James Hamilton at Econbrowser asks the remarkably sanguine question: CDOs: what’s the big deal?
He does a nice job looking at many of the issues this topic raises. However, I still have some unanswered questions about CDOs:
Thus, I will respond to the Prof in true Socratic method by asking more questions. These raise the possibility that not only CDOs may be a big deal, but we don’t have a clue how big a deal it may be — Modest? Gargantuan? Ginormous?
Here are my 10 questions:
1. What would have happened had Bear Stearns simply let their two funds, High-Grade Structured Credit Strategies Fund and High-Grade Structured Credit Strategies Enhanced Leverage Fund, dissolve?
2. If CDOs are not priced to market, what are the actual values of these holdings?
3. How levered up are the funds that own the bulk of the CDOs? 10-to-1? 20-to-1? More?
4. How many Hedge funds are or have been taking quarterly or annual performance profits, based in whole or in part, on hoildings that have been marked to a theoretical value ("Mark-to-Model") versus an actual value ("Mark-to-Market")?
5. Liquidity has been a driving force behind M&A activity, share buybacks, and leveraged buyouts. Might the CDO situation somehow impact liquidity?
6. Might a liquidation in a CDO/illiquid derivative fund spread to other asset classes?
7. How widely held are the toxic CDO tranches in funds that are self-decribed as "conservative" or "risk averse?"
8. How accurate are the major ratings firms (Moodys, Standard & Poors, Fitch) assessment
of these products. Are these outfits arm’s length objective raters, or
are they merely corporate whores who play for pay?
9. After the final chapter is written on CDOs, what might the total losses on the $250 Billion in quarterly CDOs that Wall Street has created actually be? 10 Billion? 100 Billion? 1 Trillion?
10. How much will systemic confidence be impacted if there is a
series of large fund failures due to CDOs? What impact might that have
on the rest of the markets?
The big deal is that we simply know so little about these issues. Wall Street has gotten better, for the most part, about managing risk. But these CDOs are increasingly looking like unknown factors, with an unknown set of risk parameters.
Hence, that is what the big deal is . . .
Another question might be what type of asset sell-offs will be required if Mark-to-Market devaluations prompt margin calls?
another big question is if the SEC will finally get involved in this mess.
selling subprime debt with a AAA rating sounds like fraud to me.
if there were more transparency and regulation in these derivative markets, this may have been avoided.
Margin calls are indeed the key.
If the prime brokers of all these hedge funds decide that the underlying holdings of these levered up funds are not priced properly then won’t the margin clerks do the selling for them? And if the hedge fund managers’ pay was somehow based on the value of the funds holdings isn’t there a built in conflict to price to the high side?
CEOs essentially setting their own pay has not worked out well for shareholders and I think allowing hedge fund managers to value their illiquid holdings is starting to smell the same way—BAD.
Margin calls are indeed the key.
Absolutely. The ultimate question is how much unsupported leverage is out there? Anyone have even the foggiest idea?
Are we waiting for the first situation that forces a marking to market?
Check the latest Quarterly Holdings report for your Money Market fund. Many of them hold not-insignificant percentages of Asset-Backed Commercial Paper (ABCP). In many instances, those ‘assets’ are mortgages, just like CDOs ….
I have a feeling once the tide goes out a lot of former high rollers are going to bare a strong resemblence to shoeman Al Bundy.
Chuck,
Mortgage backed securities are not necessarily a bad investment. When used properly they have been a relatively safe way to invest for many years. CDOs are a special case because of how they have been packaged and because they are illiquid.
1. Assets would have sold considerably below intrinsic value because there is just no liquidity in this market right now. Investors would have 100% losses and creditors likely would have as well. One of the reasons I think the market is so thin is that there is a lack of analytical ability to assess the risk of these cdo bonds on the buy side now that people realize the Rating Agencies models are bunk.
2. Some should be worth Par, some closer to the present value of some IO stream of cashflow because writeoffs will be 100% of principal. Building off point 1, it’s very difficult to sift through the cdo’s and determine which ones are money good in most scenarios (as investment grade paper should be)
3. That’s difficult to answer. Although I wouldn’t be suprised if some players were similarly levered. Leverage is also a ambiguous term in this sector. A cdo equity investor can be up to 25×1 leveraged from a structural perspective to the underlying assets (say the tranche is 4% thick.. if the collat takes 4% losses the tranche takes 100%… 100/4 = 25x leverage to losses). But a cdo equity investor can then go and get short term repo financing in order to get more traditional financial leverage.
4. The idea of Mark-to-Market in the abs cdo space is laughable to me. These assets weren’t created with the intention of frequent trading, and even if they do trade with higher frequency than intended, the market will never be liquid enough to create an accurate price point such that the market view accurately reflects the present value of future cashflows at the appropriate discount rate. The reason is that CDO debt is almost uncomparable to any other piece of debt. They have to valued individually. Some would argue there is some element of comparability, but there are just two many moving parts involved for me to square with that statement.
5. ???
6. ???
7. CITI put a piece out on this just yesterday. It details holders by fund type, geographical distribution, and the assets they hold (senior, vs. mezz, vs. equity).
8. Their accuracy is laughable at best… I could go on and on about this one…
9. It’s hard to say, but near term mark downs could be drastically different than ultimate losses, either to the upside or downside… there’s just that much uncertainty. If you are trying to do a calc… make sure to break out cdo issuance into different assets classes… we are only talking about ABS CDOs here.
10. ???
thanks for the Xreference
i’ve posted the prof for a trial run on my yahoo
but he really just confirms the inherent fundamental dangers of this ripple effect becoming the tsunami in its final form
your point about other asset classes being dramatically (my word) sold off to fill (if possible)the black hole
by the way history will note you were one of the first to pick up on this weeks ago
rgds pcm
How do investors in other leveraged CDO funds feel when they hear about Bear Stearns, who is one of the most talented players in the structured products market, blowing up two CDO hedge funds? I’m guessing they’re pretty nervous. Many probably would like their money back right about now.
To me, the biggest immediate risk is redemption notices from nervous investors hitting the fax machines of CDO managers. The manager can suspend redemptions for a while, but he will eventually be forced to sell this stuff. If bids are as scarce as widely reported, realizable value will be far less than where the stuff is being carried.
So, that’s when the margin calls go out and potentially trigger forced selling of other financial assets to meet the margin calls. And after all the selling, if they still can’t cover the margin debt, the primes will be holding the bag. And the balance sheets of the primes are levered and stretched beyond belief, so their creditors are probably getting nervous.
The trick is to stop the CDO investors from redeeming. It’s not difficult for me to imagine a selling stampede across markets if they don’t.
1. I’m surprised Bear didn’t just let the funds wind down…assuming they could sell all the securities without more than 100% loss of capital, then the fund investors would have borne the losses, bear would have lost nothing (except for what i believe was its own small investment in the funds), and the prime brokers wouldn’t have lost anything. I am guessing bear stearns was also a prime broker to these funds? then that would suggest to me that the funds lost more than 100% of their capital. ouch…
2. it’s not the case that “CDOs are not priced to market”, it’s up to any particular fund/bank to mark its securities where it sees fit. for example, banks and insurance companies often mark investments (not just cdo’s, but bonds/equities) at the purchase cost, and from time-to-time may take impairments against them, although usually only if that asset is not performing (ie. a loan not paying)…they typically don’t care about mark-to-market. but for hedge funds, since they are using prime brokers, those prime brokers will want as accurate a valuation of all securities in the fund, so they can correctly evaluate exposure. what i suspect is happening is that the market fell so quick for ABS CDO’s, and they are so illiquid, that many funds didn’t mark them down fast enough.
3. Leverage in a CDO depends on the tranche you are in…the AAA tranches for example are carry less exposure than if you individually owned the underlying instruments, so are “safer”…but what I’ve seen is hedge funds then leveraging up those tranches, and so exposing themselves to mark-to-market (MTM) risk on these tranches, even if ultimately there are no defaults at the AA or AAA level.
4. Repeat 2…any hedge fund investments would be MTM by prime brokers…but I am sure there was plenty of leeway on those marks whilst times were good and nobody was really concerned with CDO exposure.
5. If you look at credit CDO’s, as opposed to ABS CDO’s which is what has been causing the problems, this tranching of credit risk and appetite for equity tranches from a variety of investors is what has driven credit spreads to record low levels. This then allows corporates to issue debt also at record low levels (since corporate bond spreads tend to trade close to where CDS levels are). For example, say a hedge fund wanted to invest 100mm in CDO equity (which pays a large yield, say something like 15% for credit CDO’s), that takes a TON of credit risk out the market, since higher up “safer” tranches are typically easier to sell…so if that equity tranche takes the 0-3% loss portion, then the total portfolio size created if all the other parts are sold (3% to 100%) is 100mm *100/3 = 3.3 BILLION. And as a long-time CDS trader, I can tell you these trades have been happening EVERY DAY for the last 5 years. An amazing amount of credit risk has been taken out of the market by investors, and this is a major reason why credit spreads are at record low levels.
6. yes
7. very…see the above note on leveraging up the AA and AAA tranches. also, on a slightly separate note, one of the things that makes me laugh with some funds is when they advertise themselves as never having suffered a default on a bond…that’s pretty much always because they puked out of it just before it technically defaulted! selling Enron bonds at 10 cents on the dollar the day before they announce bankrupty for instance, and then still claiming you’ve never had a default, even though you’ve basically lost just as much money! investors love that line.
8. rating agencies accurancy will never be known, as its all just subject to probabilities. but basically the CDO game exists because the ratings agencies are being arb’d…you can tell from market spreads that the market certainly thinks they are wrong…CDO tranches with the same ratings as corporate bonds ALWAYS trade wider!
9. CDO’s only generate losses because the underlying instruments have losses (say, like sub-prime mortgages). Those mortgages would have had the losses anyway, so all you’ve done with CDO’s is transfer that risk to counterparties who want that risk. Yes, you can argue that the underlying mortgages (in this case) wouldn’t have been approved if it wasn’t for mis-pricing in the CDO market, over-generous ratings agencies, and downright fraud throughout the whole process, from the borrower to the bank to the hedge fund manager to the hedge fund lenders!
10. as per above, I think it impacts liquidity for a while…but ultimately CDO’s are here to stay, and are a good thing like any derivates…they allow the transfer of risk to those able to bear that risk. Also I would just point out again the CDO’s can have many different underlying risks in them, it just happens that the current “crisis” is in asset-backed (ABS) CDO’s, and specifically in sub-prime ABS CDO’s (although it does look like it’s spreading to other areas of the ABS market).
Regards,
the cds trader
“8. How accurate are the major ratings firms (Moodys, Standard & Poors, Fitch) assessment of these products. Are these outfits arm’s length objective raters, or are they merely corporate whores who play for pay?”
I pick door #2.
Question # 11
It is the end of the quarter. How many of the accountants are going to continue to go along with the fiction that there are no losses requiribg markdowns?
hello cds trader,
lets see if i’m moreless correct here
we have what most claim to be highly illiquid, quite heterogeneous instruments termed collateralized debt obligations yet these are in fact market-to-market (?) rather than marked-to-model with agency ratings as a ‘price’ input and people like lombard street research’s charles dumas stating that ‘we don’t know what the value of this debt is…’
“the cds trader” – Thanks for the info., much appreciated.
In the reading that I’ve done on CDOs for the last few years, it seems that the riskiest CDO tranches are “planned to fail” from the outset, but this needed in order to offer higher returns to the A/AA/AAA tranches.
What I’ve never understood is why there’s a belief that there will always be buyers for these “First Loss” tranches? As you say in #9: “transfer that risk to counterparties who want that risk”. What happens if there are absolutely no buyers for BB/BBB/Equity/Mezzanine tranches?
Chuck: “What happens if there are absolutely no buyers for BB/BBB/Equity/Mezzanine tranches?”
Answer: tighter mortgage lending standards.
According to a Bank of America analyst, between 2007 and 2008 $836.5B in subprime ARM mortgages will reset.
How do you rid yourselves of the glut of inventory of homes for sale while tightening credit at the same time?
CDS trader: “Those mortgages would have had the losses anyway, so all you’ve done with CDO’s is transfer that risk to counterparties who want that risk.”
Is it that they wanted the risk or were stretching for yield?
Here’s my question – about #13, I guess:
How can lenders take CDOs as collateral – as Merrill seems to have done with Bear? You have a guaranteed huge loss in precisely the only situations where you might conceivably want claim your collateral. It makes the Japanese banks reliance on real estate in the late-1980s look prudent by comparison.
winston, i’m sure plenty of buyers of riskier tranches of CDO’s (both ABS and credit CDO’s) were/are stretching for yield, but ultimately derivatives of all forms, including CDO’s, are good for the system in that they allow efficient transfer of risk. notice that there is very little talk of interest-rate swaps being a bad thing, although just because credit derivatives and CDO’s are a new instrument, they are tagged as “weapons of mass destruction”. maybe this sub-prime blow-up will give everyone a “caveat emptor” warning, much like the Orange County blow-up did for interest-rate derivate investors. In years to come we’ll look back and see that the main problem in this whole episode was just plain and simple excessive risk taking.
“they allow efficient transfer of risk”
I’ve heard this ‘party line’ more than a few times. It carries with it a few unspoken assumptions. Let me try to uncover one assumption by this question:
If *absolutely nobody* will buy BB/BBB/Equity/Mezzanine tranches, as constructed in the last 5 years, can a CDO actually deliver much more than a AAA-rated bond from IBM or GE?
It seems to me that if you pull all of the sure-to-fail garbage out of these things, you’re left with plain-old bonds. Without a sucker willing to flush his capital down the toilet, what is there?
I am a CDO manager, so here are some points of clairification, in no good order:
1) Performance fees on CDOs is usually based on cash flow paid to the equity holder. So while M2M never comes into play, the only return a manager gets paid on is actual return.
2) Most CDO debt is AAA rated, and this is usually held by banks and insurance companies. It will take a truly massive wave of sub-prime defaults, far greater than what’s going on now, to cause AAA tranches to fail. Be careful when you see media reports that say “The CDO market size is X” and think that this all high-risk securities.
3) ABSCP held by MM funds is nothing like CDOs.
4) Figuring out the real leverage at a hedge fund with CDOs is nearly impossible, because the CDO strucutre itself has tons of leverage. The degree to which any CDO tranche is levered changes over time.
5) AAA CDOs are pretty liquid in general, probably similar to a smallish corporate bond. But as soon defaults in a CDO start coming in high, liquidity for the junior tranches dries up. The only people who want to bid are vulture bidders. So the value of a junior CDO tranche can very rapidly go from par to $10.
6) There is no such thing as a redemption or margin call for a CDO manager. CDO debt is sold to the public, just like a normal corporate bond. The only thing a holder of CDO debt can do is sell it.
I wrote a intro to CDOs for anyone who is interested:
Sorry… that link is: http://accruedint.blogspot.com/2007/03/how-does-cdo-work.html
Another question might be what type of asset sell-offs will be required if Mark-to-Market devaluations prompt margin calls?
Many holders will be required by covenants to sell when the paper is re-rated. Bear and its co-conspirators acted to prevent ‘mark-to-market’ for this and the other reasons.
Looks like most of these questions have been answered – or are in the process of being answered. “Baneful Night bore Nemesis, too, a woe for mortals…”
~~~
BR: In Greek mythology, Nemesis was the goddess of retribution.