Market crash nobodies fault

[my_wan]

Most of us has heard about how shocked Greenspan was at the market crash. This sent me on a paper trail of the theory behind mortgage derivatives. It starts with an equation invented by economics Nobelists Robert Merton and Myron Scholes. I'll try to explain what's behind this equation.

The problem is how to evaluate risk of a stock option, a classic problem in economics. What this equation does is to avoid the problem of risk premium value of an option by a very simple mechanism. Suppose you own stock S. You can essentially remove your risk of owning this stock by writing two options for this stock. In essence betting that the stock will both rise and fall. By doing this the market value should be such that the return is zero, or more specifically equal to that periods risk free treasury bill return. It didn't remove the risk it just recognized the risk was already accounted for in the pricing of the stock. It is actually quiet a beautiful method and worked quiet well in the risk assessment of stock. So why did it break and make Greenspan look like a bumbling fool?

The market qualifies for what is called a self referential system. Self referential system are at the crux of many modern day paradoxes. It's even behind an ingenious interpretation of Quantum Mechanics. Consider an absurd example of someone finding a full proof method determining what a stock will do 24 hours in advance. You get ultra rich and world famous for your method. However, now everybody knows how this method works and the no risk means that investors will change market dynamics such that there is no return on the method because it is risk free. Market feedback does not allow risk free money. So what changed when these derivatives based on this equation went on the market?

Notice that the central assumption of the equation was that the stock valuations set the zero-sum point of the game. Perfectly true and valid assumption at this point. However, these same stock are now packaged into derivatives and these derivatives now determine stock valuations, not the stock themselves on which the assumption of the equation depended. Without the underlying stock setting the derivative valuations the equation begins using the value of itself to try to determine the value of itself. It becomes totally insensitive to any actual value of the packaged stock. The risk free zero-sum approach to determining risk valuation then creates a humongous over-capitalization of the mortgage market exacerbating the situation. It couldn't do anything but stretch until it broke no matter how much care was given.

Of course you can still blame Nobelists Robert Merton and Myron Scholes. They have been un-repentantly sloppy in so many economic issues, especially with regard to feedback mechanisms. They attempted to use the market feedback mechanism against itself and lost. The sad thing is now governments are questioning the legitimacy of the free market itself.

 

[Francesca R]

Sorry, not at all sure what you're trying to say

 

[my_wan]

Sorry, not at all sure what you're trying to say

Hmm... I'll try to articulate. Fundamentally the Black–Scholes formula was the formula used to value the derivatives behind the mortgage crisis. It was the inability of this formula to do what it claimed that caused the crash. You can look at the many parts that seemed responsible, like subprime lending, regulation, etc., but this is the formula that was fundamentally being depended on to manage these risk, with or without regulators. Essentially my claim is that it did not manage the risk it just made the risk invisible. The market therefore took on more and more risk with the illusion that it was well managed. In fact there was a Nobel Prize supporting the notion that these risk were well managed.

All the arguments about corrupt banks, etc., doesn't hold all that well in the big picture. So what I'm looking at here is what went wrong with that formula that made it ok to take on such huge risk.

The Black–Scholes formula is based on the idea that you can remove risk of owning a stock by betting that the price will both rise and fall at the same time, selling two options one "short selling" and one "going long" at the same time. In fact this does remove or significantly reduce your risk of owning that stock but at a zero or greatly reduced profit or loss. If a profit is anything other than zero (actually equal to a treasury bill return) then that is considered market inefficiency and should be priced out of existence because it is risk free profit. In essence a derivative we hear so much about is a group of stock which has been grouped and priced in such a manner that you are betting that the individual stock that it's made from will both rise and fall. A derivative is then considered a risk free or greatly reduced risk investment. That's why you seen bankers in senate hearings in 2005 claiming there is no risk.

In fact this hedging of bets scheme works on a small scale when the market trades the underlying stocks, i.e., components of the derivatives. In fact it will continue to work when derivatives are traded as long as the derivatives market remains a small part of the market for the underlying stocks. This is because the derivatives can adjust their value on the day to day variations of the traded value of the underlying stock. Once the derivatives market exceeds the market for the individual stock then the derivatives market is setting the market price of individual stock rather than the other way around as the formula intended. The derivatives are then setting the market value of themselves. Even worse is the illusion of being risk free with all changes in actual risk being totally hidden by the lack of a market for the underlying stock. The last term in the equation assumes a stochastic process so without adjustments to actual risk from underlying stock the equation becomes a purely random walk into oblivion.

If you think you have some idea but have questions it would make it a lot easier to respond. If it is still incomprehensible at least try to mirror what you don't get so I can better guess what points to articulate. Perhaps I can model this effect using a thermodynamic model and write a computer program to simulate it. Of course you can just say I don't get it if it fits.

 

[Just thinking]

The walk into oblivion ... that I understand.

 

[Scottch]

The promulgation of your descriptive thought is enticing. However, I further depict your argument as a broke illustration of a bursting consideration by using immense vocabulary to impel your suggestion. My inclination is that you would bring into play supplementary customary expressions to explicate what you are chatting about. I am anxious that, for the reason that you deem you are intelligent by using physically powerful words, you are in point of fact marginalizing yourself from the wide-ranging communal and being elitist and pompous. I anticipate your comments (not you) should go to meet your maker in a conflagration of internal combustion parts of organic matter.

Scottch

 

[mhaze]

Hmm... I'll try to articulate. Fundamentally the Black–Scholes formula was the formula used to value the derivatives behind the mortgage crisis. It was the inability of this formula to do what it claimed that caused the crash.....

Black Sholes was behind the 1987 crash.

The very theory underlying all insurance against financial panic falls apart in the face of an actual panic.

Nobody learned.

 

[dudalb]

The OP needs to get a copy of Strunk and White's "The Elements Of Style", fast.

 

[my_wan]

The promulgation of your descriptive thought is enticing. However, I further depict your argument as a broke illustration of a bursting consideration by using immense vocabulary to impel your suggestion. My inclination is that you would bring into play supplementary customary expressions to explicate what you are chatting about. I am anxious that, for the reason that you deem you are intelligent by using physically powerful words, you are in point of fact marginalizing yourself from the wide-ranging communal and being elitist and pompous. I anticipate you go to meet your maker in a conflagration of internal combustion parts of organic matter.

Scottch

Ok ok I get it.. LOL

Let's try this analogy. I flip a coin and bet Jack $5 bucks it's going the be heads and bet Bob $5 bucks it's going to be tails. Zero-sum game since I'll come out losing as much as I win. If I don't it's because I made different bet with Jack $6 and Bob $4 thinking the odds was better for heads but wasn't willing to "risk" the whole $5 bucks on the bet. Stock prices don't really have 50/50 odds of moving up or down. Market coins are biased to one side and change the bias some from day to day. If the stock market values heads the sames as tails then the market is giving odds of 50/50. If it values heads at 60% and tails at 40% then the market odds are 60/40 in favor of heads. If market odds are 65% and 40% (65/40) then it's possible to make instant risk free money which the market will adjust out of the price real quick from people taking advantage of it.

So let's make a derivative out of our mortgage coin tosses. Our odds on our coin can change from day to day. Going market prices say it's got a 60/40 chance of going up. We have our bookie (Wall Street, investment bankers, etc.) package together 100 coin tosses to a group (a derivative). Now we can control exactly how much risk we want to take because we are averaging over so many coin tosses. The next day the individual coin toss market adjust the odds 59/41. We adjust our derivative prices accordingly. So far so good. Now we see it works so well that we package all coin tosses into derivatives. Problem is that now there is no way to track the market value of the individual coin toss because the individual coin toss market doesn't even exist anymore. Only nobody notices because everybody is assuming that the derivative value/100 is the market value of the individual tosses. Even if a few coins started coming out two headed the derivative wouldn't notice because it's so tiny in comparison and the derivative is pricing itself without any help from the non-existent individual coin toss market. Nobody is even looking at individual coin tosses anymore to see what the real value is. The demand for these nearly risk free derivatives then drives the prices for derivatives that (inappropriately) determine the supposed value of individual coin tosses. The formula was designed to price derivatives from coin tosses, not price coin tosses from derivatives. When the value of the individual coin toss starts disagreeing with the value of the derivative bad enough, like when the forclosure rate is zooming up without any indicators from the derivatives, everybody wants out, it crashes, Greenspan goes: 'Duh.. I don't get it', and the rest is history.

If the derivatives had actually been able to track the real risk then mortgage money provided by investor would have been far more limited so that capital wouldn't have been available for such massive subprime loans. It was only the fact that the Black–Scholes formula promised nearly perfect risk management, when in reality was blind as a bat to any risk, that the money kept pouring into subprime loans.

 

[my_wan]

Black Sholes was behind the 1987 crash.

The very theory underlying all insurance against financial panic falls apart in the face of an actual panic.

Nobody learned.

This is interesting. I haven't had time for a careful read yet but I definately intend to track the paper by Taleb mentioned here.

 

[Gazpacho]

It was only the fact that the Black–Scholes formula promised nearly perfect risk management

The Black-Scholes formula never promised this. It only provides a relationship between statistics. You remind me of Proudhon going on about logarithms and how they're a tool of ruination

The problem is not the Black-Scholes formula. It is the persistent belief in a perfect investment, the ignorance of the difference between what is expected and what will happen. Basically the belief that any model is reality, rather than a tool for understanding reality that, like any tool, has to be inspected from time to time.

 

[mhaze]

Ok ok I get it.. LOL

Let's try this analogy. ....

For those who do not have the light of understanding on this, here is an analogous sham.

Assume prices never change. A convenience store builds an extra room, and fills shelves with envelopes, more than the number of items the store has. Each envelope contains a promise to provide several of the store's items at a future date. Trading at the store, cash flow and profit skyrocket, because people buying like the envelopes.

The obvious problem is the store can sell more than it owns.

Now presume not only that prices change locally due to demand, but that the envelopes contain a mix of promises to provide product if certain conditions on product current value are met. Demand and price are coupled, and envelopes come to drive demand.

The death spiral.

Oops, they say. We never thought of that - give us a bailout.

 

[my_wan]

Black Sholes was behind the 1987 crash.

The very theory underlying all insurance against financial panic falls apart in the face of an actual panic.

Nobody learned.

This is quiet a bit different from the claim I made here but this guy makes a powerful case. I still haven't got his main mathematical paper but I read a lot of what his critics had to say. I'll find his paper soon.

Taleb's argument is a far more general than mine. It deals with volatility, rare events (Black Swans), and the non-linear sensitivity of the entire system to those events. He then considers that is is not just risk that must be considered but risk times consequences. I think his argument could be cleaned up but I'm having trouble cleaning my argument up. His powerpoint presentation seemed to endlessly drum out the obvious basics. The analogy that occurred to me wrt Taleb's argument is to compare Black–Scholes formula to the linear wave model. Based on this linear wave model rogue waves commonly report by mariners were thought to be myth until the Draupner wave in 1995.

My claim is limited to the notion that when derivatives supplants the underlying stock as the commodity that set the price of those underlying commodities it breaks the individual stock pricing mechanism on which the formula depends. It is in principle possible to fix this by tracking enough information about the individual underlying stock and making it transparent to the investor. However, that breaks the utility of the derivative to a large extent. It would make some derivatives more valuable than others even though they contain the same type of underlying stock, especially with respect to mortgages. In the way it's done now the underlying stock is considered a representative random sampling, whether true or not.

 

[my_wan]

The Black-Scholes formula never promised this. It only provides a relationship between statistics. You remind me of Proudhon going on about logarithms and how they're a tool of ruination

The problem is not the Black-Scholes formula. It is the persistent belief in a perfect investment, the ignorance of the difference between what is expected and what will happen. Basically the belief that any model is reality, rather than a tool for understanding reality that, like any tool, has to be inspected from time to time.

Don't know about Proudhon. I don't read on anarchist though I presume it had something to do with his dislike of compound interest or usury as a political position. That alone makes it a strawman in my view. I'm talking about a market mechanism and "the difference between what is expected and what will happen" and I'm inspecting this tool. Within the confines of my claim I'm not even saying the tool is not valid. I'm saying the validity of the tool breaks down in the limit. The limit being the when the stock is exclusively marketed and/or priced by the derivatives. The breakdown starts slowly enough and gains speed as the limit is approached.

The Black-Scholes formula does in fact promise to finely tune the control of risk. "Promised nearly perfect risk management" does not mean risk free or "perfect investment". It means that people can set the level of risk they are willing to take for the potential profit. Greater risk higher profits and losses. Derivatives were theoretically supposed the represent the lowest possible risk and return levels. It's basically meant to be a different version of diversification. I think the notion that people have a "persistent belief in a perfect investment" is a mythology you created in your theory about people. I'm sure you can find plenty anecdotal cases Vegas and even some on Wall Street but they certainly don't drive the market. I stand by my claim that there are endemic problems with the Black-Scholes formula. The value problem of derivatives is a classic economics problem.

 

[mhaze]

This is quiet a bit different from the claim I made here but this guy makes a powerful case. I still haven't got his main mathematical paper but I read a lot of what his critics had to say. I'll find his paper soon.

Taleb's argument is a far more general than mine. It deals with volatility, rare events (Black Swans), and the non-linear sensitivity of the entire system to those events. He then considers that is is not just risk that must be considered but risk times consequences. I think his argument could be cleaned up but I'm having trouble cleaning my argument up. His powerpoint presentation seemed to endlessly drum out the obvious basics. The analogy that occurred to me wrt Taleb's argument is to compare Black–Scholes formula to the linear wave model. Based on this linear wave model rogue waves commonly report by mariners were thought to be myth until the Draupner wave in 1995.

My claim is limited to the notion that when derivatives supplants the underlying stock as the commodity that set the price of those underlying commodities it breaks the individual stock pricing mechanism on which the formula depends. It is in principle possible to fix this by tracking enough information about the individual underlying stock and making it transparent to the investor. However, that breaks the utility of the derivative to a large extent. It would make some derivatives more valuable than others even though they contain the same type of underlying stock, especially with respect to mortgages. In the way it's done now the underlying stock is considered a representative random sampling, whether true or not.

Interesting, I was also considering wave action similarities. By the way, my little convenience store case is of course a generalized example.

Possibly another good example would be kids playing with overpowered 4 wheel drive trucks in mudholes. Will they never get stuck? That's what the truck salesman told them. Will that little winch pull them out?Anything to attach it to ?

As long as the freak wave action doesn't approach those of Lituya Bay we are doing - fine...

 

[my_wan]

For those who do not have the light of understanding on this, here is an analogous sham.

Assume prices never change. A convenience store builds an extra room, and fills shelves with envelopes, more than the number of items the store has. Each envelope contains a promise to provide several of the store's items at a future date. Trading at the store, cash flow and profit skyrocket, because people buying like the envelopes.

The obvious problem is the store can sell more than it owns.

Now presume not only that prices change locally due to demand, but that the envelopes contain a mix of promises to provide product if certain conditions on product current value are met. Demand and price are coupled, and envelopes come to drive demand.

The death spiral.

Oops, they say. We never thought of that - give us a bailout.

It sounds to me like what you are talking about is "naked shorting". This did in fact occur a lot but as of September 2008 it is prohibited. It does in fact exacerbate the situation, and depresses stock prices in general, but by no means caused the crash. Rather than driving demand it reduces demand and prices because it makes it appear that more stock is on the market than what actually exist.
http://www.forbes.com/2007/02/02/naked-short-suit-overstock-biz-cx_lm_0202naked.html
Otherwise I'm not sure what particular market mechanism your anology refers to.

 

[my_wan]

I was reading comments on Taleb's work, http://www.fooledbyrandomness.com/ayache2.pdf, and this quote exactly mirrors the main issue I have taken with derivatives.

You need a certain context to dynamically replicate and price a certain derivative (for instance, you need a stochastic volatility model, consistent with the vanilla smile, to dynamically replicate and price a barrier option), and the pricing/trading of that derivative soon forces you to overstep the context (as soon as it trades, the market forces drive the barrier option away from the value of its replicating strategy).<snip (bolding mine)>

In fact my claim is that once the derivative defines the market value this way the "stochastic volatility model", dominates the price variation rather than market risk/value adjustments. This means that the apparent reduction in volatility is an illusion. Any adjustments from actually market risk/conditions can come but months or even years after the fact. Even then only when the assumed risk and real market risk diverge to the degree it can no longer be ignored, like when people start defaulting in mass as in the mortgage case. That's why so many seen the markets warnings even before 2005 but market variation was hidden in the shear size of the derivatives with variation determined by the idealized stochastic model.

 

[mhaze]

Do we have approximate numbers on dollar value and dollars leveraged on the questionable derivatives, relative to the market values of the stocks underlying them?

Isn't it fair to say that while Black Sholes implies one way cause and effect A->B, in reality we have had cause and effect running A<->B.

 

[Darth Rotor]

My Wan, my take from your OP is that this self-referential system (Godel wept, the crash is not his fault) seems to have taken on the characteristic of economic woo among a significant percentage of those in the financial, and financial services, industry.

Am I close?

DR

 

[DanishDynamite]

Most of us has heard about how shocked Greenspan was at the market crash. This sent me on a paper trail of the theory behind mortgage derivatives. It starts with an equation invented by economics Nobelists Robert Merton and Myron Scholes. I'll try to explain what's behind this equation.

The problem is how to evaluate risk of a stock option, a classic problem in economics. What this equation does is to avoid the problem of risk premium value of an option by a very simple mechanism. Suppose you own stock S. You can essentially remove your risk of owning this stock by writing two options for this stock. In essence betting that the stock will both rise and fall. By doing this the market value should be such that the return is zero, or more specifically equal to that periods risk free treasury bill return. It didn't remove the risk it just recognized the risk was already accounted for in the pricing of the stock. It is actually quiet a beautiful method and worked quiet well in the risk assessment of stock. So why did it break and make Greenspan look like a bumbling fool?

The market qualifies for what is called a self referential system. Self referential system are at the crux of many modern day paradoxes. It's even behind an ingenious interpretation of Quantum Mechanics. Consider an absurd example of someone finding a full proof method determining what a stock will do 24 hours in advance. You get ultra rich and world famous for your method. However, now everybody knows how this method works and the no risk means that investors will change market dynamics such that there is no return on the method because it is risk free. Market feedback does not allow risk free money. So what changed when these derivatives based on this equation went on the market?

Notice that the central assumption of the equation was that the stock valuations set the zero-sum point of the game. Perfectly true and valid assumption at this point. However, these same stock are now packaged into derivatives and these derivatives now determine stock valuations, not the stock themselves on which the assumption of the equation depended. Without the underlying stock setting the derivative valuations the equation begins using the value of itself to try to determine the value of itself. It becomes totally insensitive to any actual value of the packaged stock. The risk free zero-sum approach to determining risk valuation then creates a humongous over-capitalization of the mortgage market exacerbating the situation. It couldn't do anything but stretch until it broke no matter how much care was given.

Of course you can still blame Nobelists Robert Merton and Myron Scholes. They have been un-repentantly sloppy in so many economic issues, especially with regard to feedback mechanisms. They attempted to use the market feedback mechanism against itself and lost. The sad thing is now governments are questioning the legitimacy of the free market itself.

I blame stupidity and greed. Stupidity on the part of the legislators and their lack of instigating sufficient overseer institutions w ith real powers. And greed in the sense of the unscrupilous who repackaged the sub-prime loans into shiny new bottles.

 

[my_wan]

Do we have approximate numbers on dollar value and dollars leveraged on the questionable derivatives, relative to the market values of the stocks underlying them?

Isn't it fair to say that while Black Sholes implies one way cause and effect A->B, in reality we have had cause and effect running A<->B.

Here's the thing. Without the derivatives then those few initial bad mortgages would have been inspected for sufficient constraints on risk and those conditions thought to cause it weeded out or sold at much higher risk. Being packed into derivatives they were a minor blimp on the total value of the derivative and where therefore allowed to propagate.

It's not cause and effect being argued, it's a case of information flow. If the information about A says that B=A*N and then we replace all A with B we no longer have the information about all A's to set the value of B. The value of A becomes B/N where we must consider all A's equal regardless of the drift in risk associated with individual members of the set of
A. Our knowledge of where the risk is changing is hidden somewhere in the averages of B. It's the same thing when we must give up on knowledge about the position/momentum of any given air molecule to talk about temperature.

My Wan, my take from your OP is that this self-referential system (Godel wept, the crash is not his fault) seems to have taken on the characteristic of economic woo among a significant percentage of those in the financial, and financial services, industry.

Am I close?

DR

Yes, that is a reasonable assertion from my claim. They do have a Nobel Prize to reject the woo claim but authority does not make right. As I continue to study this it has become apparent that it was well known that the price valuation scheme was suspect as soon as any derivative actually went on the market. It continued for no other reason than the fact that it was profitable at the time.

I blame stupidity and greed. Stupidity on the part of the legislators and their lack of instigating sufficient overseer institutions w ith real powers. And greed in the sense of the unscrupilous who repackaged the sub-prime loans into shiny new bottles.

Stupidity is subjective. We depend on greed for a free market to work efficiently. We also depend on this greed operating within the confines of an agreed set of rules and anti-trust laws to avoid a virtual dictatorship, like Ma Bell used to be. You can call those that repackaged mortgages into "shiny new bottles" (actually a GSE or two) unscrupulous but there was a Nobel Prize lending legitimacy to the practice. Traditionally market forces defined what was and wasn't a sub-prime mortgage on a per case basis. These same market forces that allowed the relaxation of the rules under the derivative scheme. Except this new scheme was blind to individual cases, only the averages counted. As long the the market demanded these mortgages as-is they would continue. Regulation may have required them to maintain greater cash reserves but wouldn't tell them they couldn't loan money to people. It was therefore inevitable regardless of regulation.

Even with the derivatives if the drift in risk happened slowly enough corrections could have been made. However, demand was so great for mortgages that to land them requires greater incentives. The teaser rates, interest only, etc., for X time was like setting a time bomb that prevented the macro-economic picture from showing the true drift in risk for X amount of time. These bombs will continue going off for this same X amount of time. However, the market demand is what determines what is and isn't acceptable risk. The ability to flip property to a new owner through easy credit is what kept these time bombs in check for the moment. It was Greenspan in 2004 that encouraged these alternative loan products.

Final note with regard to regulation. These mortgage derivative markets were created by Fannie Mae and Freddie Mac. Neither of which is a private business. They are GSEs (government sponsored enterprise). They were created by the government with lots of special privileges that no private financial institution had. They had a fed rate credit line and regulation of private institutions did not apply to them. They were essentially a government created monopoly. None of the regulations on the books for private financial institutions even applied to them. Private financial institutions and individuals could however invest in and insure them thus leaving them and the whole system exposed to the risk. On the other hand the individual mortgages that they underwrote were generally less risky than the average mortgage.

In theory derivatives are supposed to be a zero-sum game. If somebody loses money it means somebody made that same money and visa versa. Only problem is that now there is nobody to pay the winner because the real estate can no longer be flipped to a new owner without heavy losses. It was this ability to flip the property that kept the higher than normal default rate manageable.
http://www.portfolio.com/views/blog...hen-mortgage-derivatives-get-included-in-cdos

Rough timeline: http://caps.fool.com/Blogs/ViewPost.aspx?bpid=98030&t=01004109049910303346