CDO's: A common sense question

This is a common sense look at the subject. I am not a Wall Streeter, just a real estate broker and investor. I don’t think the sky is falling. Correct me when I go wrong on this simplistic analysis. And I am assuming that these CDO’s consist of first mortgages. If someone has made CDO’s out of aggressively written subordinate notes then they are just paying a high tax on their extreme stupidity.

A Collateralized Debt Obligations (CDO) is just a large group of mortgages. It gives the holder of the instrument a flow of cash, primarily from principal and interest payments on the individual mortagages which are held. A certain percentage of these notes are expected to be in arrears each month and a certain percentage of the notes are expected to be in every stage of foreclosure. Foreclosures must be expected and must be factored into the price which was paid for the instrument (a bag o’ mortgages).

According to the news, the problem is that these mortgages are going into default at a higher than expected rate. So what? Defaults are higher because the market has slowed down/backed up. Did the Wall Street geniuses not consider that the market could slow?

I will use figures recently given by two frequent posters on the board; Bluemoon06 and petemfa. According to Bluemoon, 14% of the market is subprime and less than 10% of the subprime is in foreclosure. So less than 1.4% of the market is subprimes in foreclosure. Doesn’t sound scary to me. I would expect it to be higher.

Now, petemfa, who has a scared-to-death buddy on Wall Street, reports that “computer models” state that some of their CDO’s have a value of only $.03 on the dollar.

Say what???

The C in CDO stands for COLLATERALIZED. How can the value of a collateralized note be only 3% of face value?

These CDO’s consist of a large number of mortgages, right? A mortgage is a promise to pay over a period of time and it is collateralized by the property. No pay, then the lienholder can sell the property to minimize their losses.

Lets just look at how the lienholder fares on one crappy hypothetical note. Say the buyer of the property pays only 10% down on his $200,000 house so the note is for $180,000. Now assume that the buyer never made a payment and the amount owed has ballooned up to $200,000 by the time the lienholder can foreclose and liquidate the property. Now let’s assume that the lienholder sells the property for $150,000. There is a loss, but it is mitigated by the fact that it is only one of a large group of mortgages and it is far less than a total loss even when only looking at that one note.

Meanwhile, on the vast majority of the mortgages, the payments come in like clockwork.

So here is a question for petemfa: Can you give me a common sense explanation of how a CDO can possibly be valued at 3% of any relevant figure (ie, face value)?


Great post! However, Petefma will not be able to respond as he is no longer a member of this site.


Thanks for your logical and calm approach to the CDO question. Let me just add this about CDO’s; these are not just pools of mortgage loans where the owner receives both principal and interest and suffers the typical late payments and foreclosures.

CDO’s can be cut up into ‘tranches’ or slices and each slice has its own charachteristics and risk. Some slices might get the principal only payments for a certain amount of time, others might get the interest only payments for a certain amount of time and still others might receive what is called the residual, anything left over after all is said and done and a few of the loans have outlived their expected lives.

Just remember a group of mortgages with X charachteristics can be sliced and diced into a hundred different CDO tranches which might be attractive to certain types of buyers. In the end it is up to the buyer to understand the potential risk of their investment and what a change in market expectations might do to that risk.

I think much of the loss we are hearing about is simply the market over correcting for volatility that they don’t quite understand. The typical corporate/human reaction to the unknown which usually plays itself out over time. This of course creates great opportunities to people and firms with cash and patience and a little understanding of the underlying securities.


I do realize that these pools are sliced into different subsets, although I must admit that the word “tranches” is a new one to me. That could be a problem for my father-in-law next time we whip out the Scrabble board.

Anyway, I was trying to take on the big picture look at the issue. If you go too deep then nobody wants to wade through all the details.

The main thing which motivated me to post is the guy who contended that these things had been devalued by 97% by a computer model. That pegs out my BS Meter.

Hi all,
To amplify on 71tr’s very good post just a bit…The tranches he refers to are normally classified - in descending order of rights to the cash flow - as senior, mezzanine, equity and junior. The senior tranche owners are generally entitled to approximately 75% of the cash flow and it goes down from there. I suppose it’s entirely possible that the bottom rung might be looking at receiving pennies on the dollar; perhaps that’s what Petemfa’s buddy was referring to.

71tr is also absolutely right in writing that the buyers may not understand what it is they’ve got but that they should, so fear and panic - leading to volatility - sets in. And fear, rational or otherwise, is still fear and consequences arising from it are real. I guess it’s like the child that swears there’s a monster under the bed - there isn’t, but try telling the screaming child that!


Steve…I’m still laughing at the scrabble comment, thanks. I know what you mean about keeping people’s attention. This stuff can get pretty dry at times. I thought the tranche comment relevant to address your BS meter which is right on. That 97% loss is likely one of the very bottom tranches that is some interest only, residual or when hell freezes over speculative investment. Tough cookies for the buyer but he should have recognized the risk.

One of the other aspects of CDO’s that most people aren’t focusing on is leverage. One of the ways these CDO’s promise much higher returns than similarly rated vanilla fixed income products is by levering themselves up. The “computer models” say that these CDO’s are worth X. The problem is that once some of these CDo’s de-lever and unwind and, there is a glut of paper for sale. The models say that the sum of the pieces of paper should equal X, but since there is so much of it out there, no one will pay X. I know, I have been trading a lot of the corporate backed CDO’s recently at a Wall Street shop. The problem is not valuation, it is liquidity.

Brian, need some help here…While I agree that liquidity is a problem for these CDO’s now - as in no one wants to buy them and be the last w/ the hot potato - I don’t quite understand how valuation is not a problem.

These pools/packages/tranches are priced for risk against the future value of the scheduled cash flows resulting in a certain yield, right? And is the underlying asset (the mortgaged RE collateral and perceived ability of a borrower to perform) not also a part of the value? Or do I have that wrong? I admit that I very easily might be wrong - I’m relying on my commercial bank experience in assigning required risk ratings to each of my loans, and simplistically stated, I essentially do what I just described above. The result is that I have a pretty damn good idea of the intrinsic value of my asset (loan) after analyzing the components.

Again, I could easily be wrong and probably am. I don’t swim in your waters…but I don’t understand how the valuation of the underlying asset(s) isn’t a problem. If you have a payment default, then by definition the value of that income stream is diminished. To wit, that French bank Paribas inasmuch stated last week it didn’t know how to establish a fair value for its holdings that had subprime paper.

Thanks for any clarification!

Before I comment, I’ll put a quick disclaimer in: I trade CDO’s, but mostly ones backed by corporate entities. I dont trade CDO’s backed by loans, ohwever there is a very close resemblance and they are definitely related. In effect, I trade CDO’s that are backed by subprime lenders, not by the actual loans themselves. This is certainly not a huge amount of my business, but I am very aware of whats going on.

Let me clariy a bit: When looking at the underlying collateral in a CDO tranche, its not hard to figureout what any loan/loan package approximately should be worth. For example, if I own XYZ loan, and a simlar loan just traded at $67, then I know XYZ is worth +/- $67. In theory, that loan should be compared to similar loans. You can do this with a good degree of accuracy with most of the underlying collateral-- theoretically, a trade in one loan should be able to generate where others trade. THe problem is that bonds/loans are only worth what someone will pay for them. For example, theoretically, the loan or whole tranche should be worth $70 because the weighted average sum of the pieces equals that, but in reality, is it worth $70 if no one will pay it? No.

CDO’s get their high returns with leverage. These CDO’s have hoovered up tons of lans/bonds and have driven prices for indiviual pieces through the roof. Investors don’t see a lot of positives in the market. There is a glut of this underlying paper for sale. and when more CDO de-lever, there will be much more for sale. WHy buy now? This one trade that may have occured, or a few, tells the models what to theoretically price as. This is what I mean when valuation is not a problem. An indiviual loan can be figured out what its worth and may trade very close to what its worth, BUT when 2000 of these hit the market, there is a problem. The real problem is that there is no/few buyers, ie liquidity. And I dont mean of a CDO, per se, but of the underlying collateral.

If a house in a neighborhood sells for $150k, that tells you that a similar house should be worth +/- $150k. However, if there was only one buyer, and none left, is that house really worth +/- $150k?

Brian, thanks for the reply and explanantion - I think I see more of where you’re coming from and I especially agree and understand that something/anything is worth only what someone will willingly pay and further, that may not be worth much if it’s only an isolated sale. Damn - long sentence!

If I understand you fully, these CDO’s are truly and simply indentures/IOU’s with a rating assigned to the entire tranche, and not actually a loan package complete w/ Note and Mortgage… In other words, the buyer is relying more on the issuing party’s credit/financial strength and rating to make good, i.e., just like w/ an everyday corporate bond issued by XYZ Corp. If that’s accurate then I can see how easy it would be for a buyer to freak out over what they bought, because they don’t actually hold the tangible asset collateralizing the issue and damn sure don’t know what the actual value of the underlying asset is…

Just today we had 2 instances that made me think of our conversation here… The first was Thornburg Mortgage s/s falling 46% causing the exchange to suspend its trading (apparently the company isn’t sure it has enough liquidity to fund its daily position tomorrow.) Further, it has stopped accepting rate locks on its loans. The second was the story of Sentinel Management Group (a futures commissions agent in IL) asking the CFTC if it could block customer redemption requests for their cash, as to not do so could cause it to have to sell securities at deep discounts or, face liquidation. This company is basically a cash-management firm and the fund it was trying to block is essentially a money market fund, albeit a more “sophisticated” one! That’s scary!!

Thanks Brian,

The actual CDO is a passthrough vehicle. It passes the cashflows through to investors. The creditworthiness of the issuer has nothing to do with ability to pay.

The real crux of the matter is this: CDO managers were able to borrow cheap money and use highly leveraged money to invest in loans and pass on high returns to investors mainly created by this leverage. Loans were hoovered into these CDO’s. Wall Street couldnt get enough of them to repackage as CDO’s. The problem is that once these CDO’s unwind, a ton of loan paper floods the market. Even though the model tells them that these loans are worth X, based on a couple trades, there are not enough buyers to support that level. The problem is what a loan should be worth based on comparable trade versus what someone is really going to pay. Models cant predict that.

Btw, thornburg was interesting yesterday. It’s bonds trading $.50 on the dollar. Even more interestong is Countrywide bonds trading like its going bankrupt. You can buy CFC bonds maturing in 4 months @ a 9%!

Rather than just plagiarize, I’ll copy and paste (and sort of plagiarize while giving credit to the writer). Hopefully this will help those who are new to CDOs, securitization, and tranches:

The mortgage crisis…

hat happened (and what is still happening) is simply leverage in reverse, or what people used to call a “run on the bank.” But… I think a great more detail would be helpful for you to understand. Please excuse the intricacies: None of this stuff is very easy to understand the first time you think about it. I’ll try to avoid using any jargon…

For nearly 10 years, as interest rates fell from 1995 to 2005, the mortgage and housing business boomed as more and more capital found its way into housing. With lower rates, more people could afford to buy houses. That was good. Unfortunately, it didn’t take long for some people to figure out that with rates so low, they could buy more than one. Or even nine or 10. As more money made its way into housing, prices for real estate went up – 20% a year for several years in some places. The higher prices created more equity… that could then be used as collateral for still more debt. This is what leads to a bubble.
Banks, hedge funds, and insurance companies were happy to fund the madness because they believed new “financial engineering” could take lower-quality home loans (like the kind with zero down payment) and transform these very risky loans, made at the top of the market, into AAA-rated securities. Let me go into some detail about how this worked.

Wall Street’s biggest banks (Goldman Sachs, Lehman Bros., Bear Stearns) would buy, say, $500 million worth of low-quality mortgages, underwritten by a mortgage broker, like NovaStar Financial. The individual mortgages – thousands of them at a time – were organized by type and geographic location into a new security, called a residential mortgage-backed security (RMBS). Unlike a regular bond, whose coupon is paid by a single corporation and organized by maturity date, RMBS securities were organized into risk levels, or “tranches.”

Thousands of homeowners paid the interest and principal for each tranche. Rating agencies (like Moody’s) and other financial analysts, believed these large bundles of mortgages would be safer to own because the obligation was spread among thousands of separate borrowers and organized into different risk categories that, in theory, would protect the buyers. For example, the broker (like NovaStar) that originated the mortgages would be on the hook for any early defaults, which typically only occurred in fraudulently written mortgages. After that risk padding, the next 3%-5% of the defaults would be taken out of the “equity slice” of the RMBS.

The “equity slice” was the riskiest part of the RMBS. It was typically sold at a wide discount to the total value of the loans in this category, meaning that if defaults were less than expected, the buyer of this part of the package could make a capital gain in addition to a very high yield. Even if defaults were average, the buyer would still earn a nice yield. Hedge funds loved this kind of security because the yield on it would cover the interest on the money the fund would borrow to buy it. Hedge funds could make double-digit capital gains annually, cost-free and risk-free… or so they thought. As long as home prices kept rising and interest rates kept falling, almost every RMBS was safe. Even if a buyer got into trouble, he could still sell his home for more than he paid or find a way to restructure the debt. On the way up, from 1995-2005, there were very few defaults. Everyone made money, which attracted still more money into the market.

After the equity tranche, typically one or two more risk levels offered higher yields at a lower-than-AAA rating. After those few, thin slices, the vast majority of the RMBS – usually 92% of the loan package – would be rated AAA. With an AAA rating, banks, brokerage firms, and insurance companies could own these mortgages – even the exotic mortgages with changing interest rates or no down payments. With the magic of financial engineering and by ordering the perceived risk, financial firms from all over the world could fill their balance sheets with higher-yielding mortgage debt that would pass muster with the regulators charged with making sure they held only the safest assets in reserve.

For a long time, this arrangement worked well for everyone. Wall Street’s banks made a fortune packaging these securities. They even added more layers of packaging – creating CDOs (collateralized debt obligation) and ABSs (asset-backed security) – which are like mutual funds that hold RMBS.

Buyers of these securities did well, too. Hedge funds made what looked like risk-free profits in the equity tranche for years and years.
Insurance companies, banks, and brokers were able to earn higher returns on assets by buying RMBS, CDOs, or ABSs instead of Treasury bonds or AAA-rated corporate debt. And because the collateral was considered AAA, financial institutions of all stripes were able to increase the size of their balance sheets by continuing to borrow against their RMBS inventory. This, in turn, supplied still more money to the mortgage market, which kept the mortgage brokers busy. Remember all the TV ads to refinance your mortgage and the teaser rate loans?
The cycle kept going – more mortgage securities, more leverage, more loans, more housing – until one day the marginal borrower blinked.

We’ll never know whom or why… but somewhere out there, the “greater fool” failed to close on that next home or condo. Beginning in about the summer of 2005, the momentum began to slow… and then slowly… imperceptibly… it began to shift.

All the things the cycle had going for it from 1995 to 2005 began to turn the other way. Leverage, in reverse, is devastating.

The first sign of trouble was an unexpectedly high default rate in subprime mortgages. Beginning in early 2007, studies of 20-month-old subprime mortgages showed a default rate greater than 5%, much higher than expected. According to Countrywide Mortgage, the default rates on the riskiest loans made in 2005 and 2006 is expected to grow to as high as 20% – a new all-time record. The big jump in subprime defaults led to the first hedge-fund blowups, such as the May 2007 shutdown of Dillon Reed Capital Management, which lost $150 million in subprime investments in the first quarter of 2007.

Since Dillon Reed Capital, dozens of more funds have blown up as the “equity slice” in mortgage securities collapsed. Remember, these equity tranches were supposed to be the “speed bumps” that protected the rest of the buyers. With the safety net of the equity tranche removed, these huge securities will have to be downgraded by the rating agencies. For example, on July 10, Moody’s and Standard and Poor’s downgraded $12 billion of subprime-backed securities. On August 7, the same agencies warned that another $1 billion of “Alt-A” mortgage securities would also likely be downgraded.

Now… these downgrades and hedge-fund liquidations have hugely important consequences. Why? Because as hedge funds have to liquidate, they must sell their RMBSs, CDOs, and ABSs. This pushes prices for these securities down, which results in margin calls on other hedge funds that own the same troubled instruments. That, in turn, pushes them to sell, too.

Very quickly the “liquidity” – the amount of willing buyers for these types of mortgage-backed securities – disappeared. There are literally no bids for much of this paper. That’s why the subprime mortgage brokers – the Novastars and Fremonts – went out of business so quickly. Not only did they take a huge hit paying off the early defaults of their 2005 and 2006 mortgages, but the loans they held on their books were marked down, with no buyers available and their creditors demanded greater margin cover on their lines of credit… poof… The assets they owned were marked down, they couldn’t be readily sold, and they had no access to additional capital.

The failure of the subprime-mortgage structure – which started with higher-than-expected defaults, led to hedge-fund wipeouts and then to mortgage broker bankruptcies – might have been contained to only the subprime segment of the market. That’s why we jumped in during late spring and recommended the higher-quality mortgage firms, such as Thornburg and American Home. We believed that the higher quality of these firms’ underwriting would prevent a similar run on the bank.
But… the risk spread because of the financial engineering.

With Wall Street wrapping together thousands of mortgages from different underwriters, it’s likely that hundreds of financial institutions around the world have traces of bad subprime and Alt-A mortgage debt on their books. Parts of these CDOs were rated AAA. Almost any financial institution could own them – especially hedge funds. Hedge fund investors quickly figured this out – and asked for their money back.
And so, in July, liquidity fears began to creep through the entire mortgage complex. Not because the mortgages themselves were all bad or even because the mortgage securities were all bad – but because all the market players knew a wave of selling, led by hedge funds, was on the way. Nobody wants to be the first buyer when they know thousands of sellers are lined up behind them.

The market “locked up.” Nobody would buy mortgage bonds. And everyone needed to sell. Suddenly even Wall Street’s biggest banks – the very firms that created these mortgage securities – were suffering huge losses, as the bonds kept getting marked down as hedge funds and other leveraged speculators had to sell into a panicked market. In this liquidation, even solid firms, like American Home and Thornburg, were trapped owning new mortgages they couldn’t sell to Wall Street. Meanwhile their banks, worried about the collapsing prices of mortgages, demanded greater collateral.

It’s a classic “run on the bank,” except today the function of the traditional bank has been spread out among several institutions: mortgage brokers, Wall Street security firms, hedge-fund investors, and banks. The real problem is that the long-dated liabilities (a 30-year mortgage) were matched not by reliable depositors, but by fly-by-night hedge funds, which were themselves highly leveraged and subject to redemptions.

That’s why even as the top executives in these firms believed their mortgages were safe and sound, they can’t get the funding they need to hold onto them through the crisis. As Keynes predicted, the lives of every higher-leveraged financial institution is precarious: “The market can be irrational longer than you can remain solvent.”

The hedge funds have no solution. Redemptions will force them to sell. They’ll continue to pressure the market, resulting in huge losses. Hundreds of funds will likely be liquidated.

Wall Street’s investment firms, if they can find additional capital to meet margin calls, might weather the storm… depending on how far it spreads. We saw a move in this direction yesterday when Goldman announced $3 billion in additional funding for its big hedge funds.
For most mortgage brokers, the party is over – goodnight. Something like 90% of them will be out of business by the end of the year.

The only chance they have to survive is very conservative underwriting (which might result in a premium for their mortgage securities) and lots of additional funding. Delta Financial, for example, is renowned for its very conservative underwriting, which requires a substantial (20%) downpayment. The company raised $70 million last week from two investors (one of which is our friend, Mohnish Pabrai) to hang on to its $5.6 billion in on-balance-sheet mortgages. The stock is up 14.5% on the news today. Will it be enough capital? It’s very hard to say. It depends on whether or not the company is able to sell some of its mortgages to raise cash. It depends on whether or not it is downgraded further and the firm receives additional margin calls.
I wouldn’t be surprised to see Thornburg take a similar step – raising funds from existing shareholders. But, for now, Wall Street remains very skeptical the firm will survive. Its shares are down another 46% today.

As analysts, what we got wrong was how far the crisis would spread. We thought by buying the most respected firms with the best underwriting, we could avoid the subprime train wreck. What we didn’t know was how far the subprime sludge had been spread via mortgage securities. The insiders at these firms made the same mistake. They assumed by operating conservatively their businesses would retain a premium price on their mortgage and better access to capital. But in a panic, the baby is often thrown out with the bathwater.

And… we have to consider one more thing. Nobody knows right now how far the crisis will spread. It could certainly get worse. As these mortgage bonds are downgraded, the financial institutions that own them must raise more cash in order to meet liquidity regulations. To hold AAA-rated paper, banks, and other financial institutions need only to maintain $0.56 in capital for each $100 of paper. But as the paper is downgraded, the amount of capital they’re required to hold goes up, exponentially. At a BBB rating, financial institutions must hold $4.80 of capital. At BBB-, they must hold $8 of capital per $100 of asset-backed securities. Thus, as the crisis worsens, the demand for capital from these firms could grow substantially.

We can’t know what will happen. And, as we can’t know, we must stand aside when our trailing stop losses are hit. As I wrote, back in early July, about American Home Mortgage:

Speculation on Wall Street is that “Alt-A” debt will be downgraded next. Most of the loans held by American Home Mortgage are considered “Alt-A” because they have adjustable rates. Even with the high credit scores of the company’s borrowers, if rating agencies downgrade the bonds it holds, the company’s solvency will certainly come into doubt. Whether this happens or not is a moot point for us: Our speculation hasn’t panned out. We should have realized it sooner… but in a few more weeks we might be very glad we got out while we could.

Porter Stansberry
Baltimore, Maryland
August 14, 2007

Paul, good article. The most interesting part, which the article does not mention, is ratings agencies Moody’s and S&P’s rating of these securities. They rated some AAA, on par with risk free treasuries. To give a good example, there was one CDO that priced a few weeks back… it was rated AAA and priced at $100. 3 weeks later? trading at $70. How can a ratings agency rate something as risky as treasuries and then it loses 30% in 3 weeks???

Well, I’m not enough of an expert to say why that is. You would think that with a triple-A rating these tranches of securities would be so far up the food chain that they would be without risk. I mean, there will always be SOME cash flows in from these mortgage pools, and assuming that the AAA stuff is first in line, they should be without risk.

Maybe the 30% haircut in valuation is an over-reaction to the market? There are fewer (or no) buyers and plenty of sellers who need to raise cash, so the bids are dropping vulture-style. It would be nice to know who is buying these securities at 70 cents on the dollar. Looks like they might very well be in for some 40% gains when the dust settles.