大象传媒

2) It raises questions about whether other banks and insurers have been over-valuing their exposure to sub-prime.
Er, if they've over-valued their exposure then doesn't that mean the pain could be less? Or am I just being financially illiterate?

Despite all this doom and gloom we have to look at a couple of the long term positives. In my opinion many of the banks' share prices rose above their their true market value and a slump is very much over-due.

However, in the long term institutions such as Royal Bank of Scotland who have such a diverse range of profit generating tools will grow stronger again. Many of the banks' shares are way below their like-for-like 2007 levels and we will find out within the next 4 weeks just how damaging Q4 07 was. I have a feeling that we will be pleasantly surprised.

Insurers will probably hide their losses by shifting the rubbish into their longer expiry funds. BUt teh truth is that banks alone could not account for the $trillion damage Sub Prime has purportedly cost.
When our own Building Soceities are now admitting damage you can bet the Insurers will all come out of the woodwork as well.
Its a total sham and the Bank of England and FSA should insist they show the damage now to get it out of the way

By over-valuing there exposure, they have actually underestimated their losses.

Yes you are. Do you work for a ratings agency?

What is a 鈥渕odified Binomial Expansion Technique鈥� ?

With regard to Post No.1 - I believe Robert is describing the 'assets' have been over valued and therefore the losses to be reported could be even greater - ie IT'S GOING TO GET WORSE BEFORE IT GETS BETTER !!!!

I suspect that "over" is a typo.

AIG sponsor Man Utd too, don't they?

I think the statement means that the insurance companies have been overvaluing the securities (CDOs) that they have been guaranteeing. Instead of valuing them at the "market price", if any, they have been valuing them, for the purposes of assessing the exposure created by their guarantee, at the price suggested by their internal financial model. This valuation is higher than the "market price". However this poses a philosophical question; does the fact that no one currently wants to buy a thing from you make it worthless? Obviously not necessarily, unless you are forced to mark it to market (excuse the pun) by some highly competent financial regulator that is no doubt doing its best!

This 'declaration of independence' by PWC needs to be monitored very closely over the next few months. Pressure from politico-economic vested interests to delay, confuse and generally cloud the issue will be huge.

The moment at which this pressure becomes generally recognised as truly "corrupt" will be a pivot point in the whole credit-crunch situation, and the first real sign of hope for the financial system.

I can see where you are coming from Simon, but i see it that if you over-value somthing it is worth less than it really is, so if u over-value a loss it follows it must be worth even less than you say it is. Or is my understanding of english bad?

Simon, I think what he means is that the banks etc have been using their own valuations for the assets they have, rather than the true market value. So they will have unexpectedly large losses to report. It's a bit like cars, everyone thinks their car is worth more than it actually is.

It amazing how many ways the money can be counted with all these re-rating and revised reports. The bottom line never lies. With accounting like this, the bottom line is looking like Spaghetti Junction.

No wonder PWC are being upfront about reporting their concerns. After Enron, auditors in general have had to raise their game.

And even if the financial troubles are fully reported, there's the wave of resultant litigation to come through.

Would I be right in thinking that the bonuses paid to staff have been based on these 'mark to model' valuations?

If so, what are the chances of the bonuses also being remarked to market, and the recipients being asked to repay part or all of their bonuses. Small I expect.

Since those receiving the bonuses have been the very people who have been creating the models to mark to, are they not committing a fraud?

#1 I think it means they have considered the debt to be more valuable than it really is, so for example debt today valued at 80% on the books at 100% would make a 拢4 loan into a 拢5 loan :( if they have leveraged this stuff, borrowed more than once against the same sub prime asset this starts to multiply quite quickly i think. Be glad if Robert had a better example :)

We know that the 'model' was clearly 'wrong' but is the 'market' valuation 'right'? The Northern Rock article highlights the vast differences between the value of a book and the potential downside risk. I don't think anyone really knows until everyone starts to lay their cards out on the table.

Personally I'm not suprised about the auditors trying to distance themselves - at least they've learned something from Enron!.

The latitude given by the financial regulators to institutions over methods of valuation has been questioned, and the ineffectiveness of the credit ratings agencies to warn of the credit crisis has been highlighted, but surely nobody should applaud PWC for not participating in any fudge of valuations - as this is the duty of an auditor and the point of an independent audit?
This credit crisis has highlighted the need for liquidity risk to be factored more realistically into institutions' risk models, and I don't think that the true extent (however bad) of the crisis will become clear until regulators enforce the requirement of mark-to-market valuations in independent audit, alongside the instituions' internally controlled model assumptions.
OR, heaven forbid, a mark-to-bid valuation.....if indeed there is a bid in the market for some of these underlying assets.

Binomial Expansion Technique = Bet.

Like gambling but more risky obviously!

Can I have a biscuit?

#2 I disagree. Banks such as RBS have a long way to go to redeem themselves.

As Martin Wolf said in the FT "A financial sector that generates vast rewards for insiders and repeated crises for hundreds of millions of innocent bystanders is, I would argue, politically unacceptable in the long run."

I'm not certain whether the intention (on the part of AIG) was malicious in "hiding" exposure to sub-prime losses.

But it seems clear that this method of estimation is fundamentally flawed.

I am genuinely surprised that in such a business-critical field, "marking to model" rather than to actual market conditions is at all permitted.

As a potential shareholder, one might study these figures as a prelude to investing in the company... If they are inaccurate (and wildly so, if the loss is $5bn), then this represents seriously misleading information.

I take the point about fear being sometimes overstated in the meedja [sic] - scaremongering, in other words (18) - but for a market that relies so heavily on confidence, transparency in the accounts is surely desirable!

The sooner this is achieved, the better.

The essential truth which Mr Preston has pointed out,is that the entire financial sector have, to a greater or lesser extent, been over-stating the value of their balance sheets.

Since the financial institutions lend or insure against the value of those balance sheets, then in future, they will be forced to do far less lending or insuring of risks.

The consequences for the public are straight forward. One, the era of cheap and easy money is over. Two, the era of cheap insurance is over.

The consequences for the western economies is a significant slowdown. This will be, in my view, a far deeper and far longer slowdown than many commentators are prepared to admit.

The reasons for this are that it will take a long time for financial instutions to rebiuld the health of their balance sheets, and to admit the extent of their losses.

Greg, whilst I think fear is playin a part here go and look how the loan books have opened up over the last few years, for corporate and personal, loan criteria have gone out of the window and the likes of Northern Rock have been lending 125% (in some cases) of asset value, that now looks both foolish. This has had the impact of driving growth in house prices. Then think about the comapnies that buy NR bonds and use those asset backed securities to justify borrowing another 125% (i am being generous i think)to buy comapanies and we get another bubble this time in the stock markets. Then look at AIG who might have said we value this at face value and insure it at face value ...... doh!

The fact that you can no longer get 9* multiples or self cert mortgages without a significant risk premium or significant equity tells you how bad this has got, lending has gone back to basics. For a laugh go and ask to insure a self cert mortgage at 90% LTV.

Now take this and apply it to private equity loans, credit cards, Car HP, etc.

The institutions have no idea what this is worth hence the fear, fear comes from not being able to comprehend a risk. If this stuff was valued clearly fear would reduce. That clarity would impact those responsible for making the decisions to lend in the first place, they have no interest in being clear as it makes them look stupid and kills careers and bonuses. PWC forcing this is going to kick start the clarity (hopefully).

if you have been involved in creating this mess better resign now, ENRON II is just around the corner :)

Is there an analogy here to the use of the Black-Scholes equations to supposedly eliminate risk in the markets? Isn't it OK in 'normal' market conditions- but didn't the inventors (who won the Nobel prize!) almost bring down the entire financial system by betting billions on the back of it and it turned out not to be robust to 'shocks' to the system? It is still being used I believe.

The essential truth which Mr Preston has pointed out,is that the entire financial sector have, to a greater or lesser extent, been over-stating the value of their balance sheets.

Since the financial institutions lend or insure against the value of those balance sheets, then in future, they will be forced to do far less lending or insuring of risks.

The consequences for the public are straight forward. One, the era of cheap and easy money is over. Two, the era of cheap insurance is over.

The consequences for the western economies is a significant slowdown. This will be, in my view, a far deeper and far longer slowdown than many commentators are prepared to admit.

The reasons for this are that it will take a long time for financial instutions to rebiuld the health of their balance sheets, and to admit the extent of their losses.

"What are the implications? Well, here are a few:

1) It suggests that the latitude given to insurers and banks by financial regulators over methods for valuing CDOs and other complex investments may have been misguided and misplaced."

Rather than 'misguided and misplaced' surely you mean't to say 'criminally negligent, compounded by lies'?

LOL. There's an elephant in the room - and it's the size of the lower spiral arm of Andromeda, yet it's been hidden from plain view until now.

What has suprised me is the de-coupling of the stock market from the interbank money markets - yeah, some losses seen on the FTSE and DOW during the last 6 months, but nothing to reflect what's going on in the financial industry - the circus keeps rolling to entertain the masses. It's only a matter of time.

Tinfoil helmet strap tightened another notch!

The debate over whether one should mark to model or mark to market is tricky. At heart is the central dilemma: are the market prices correctly valuing a particular financial instrument or are they too illiquid and too irrational. Accounting standards say that if the market is illiquid then a company may choose to mark to model if they can demonstrate that the model inputs are externally generated. This is a grey area and understandably so.

I think another key consideration is big bath provisions. This is an ideal time to bury bad news and over-provide for losses. Then when markets pick up, banks can release those provisions, and hence make their future profits look much better.

Looking at the end of the AIG statement, it's clear there's more bad news to come, as AIG states that it has not yet worked out the full losses at its subsidiaries ("AIGFP"). And, readers may raise a wry smile when the statement talks about the "BET model using the Monte Carlo simulation". Who said the markets were a casino?

Re posts 1 & 18
AIG's 'exposure' is their undertaking to cover their clients' losses on debt obligations their clients hold(debt obligations issued by the NRs of the world). AIG had placed a value on those debt obligations that was too high ('overvalued') and when things started going pear-shaped they faced increased losses.
Robert P is right but did not explain the situation very clearly!

By conspiring to overstate their balance sheets these people no doubt sought to inflate their own personal reward.

In other words they have sought to obtain a pecuniary advantage by issuing false statements. In the old days we called this fraud.

The question comes down to whether you measure an asset as, say for example, 10 x likely annual revenue; or as what someone will pay for it.
The problem with CDOs is that "likely revenue" is a matter or opinion.

To be sure, some (but not all) of the debt will go bad and that will change the distribution of risk. This leads to the question of whether a large increase in the rate of default will continue, top-out or return to mean.

Similar principles apply to market value CDOs, are the prices temporarily depressed (i.e. the value is greater than the price), in terminal decline, or at their true level? AIG's model says the former, PwC says the latter (being auditor's paid to be conservative).

The AIG statement basically says they differ in their opinions but PwC's goes on record.

PwC have distanced themselves from the reporting for the year to 31 December 2007, OK. Had PwC signed off to the internal controls in 2006? If so what has changed in the year?

A situation made more complicated by:

1) In many cases the interest is still being paid on the bonds.

2) You could maintain that the true value will only become visable when the bond matures in say 10 years time and you see how much you get back.

If the interest is still being paid then you could argue that the bond is still good and worth what you paid for it.

Rather like negative equity, it only becomes "real money" if you have to sell and fund the difference between what you sold for, and what you owe.

It comes down to the difference between what you think something is worth to you (model), and what you could sell it for right now (market). There is often a large difference between those two things.

The fetish for marking to market makes the implicit assumption that "the market is always right", which is as extreme, dogmatic, and likely to be flawed, as insisting that the model is always right.

The market for CDOs is currently highly dysfunctional - if indeed it ever really existed - but the regulators are insisting that it is used as the benchmark for valuation, precisely when that is most damaging to the financial system.

One point to make is that its very hard in practice to mark to market for many illiquid (i.e. not often traded) securities. There seems to be the inference that some unhealthy thing has happened by using models. This is modern finance. away from the major publicly traded securities the use of models is often the only real measure of a security. when the market moves the re-pricing (market price) is not necessarily a better value. its true that there is now 'proof' that the asset is worth a different price, but if you cant / dont trade it at the time, you still have a dis-connect between market and 'actual' value to the firm. there has been a lot of sloppy poorly thought out practices, and there will be a lot of red ink. however, criticising the techniques themselves is naive and mis-informed. maths is a tool; risk management a culture and way of looking at the world. its that culture that needs to looked at and worked on.

"But, for me, what is most important is the declaration of independence by PWC, the auditor. PWC has made it clear that it will not participate in any fudging of the scale of the sub-prime disaster 鈥� even where full and painful disclosure has the potential to undermine market confidence.

So in the coming few weeks the very worst of the losses from banks and insurers should be on display for all of us to see in their gory detail 鈥� and there may yet be further unanticipated horrors."

I wonder if your confidence in PWC's ability to assert their independence is well placed. I'd suggest that there are two major thoughts in the minds of senior management in the financial sector:-

1. How can we delay disclosure of our losses until the green light has been given for a general stampede, during which it may be possible to portray the write-downs as a general market adjustment, and not as the "fruits" of management mistakes and incompetence.

2. When this stampede occurs how can we overstate our write-downs to the maximum, thereby giving us the greatest opportunity to "assist" the post-fallout results through the release of unnecessary provisions.

Have PWC and their peers really got the muscle to prevent managements from doing these things?

Dear Mr Peston, I believe that you should be reported to the Plain English Society for this article.

Having read through your piece, and then all the comments, it seems that either you don't know what AIG were saying or your readers don't understand what you are saying... or both.

Can I try to summarise:

AIG have done their accounts and made provision for insurance losses relating to the credit crunch fiasco.

PWC have taken a look at the figures, burst out laughing, and told AIG to get real.

PWC reckoned they had over-valued their policy assets, therefore under-valuing their potential losses.

AIG had to raise their hands and admit they had screwed up, and now the stock market guys are pooping themselves..... yet again.

It seems to me that a new type of insider trading has been taking place, with it being in too many individuals, businesses, banks and governments interest to pretend that the good times would roll on for ever and that no-one would ever want to call in or to check the vacuous and complex promises that everyone was making.

Some banks seem to have got out of passing around irredeemable debt when they thought they had pushed their luck to the max, whilst others got greedy and were simply left holding everyone elses' worthless pieces of paper.

It doesn't seem much more complicated that this to me.

If you keep an economy afloat by encouraging people to buy houses that you know you cannot afford, or by allowing banks to lend money to the poorest people around, then governments and regulators must have known what they were letting happen but decided it was a price worth paying.

Just who is this Robert Peston?

(i) This is not a cash loss, but an adjustment to the balance sheet based on a modification of the fair value. The FV issue is still contentious.

(ii) The BET model is taking into account credit market spreads and therefore when there is no market or the market declines, there is a problem. So Mr. Peston is wrong. This IS marking to market but using a model. There have been calls to mark to models without reference to the credit markets, because of the vagaries of the credit markets.

(iii) I find this blog overall very depressing. There is never any balance. CDO-bashing is wrong. CDOs it have funded the econmomies. More balance on this aspect rather than bleating "toxic waste" "toxic waste" "toxic waste" is not balanced reporting.

The fact is without CDO products many many people would never have been allowed credit. And now, the anglo saxon style economies cannot cope as a credit junkies having their drug-supply (credit) lines cut. This is a 10 year addiction of enormous proportions.

The solution relies on bringing certainty to the credit markets not only the high quality credits but across the entire market -- or thousands of poorer people will be denied credit. That is not good.

More blogs on market reform would be welcome, but it must be balanced as CDOs etc do an awful LOT of good - and as an expression of what the capital markets are there for anynow. The captial markets are there for one thing only: to disperse risk. They have done this extremely efficiently for over 10 years and allowed millions of the poorer - the honest poorer -- the power to develop their own budgeting and lift their aspirations.

To turn off the CDOs would be like asking us all to go back an live in caves. Unless one is the most extremist of environmentalists, no one would do that. So stop the extremism when it comes to CDOs.

CDOs ARE for the good of EVERYONE.

it's about time the insurance companies and asset managers started to get a mention. The banks who are supposed to be the experts have been holding on to the best quality assets and these have turned out to be rubbish. so it follows that the absolute rubbish would have been sold to their customers. I wouldn't be surprised if I was to look into my pension providers fixed income portfolio and find that they had any number of oddly named bonds that turned out to be CDO's linked to US Subprime assets.

Any person or company that has been deliberately muddying the waters to cover up for it's own greed, avarice or incompetence (or as is increasingly the case a combination), any bank or lending house or government that has gone along with this, thoroughly deserves what's almost certainly going to happen.

When you think that last year, the top five US banks paid out $38bn in bonuses to people who essentially created this problem, it makes you wonder if they should give it back!

This information translation shows clearly what AIG and others have hidden from their disclosures for some time.

Those who define the measuring criteria, somehow meet it...always.!

Should not there be an assumption by regulators that complicity and creative accounting are the norm and therefore independance, of all financial calculus should be defined by the regulator.

It is the regulators role to protect the public, whether they be misguided investors or interested by standers. Removing the power of these corporations to define the rules has to be the top priority.

On the subject of Black Scholes etc., at the time of the equity market crash of 1987, I was working in New York and was interviewed by a member of the President's Commission that looked into the role of programme trading in the crash. We discussed the difference between the anticipated probability of the event occuring prior to the crash as implied by option prices (1 in 5 million +) and the actual observed probability (one in about 25,000) based on the actual number of such events since trading on the NYSE began. We all think air travel is safe, based on the long odds of death. Would we get on a plane if the odds were actually as different as the financial models were from financial reality?

Robert, nice to have your public vote of confidence in PWC. Mine too, they are a smashing firm. Nevertheless, as with Andersen's in Enron, accounting corporations are by definition themselves compromised. Problems keeping to their own side of the fence... losing their bearings... inevitably having to save face. and that's why regulators are so vastly superior in the field. Impartiality. There's muscle in it. read robustness. Regards

Whether "Marking to Market" is preferable to "Marking to Model" depends on how sound the models are. We know that markets tend to over-correct on both the up- and down-side so there is some sense in using models based on sound maths or long-run statistical data in preference to "Marking to Market" (particularly when market participants can, and do, manipulate the market at key times for this purpose). The problem is that neither bank managements nor financial regulators (nor financial journalists, for that matter) understand swap and option modelling/valuation techniques.

#37 Thank you - an excellent summary!

Post 27 : observer : 1.01pm

big bath provisions

You beat me to it!

I agree with Seamus that you should be reported to the 'Plain English Society'..

"There will be particular concerns about those financial institutions which have tended to mark to model rather than to market (you know who you are)."

Que ?

1: The binomial expansion technique assumes that defualts of the underlying reference entities in the CDO are independent from one another (a gross simplification, since defaults are correlated over time) and they have the same probability of default (another simplification) meaning that they can be modelled using the "binomial probability distribution". (Google it)

However modifications can be made to this technique.

One thing to point out is that the m-BET used market prices in deriving default probabilities, however due to the assumptions of this model, obviously not all effects were included leading to a decline in the value of the super-senior swaps not predicted by the model.

AEG's statement basically boils down to the fact that they believe PwC estimate to be overly conservative, but the auditor's word is what goes on record.
The issue with "complex" investment vehicles is that without sufficient research (of the particular product) they are essentialy "lucky bags". Too many of them were bought for their return without due care and attention to the risks "The property markets are going up - just let the money role in".
I would defy anyone to work out the value of a hybrid mezzanine CDO cubed (with a CIO element). The people bought these things on behalf of other people did so either negligently or with flagrant disregard,
Don't get me wrong there is nothing wrong with transparent CDOs (spreading risk and returns), but some of the stuff is akin to a wholesaler sticking malt whisky labels on moonshine; then the retailer claiming they could not tell the difference when people go blind.

Don't AEG make fridges? Don't tell Peston but its worse than we thought... even white goods manufacturers are exposed to sub-prime losses.

The market price of an asset is the best price anyone will pay for it - financial institutions have to mark their assets to market so that we know how solvent they are. If no one will buy an asset at any price, it's worthless, full stop. Only chaps in the City (and Nobel Prize-winners) don't understand this.

Robert, there is nothing wrong with marking to model. Because the market for super-senior CDO tranches has frozen, it is in the ONLY way to value them. Also, if you'd read the AIG statement properly you would have seen that actually marking-to-model gave a massive writedown of $5.9bn. The whole problem is that AIG's management then decided they didn't like the model and added their own, completely arbitrary, "benefits" to the pricing, which reduced the writedown to $1.6bn. PwC didn't have a problem with the model - just the management changes to it afterwards.

Robert I read and fully understud your report. It makes very clear sense.
A very informative and wonderful insite into the crazy world of investment finance.

Bravo

Monte-carlo simulation is a recognised method of assessing the impact of posssible errors in the data (and sometimes model parameters). Each datum (yes, sing. of data!) is assigned a likely error range and the whole calculation is repeated many hundreds of times with random variations in all the data over their permitted error ranges.
If the results are well concentrated the model is consistant (this does not mean it is right!). But if the model is sensitive to a critical parameter or datum, the results may diverge wildly which means prediction may not be reliable. This is used in weather prediction and in engineering.

There are several theoretical flaws remaining of which these are just a few:-

1)Mathematical models represent beliefs and assumptions about a system, not the system itself. One of the assumptions in Monte-carlo analysis is that the error margins are realistic and statistically independant, which may not be true. But the whole structure of the model is also an assumption.

2)Economics involves the collective consequences of many individual decisions so the beliefs of the individuals are important, and may change over time. No economic model is capable of being 'True'.

3)When a model is used to make decisions, it becomes part of the system itself. Which means an accurate model must include both itself and the consequences of peoples belief or otherwise in the way it and the system operates. Needless to say few models even attempt to do this properly, which is why no economic model lasts very long before being abandoned.

Surely the whole problem with the use of models is when there is insufficient information to accurately set the parameters of that model. There is far too much extrapolation based on past market behaviour. Binomial distributions are pretty basic maths. Predicting mean future levels of mortgage defaults based on what has happened in the past is much more tricky. If you get that badly wrong then you are going to lose an awful lot of money, no matter how "sophisticated" your model may be.

The risk of this happening is greatly compounded when there have been dramatic changes in the nature of the market. For example when the US money market suddenly starts lending huge amounts of money to sub-prime borrowers who have never been lent to on that scale previously. There was just no evidence as to how they might behave if money tightens up.

Similarly UK house prices have never been as over-valued and unaffordable as they are now. We are in uncharted waters, so who really knows what level of defaults could be seen if there is a major house price crash? Extrapolating from past, smaller housing bubbles might not give the right answer.

And the extensive use of CDOs to "spread the risk" is what has allowed lenders to feel they could expand rapidly and yet safely into new areas without waiting for true empirical data on default levels to emerge. How wrong could they be.

Binomial distributions describe the outcome probabilities of a series of individual events each of which has one of two outcomes.

Basically what's the odds of getting the best of three when tossing a coin :-)

If you remember drawing Pascal's Triangle from school then you've expanded it as well

High finance eh?

#39 - encouraging people to borrow beyond their means is not good.

Soap Box corner time again for Robbo.

1 - ALL auditors are going to give the most SEVERE treatment of all structured assets following the collapse of AA after Enron.

2 - By their very nature these assets are difficult to value, and given the volatility of the current market, and absolute lack of transparent pricing today (as there is no open market), it is nigh on impossible to give a known, concrete value. AIG is one of the world's largest insurers, with actuaries and consultants coming out of their ears. So it will not be as if someone dreamed up a number. It simply happens to be a market that is baying for blood and an auditor who has to appear whiter than white.

Mr Peston, I suggest that if you are such an expert, why not fly over to the US and show them how it all should be valued?

Everyone is nervous and It's a mess with the volatility in the markets. AIG seem to be saying she'll be alright, but there will probably be some bad news.... not reassuring and I think contradicting.

Generally what we have to remember is that what goes up might come down... sometimes crashing down back to earth!

#59 - this is #39

I did not say encouraging people to over extend is good. BUT One of the biggest advances has been micro-credit. It is even a Worldbank/UN program. How is that funded? Tea party plate collections in middle England, or the capital markets?

This is why i really dislike Pestons hysteria.

Most people in the UK have over borrowed by falling prey to the housing bug. As a result, UK housing is vastly overpriced - because there is a trend to beleive one can be a millionaire by moving up the property ladder. Greed. Not bank greed. Just a bank taking the opportunity associated with people's greed.

But I think what really annoys me about sweeping generalisms is the lack of awareness of what the market is about, and now everyone is an expert. As Rockefeller supposedly said 鈥渨hen your shoe shine boy passes you a tip, it鈥檚 time to liquidate鈥�. And Rockefeller avoided the 1929 for that very reason 鈥� he liquidated his positions the day his shoe shine boy suggested a stock move. Why? Because it was indicative that the stock market was trading close to sentiment only 鈥� ad he was right - and this is what we have here. And the sentiment is wrong: CDO bashing is wrong.

Most students in the UK have over borrowed. Indeed but for the CDO market, of themselves, many student loans would be so unattractive in the capital markets many children would be denied the option to go to university. So see CDOs in their context. Banks loan to students with no guarantee at all that student will even get a a job, never mind a well-paid job.

And what makes that possible? The capital markets.

The capital markets ARE there to disperse risk - NOT to make moral calls as to whether john and jane doe can be trusted to not over-extend themselves for anything other than a new designer handbag.

Stop blaming banks, stop blaming CDOs. And undersand the responsibility for budgeting lies with the individual. Not the nanny state and not nanny state banks. And definately not the capital markets.

Re post 61.

We saw an interesting comment from the CFO of Goldman Sachs the other day where he said credit markets were priced as if we were in a very, very severe recession.

I think the absolute worst has already been built into markets, although bad news such as AIG / FGIC, LBO problems etc make it even more depressing.

On the brighter side, reading Buffett's transcript on CNBC the other day, it would almost look like the Sage has started to call a bottom.

I'll just like to point out a couple of things about financial models.

It is important to recognize that there will always be necessary assumptions made to build a usable model. Financial markets are unlike physical laws; financial markets are driven by market participants and they do not obey universal physical laws.

A lot of people (even practitioners) are very confused by terminology like 'prices calculated from expected value' or 'expected default rates'. The whole point of modern finance is that we have developed mathematical techniques to REPLICATE prices under certain assumptions. It is a wrong to believe that models are purely statistical - ie, the price of a derivative is somehow the empirical expectation of a contingent claim etc. There will be some statistical assumptions underlying the models, but prices are calculated based on hedging arguments that would replicate the price to some sensible variance. The underlying models may become unusable, and that is why we still employ quants and traders to adjust for new realities.

Now, the hard part is to understand when the assumptions break, and how these assumptions affect the price of the option. I think banks/institutions should be entitled to mark to model if they can demonstrate that they are able to replicate the price of the instrument. It is also false to assume that that the market is fully efficient, and if one market participant is able to replicate the price of the instrument, then the existence of arbitrage would require the market price to converge very quickly to the replicated price. Other participants may not have the same access to the replicating portfolio, or they may encounter higher transaction costs, or the structures required to trade the replicating portfolio is not available.

I am not arguing that there is always a perfect replicating portfolio/strategy behind every price. In some cases, it is quite possible with acceptable variances. The insistence to mark to market in all circumstances is a conservative lazy man's refuge.

BTW, if anyone cares to check, you'd might like to look at the US listed equities options official market prints everyday. You'll find problems with inconsistent synthetics and boxes all over the place even on very liquid options with tight spreads. On more illiquid options with wide spreads, even the idea of mark to market has significant problems.