I've followed with interest the headlines you've posted about CDS contracts... though I'm not sure about your conclusions.

I can see why, to an individual institution, these contracts are dangerous (for both parties) but I don't see why there should be any massive (predictable) implication on a macro level. I can imagine consolidation of banks, but not systemic failure of the financial system- or, even... an inflationary effect.

I am curious, however, to pick brains:

How do CDS relate to Synthetic CDOs?
Am I right in dating the latter to 1997-ish?
Would it be fair to suggest that CDS facilitate greater leverage when investing in debt?
Do we have any estimate of the exposure to UK banks to CDS (on either side)?

Thanks for that, I´ve had a strange feeling that a huge spectre has been looming and that fits the bill perfectly.

the systemic failure comes from the fact that there has been a catastrophic failure in the performance of the ratings agencies.

it wasn't necessarily that subprime loans were given out, usually they would just be charged more to off-set the risk, it was that they were rated as AAA.

unfortunately, it wasnt just the subprime mortgage debt, it was EVERY SINGLE CREDIT ASSET CREATED over the past several years that has been wrongly rated.

from insurance, to CDO's, CDS's, MBA's, Commercial paper, even in the case of Iceland, government bonds.

Can someone clear something up for me please? Can you please clarify how CDS's fit in to the grand scheme of things. I understand what they are purely by definition, but I don't understand who the parties involved are. One thought I had was that possibly the CDS's are another name for monoliners insuring other monoliners? Feel free to flame me!

http://news.yahoo.com/s/nm/20080219/bs_nm/usa_banks_fed_dc_2

Banks in the United States have been quietly borrowing "massive amounts" from the U.S. Federal Reserve in recent weeks, using a new measure the Fed introduced two months ago to help ease the credit crunch, according to a report on the web site of The Financial Times.

Everyone, watch out for this massive wall of freshly created money coming along your way.

Maybe I should turn these questions into a separate thread?
I'd like to add the question "Exactly how do CDS relate to leveraged positions in credit?"



First, my apologies "Mr Nice" for my apparently aggressive response to your calling CDOs "like mini banks" - I noted the same description in Despatches, and now think I understand what you meant. I think we were arguing over a trivial issue of syntax rather than anything substantial...

I agree that the ratings agencies are now a joke, and that they can no longer be trusted... but I don't see how this necessarily leads to systemic failure. Sure, there'll be a lot of upheaval - but not necessarily a catastrophe.



I can't flame you WP, because I've very similar questions. We both understand the function of a CDS, as a form of insurance - which, incidentally, I note, can be held for the benefit of a party other than the party suffering the losses (which, I think, would be illegal if it were another form of insurance - such as insurance against fire or theft.) So, aside from the moral questions that arise from such a contract - I believe a CDS to be OTC (Over The Counter) - and, as such, is a binding but unregulated agreement between, typically, banks. I am aware that some CDS contracts are of a standard form and are traded openly - but that others are bespoke and, likely, are not straightforward from the perspective of risk analysis.

I think (someone will hopefully correct me if I'm wrong) that CDS differ from monoline insurance in so much as CDS insures against individual default events - whereas monoline insurance insures against the value of bonds? If Joe Public defaults on his mortgage in February, a default event occurs and the CDS pays out the money that Joe would have done... but... the monoline insurance only pays out when the expectation of future defaults is so bad as to affect the face value of the bond. As I venture further outside my sphere of knowledge, it seems that a CDS would be a great way to synthesize bonds which appear to have no default risk... and, if the CDS contract has a lifetime shorter than the bond it protects, it offers an opportunity to defer proper assessment of long-term risk.

I would welcome the views of others on these ideas... (if someone shouts, I'll start a new thread with this...)

I hesitate to offer any explanation because I'm no expert and listening to me would be more like engaging in a game of Chinese whispers than educational. I now have a rudimentary knowledge/understanding, but not sufficient to teach others. I can express an opinion - and maybe I'm right, maybe I'm wrong - but instructing another is a different matter altogether.

Truth is, before I came to HPC I'd never even heard of a CDO or a CDS. However, there have been some really good links posted that give very good explanations - only I can't remember where or what they were. If you have the time, read the following blog - it goes a long way to explaining derivatives of every kind. To get the full picture you really need to trawl back months - I did. Very enlightening.

Sudden Debt

I think that experts in this domain are few and far between. I suspect that the details are closely guarded secrets.

I'm interested to know what people here think... whether it is right or wrong.

I'm especially interested to discover:

Exact relationship between CDS and Synthetic CDOs.
Statistics on CDS/Syhtnetic CDO issuance - by currency.
An exposition on how CDS relate to leveraged positions when investing in debt.

The auction seemed to go well - can you elaborate on whu it didn't work?

I would suggest that there are a larger number of people with knowledge in this area than you suspect. if you are interested in the subject there are lots of books on Amazon that are as detailed/high level as you require. I have a section on my blog dedicated to credit derivatives and my dealings with them over the last few years - I will post link up when I fix my web server

a quick example:
if someone crashed into me, and wrote off my £10,000 car, but my insurance company only paid me £3,000 because they were having to pay out lots of other write-offs, I would say that my car insurance hadn't worked.

"This arrangement was done at just over 36 cents on the dollar"

The recovery rate is assumed to be 40% for senior debt (it differs from company to compnay but traders tend to assume blanket 40%). The fact that the auction was at 36% indicates that it was pretty close to what people were expecting, unless I am missing something?

I think I'm the one missing something here! The way Goldfinger put it, it was a simple case of 2/3 of their "insurance" cover not being paid out when Delphi defaulted on their bonds.

But also, how can the default swap cover be ten times the value of the bonds in the first place? (and hence why this whole market gets quoted as being something silly like $45 trillion). To use my stupid analogy again, that's like ten people people all having and claiming on insurance when one car gets smashed. Either you own the bond (and therefore want insurance on it defaulting) or you don't?

In this post from somewhere else, the Merril Lynch - Credit Derivatives Handbook is available in 2 parts, 300 pages of pure unadulterated derivatives not sure if it would answer your questions though as I haven't read it all.

US Quarterly Reports on Bank Derivatives Activities:
http://www.occ.treas.gov/deriv/deriv.htm

Shutty,

I was slightly wrong in my original numbers - my excuse is that I'm on holiday so my brain is switched off!

The following link mentions the Delphi default - Page 46

http://www.bis.org/publ/qtrpdf/r_qt0606d.pdf

"Under certain circumstances, a shortage of
deliverable debt can drive up the price of such paper beyond the level that
might otherwise be justified by the expected size of repayment. In the case of
Delphi, the settlement price of 63.5% (and an average CDS recovery price of
53.5%) was considerably higher than the settlement prices of other firms from
the same sector or than rating agencies’ estimates of the ultimate recovery
rates on Delphi’s debt."

So the effective recovery rate was 53.5% against an average assumption of 40%. This would mean that the protection buyer would not get back as much as he expected, but as I said earlier, recovery rates differ across all sectors.

My original point about this was that the auction went without too much of a problem.

The reason that the notional (for some names) is greater than the outstanding bonds is because CDS's are now used as a trading instrument (the document states this), rather than just for hedging credit risk. You also get situations where there is a CDS market with no underlying bonds (Nokia was one name I recall) - the assumption being that bonds will be issued at some point in the future.

I am interested... I'm even willing to read marginally relevant text books to find out more.

Don't really know what i'm talking about but here goes.

Imagine a horse race with prize money. The horses race for the money, say £1000.

The punters make their bets, but with IOUs. The value of these bets could be anywhere in the range of £0 to £infinity. Just for good measure, imagine the punters placing more IOU bets based on their hoped-for winnings. This way, the betting could amount to $45 trillion, or any other silly number.

CDS are side bets, like all the other financial engineering, used to "reduce" risk, but at a lower cost than paying the correct price. In other words, something-for-nothing, perpetual motion, big mac with omega-3, too good to be true.

All depends on the IOUs being good.

ACA looks toast





http://ftalphaville.ft.com/blog/2008/02/20...-bond-insurers/

KKR subsidiary having problems. Apparently they are offering creditors the option of taking the underlying mortgages in lieu of payment.



http://www.bloomberg.com/apps/news?pid=206...&refer=home

Oh my God - we now have "CPDO's". You heard it here first. Only a matter of time before C3PO appears followed by Luke Skywalker if we're lucky.

http://bloomberg.com/apps/news?pid=2060108...&refer=home

Anyway - this is what Bob Diamond seemed to indicate Barclays have been doing...buying lots of hedges... against mounting losses at ever increasing cost.

I am sure they are very keen to hold mortgages.

CPDO = Constant Proportion Debt Obligation

Stunning. Standard Chartered have decided not to rescue its SIV and put it into receivership instead.

http://news.bbc.co.uk/1/hi/business/7254981.stm

I am still troubled by Barclays' surprisingly good results.

As I have said elsewhere, Barclays Capital are known to have grown strongly on the back of the structured credit markets over the last few years. But they haven't solved the fundamental problem of poor sub prime families defaulting on their mortgages. So Barclays has apparently shifted the problems onto someone else somehow.

When questioned about the results, the silver tongued Bob Diamond stated:

"We have no risks we didn't know we had July/August last year (that is supposed to provide comfort). But that doesn't mean we don't have to manage those risks (good). We have divested ourselves of some of the risks (poss. sold some to unwitting suckers like our clients and customers) and others we have hedged (we just hope we are going to be covered for this lot when the time comes to pay up). "

So WHERE ARE THE BODIES BURIED Barclays?

Where they could they have sold the bodies to grave diggers. However, given the deterioration in the markets, it is safe to assume that much remains that could possibly come back to haunt them... these bodies have presumably been sprinkled with CDS perfume to stop them smelling over the audited year end results.

I'm assuming that most CDS's have a fixed term (can anyone confirm). And that this fixed period is less than the term of the mortgage or corporate bonds being insured and would therefore require renewing from time to time. So if things continue to deteriorate as they are, then the price of those CDS's will continue to rise and there could even reach a point whereby nobody actually wants to offer CDS's at any price. In this scenario, Barclays could look very poorly indeed and is probably a greater risk than counter party risk itself.

Does anyone have any inside info?

Port Authority of New York Bonds Reset at 8% (from 4% - 5.70% previously)

Feb. 20 (Bloomberg) -- Interest rates on $100 million of bonds issued by the Port Authority of New York and New Jersey were set at 8 percent in a weekly auction after surging to 20 percent on Feb. 12. ...

Rates had soared from 4.3 percent when too few buyers bid for the so-called auction-rate debt and Goldman Sachs Group Inc., which runs the auction, refused to put up its own capital to buy unwanted securities. That caused the yield to be set at a level predetermined in bond documents....

http://www.bloomberg.com/apps/news?pid=206...&refer=home

No inside info, sorry. And if I did I'd certainly not be printing it on here. But there is another issue burbling around of relevance. I was reading some discussion in the CDS market this morning, and in particular that documentation in the estimated $45trn CDS market is "somewhat optional", with 13% of trades unconfirmed in December, and an example of one party in a resale transaction being unidentified and unknown 30 days later.

So, suppose you are the auditor. You have to sign off accounts where a large exposure is hedged by a CDS with little or no paperwork and where the credit quality of your hedge is unknown, and most likely substantially below the rating of the asset you are supposedly insuring...

You might even be prime broker lending the capital to the hedge fund that provides you with the credit protection, so you'd be calling on the capital you provided to protect your capital in a fund whose investors have borrowed your capital to invest.... in the words of Austin Powers, "oh er I've gone crosseyed".

I'm not asking for illegal insider share dealing type info - merely expertise relating to the CDS question highlighted in bold...are they rolled over frequently against a reference asset and if so how frequently would be typical? If you don't know it doesn't matter. Thanks for taking a look anyway.

not an expert on cds, but i think I recall being told typically fixed term 5 years. no roll-over, would have to be a new contract.

i don't have any inside info, but looking back over the last few months there are 2 stories regarding barclays, that have been posted on here (can't find them now).

1, the trader who put much of the cdo stuff together at barclays went missing - has he resurfaced yet? could it be that he's the only person that knows how deep the sh1t is?

2, if i remember rightly, someone posted a link that suggested that barclays were somehow involved with AMBAC and MBIA (insuring them?). could it be that barclays exposure is via MBIA & AMBAC? - thus the the main pain is yet to come.

I believe Barclay's was also the issuer of swaps / under-writer for Granite (with Lehman and Merril).
Not that that is entirely relevant, but funny how insular it all is.

Think Goldman Sachs were the advisors...


some gumpf from the prospectus:
"The offered notes will also have the benefit of the following derivative instruments: (i) the Funding 2 basis rate swaps entered into
between Northern Rock plc, as basis rate swap provider, and Funding 2, and (ii) the issuer swaps to be entered into between the issuing
entity and each of Barclays Bank PLC, as the issuer swap provider in respect of the series 2007-2 class 4A1 notes, and Deutsche Bank AG,
London Branch, as the issuer swap provider in respect of all other classes of offered notes.
Neither the Securities and Exchange Commission nor any state securities commission has approved or disapproved of the offered
notes or determined that this prospectus supplement or the accompanying prospectus is truthful or complete. Any representation to the
contrary is a criminal offense.
The underwriters or affiliates of certain underwriters will pay and subscribe for the offered notes at the price per note stated in the
table above. The proceeds (net of management and underwriting fees and selling commissions) to the issuing entity from the sale of the
offered notes is expected to be approximately $4,503,450,600....

Not that I understand half of it.

"Remarketing Bank Barclays Capital Inc.
Conditional Purchaser Barclays Bank PLC
Remarketing and conditional
purchase arrangements
The series 2007-2 class 4A1 notes have the benefit of certain
remarketing and conditional purchase arrangements
provided for in the remarketing agreement and the
conditional purchase agreement. Subject to the satisfaction
of certain conditions, the remarketing bank will be obliged to
use reasonable efforts to identify third party purchasers to
purchase tendered notes on each transfer date, from (and
including) the transfer date occurring in April 2008 up to (and
including) the transfer date occurring in April 2054, prior to
the redemption in full of the series 2007-2 class 4A1 notes.
If the remarketing bank is unable to identify third party
purchasers for all tendered notes on a transfer date, then the
remarketing bank will give notice to the conditional
purchaser, who will, subject to certain conditions, be obliged
to purchase all of the unremarketed series 2007-2 class 4A1
notes. The conditional purchaser will be obliged to purchase
unremarketed notes on each transfer date up to (and
including) the transfer date occurring in April 2054. See
"Remarketing agreement and conditional purchase"

I too think they are hiding the real bad news, just like hotairmail said, somethings not right....

it HAS to come out in the end, but they are doing a very good job at putting it off, which is also just doing a perfect job of inflating the sh1t when it comes out.

we WILL be the last ones laughing.

ABX indices hit many new lows today. That's got to hurt.


http://www.markit.com/information/products/abx.html


20-Feb-08 Overview

Index Series Version Coupon RED ID Price High Low
ABX-HE-AAA 07-2 7 2 76 0A08AHAD4 63.69 99.33 63.69
ABX-HE-AAA 07-1 7 1 9 0A08AHAC6 66.75 100.09 66.75
ABX-HE-AAA 06-2 6 2 11 0A08AHAB8 79.84 100.12 79.84
ABX-HE-AAA 06-1 6 1 18 0A08AHAA1 93.48 100.38 90.09

(that's just the "good" stuff)

Looks like the plans announced by Bush and the emergency cuts haven't helped the defaulting home owners whose mortgages are wrapped up in these notes.

I wonder if any of the notes will hit zero (ie. the expected recoveries being below the expected cost of administering the notes).

Dresdner rescuing its SIV (unlike STANDARD CHARTERED - see earlier) but no 'direct exposure' to CDO's.

SIV assets shrinking dramatically...will restrict banks' ability to lend.

http://www.bloomberg.com/apps/news?pid=206...&refer=home

Oook:

ABX-HE-BBB 07-1 7 1 224 0A08AIAC4 12.69 98.35 12.69
ABX-HE-BBB- 07-1 7 1 389 0A08AOAC1 12.13 97.47 12.13

Soo.. if this toxic BBB stuff has been recombined to give more AAA stuff.. which I believe has been happening (feel free to correct).. perhaps those AAA things ain't.

TYhis is going to end badly, isn't it?

CDOs are typically denoted as high yield or mezzanine = the mezzanine CDOs are what you are referring to, load of low grade tranches added togther, wave a wand, and hey ho, most of it is relabelled AAA. Or was, now its toast.

Here are the credit derivative related postings on my blog

http://www.noelwatson.com/blog/CategoryVie...erivatives.aspx

Feel free to ping me an email if you have any questions - and if I don't know the answer, I will find someone that does.

I recall the CPDO rumours being discussed here a few weeks back - as far as I am aware, it is still rumours. There were people doubting CPDO's from Day 0 - there was a good article in Wilmott a year or two ago.

excellent stuff thanks.

Your messages here are "turned off" - and I didn't see an email address on your blog. I was going to send this privately...

I guess it all depends (for CDS at least) on whether Barclays have been buying or selling protection. If they were net buyers when spreads were low, they will be laughing now.

For single name CDS, typically the most liquid term/tenor is five years, with contracts maturing four times a year. So if I bought five year protection tomorrow, it would mature on 20th March 2013. The indices trade on 3,5,7 and 10 year maturities, with the same index rolls on 20th March and Sept

This sounds to me like selling on margin at every layer, icing on the icing on the icing on the icing, ad infinitum, on a cake that may consist of not much more than a single raisin. Or am I being too cynical?

I think it may be this article

http://www.nytimes.com/2008/02/17/business...amp;oref=slogin

I am intrigued where these numbers are coming from. When sell side (typically market makers in bank) trade with other banks, it is done electronically and settled at the end of the day on DTCC

http://www.dtcc.com

and I was sure that other trades with a bank are also settled promptly - also on DTCC. Maybe someone has a link with further granularity on the numbers - it could be that hedge funds are typically less well organsised

I think I have turned on messages on this forum - email on blog is just below disclaimer

Anyway, if I were to buy two books, they would be

Tavakoli

http://www.amazon.co.uk/Collateralized-Deb...9490&sr=8-2

The first sentence on inside cover is "What is a CDO?", and the rest of the book goes into depth on several issues, including problems with AAA ratings.

Chaplin

Probably the best book on how credit is traded in the real world.

http://www.amazon.co.uk/Credit-Derivatives...9876&sr=1-1

I have used the book most frequently at work - it even goes into getting a simple gaussian copula model for pricing CDO tranches - something I couldn't find in any other book

http://www.noelwatson.com/blog/PermaLink,g...e9b697ce1c.aspx

Thank you.... I've extended my reading list... and have also found your email address... I'll set to reading your blog soon. :-)

(HPC is still telling me that your messaging is turned off... though, it isn't a concern for me now...)

Excellent way of describing things. Just picture a whole load of investors standing around wondering who get the raisin...

A more recent auction - recovery rate was ~40%

http://www.reuters.com/article/bondsNews/i...928853020080219

http://www.creditfixings.com/information/a...s/cds-2008.html

Markit do good daily updates for those that don't already read them

http://www.markit.com/information/news/commentary/cds.html
http://www.markit.com/information/news/com...redit_wrap.html