interested in originating neg am loans, at precisely the time (the top of the RE market) when it wasn't such a hot idea. This had no particular effect on the credit quality of World's loans, but it sure did result in some bids. The owners of World, not having been born yesterday, took the money Wachovia was waving at them and ran with it.
Could World get the same bid today for its neg-am portfolio that it got a year ago? I doubt it. Has the credit quality of that portfolio gotten substantially worse since last year? I doubt it.
A whole bunch of people have gotten themselves confused about how liquidity works. The World portfolio was illiquid for a long time, then suddenly very liquid for a while, and now I suspect it'll stay on Wachovia's books for a loooong time, just as it stayed on World's books for a long time. But you can't map that run up and run down of the liquidity of that portfolio to its credit quality in any direct fashion.
The interesting question is when all these goofballs who wrote a bunch
of neg-am loans at the wrong time for the wrong borrowers discover that they are not World Savings and there is no Wachovia standing around to throw money at them.
I was also stunned at Morgenson statement about rating agencies marking to model. The rating agencies model the probability of loss to particular bondholders due to defaults and losses on the underlying collateral, not due to market spread-widenings or other fluctations. I thought she actually knew something.
Where rating agency models did miss was on their default curves. Some of these sub-prime loans went into default immediately, rather than following the seasoning curves estimated and assumed by the rating agencies.
Neal quoted this article yesterday, but it's worth a full read
CDO Boom Masks Subprime Losses, Abetted by S&P, Moody's, Fitch - Bloomberg.com
Well, now that I've been proven irrefutably wrong, I'll disappear. You know, the Sebastian Model.
Tanta
I don't see what the Bear problem has to do with either the rating agencies.
My understanding is that these hedge funds achieved leverage to illiquid but senior CDO tranches through margin debt.
The problem is that the subprime loans in the underlying pools have performed badly, so these highly rated securities aren't worth as much as everyone thought. In response, the investment banks have marked down the value of collateral, and asked for more margin.
You don't even need to mention the agencies here. Neither the investment bank or the hedge fund cares about the rating.
The interesting question is when all these goofballs who wrote a bunch
of neg-am loans at the wrong time for the wrong borrowers discover that they are not World Savings and there is no Wachovia standing around to throw money at them.
Well they only have a few options... none of them will prove out 'immediately'.
If the goofballs want to be able to make this stuff more liquid... the loans backing up the securities will have to perform & cashflow as per plan... the goofballs will have to provide visible evidence their product performance actually matches the model upon which the economics were based.
If so - then folks will wave money at them also.
I doubt they will wave money just because the models say the products SHOULD perform. I would hope we've moved beyond that.
But ya just never know.
Where rating agency models did miss was on their default curves.
Absolutely. I would go farther and say that this suggests that there may well be a big distribution problem looming, not just a timing problem.
In other words, what caused all those too-early defaults? Well, all the preliminary analyses agree: no docs and no down is what let all those doomed flippers and outright fraud cases and so on into the mortgage door.
So those loans "blow early." They also "blow often" if they're 45% of new production.
The problem is that the subprime loans in the underlying pools have performed badly, so these highly rated securities aren't worth as much as everyone thought. In response, the investment banks have marked down the value of collateral, and asked for more margin.
But is that "the problem"? What does "performed badly" mean?
If it means "realized losses," how many realized losses of senior tranches of mortgage-backed bonds have we seen lately?
If it means "current delinquencies and projected lifetime realized losses," where would we all be getting that information except the rating agencies?
I most certainly am not arguing that the only "surveillance" of the performance of these bonds should be the rating agencies. I am merely observing that apparently they are the only ones doing it.
"If it means "current delinquencies and projected lifetime realized losses," where would we all be getting that information except the rating agencies?"
Don't you believe it would be fair to assume that the third party pricing models have changed in some fashion?
ie the alledged national financial markdown?
more from bis-
"The ratio of banks' exposure to real estate as a proportion of total lending is about two thirds in Spain, slightly less than half in Germany, about one quarter in Britain and well over 50 percent in the United States and Australia, BIS data showed."
Banks face stern test when credit cycle turns : US Business
But is that "the problem"? What does "performed badly" mean?
=> This is not a hard question. The delinquency and loss curves for the 2006 vintage for the industry are the worst ever recorded. That information comes straight from the performance reports from the trusts, not the agencies.
It doesn't matter that the AAA trache has not and probably will never see losses. The point is that due to today's view of current and expected losses it is now a AA or A rated tranche. That means it is worth less, and when you buy on margin, it means either you put up more capital or get liquidated.
You know it's Sunday... and we've been blogging here all morning... and no one's gone to church to pray for a healthy real estate market.
No wonder things are going to hell in a handbasket.
I learned this distrust of models as a chemical engineer where we rigorously modeled chemical plants - fairly easy to do compared to complex unquantifiable open ended business models.
Exactly. I've done some work with modeling electronic systems. It all depends on how detailed the model is. You can have a model that seems to have performed quite nicely in the past... but then some unforeseen event happens that renders your model useless. A model is only as good as the assumptions that go into making it.
--
Thanks to NC, CR and Tanta.
Gretchen is one of my favorite reporters but she, like all almost all Americans, including CR and Tanta (whom I like), suffers from putting all the emphasis on systems, models, etc., and not on the people who control the system and models.
Good systems produce good leaders, relatively speaking. Bad systems bring bad people to the positions of power. Similarly, economic models are only as good as the people, especially, the managers of the businesses that produce models, behind them. Democracy is a good political system even when it re-elects GW Bush? Free Market is a good system to distribute fruits of labor commensurate with effort and risk even when tens of members of criminal gangs of bankers and financiers got higher compensation, each, than the TOTAL LIFE TIME compensation of all the 3,000+ soldiers who have dies in Iraq, without any risk to life or their own capital?
In America of the past dozen years or so the economic system, the political system, and the economic models of risk, are all being controlled by what I have termed the Criminal Gangs of Bankers and Financiers of New York City. Goldman Sachs is more evil than any of the Mafia gangs of the past. The difference is that Goldman Sachs owns shares in the American govt!
It is NOT the system, Stupid! And it aint the Models, Stupid!! The systems and the models are in the hands of an organized gang of criminals who suck the financial blood out of the men who work and risk their lives. These men are evil in case an American has a hard time in identifying what evildoers look like.
All organized human endeavors are subject to be controlled by a group of evil men. In America, we have reached that fateful point.
Jas
Speaking of Models, Weather models can't predict the weather 2 weeks in advance. Do you really think climate models can predict climate 50-75 years from now?
--
"Exactly. I've done some work with modeling electronic systems. It all depends on how detailed the model is. You can have a model that seems to have performed quite nicely in the past... but then some unforeseen event happens that renders your model useless. A model is only as good as the assumptions that go into making it."
I have some background is systems theory. I was going to do my Ph.D. work in non-linear systems but I got more interested in digital signal processing. Anyway, when things breakdown linear system theory doesn't apply. Limit Cycles are one bad beasts and that is where the US economy is heading. All these self-serving modelers will hit the walls of the limit cycle and circulate in hell.
The only risk worth discussing is the Uncalculated Risk! Calculated Risk will be down the toilet. No offense to CR.
American economy will be stuck is the limit cycle from hell from which it will only come out after the political system that supports the criminal gangs of bankers and financiers is toppled. It will happen during the lifetime of many here.
Jas
--
The secret of success of the bankers and financiers of New York City:
Push all the risk into the future and collect the rewards now.
What a system!
Jas
Jas, chill out! While I agree that some men/women are evil, most are just plain stupid, "doofuses" if you like. Doofuses overcome by greed and avarice, looking for the latest "story," which is what Wall Street sells.
Wall Street, hell, it's what most people sell.
I think it might be better as Incalculable Risk
One of the quality of this board is to keep comments on subject
Tanta,
It seems that you are confusing everyone with CDO definitions. Everything that you so far defined as CDO should really be called an "ABS CDO" (i.e. a CDO of various ABSs). There are other types of CDOs that DO hold mortgage debt directly and as a matter of fact, it is common to refer to CLOs and CMOs as speficic types of CDOs, even though they have one of degrees of tranching difference (which Tanta explained very well). In other words, a CDO^3 in corporate loan world has the same degree of tranching as CDO^2 in the mortgage/ABS world.
So everything you said only applies to ABS CDOs (whether they are real cash CDOs or synthetic).
What I'm really wondering about is how does the liquidity of risky tranches of CMOs and CLOs compare to that of ABS CDOs? In other words, is the Bear Stears crisis enough to cause concern in the world of CLOs? That's the $64,000 question in my mind because corporate lending is on steroids right now.
Jas
Have you ever heard of Carl Schmitt and Leo Strauss, they are the evil heart of darkness that guides this administration. Carl Schmitt believed in his idea of the CONCEPT OF THE POLITICAL, where politics is not about debate but life or death. Leo believed in eternal war as a way of controling the masses, and creating an uber executive branch of government.
This administration will create a new reality for us all by going to war with Iran.
I lean more with Jas on this one.
Can anyone look at the face of Dick Cheney and not see an incredibly mean, evil man??
I'm serious about that. How about Henry Paulson?
I truly see a disturbed, lying crook. And that's not because he is a doofus. He is pure evil.
Wayne,
I am familiar with Leo Strauss, the intellectual father of the Neo-CONS.
Most of the evildoers in America are cut from the same cloth. Neo-CONS have total control over W's mind (I apologize to all Americans for voting for W and supporting him with money and efforts in 2000; I recognized my mistake three months after W took office).
I was born ignorant and I am sure to die as a grossly ignorant man. I just try to lessen my ignorance a bit by bit. It is the people with blind faiths that I see problem with.
Jas
--
"I truly see a disturbed, lying crook. And that's not because he is a doofus. He is pure evil"
For the past few decades the conditions have been such that only evil men can reach the top of the economic and political power in America.
Jas
--
In one sentence, Why Are Americans Soooooooo Stupid and for How Long?
Because, thanks to the free market, there is lot of money, as in several trillions of dollars, in it and the stupidity will cease only after all the stupid money has been sucked, i.e., 80%+ of the American households have very little networth left to speak of!
Jas
"Speaking of Models, Weather models can't predict the weather 2 weeks in advance. Do you really think climate models can predict climate 50-75 years from now?"
Sure I do. Seeing as it is aggregate behavior and not specific behavior. Models are far better at aggregate behavior (which are highly tolerant of simplifying assumptions) than they are at specific behavior (which is where the butterfly effect kicks in).
Models are not completely worthless. Otherwise people wouldn't have started using them. They just have their limitations. The problem is that our math illiterate population doesn't know what those limitations are. So we get ridiculous statements like the one quoted above.
Jas
I think your message would be more easily received if you just simply described the facts. The facts alone tell the story surely? I wonder if most people even know?
In a nutshell, from the JPM link above; this is why CDOs exist:
"In an arbitrage CDO, equity tranches allow non-recourse term financing of the underlying assets. If the CDO assets perform poorly, debt tranche holders have no recourse...this is in contrast to the repo market, where the creditor has recourse to the borrower."
What this means: a CDO comes levered 10:1, and if it fails the hedge fund is not on the hook for any more than its investment. Think of it as
LEVERAGE IN A BOX
except it does not stop there. The hedge fund takes on another 10:1 leverage through repo financing from its prime broker. Total leverage on the equity tranche: 100:1. That is why they are so toxic.
Worse than I thought, really.
I guess we get some remit numbers tomorrow,no?
Tanta,
I was curious to see how these CDO things were actually valued. I found a presentation at
404 Not Found
which I interpret as follows
How a single security CDS is valued:
Value = Premium_PV - Loss_PV
- PV: Present Value
- SUM_OF: the summation function
- TIME: number of days
- DF: Discount Factor (for some future TIME)
- SF: Survival Factor (probability of creditor making payment at TIME)
- R: Recovery value
- ICS: Incremental Change in SF (for a given TIME)
- Premium_PV = payment * SUM_OF( TIME * DF * SF)
- Loss_PV = (1-R) * SUM_OF(DF * ICS)
Think of the survival factor as coming from a seasoning curve. If actual losses are higher, your Premium_PV decreases and the Loss_PV increases a double hit.
CDOs can be valued in a similar manner but, since there are many securities in the underlying pool, you must consider the degree of default correlation between these securities (potentially a really hard problem). Also, since CDOs are tranched (is that really a verb?), the difficulty of the calculation is further increased.
Two questions: 1) Do I have this right? and 2) Absent knowing EXACTLY what is in the CDO right now, how the hell can ANYBODY tell what its worth?
TIA
FWIW-
Had brunch this morning with a friend who is best buddies with a guy who worked at Brookstreet. So this is purely for entertainment value.
Apparently the salesmen didn't have a good grip on the risk of the investments they were selling - so what they assumed was moderate leverage wasn't. One of this guys' clients went from $150k to -$300k overnight!
It seems that you are confusing everyone with CDO definitions.
That wouldn't surprise me.
I personally have never used the term CMO and CDO interchangeably. I am not aware that this is uncommon.
Basically, if it's a tranched bond backed by an undivided interest in a brain-dead pool of mortgage loans or mortgage-backed securities owned by the trust that makes cash-flow distributions to the bondholder, then it's a CMO.
You want to say a CMO is a type of CDO? Fine. To me that just confuses the issue.
I use CDO to refer to a securitization that cannot be considered a CMO, either because 1) it doesn't own all the underlying assets (it's "synthetic" or "hybrid") and/or 2) the asset pool isn't brain-dead (it's managed) and/or 3) the asset pool isn't all mortgages.
I thought that was actually less confusing than considering the CMO to be a type of CDO. Actually, I usually just avoid the term CMO entirely; since most of the recent CMOs are REMICs, I use REMIC, which is harder to confuse with CDO or CLO or CBO.
The fact of the matter is that the press keeps telling us the BS hedge funds' collateral has something to do with subprime mortgages. That would certainly imply that what they hold are CMOs of some sort. But I can't quite figure out, then, how come it's also supposed to be so illiquid and hard to mark.
Is the CMO market that illiquid? Really?
I guess I just assumed--generally a mistake, but there you go--that all this hoo-ha over the illiquidity of the stuff suggested that we were probably talking about hybrid CDOs, not (relatively) boring old CMOs. The snippet from the MER auction list that got published last week sure didn't look like CMOs to me.
I haven't the faintest idea what any of this might mean for CLOs. Once we wander too far off the mortgage reservation I get lost easily.
Per the Bloomberg article I quoted yesterdy and JBL kindly linked above: 25% of the lowest INVESTMENT (not the bottom of the barrel) grade CDO's go bad-and that was during the up years in real estate.
Figure this out-if 25% get wiped out what rate of return do you need on the surviving to make a 10% return? If 50% go bad, what rate of return do you then need? Do you think those returns are achievable in the current market?
As for CDO^2 and CDO^3, read the article, there is no current way of modeling them. How do you think those will work out?
That's why it's pennies on the dollar at the fire sales. It's not only that people don't want to have actual value known, it's people having no way of evaluating actual value.
It the modern "fish tank" at the carnival. You mostly end up disappointed.
Interesting read from Barrons.
As of March 31, 2007, MBIA had insured $5.4 billion in subprime mortgage-backed securities, or MBS, which receive cash flow from underlying pools of mortgages. Here, MBIA probably faces negligible risk, since it generally insures higher-rated classes, or tranches, of MBS, including the triple-A bonds that have first dibs on the flow of mortgage interest and principal payments cascading down the payment waterfall.
But MBIA could be in significant peril as a result of its guarantee of a class of subprime-mortgage derivatives called collateralized debt obligations, or CDOs. These instruments represent a degree of separation from underlying mortgage-backed securities, since CDOs are constructed from individual classes hived off from different MBS or, in the case of "CDOs-Squared," classes of other CDOs.
On its Website, MBIA said that as of year end, some $2.4 billion of the total $7.7 billion in mortgage CDOs it has insured were backed by subprime mortgages. The $2.4 billion of CDOs, in turn, is a mix of "mezzanine" and "high-grade" CDOs, according to the company. Mezzanine portfolios are comprised of MBS tranches rated no higher than triple-B, while about two-thirds of high-grade CDOs are made up of single-A tranches.
1) Do I have this right?
I'm touched you think I can answer that question.
and 2) Absent knowing EXACTLY what is in the CDO right now, how the hell can ANYBODY tell what its worth?
Good question! And if we don't know what's in it, how the hell do we know that its value has deteriorated because of subprime mortgage losses?
There has been a lot written on by others, fleck included that these securities cannot be marked down unless a downgrade occurs, I am going to take issue with this after careful thought due to what apparently happened at brookstreet, National makred these securities in question to market based on their pricing model if what has been reported is true. That said, they are marking to market based on trades occurring in my opinion. Currently, this is possibly just a few different securities that have been effected, this would also lend credence to the aim to not allow the auction securities trade at reduced values in the BS case (due to the realization that this was in fact taking place at National) which would have set off a potential reaction that would not have been pleasant.
In my opinion, all of the information that has been brandied about concerning waiting for the agencies is incorrect.
The question now is when is the garbage coming to market and at what price, because in my opinion, with or without the ratings agencies help, it is going to be marked, thus the reaction recently in the marketplace.
This is not over and may not be for quite sometime.
Compounding the issue is that many deals that have been announced recently need to be funded, the credit markets are moving quite fast recently and this could shape the environment in coming weeks dramatically, if one of these deals falls apart, it could be rough sledding.
I see... well basically, in corporate land, a CDO is a pool of loans or a pool of CDS on loans. I think that's why there's confusion. People from the corporate lending world will think that a mortgage CDO is an instrument that buys whole loans or something. anyway...
"I think your message would be more easily received if you just simply described the facts. The facts alone tell the story surely? I wonder if most people even know?"
Thanks.
Facts behind certain generalizations, or general observations, are easy to find if one makes the minimal efforts. For those who don't like certain conclusions no facts are enough.
I presented REAL FACTS, i.e., data, to CR, a man I like and respect, about the Fundamental Demand for house-dwellings but he is married to his estimates. I say, why estimate when you can count?! Especially, when someone has been doing the counting, via regular surveys, for you and reporting regularly. CR doesn't want to know that number of Occupied Housing Units could go down even when the population and households go up. Is it a theoretical impossibility? Hell, no. It is an observable and easy to verify fact that when costs skyrocket, or economy weakens, the demand for housing units that are occupied is pushed down. He is a perfect example of practitioners of the Dismal Science. It is not personal; it is strictly business.
FACTS:
Increase in Vacant Units, Year Round, over the past 13 Quarters = 2 million!
Units Completed, over the same 13 Quarters = 6.2M.
For those who are arithmetically challenged the actual demand during the past 13 Quarters was close to 1.3M. When a yuppie woman in her 30s jointly buys a home with her father (at the absolute peak for the neighborhood) and moves in with him there must be a compelling economic force at play, no? Poor woman didnt have enough space in the 2-car garage for her Beemer! I put very little faith in theories and lot of emphasis on facts and the best available data.
If the recession begins in 2007 and depression in 2008, as I predict, the demand for 2004-2010 would be below 0.5M a year. At the end of 2010, in that case, there should be 7M more Vacant Units, Year Round, than there were at the end of 2003.
Sorry, facts dont matter to believers! America is a nation of true believers more so than Saudi Arabia!
Jas
well basically, in corporate land, a CDO is a pool of loans or a pool of CDS on loans. I think that's why there's confusion. People from the corporate lending world will think that a mortgage CDO is an instrument that buys whole loans or something.
probert, doin his part to confuse people.
Getting back to "models," I don't really think the problems are in the models per se, but the assumptions that go into those models. How are you going to value a CMO or a CDO for that matter without using a computer to project cashflows and run them through payment rules? You can't just use a price from a recent trade in the market because there are so many other factors that go into it such as who originated the collateral and the historical and future performance expectation. You can, however, use a credit spread that you've seen on a bond of the same rating and similar collateral, and that's what is done. Inputs are default/loss assumptions, prepayment assumptions, index levels, a yield curve and this credit spread which is basically a measure of the riskiness compared to treasuries or LIBOR. Traders come up with these numbers based on where the market is, type of collateral, originator and historical performance of that collateral, expected performance, etc. What we have been seeing since the beginning of the year is that the default/loss curves used were way off. What we are seeing now is that credit spreads are widening very quickly, and that could cause a mass re-marking of levered portfolios which is far more dangerouse to the financial system.
--
"Getting back to "models," I don't really think the problems are in the models per se, but the assumptions that go into those models."
That is precisely where the common human weaknesses come into play.
ALL COMMENTARIES ON ECONOMICS, INVESTMENTS, AND POLITICS IS ABOUT HUMAN BEHAVIOR AND NOT ABOUT VARIOUS THEORIES. Guess, who create the theories?
Quote: "Every bubble has its academic shield."
The most profitable business in America is the Propaganda Business. Those who control the propaganda machine control the economy and the govt. Needless to say that they also have a iron grip on the minds and money of most Americans.
Jas
--But MBIA could be in significant peril--
no,no,no
No one is in peril... this is America , the can't lose, won't lose, wish upon a star , everyone always wins economy....
And for proof of that , look no further than the OCC report from the fed... Pg 5
Credit losses from derivatives contracts are nearly always quite small, if not ZERO...
see ,the fed has said soooo
no ONE CAN LOSE, we forbid it...
http://www.occ.treas.gov/ftp/deriv/dq406.pdf
18 cents on the dollar, it could be worse....
how about 1 cent on the dollar?
Topic Galleries -- South Florida Sun-Sentinel.com
Mason-Rosner on MBS and the inherent risk-
Link
Back in the day, when straight corporate and municipal debt was most of what the ratings agencies assessed, you could usually count on two things. (1) Ratings were highly correlated with the ratio of cash flow coverage to interest payable; (2) Ratings were always slow to change and followed the market. Still, the agencies were slow if straight and all bond market professionals knew the game.
The road to hell began when the agencies got a hand in designing structured finance so that ratings targets could be met -- that made them consultants to issuers in a more blatant way. And the descent continued from there.
Judge Drain is going to have a full Calendar
I see... well basically, in corporate land, a CDO is a pool of loans
I would like remind the readers that these are not just a pool of loans, but rather a pool of liar loans. And the value of the collateral backing them is worth 90%...89%...88%.......86%............
(cynical man) Middle aged men in a fistfight over a pretty young woman dryfly? Now that's stupid. They should have reached into their pants to see who had the bigger credit limit. (/cynical man)
"The road to hell began when the agencies got a hand in designing structured finance so that ratings targets could be met -- that made them consultants to issuers in a more blatant way. And the descent continued from there."
So who's going to be Arthur Andersen?
(maybe nobody because there are even fewer ratings agencies than big accounting firms?)
Not sure how much of this I follow, but it seems reasonably evident that valuing or rating these issues necessarily entails some highly uncertain assumptions about the liklihood of timely repayment of the underlying debt, and all this talk of "models" is designed to obfuscate the subjectivity of the inquiry. In view of the mass of anecdotal information coming to light about the misconduct of the originators of many of these loans, one might wonder if the rating agencies have not undertaken, for a buck, to rate the unratable.
lama said: "Well, now that I've been proven irrefutably wrong, I'll disappear. You know, the Sebastian Model."
Meanwhile, no recession, no 400k-600k in residential construction job losses, virtually unchanged existing home sales from last month, and no stock market collapse.
Sebastian