That's exponential growth - it must be nearly doubling every year

Exponential growth that will end with a collapse.

Could you explain why please

When you say the markets move to fast what is wrong with (for example) using risk from prior day and adjusting your intraday P+L based on spread movements*DV01. This is what some banks do. I can't see a problem with this?
Agree that some of the modelling for the more exotic stuff is flawed, but I would be interested to have an example of a reporting black spot

Could it be the numnber of corparate CDS with protection on them has gone up to times in four years and hence the CDS written per name hasn't changed a great deal?

I think I can explain the logic of that (and I suspect I know why you baulked.)

If net CDS are growing exponentially, then either debt is growing at least as quickly - or there is greater profit in a collapse (by collecting insurance) than in the debt itself.

As we've discussed before, we don't have figures for net CDS - only total nominal value... and we assume that one institution is laying off debt to another... but this implies that each year new backers still need to be found... and if growth in issuance is exponential, you'll very quickly run out of assets to back the insurance... unless the issuance becomes cyclical... which, again would ultimately result in a collapse - since when a default occurred, an institution who thought they had laid off the risk had also taken it on again.

The usual logic for any exponential growth is that it will break. At some point it will not be possible to grow exponentially - and then the consequences are usually abrupt. Maybe this is being managed - maybe all the issuance is being carefully managed and there are strategies beyond CDS when there are no credible insurers left... but I doubt it.

The only question I see is to ask if something else will collapse before CDS expansion becomes impossible.

"CDS are growing exponentially, then either debt is growing at least as quickly - or there is greater profit in a collapse (by collecting insurance) than in the debt itself."

I assume by debt you mean the underlying bonds (assuming we are talking about corporate). Could it be that the number of corporates that have an active CDS market traded against them has increased?

"but this implies that each year new backers still need to be found"

or the same banks have traded more with each other. Don't forget that the market is still in its infancy, so say you did a trade 3 years ago on a 5 year product, it would still have 2 years to run, and the gross notional will remain. In the meantime you have continued trading so I'm not surprised that the gross values have increased

"and if growth in issuance is exponential, you'll very quickly run out of assets to back the insurance... unless the issuance becomes cyclical... which, again would ultimately result in a collapse - since when a default occurred, an institution who thought they had laid off the risk had also taken it on again"

When Delphi defaulted, there was a greater notional of protection written than underlying bonds. While the recovery rate was higher than expected for the index auction, the auction itself went without a hitch. I'm not sure why an institution would find they had unwittingly taken on risk again - are you talking counterparty risk?

It certainly could... but, if the underlying bonds have been growing more slowly, pretty soon all will have a CDS against them... Real numbers might help here, of course, I don't have any.



Yes, absolutely, there are lots of ways in which the derivative backing could become more complex... but again, while I hope I'm not making too much of a simple point, eventually the "efficiencies" found by increased derivative trading will expire and that in itself will cause a change in the direction of the market... both for credit derivatives and the credit they promote.



I think I'm pointing at things from a greater distance with a lot more hand waving than you are. At the simplest level I'm saying that exponential growth in any system is an indicator of (fairly imminent) change - and if I go even bigger picture - I can say this for sure because we've limited resources - which implies a bound on expansion both from supply and demand. Exponential growth in any context should scream warning sirens... even if the nature of any problem is unknown. Once exponential growth is established, it is necessary that either it will be changed abruptly by an external shock - or it will cause a shock itself... it is only a matter of time.

Several things remain unclear - and here your more detailed analysis works better... because we have no idea how long this matter of time might be - or if the shock will be in Credit or Credit Derivatives. I think you're arguing that a potential shock arising as a consequence of credit derivative expansion is a long way away... in fact, so far away that it is all-but inevitable that an external shock will halt credit derivative expansion before it becomes the problem itself. This is a very difficult call to make - and I respect your fine-detailed analysis that suggests that there may be no imminent threat from the expansion. I don't, however, think your arguments are sufficient... I think we need to establish (over a significant period) the following:

* Rate of Net expansion in exposure to credit derivatives. (Yes, I know this would be hard/impossible to discover...)
* An estimation of the impact of credit derivative expansion on credit issuance. (Also hard - though I believe I found a compelling informal correlation between the two looking at broad money in the UK...)

Until we can quantify the above, all we've established is the potential within an unregulated market for CDS to have been used to transfer credit risk to those with insufficient assets. I'm a cynic - where I see opaque systems I assume abuse.

I am not disagreeing that there may be trouble looming, but

a). You or I or the rest of the world proably don't know what it is - did anyone say the housing market was going to plummet beause the banks stopped lending - I for one thought it would be driven by the consumer
. I think we need to all understand what the potential problem may be (as we are doing now, rather than cutting and pasting articles and highlighting random sentences!)

"Exponential growth in any context should scream warning sirens"

Agree

"I think you're arguing that a potential shock arising as a consequence of credit derivative expansion is a long way away"

I'm not saying that. What I am saying is what has changed from 2 years back when Delphi defaulted. Are we worried that more companies are going to default in short succession, is it the fact that new players have entered the market - we need to understand what is different from a couple of years ago

* An estimation of the impact of credit derivative expansion on credit issuance. (Also hard - though I believe I found a compelling informal correlation between the two looking at broad money in the UK...)

It depends on how far back you are looking. Bloomberg shows M4 YOY to be 18% at end of 1989 dropping to 3% in 1992. Correlation between M4 and house price growth may be a different matter

a ) I certainly don't - but I can be analytical at times. I rather hoped that you'd have better raw data than I do.
b )



The only thing we know has happened is that the total nominal derivatives have grown exponentially. Even if all we are seeing is more complex hedging strategies, this implies an additional risk (for abandoning the KISS principle) - even if we can't quantify this risk from public statistics. One thing that is different from a couple of years ago is that either some participants have much greater exposures, or many more participants have exposures... this added complexity (however it arises) definitely poses some kind of risk.

Your M4 seems far more substantial than mine was - I noted inflection points at 1997 and 2003... I understood that 1997 was when Synthetic CDOs went mainstream (and I was sure 2003 was relevant for a derivative reason too) - but, as I'm sure you will point out - I'm aware that other relevant things happened in those two years. The inflection points for M4 seemed to coincide with the inflection points for HPI... (maybe I'm clutching at straws here....)

I am naive about financial risk management and derivatives trading. From my layman's perspective, I see the following

- everyone (esp Bernanke) compares this crisis to the Great Depression. But the intertwined complex web of derivatives contracts makes this different qualitatively in some crucial aspects
- the importance of derivatives in this crisis seems underlined by the rumours swirling around the Bear Stearns bailout (done to prevent a chain reaction of counterparty defaults stemming from derivatives)
- no-one seems to know what the effects of derivatives will be. The greatest investment minds (eg Soros, Buffett) compare them to WMD

Given all this, why are the regulators/ central banks not insisting on full disclosure of ALL outstanding derivatives contracts (including the OTC variety) as part of an attempt to defuse the (as yet) unexploded financial bombs that may threaten the global financial system? Surely all the contracts are precisely specified so their effects, however complex, could be worked out

I'm not sure about this... I know that derivatives called "derivatives" that can be modelled by Black Scholes and synthesised from credit and frequent trading are fairly new... but, if we take a step back, derivatives are no different, really, than bets placed at Ladbrokes. I vividly remember a story from when I was at Primary School (I can remember the teacher talking about it) - and (I apologise for this being rubbish - I was <11 years old... so, pre 1985... at the time and wasn't paying proper attention) the general gist was that "a long time ago" ™ "somewhere" ™ everyone was doing rather well - lots of productivity; lots of investment and progress year on year... until, at some point, the investors realised that they could make greater profits by betting about which businesses would do best - rather than investing in the businesses. A successful year's investment of 100 coins might yield 20 if the right investment was picked - or loose 10 if the wrong one. Gentlemen investors realised that they could put their 100 coins to better use betting against other gentlemen - since in a year a successful 100 coin bet would give a much better yield than a 100 coin investment. Unfortunately, starved of investment the businesses failed - everyone - rich and poor - went hungry.

Were "Gentlemen" making bets rather than investing prior to the Great Depression? If so, to what extent can we view derivatives as being the modern equivalent?



The regulators can't manage to regulate on Eric's "LIAR LOANS" - let alone an over-the-counter derivative market where innovation happens orders of magnitude faster than civil servants move.
The derivatives could be worked out - in principle - if the participants knew what their counter-parties had... but that sort of information is business sensitive, so kept secret.

"I rather hoped that you'd have better raw data than I do. "

I have access to certain raw data but does that paint an overall picture? Probably not. For example

I am on the sell side flow desk, so do I know what hedge funds are up to? Not really, because they will trade with other counterparties other than my bank. The same goes for otherbanks - how do I know what their exposures are? I don't.

So all I can really do is say for a typical bank, we trade x names, x indices, and theoretically we would have x exposure if x entity defaulted. I would be very surprised if other banks/hedge funds had radically diffeent strategies, although the monolines seem to have sold a lot of net credit protection.

Assuming one of the perceived problems with Bear were the CDS exposures, it may be that they were net neutral in their CDS positions, but to unravel it with several hundered other counterparties would have beena nightmare - hence your comment about more players adding to the complexity

As you say, derivatives have been round for centuries

http://en.wikipedia.org/wiki/Derivative_(finance). Even equations very similar to Black-Scholes have been around since the turn of the century

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1012075

I think the problem may be in the growth of the OTC market, and also the more complex types - getting an option value based on an underlying is easy, getting a CDO valuation is a lot more tricky

I know that you're better connected to sources of information than I am.

Your comment about a perspective at an individual bank is very interesting - though probably not for the reason you thought. Let me wander a bit...

I am absolutely fascinated (and have been for many years) with the idea of emergent behaviours... systems in which seemingly innocuous local behaviour gives rise to surprising global behaviours. There are numerous examples in the context of AI - for example 'swarm algorithms' - where seemingly innocuous individual behaviours as a collective exhibit some unexpected function. I suspect that this might be an analogy for markets - where, for all the local genius, might result in an emergent catastrophe. One of the most simple examples is where balance sheets reference other balance sheets - and risk/value are calculated for each balance sheet - assuming they form a directed acyclic graph of dependency... when in fact they don't. As the system becomes more complex, I expect, it becomes ever more difficult to analyse... partly because it is easy to overlook flawed analysis when there's a seemingly credible happy outcome that pays huge bonuses - but also because I don't have blind faith in the honesty of individuals... and I expect more fraud where it would be hard to detect/prove.

It is my impression (not based upon experience, you understand) that much of risk analysis draws on the principles of calculus - where prices and their first and second derivatives are calculated (ignoring higher orders) relative to "interesting" external metrics such as time, interest rates, exchange rates, asset/future prices - etc. In a sense this kind of analysis is very impressive - and would likely work brilliantly if only one participant adopted the scheme. My gut instinct is that, for all the apparent rigour of the approach, this doesn't eliminate risk from a system-wide perspective... it just hides it. For a single adept market participant hiding the risk on a competitor's balance sheet is a winning strategy... but as more and more participants mimic this approach, the risk can't be eliminated everywhere. The additional complexity - for all participants - gives rise to risk likely residing with the least adept... who, likely, have the least to loose - and hence can't be held accountable for their losses. I don't believe that Banks have been positively productive in booking their huge profits in recent years... I think that these profits have, to a large extent, arisen where risk has fallen between the cracks in contracts. One possibility might be default risk where derivatives are traded on margin, another might be that an index proves to be a poor metric for a measure it is thought to closely approximate.

The only way I can see to get any grasp on the real risks is to establish the right sort of system-wide statistics. As soon as we're in a situation where hedge funds, pension funds and banks's funding, solvency and liquidity are interlinked, which I believe we are, then it is effectively impossible to base useful predictions on the perspective of just one bank - typical or otherwise. Of course, as complexity builds, this kind of prediction becomes more and more difficult... I'd not be surprised if it was all but abandoned today. I see, from a distance, that banks have pressed for every optimisation of profit possible - which, in turn, has lead to thinner and thinner margins - which, eventually, means that some trivial unexpected deviation will be enough to cause the entire system to collapse. While I admit that there will be will to forever intervene to prevent a crash - eventually a crisis will either be missed or be too complex to resolve politically.

My interpretation is twofold. First, the risks of default must lie with someone other than the borrower. Second, even if the lending is hedged properly with viable underwriters, there remains a risk that an organisation may be overweight buying CDS - just as Goldman were - and discover that there's a profit to be made from bankruptcies. If an organisation is in such a position but is stressed with huge unknown risks itself, it might decide to take a hugely risky step of sabotaging its own customers for the default-swap payout. For an individual bank (hedge fund, etc) I expect the greatest risk is that it may overlook the inter-dependency of its exposures. As long as any market participants trade on margin, I don't see an account of risk from an individual institution's perspective is going to be all that helpful.

Noel and A. Steve,

Fascinating and informative debate between you two. Just wanted to say how much I appreciate it.

Thanks.

My gut instinct is absolutely with Warren Buffet on derivatives. I also think they raise numerous ethical questions that drill right down to fundamental philosophical questions of value, worth, property, business, trade and profit.

Where I bumbled on about my recollection from the 80s in primary school - if there was a point - it was that I don't think that the problem is an unprecedented growth in an OTC market - since, I doubt very much that "a very long time ago" regulation of 'bets' was any better than it is today. If we had an adept historian, I think it would be interesting to draw comparisons between then and now - if we ever discover when this "then" actually was. The economy will have been smaller - but, that would have been offset by the concentration of wealth among the aristocracy...

ZCZC CGNAOGLD2 ALL
TTAA00 KNHC DDHHMM
BULLETIN
MAJOR FINANCIAL MELTDOWN ADVISORY - UPDATE
NWS TPC/CGNAO
FRI AUG 3 23:08:14 CEST 2007

...POTENTIALLY CATASTROPHIC CATEGORY FIVE FINANCIAL STORM
INTENSIFYING TO UNPRECEDENTED LEVELS
...SUSTAINED SELL-OFF THREATENING THE CURRENCY MARKETS AND
THE INTEGRITY OF THE INTERNATIONAL MONETARY SYSTEM
...SIMILARLY SEVERE CREDIT MARKET CONDITIONS WERE LAST SEEN
AT THE ONSET OF THE GREAT DEPRESSION IN 1929

A FINANCIAL CATASTROPHE WARNING IS IN EFFECT FOR THE WORLD
CURRENCY, STOCK AND BOND MARKETS.

THE CREDIT CRUNCH ORIGINATED IN MORTGAGE BACKED SECURITIES AND
RECENTLY SPREAD TO THE LBO AREA CONTINUES TO GATHER DESTRUCTIVE
FORCE. CONTAGION IS NOW SPREADING TO CORPORATE BONDS AND IS
FORECAST TO EVENTUALLY AFFECT THE SOVEREIGN BOND MARKETS.

LIQUIDITY SURGE - THE FEDERAL RESERVE OF NEW YORK AND OTHER CENTRAL
BANKS AROUND THE WORLD ARE EXPECTED TO SOON UNLEASH A WALL OF
LIQUIDITY IN THE ATTEMPT TO STAVE OFF THE UNAVOIDABLE COLLAPSE OF
MAJOR INVESTMENT BANKS AND ULTIMATELY OF THE BOND MARKET.
SUCH ATTEMPTS ARE UNLIKELY TO BE SUCCESSFUL.

IN ANY CASE PREPARATIONS TO PROTECT CAPITAL BY SELLING ALL PAPER ASSETS,
IN PARTICULAR THOSE DENOMINATED IN US DOLLARS, AND BUYING HARD ASSETS
SHOULD BE RUSHED TO COMPLETION AS SOON AS POSSIBLE.

http://www.webofdebt.com/articles/banking-bailout.php

Magic.

Time for a black swan I think.

After the AMBAC/MBIA downgrades, another crisis upshift coming.

Exponential growth of the amounts involved, this is the mark of the derivative beast.

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

June 6 (Bloomberg) -- Royal Bank of Scotland Group Plc may have less capital than its largest peers and go back for more after raising 12 billion pounds ($23.5 billion) in Europe's biggest rights offering, according to Deutsche Bank AG.

Hadn't heard about rescap, but here we have latest news.

http://www.iht.com/articles/2008/06/05/business/05gmac.php

These are huge sums, $60bn, and we know that B&B are compelled to pick up some GMAC mortgages.

Maybe the UK interest in the B&B has led people away from looking into GMAC?

Cerberus capital already own 51% of GMAC, and had to come in with fresh money.

This does rather all tie in with the idea that the FED orchestrated the TPG investment into B&B.

The difference between BS and the Wellington rumours, is the BS rumours were true.....

This time however, its a bit like the Enron explosion, the fraud and mis-management are obvious to anyone who can read an annual report.

More players do add to the complexity of the derivatives market, yes, plus what really increases the complexity is what some one said above the way the contracts are written and how they are carried on.

Can you clarify please?

I assumed Crash 2006 was referring to the fact that OTCs are not a standardised contract - and by "carried on" in inferred "hedged".

For CDS single name, indices and tranches, contracts are standard (for the vast majority of deals that I am aware of)

http://www.isda.org/2002masterprot/2002masterprot.html

http://www.isda.org/2002masterprot/docs/2002Protocol.DOC

http://www.isda.org/2002masterprot/masterprot_hg1.html

Can we quantify either the nominal outstanding for these standardised CDS contracts, or - preferably - a more detailed picture?

Are these standardised contracts traded OTC, or in a market? I presume that these standard contracts are not interchangable - but rather represents a cookie-cutter approach to defining a contract? If so, what impact on risk assessment have the contract's parameters?

After a bit of digging around it seems that there is very little that doesn't use this documetation - of course this may have been different in the past. This documentation is used over the inter dealer brokers. The documentation will differ from region to region. I would need to go and have a look at the ISDA documentation to find out more

ISDA wikipedia link

http://en.wikipedia.org/wiki/International...ves_Association

The "close-out netting provisions" sound extremely relevant... though I wonder if the conflicts over restructuring events might be a red-herring.

I guess we need to establish how "close-out-netting provisions" take account of counter party risk.

It's all kicking-off tonight!

Looks like we're pretty close to the end game.

we're still a year or two away for the wider economy methinks.