"The range of derivatives contracts is limited only by the imagination of man (or sometimes, so it seems, madmen)." -Warren Buffet

January 2, 2008 1:44 PM

A primer on the global derivatives market, the City of London, and the credit crunch:

"In 2003 the total size of the world economy was $49,000,000,000,000. The total size of the derivatives being traded was $85,000,000,000,000. In other words, derivatives today are worth far, far more than the total economic activity of the planet. More than $1,000,000,000,000 of derivatives are bought and sold every day. Every single thing that can be traded through derivatives, is."

posted by anotherpanacea (30 comments total)

COmparing derivatives against the world GDP is a bit deceptive, because derivatives deal in things which already exist, including things from previous years, and decades, and centuries.
posted by Steven C. Den Beste at 2:12 PM on January 2, 2008

Steven, can it possibly have escaped your notice that GDP and GWP also include things which already exist? Why bother commenting if you don't know what you're talking about? If I sell you my car, it's the same car I bought three years ago, which was built the year before. The sale gets added both times, because GDP is better described as a measure of the size of the economy than of its yearly productive capacity.

I'm not knocking derivatives which are fundamental to managing risk, but like most half-way intelligent folks, I'm also troubled by their opacity and the tendency of that opacity to encourage stupid risk-magnifying moves like those of LTCM, Northern Rock, or the whole subprime mortgage fiasco.
posted by anotherpanacea at 2:26 PM on January 2, 2008

...can it possibly have escaped your notice that GDP and GWP also include things which already exist?

Of course not. But they don't include all the things that exist.
posted by Steven C. Den Beste at 2:38 PM on January 2, 2008

One of the ironies of the LTCM collapse is that the predictions on which they based the trading positions that caused them so much trouble turned out to be pretty much spot on. Didn't stop the market moving against them inbetween times though.
posted by pharm at 2:39 PM on January 2, 2008

Well, drunk sailors tend to be spending their own money,’ Tony said. ‘By contemporary standards they’re quite prudent.’

Incredibly prudent, the way it was till it became fashionable to spend without knowing how to read a contract and how to calculate , even with the help of a calc, how much 15% is over your debt , not mentioning that little fact that one person payment don't necessarily reduce the debt at the same rate at which it pays the interest..making the debt a lot more expensive then one would think.

But the opacity of these instruments makes me wonder , quite a lot, expecially of the competence of the people with advanced degree in maths that could probably change the form of an equation long an arm, but may overlook that risk profile because the risk information may not be expressed adequately in the equations. I do have a friend with such a degree, but I just don't want to embarass him with question such as "did you guys realize maybe you just didn't appreciate what you were doing ?" . But what's the variable for "pass the bucket" ?
posted by elpapacito at 2:40 PM on January 2, 2008

One wonders how much credence we can place in GDP as a meaningful (let alone tangable) measure of anything.

That said, I've been wondering what percentage of GDP ends up as assets of people involved in the financial sector. It would be nice to know how much money is being used to feed the system, particularly if we consider the market as a means to an end. Like the derivative-to-GDP ratio, it wouldn't be a useful number so much as an interesting statistic.
posted by FissionChips at 2:57 PM on January 2, 2008

In many cases the notional value of a derivative is drastically greater than the actual value at risk, especially with vanilla transactions like interest-rate swaps. See this explanation from the Dallas Fed.
posted by mullacc at 3:31 PM on January 2, 2008

It's mentioned in the article and I have mentioned it before on here but if you want to read further on this topic Phillip Coggan's The Money Machine is a fantastic book on the City. Well worth reading.

A nice link too. Thanks.
posted by ClanvidHorse at 3:31 PM on January 2, 2008

If I sell you my car, it's the same car I bought three years ago, which was built the year before. The sale gets added both times, because GDP is better described as a measure of the size of the economy than of its yearly productive capacity.

I'm sure a real economist will come along shortly, but I'm a fake economist and I'm pretty sure that you're wrong. Stuff that's already made doesn't get counted.

The whole thing is a bit ridiculous anyway, of course, because painting my house counts toward GDP when I pay you to do it, but not when I do it myself. Which is weird.
posted by Kwantsar at 3:40 PM on January 2, 2008

Mullacc -- thanks for that critically-important qualification. Discussions that compare the notional amount of derivatives contracts with some tangible non-notional economic figure are in trouble from the get-go.

To say that the derivatives market is larger, or even comparable in size to, the whole economy, or even the whole cash securities and loan market, is wrong by pretty much every meaning that one would set by such size comparisons.

This is not to say that there aren't huge amounts of derivatives nor that there aren't likely to be some institutions with too much derivative exposure. However, it's interesting to note that virtually all of the 2007 credit crunch damages has been inflicted on people who had too little derivative exposure, not too much. The big winners (hedge funds who were mortgage bears) took on vast derivatives positions and won; the big not-loser (Goldman Sachs) avoided being a loser by offsetting the huge credit positions their leveraged finance and asset-backed origination businesses required with equally huge derivative hedges.
posted by MattD at 3:46 PM on January 2, 2008

Ha, when I began reading that this thread was at 0 comments. It is absolutely worth reading the whole thing, what a fantastic read (and not in the slightest about GDP, folks!).

Also, and almost uniquely in the non-financial press recently, Lanchester's piece doesn't say THE SKY WILL FALL. But it does, in a beautifully polite fashion, point out precisely how low the clouds are.
posted by bonaldi at 3:47 PM on January 2, 2008

And here's a cite for my claim.

bonaldi, I know that the article wasn't about GDP, but nonsense ought to be called out wherever it lies.
posted by Kwantsar at 3:54 PM on January 2, 2008

I was all ready to get my high horse about the article being nothing but a bitter polemic (god that intro is classic LRB) but as a description of derivative products it's ok.

However, as I ploughed on I realised there are just too many things left out or distorted -- I understand now how MeFi scientists feel in every thread about the latest physics breakthrough. Non-specialist journalist with world view to push + complicated subject = a shoddy article.

The most annoying part of the article is not derivatives related though, it's this disparaging remark: "Maths PhDs are all over the place in this business". Why on earth should finance be simple? In which other technical field is it preferable to have less qualified rather than better qualified people?
posted by patricio at 4:19 PM on January 2, 2008 [1 favorite]

But the trade in derivatives was hampered by one big thing: no one could work out how to price them. The interacting factors of time, risk, interest rates and price volatility were so complex that they defeated mathematicians until Fischer Black and Myron Scholes published a paper in 1973, one month after the Chicago Board Options Exchange had opened for business. Within months, traders were using equations and vocabulary straight out of Black-Scholes (as it is now universally known) and the worldwide derivatives business took off like a rocket.

This is wrong, wrong, wrong, wrong. I keep feeling like I am grinding an axe when threads about this come up, but the point bears repeating. Derivatives have a long, long history. Derivatives have been recorded since the time of Thales of Miletus , the philosopher who was asked if he could make money as well as he philosophized. The story goes: he bought oil-presses when they were cheap, during the winter, and liquidity was low. It was a good year for olives, and he made a lot of money on the derivatives. I believe there was another component of the story about him doing a second order derivative, but I cannot remember it and the article linked does not reference it.

But of course that is sort of cheating, as it is not the derivatives trade we knew it. Indeed in Amsterdam (as again the article linked to) there was a healthy, sophisticated derivatives trade going back to the 16th century on grain and commodity prices. This continued up until the 19th century with equities, even though there was only one company being traded on the exchange. Anyone who has read a translation of Bachelier's thesis knows that there was future and foward swaps going on in the back alleys at the turn of the century. These were just as sophisticated as the derivatives we know now.

Even closer in time, Nelson's book (ABC of Options and Arbitrage, 1903-4) which I just finished reading, talks about arbitrage between London and NYC and gives the amount of message of up to 300 a day over telegraph wires. That's really phenomenal given the time. Even more interesting, this is before Merton-Scholes mind you, what is known as the "delta" was priced at 50% for options that were at-the-money. This may seem like greek to someone who is not familiar with option pricing (a little pun), and there are some very beautiful equations that explain this, that 50% is the asymptote for at-the-money options, but sophisticated traders who no doubt used very little calculus and should have had no knowledge of Brownian motion were already pricing options like this. To argue that derivatives trading as being sophisticated and new is simply absurd. The academic theory behind it is indeed complicated, as is the theory on how to fly a plane. But the Wright Brothers did not need to know anything about fluid dynamics to intuitively come up with a functioning flying machine.

Now derivatives trading did not really take off until after WWII, and granted a lot of experience was lost, but it is certainly not a modern invention. Sure it collapses, as do other markets, from time to time. There are often very spectacular collapses, but it seems that a highly evolved financial system will always have derivatives trading, in fact I would even give derivatives trading as a sign of a robust economy. It shows sophistication in the market of the unknown.

I sort of skimmed through the article, and I printed out a copy to read later tonight. I think I got the jest of it, and it does have some good points beyond being scary about the derivatives market (the gap between the well-to-do and the have-nots in modern cities).

I'm in somewhat of a rush, so I'll throw one more point about the history of derivative trading: without being conspiratorial it has long been, since at least de la Vega, and probably going back to the markets of Renaissance Italy, strongly associated with Jews. It has often been relegated to back alleys as it was seen as speculation and gambling, and much like a lot of finance, was quickly scorned as something esoteric and Semitic. Of course we don't see that anymore, at least not in a periodical like the London Review of Books. I think the same otherness is still there, that it is something people make a lot of money in and is often not well understood. Derivatives traders sort of shoot themselves in the foot with this, because I will admit they almost try to make it hard for an outsider picking up a text (say Hall's pedagogical book).

I am not saying that derivatives traders don't act unethically, they do, and I am not saying it shouldn't be regulated. But poorly written articles like this, that seem to pose it as the bane of the financial system, really undermine a lot of stability and trust that derivatives contracts present. To me derivatives, looking at it holistically, is the markets way of preserving itself. It requires trust, a high degree of sophistication and understanding of how markets react.

Oh, and one more thing: there really is no thorough treatment of the history of derivatives trading. Even the great text books, but Rubinstein or Natenberg, don't really give a good overview. That's not to downplay the contributions of modern option theory, but there's no real good history of high finance. It is very, very hard to research (I have a lot of hard time even finding texts pre-WWII talking about it), and when I do stumble upon it, it is usually in a language that I don't understand.
posted by geoff. at 4:21 PM on January 2, 2008 [15 favorites]

hmmm that was a pretty weighty article, i was very glad i read it though - one of those things you don't really see in the news - like european human rights laws i dont think i ever saw a single newspaper article actually telling people what the rights laws meant.
posted by sgt.serenity at 4:46 PM on January 2, 2008

Just to clarify what other posters have been saying, the "amount" of a derivative is often far greater than it's value.

Let's take a type of derivative called an "interest rate swap," for example.

In our example, for five years, Alice agrees to yearly pay Bob 5% and Bob agrees to pay Alice 1 year LIBOR, capped at 8%. The amounts are calculate with respect to a $1,000,000 "notional principal amount."

The principal is "notional" because, unlike in the case of an ordinary loan, neither party actually paid $1,000,000 to the other at the beginning of the swap, and neither party will repay $1,000,000 at the end. It's simply the amount by which their period payment obligations are calculated.

Since Alice and Bob are both paying each other dollars, they might as well net their yearly payments, so each year, only one (or maybe neither of them) will make a payment.

Let's say that 1 year LIBOR is 5.2%, so Bob will have to pay Alice 0.2% of $1,000,000 in the first year, or $2000. In year 2, 1 year LIBOR falls to 4.7, so Alice owes Bob $3000.

As should be clear from the example, describing $1,000,000 as the "amount" of Alice and Bob's swap distorts the picture. If LIBOR drops to 0%, Alice owes Bob $50,000/year, and if LIBOR climbs above 8%, Bob owes Alice $30,000/year. The total payments made, and the amount at risk, will never approach $1,000,000.
posted by "Tex" Connor and the Wily Roundup Boys at 5:04 PM on January 2, 2008 [4 favorites]

I believe there was another component of the story about him doing a second order derivative, but I cannot remember it and the article linked does not reference it.

Possibly the second order is that he took out leases on the presses rather than on the olives themselves (as others were doing)? More clever, philosophical if you like, than simply buying options on the fruit.
posted by IndigoJones at 5:31 PM on January 2, 2008

"Tex" Connor and the Wily Roundup Boys writes "describing $1,000,000 as the 'amount' of Alice and Bob's swap distorts the picture"

True, as the concept of "notional amount" could be easily be misinterpreted as "nominal amount" which is the value that should be returned to the owner of an obligation that decides to cash back the obligation at t=expiry, instead of selling it before at t=x

If my memory serves, a lot of derivative trading happens in compensation room, which regulate the respecitve position of traders on a daily (?) basis, but I am not aware of the cap limits to broker exposition ,in the event he "runs out of liquid" so to say. In this kind of trading, of course, the notional amounts are never exchanged but there is a compensation like the one you described, of all position at a given time. Yet, there must be a mechanism that prevents any player from having uncompensated positions, the details of which I am not aware of.

One could also include a limitation to the contract, so that in the event of a put/call option on stock , it is made clear that the stocks are never to be exchanged, but only the difference between the quotation price at t=0 and at t=exp.

Yet what about unhedged positions ?

By then account 88888 was £23 million in the red. In order to bring in enough option premiums to fund his losses, Leeson took to selling completely unhedged positions. When the market went against him the losses hit £50 million.

So at the end somebody must, obviously, take the bucket if the position aren't properly compensated. Clearly, it's part of the risk of an investment bank , not to let its employee bet the house irrationally, but what if the risks are somehow hidden from scrutiny , for instance by misrepresenting the situation in the balance sheet ? I can't imagine national banks constantly refinancing losing banks , thus socializing the cost ?

Or can I ?
posted by elpapacito at 5:43 PM on January 2, 2008

So when a market becomes less liquid, buyers and sellers demand wider spreads on the assets they are trading. Now I've heard that CDOs backed by subprime mortgages have traded with wider spreads, but is this true of CDOs backed by other assets, too -- ABS, CDS, etc.?

Is there anyone out there who can direct me to publicly available bid/ask spreads on ABS and CDS?

And can someone tell me why widening spreads on CDOs backed by mortgages would cause CDOs backed with other assets to become more risky?

Enquiring minds and all.
posted by A-Train at 6:06 PM on January 2, 2008

i liked geoff.'s thoughtful comment, and the part i liked best was

i think i got the jest of it...

chronic amusement deficit sufferers like myself are always desperate for more jest.
posted by bruce at 6:36 PM on January 2, 2008 [1 favorite]

You'll generally not find publically quoted bid/asks on specific asset backed securites. There are a few indices though at markit
posted by JPD at 7:25 PM on January 2, 2008

Bah. It serves me right for getting snotty: used goods are excluded from GDP calculations. Anyway, the article isn't about that, so enjoy it.
posted by anotherpanacea at 7:40 PM on January 2, 2008

Derivatives are the best way to split a pie into more slices than will fit in one pie.
posted by HTuttle at 8:35 PM on January 2, 2008

[...]
posted by geoff. at 7:21 PM on January 2

My, that was interesting! Thanks, geoff.
posted by Civil_Disobedient at 9:27 PM on January 2, 2008

the opacity of these instruments makes me wonder , quite a lot, expecially of the competence of the people with advanced degree in maths that could probably change the form of an equation long an arm, but may overlook that risk profile because the risk information may not be expressed adequately in the equations.

Me too, but I think that even people with advanced degrees can be victimised by their monkey brains if there is a sufficiently compelling vision of bananas to come.

I get the impression that there are several forces at work here.

First, no matter what the maths dictates as prudence, people start to ignore it when their intuition tells them to. You're making great money, everything's gone well so far, the outlook is rosy - why not just make a little extra bet with a little extra money?

Second, to the extent that people do let statistical work guide their behaviour, they often place too much faith in its accuracy - they trust that the model is correct, and they trust that the data used in the model is correct. (What good is a pricing model for a derivative based on a bundle of mortgages correctly if your information about the soundness of lending practise is wrong, or if your assumptions about interest rate movements are wrong?)

My holiday reading has been N Taleb's Fooled by Randomness and The Black Swan, both of which books deal (rantingly at times) with the psychology of traders and the misuse of inappropriate statistical approaches to trading.
posted by i_am_joe's_spleen at 12:55 AM on January 3, 2008

i_am_joe's_spleen writes "psychology of traders and the misuse of inappropriate statistical approaches to trading."

Damn you put psychology and statistic in the same book about business and just made the procastination inducing trinity for me ! Damn !
posted by elpapacito at 2:40 AM on January 3, 2008

Steven, can it possibly have escaped your notice that GDP and GWP also include things which already exist? Why bother commenting if you don't know what you're talking about?

Why bother commenting if you can't even parse the other comments?
posted by yerfatma at 4:30 AM on January 3, 2008

JPD: yeah, I forgot about MarkIt.

This is the first time I've heard someone expressly call out CDS as a likely site of credit squeeze problems (as opposed to subprime mortgage-backed CDOs).

As the article also points out, "What cannot be gauged is the fallout from a slather of defaults or an increase in risk-aversion that dries up liquidity through the markets. In that event, derivatives could take a massive hit, perhaps diminishing the demand for and the price of the hard assets in the process." Corporate defaults are likely to increase from their historically low levels of the last two years. Also, given the massive hits that various investment banks and global banks have taken (and will continue to take) from the sub-prime mortgage mess, it is likely that there will be decreased liquidity in the coming months. Hence, there is a strong probability that the CDS market will be going through a period of stress in the months to come.

The key question will be just how much stress the CDS market will experience and how it affects something known as "counterparty risk". TCM suspects that CDS shakeout will be far worse than what the market consensus currently is. Thus, add the CDS market shakeout to further price declines in the CDO market, increasing defaults in the Alt-A and Prime mortgage sectors, the growing acceptance of "jingle mail", the on-going meltdown in the CMBS sector, the ticking time bomb known as Option ARMs and uncertainty about the leveraged loan market.

posted by A-Train at 2:42 PM on January 3, 2008

Bah. It serves me right for getting snotty: used goods are excluded from GDP calculations. Anyway, the article isn't about that, so enjoy it.

Cool! I have even less effect on the economy than I thought I had. Maybe I'll go paint my house now :P
posted by Chuckles at 11:20 PM on January 3, 2008

A-Train -- I don't follow the analysis in that article at all. Bank to bank CDS exposures are subject to regulation and all collateralised under their standard ISDA arrangements, so saying "sort of like having a $US40 trillion insurance market with no funding set aside for losses" is simply not true.
posted by patricio at 6:23 AM on January 4, 2008