June 29, 2007
CDOs: what's the big deal?
Here are my two cents on concerns about possible systemic financial problems.
There were some very useful discussions yesterday on collateralized debt obligations, financial instruments in which an institution creates a new series of assets by repackaging the cash flows received from a given pool of underlying assets (some of which might themselves be CDOs). Felix Salmon tries to categorize calmly the various concerns in terms of possible defaults on underlying subprime mortgages, defaults on the part of the institution that issued the CDO, and mispricing that could result from "fire sales" of such assets. Felix acknowledges a potential cycling feedback between these three developments, and concludes:
So there is cause for concern, to be sure. But there isn't cause for panic.
Saskia Scholtes and Gillian Tett, writing for the Financial Times, characterize the problem as an absence of independent arms-length valuation of the CDOs, so that holders do not recognize or warn their creditors of the magnitude of already-realized losses. Naked Capitalism sounds a related theme, emphasizing the lack of public documentation and transparency. Tanta suggests that references to the "illiquidity" of such assets are euphemizing the reality that their objective market value is far below the book valuation.
The benign view of CDOs is that they represent an important technological improvement that allows for better pooling of risk. I would characterize the main concerns as centering on whether in the process existing financial arrangements have accurately priced aggregate risk.
First let me clarify what I mean by aggregate risk. Some risks are inherently undiversifiable. One can understand that point most clearly with Robert Lucas's elegant asset-pricing model. Suppose that the entire economy consisted of one big potato farm. All financial assets would then ultimately be nothing more than claims against future potato production, and there is no way they can credibly promise to deliver more potatoes than are actually available. If there is a bad harvest, people will be hungry, and no clever set of financial instruments can possibly insure you against that.
An example of the kind of aggregate risk I have in mind is thinking through what would be the consequences of, say, a 20% decline in average U.S. real estate prices, and what that might mean for default rates.
Let me next clarify what I mean by mispricing of this risk. I am not concerned about whether those who are bearing the aggregate risk (i.e., setting themselves up to be the guy who has the fewest potatoes when times are bad) earn a higher expected return to compensate them for the risk. Rather, my concern is whether they may have invested in assets with a negative expected return. For example, if you purchased an asset with an 85% chance of a 15% return (which you'll earn as long as a recession does not occur), and a 15% chance of a -100% return (you're wiped out if it does), then your expected return would be -2.25%.
The specific kind of example that comes to mind is New Century Financial Corporation, which was pushed into bankruptcy from a substantially more modest aggregate shock than the one I am concerned about here. The concern about mispricing is that if loans were extended that should not have been, the magnitude of both the real estate boom and its subsequent bust are amplified substantially relative to what they would have been with accurate pricing of aggregate risk.
But who would be so foolish to have invested hundreds of billions of dollars in extra risky assets with negative expected returns? The logical answer would appear to be-- someone who did not understand that they were accepting this risk.
So here is the heart of the question that troubles me-- who is the residual bearer of risk if there is a substantial decline in U.S. real-estate prices?
To the extent that the answer might be the taxpayers who are liable for government employee pension commitments (, ) or a perceived public fiscal commitment to the government sponsored enterprises, then I think a reasonable case can be made that the bearers of the risk were indeed insufficiently informed.
If there has been some mispricing of aggregate risk, then the magnitude of the economic consequences of the housing cycle are going to be greatly amplified. Indeed, the reason I have been developing increasing concern about this possibility is that it looks to me like we are already seeing evidence of just such effects. The ongoing woes of the housing market are clearly now reflecting something much more than a simple response to higher interest rates. But all this has happened without any significant undiversifiable shock, such as an actual recession or broad decline in real-estate values. What will happen when such a shock finally does arrive?
If any of our readers can see the answer to these questions more clearly than I can, I would welcome your insights and reassurance that all is well.
Posted by James Hamilton at June 29, 2007 01:57 PMdigg this | reddit
I can't give you much comfort, but perhaps I can add to the discussion.
A friend wrote the following: Regarding the real estate run-up - Circa 2000/2001, the Greenspan Fed saw flat earnings growth and decreased demand for commercial/industrial loans in the corporate economy and eyed the balance sheet of the homeowner in order to goose consumer spending, affecting allocation of capital at the margin. With falling interest rates, consumers were able to expand their overall debt burden while focusing on the lower payment structure. With the Fed pushing down short-term rates, this enabled the financial sector to engage in borrow-short/lend-long practices that were very attractive from a risk-reward perspective.
My addition to this is that over the past few years the FED also provided an increase in the money supply to help expand the housing "bubble" they created with their interest rate policies.
Now in true FED form what they gave with one hand they are taking away with the other. The FED has tightened the loan market and the previous boom they created is now becoming a serious bust.
And as they do so often they have painted themselves into a corner. They fear the inflation they created so they hold interest rates high pricking the very housing "bubble" they created.
I have no confidence that the FED will do the right thing and lower interest rates. Bernanke is faced with a Democrat congress that is chomping at the bit to drag the economy down farther with bad fiscal policy and he has almost said this explicitly.
If we come through this without a serious downturn I will be surprised. There are just too many interests competing to take the economy down.
Were it not for the Supply Side effects of the Bush tax cuts we would already be in a severe recession.
Posted by: DickF at June 29, 2007 02:45 PM
On a contrary note, I don't see the potential for a systematic economic collapse sparked by CDO's and housing price deflation.
CDO's are a new, relatively small and incredibly complex set of instruments. Small, somewhat isolated debacles don't usually generate enough external costs cross-institutions and cross-markets to get an attention grabbing contagion going.
And owner-occupied housing with 10%+ equity down are resilient investments - most people just aren't going to walk out of the house they own if they've sunk 10% equity or more into it. People need a place to live as well as an investment and they can always convince themselves that values will rebound, someday. The problem for this most significant cohort of homeowners is making the payment on their mortgage - as long as they have jobs they can do that, UNLESS they've done some crazy low-low starter mortgage to get in the house and then the variable kicker spirals upward due to some financial problem . . . . .
I'd recommend watching credit spreads - if/when those spike up 100-200 bp in short order, watch the hedge fund avalanche start to snowball THEN.
Posted by: Anarchus at June 29, 2007 05:22 PM
A young man swims the Rio Grande east of El Paso.
A week or so later he has made his way to his cousin's little shack in Fresno.
The next week the cousin introduces the new arrival to the head gardener at an apartment complex in Merced, where the cousin also works as the "second handyman." The new arrival gets the job, paying $9.50/hour "cash money." And now he spends his days walking around the complex with a leafblower on his back.
A few months later he picks up a flyer someone had placed on a door in the complex (one of many).
It reads, "you too can make a million dollars in real estate." He asks his cousin about it. The cousin tells him that he has a friend who claims to have bought a house in Modesto a year ago with no money down and that the house now has $100000 in "equity." The cousin tells him that he has been thinking about trying to do this too.
The next week, both the new arrival and the cousin go down to the address on the flyer, where a realtor and a mortgage broker give them a presentation, much of which they do not understand. But what they do understand is that they do not need any money and that they are both going to become very rich. And after all, that's why they came to America, right? Right!
Two weeks later they return to the office and sign a series of papers, and at the end of a subsequent two weeks each becomes the owner of a $400000 brand new 1500 sqft home in a subdivision in Ceres. And the two houses are even accross the street from each other!!!
The date is November 1, 2005.
Now I can go on with the second half of this story but the majority of the readers here know exactly what it will entail, so I will just say this...
the only way this works out for the new arrival and the cousin and for the rest of us is if the Greenspan/Burnanke put is alive and well and if the immigration reform ploy rises from the dead once again (and this time actually passes). What is needed is much more debt and many more warm bodies...and quickly.
Posted by: esb at June 29, 2007 05:27 PM
It may be a appropriate to post here also a comment I posted on macroblog yesterday.
I believe the ultimate price drop in housing will be much more than 20% -- probably more than 30%.
What we're seeing now is just the leading edge of the mortgage default wave, and foreclosures are not yet impacting market prices in most areas. But the supply glut alone is going to force prices far down.
Take Arlington Heights, IL for example. There are now 35 homes listed at prices over $1 million, with five more at $999k or $995k. The number of sales recorded as of May 11 in that price range? Just one. Between $900k and a million? Three. The corresponding numbers for all of 2006? Six and eleven.
Shall we do a little gedanken experiment?
Suppose the Top 40 listings on the MLS do finally sell this year, for the same prices as the Top 40 actual sales of 2006 (though it's clear even that would be optimistic). Let's line them up below and see the deltas line by line. We find that the average haircut off the asking price would be $350k, and is $250k even down at the 40th line -- one of the $995k homes would have to go for $750k. The average price drop is 28%
And this would be just the impact of oversupply -- foreclosures aren't even in the picture yet.
2007 Asking 2006 Sale Delta
$2,199,000 $2,478,000 ($279,000)
$1,899,900 $1,160,000 $739,900
$1,690,000 $1,100,000 $590,000
$1,580,872 $1,100,000 $480,872
$1,549,000 $1,040,000 $509,000
$1,499,000 $1,040,000 $459,000
$1,499,000 $994,000 $505,000
$1,490,000 $992,500 $497,500
$1,449,000 $955,000 $494,000
$1,425,000 $950,000 $475,000
$1,350,000 $950,000 $400,000
$1,350,000 $930,000 $420,000
$1,299,900 $915,000 $384,900
$1,299,000 $905,000 $394,000
$1,290,000 $900,500 $389,500
$1,274,900 $900,000 $374,900
$1,250,000 $900,000 $350,000
$1,250,000 $889,000 $361,000
$1,249,000 $880,000 $369,000
$1,229,000 $880,000 $349,000
$1,199,900 $875,000 $324,900
$1,199,000 $871,500 $327,500
$1,198,872 $865,000 $333,872
$1,195,000 $860,000 $335,000
$1,185,000 $855,000 $330,000
$1,185,000 $850,000 $335,000
$1,175,000 $850,000 $325,000
$1,149,000 $835,000 $314,000
$1,149,000 $825,000 $324,000
$1,125,000 $825,000 $300,000
$1,099,000 $810,000 $289,000
$1,089,000 $805,000 $284,000
$1,059,900 $797,500 $262,400
$1,049,000 $796,000 $253,000
$1,025,900 $787,500 $238,400
$999,000 $775,000 $224,000
$999,000 $774,000 $225,000
$999,000 $772,000 $227,000
$995,000 $762,500 $232,500
$995,000 $750,000 $245,000
Sum of deltas: $13,993,144
Average delta: $349,829
If someone is about to contend that these homes are owned by rich people who will be able to wait forever to get their wishing price, note that the MLS photos show more than 70% of these homes to be vacant, and most are new construction on teardown lots.
Posted by: jm at June 29, 2007 07:12 PM
The mispricing of aggregate risk occurs because model used to price risk do not factor in leverage.
Instead, they take the historical volatility of model factors as an input. The "output" volatility of returns is then used to calculate optimal leverage.
Leverage creates more demand for securitizations, lower spreads, greater credit availability, more refi's, less defaults, LOWER volatility. The model then advises HIGHER leverage. The feedback loop runs again, resulting in yet HIGHER leverage.
The problem is that the models don't take into account the effect of the increased leverage on the volatility of model inputs.
A short way of saying this is, "house prices are sticky, regardless of the leverage of creditors."
I'm sure you can do a better job of explaining why the above statement cannot be true.
Posted by: David Pearson at June 30, 2007 07:18 AM
I think the problem is that we have no clue as to How big a deal these may be.
I have some questions here :
Posted by: Barry Ritholtz at June 30, 2007 09:47 AM
Dr Hamilton (or Barry R)... anyone have any idea how much 'unsupported' leverage out there? Isn't that also part of the key?
1) The underlying collateral values (and will they hold)? RE & mortgages primarily...
2) How much leverage is tied to that collateral (in question 1)?
Does anyone even have a WAG on either?
Posted by: dryfly at June 30, 2007 11:10 AM
Well, Barry Ritholtz did a far better job than I could at laying out the reasons why this is a big deal.
But I would add that if you don't think it's time to panic, one may infer that you do not hold any of these investments. Of course, I suppose that for holders, the time to panic is past, and it's time to do the laundry.
Posted by: Charles at June 30, 2007 11:59 AM
See this article.
Posted by: Bonddad at June 30, 2007 02:10 PM
One unpleasant factor is the extent to which CDOs invest in each other. Each CDO has a difficult to move tranche that offers high yields and all CDOs want higher yields so even without collusion other CDOs are the perfect customer for the difficult parts. The result is hidden leverage and the results can be painful and similar to the recent UK split capital trust debacle and the Goldman Sachs/Shenandoah/Blue Ridge afair of 1928.
If so the effect on CDO prices will be more severe and long lasting than the actual mortgage losses would suggest and funding by some kinds of CDO would likely dry up for a time and remain more expensive.
Posted by: Jack at July 1, 2007 12:06 AM
From 04 to 06 I was involved in many Subprime, ABS, ABS CDO transactions and I am sure that most of the buyers (long/short) of these risks had no idea of what the hell they were getting into. I remember a Japanese investor that did not even understand English buying at the equity tranche of an ABS CDO,.... Keep in mind most ABS market participants are bond traders at best and have no idea of the derivative nature of the business.
There is now an active market on Subprime Correlation risk (TABX), and I am more interested in seeing how much and how well people are engaging in this product. How do you determine the default correlation of a New Century pool of loans with a Countrywide pool of loans...And yes it's the same cash bond traders that are gaussian copula to trade these instruments.
Posted by: Zar at July 1, 2007 08:49 PM
I think its important for people to understand that CDOs are cash flow instruments (usually anyway). The CDO collects all the cash flow from the collateral portfolio, then distributes it among the tranches. A M2M decline in the collateral portfolio usually doesn't trigger anything.
I'm reading a lot on the blogosphere and MSM that doesn't seem to grasp the consequence of this, and/or is mixing up the concept of M2M risk of the CDO debt holder vs. M2M risk of the CDO transaction itself.
The lack of M2M in a CDO means that a collateral bond can go from par to $10 without anyone taking a loss. But obviously if the bond is trading at $10, default is imminent. In essence, the CDO waits until the default actually happens, and as a result losses can suddenly be passed through all at once.
Also, junior CDO tranches tend to default very suddenly, particularly in more highly leveraged RMBS deals. Its not uncommon to see a deal where a tranche originally rated BBB does fine at 4% defaults but is worthless at 6% defaults. That's why its so hard to figure the value of these tranches, because any slight variation in default rates has a monstrous impact on value.
Posted by: TDDG at July 2, 2007 06:38 AM
This is the first set of exchanges about CDOs that I have found on the web that appears to be both measured and informed. I come from a background outside economics and I am based in the UK. I have a number of basic questions relating to the potential scale of this issue as it affects Europe in general and the UK in particular. Is there any way , any source that maps the amount of money backed by CDOs against the amount of money that is backed by other forms of commercial bond? And is there any way of finding out the level of exposure held by the different parts of the banking system?
It strikes me that the proportion of cashflow at risk is the critical issue and there appears to be no obvious way of finding this out
Posted by: stephen Gage at July 2, 2007 12:51 PM
James Hamilton wrote: "But who would be so foolish to have invested hundreds of billions of dollars in extra risky assets with negative expected returns? The logical answer would appear to be-- someone who did not understand that they were accepting this risk."
I disagree. Who indeed? Why, simply every lottery ticket buyer, every gambler in Las Vegas, every speculator on Wall Street, and every dreaming asset manager who thinks he can play the game and get out while he's ahead.
Posted by: Katy at July 4, 2007 04:19 PM
Excellent coverage and comments with your view & reference to Bonddad & Ritholtz views.
Average Collateral Pool Exposure To Subprime RBMS by vintage:
Cash Flow & Hybrid Mezzanie CDO of ABS by year 03:41.2%; 04:44.5%; 05:52.4%; 06:70.6%
Synthetic Mezzanine CDO of ABS by year 03:37.1%; 04:48.5%; 05:64.5%; 06:77.1%
Hit the link for more, if you dare.
Posted by: The Nattering Naybob at July 12, 2007 05:18 PM