Financial instruments behind the global meltdown cannot be understood

[bascule]

http://www.freedom-to-tinker.com/blog/appel/intractability-financial-derivatives

 

An interesting convergence of computer science and economics: researchers at Princeton studied Collateralized Debt Obligations (CDOs), securities which combine hundreds of mortgages into a single package, which were implicated as one of the primary causes of the global financial meltdown:

 

Trading in derivatives brought down Lehman Brothers, AIG, and many other buyers, based on mistaken assumptions about the independence of the underlying asset prices; they underestimated the danger that many mortgages would all default at the same time. But the new paper shows that in addition to that kind of danger, risks can arise because a seller can deliberately construct a derivative with a booby trap hiding in plain sight.

 

It's like encryption: it's easy to construct an encrypted message (your browser does this all the time), but it's hard to decrypt without knowing the key (we believe even the NSA doesn't have the computational power to do it). Similarly, the new result shows that the seller can construct the CDO with a booby trap, but even Goldman Sachs won't have enough computational power to analyze whether a trap is present.

 

In hindsight, it would seem the computer models being used to understand CDOs were woefully inadequate, and worse, it may be an intractable and exploitable problem.

 

I'm not going to go out on a limb and say the government should ban CDOs, but I think it's clear they need to be far more stringently regulated.

 

Sadly, it seems most of Congress thinks the bailout worked and there's no need to enact further legislation to prevent the problem from happening again.

 

Le sigh.

[Sisyphus]

So I guess "never buy anything you don't understand" is no longer conventional wisdom for investing.

[bascule]

So I guess "never buy anything you don't understand" is no longer conventional wisdom for investing.

 

The problem is at the time they had computer models which they thought were capable of predicting their behavior.

 

Turns out... not so much.

[JohnB]

The problem is at the time they had computer models which they thought were capable of predicting their behavior.

 

Turns out... not so much.

 

Go on. Admit it. You're giving me a straight line, aren't you?

[JillSwift]

I find the most interesting aspect of this is that for every predicting simulation out there there is a new aspect to the market - the results of that simulation, acted upon by the firms using it. The more of them there are, the more it changes the market. The more each tries to figure in the results of other simulations in one's own simulation the greater this effect becomes.

 

In short, the more you attempt to predict and the more inclusive of the data you are, the less likely you are to be right. It's like Heisenberg's uncertainty principle also happens to work for financial markets. Perhaps for the same reason; complex, overlapping probabilities.

[bascule]

Go on. Admit it. You're giving me a straight line, aren't you?

 

It just goes to show economics isn't a science.

[abskebabs]

It just goes to show economics isn't a science.

Coming from someone who unthinkingly accepts the conclusions of the central bankers who utillise the unrealistic assumptions he criticises, I don't think I could taste a richer portion of irony. Seriously I'm stuffed.

[bascule]

Coming from someone who unthinkingly accepts the conclusions of the central bankers who utillise the unrealistic assumptions he criticises, I don't think I could taste a richer portion of irony. Seriously I'm stuffed.

 

Glad to see you have nothing of substance to offer to the conversation but only resurfaced to hurl insults.

[abskebabs]

Sorry, I couldn't resist. If you'd actually responded rather than ignored and sweeped over the points I'd made in your previous thread, then I'd probably invest more time.

[bascule]

Sorry, I couldn't resist. If you'd actually responded rather than ignored and sweeped over the points I'd made in your previous thread, then I'd probably invest more time.

 

Well perhaps you can tell us your thoughts on CDOs.

[abskebabs]

Well perhaps you can tell us your thoughts on CDOs.

You can create all the complicated rules and regulations you'd like to hamper the creation of contracts on a marketplace without ever realising or tackling the root issue.

 

In 2001, a tech bust was experienced and reinflated. In 2008 a housing bust was experienced. You cannot predict which specific sector will undergo an artificial boom, but there are a few strong indications of if and where it may occur, given:

 

1)Interest rates were set consistently and extremely low during this period by Fed chairmen Alan Greenspan, producing credit expansion at a rate that has been unparalleled in modern history.

 

2)The credit expansion and inflation distorted economic calculation and lending rates were artificially lowered, allowing for an unsustainable expansion of the economy's capital structure; as submarginal and wasteful investment projects were engaged in and losses not felt due to the artificial price adjustments occuring as a result of interest rate manipulation.

 

3)The effect of moral hazard, given the effect of the "Greenspan put", and more recently the comprehensive bailouts. Investment banks have been and were given consistently the message that the government and central bank would bail them out no matter what happened. Naturally, if I am allowed to keep my profits without having to account for my losses this will encourage me to undertake more risky behaviour. Ironically you yourself have endorsed such practice.

 

4)Political pressure to coerce many mortgage companies to extend loans to people they were well aware had no abillity to pay back. Fannie and Freddie just represented the monolothic government agencies also charged to help bring this into reality.

 

Given the above, and the unparalleled distortionary effects of the credit expansion and inflation (increase of money supply) produced as well as moral hazard encouraged in the past few years, it is unsuprising that one effect, although quite extreme was the creation and adoption of collateralised debt obligations given the consistent artificial increases continually occuring in the housing sector.

 

Without dealing with the central bank, you are not touching the elephant in the room.

[JohnB]

It just goes to show economics isn't a science.

How very true.

 

Jill, I think that is one of the more interesting and profound thoughts on economic modelling that I've ever read.

 

A point would be that your model couldn't predict what their models results will be. This would mean that while their actions would follow a logical course, because of the unknown output of their model the course would be essentially semi random in nature.

 

In short, the more you attempt to predict and the more inclusive of the data you are, the less likely you are to be right.

I do believe you are right.

[bascule]

In 2008 a housing bust was experienced. You cannot predict which specific sector will undergo an artificial boom, but there are a few strong indications of if and where it may occur, given:

 

[...]

 

Given the above, and the unparalleled distortionary effects of the credit expansion and inflation (increase of money supply) produced as well as moral hazard encouraged in the past few years, it is unsuprising that one effect, although quite extreme was the creation and adoption of collateralised debt obligations given the consistent artificial increases continually occuring in the housing sector.

 

If I'm reading this correctly, you see the creation of CDOs as an effect of the financial crisis and not a cause.

[JillSwift]

Jill, I think that is one of the more interesting and profound thoughts on economic modelling that I've ever read.

*blush*

 

A point would be that your model couldn't predict what their models results will be. This would mean that while their actions would follow a logical course, because of the unknown output of their model the course would be essentially semi random in nature.

And that's the ultimate conundrum right there. The best you can hope to do is sample the actions of other firms and guage it against the market, and include that in your own models. And that means introducing further overlapping probabilities (plus the problem of unknowns like when they change their models). The more data you gather, the more chaotic it is. Unless you can get real-time data on businesses' intentions and internal conditions, of course. But the SEC rather frowns on that.

 

They must know this, so I wonder why they keep building predictive simulations.

 

 

I do believe you are right.

I wish I could prove it. With proof perhaps we could take a little starch out of the market.

[Sisyphus]

 

They must know this, so I wonder why they keep building predictive simulations.

 

I'm not convinced they do know that, actually. But in any case, they did make some people very rich, at least for a while. There is always going to be someone willing to use ever more complex formulas for better short term profits at the cost of long term chaos in one's own investments and the economy as a whole. And there is always going to be a market to support such people. Humans in general are not rational or future-oriented enough for it to be otherwise, and the humans on wall street self-select for personalities that exaggerate the problem. Not every investor is that way (Warren Buffet is an example of a successful antithesis), but not everyone has to be to ruin it for everyone.

[abskebabs]

If I'm reading this correctly, you see the creation of CDOs as an effect of the financial crisis and not a cause.

 

Not necessarily, more that their widespread use with regard to housing debt was exacerbated and made a major problem due to the effects of major inflationary credit expansion on the part of the central bank, as well as its encouragement of moral hazard and risky excesses to be taken by banks in light of its bail out policy.

 

In an unhampered market where market actors would have to account for their losses, and interest rates could not be artificially lowered to allow the funding of exhuberantly wasteful projects, they would not be an issue in the same way they have been with the present crisis.

[bascule]

Not necessarily

 

I just note they weren't listed among your "indications of an artificial boom"

 

more that their widespread use with regard to housing debt was exacerbated and made a major problem due to the effects of major inflationary credit expansion on the part of the central bank

 

I'm curious if you see the converse happening... the housing bubble was exacerbated by the creation of CDOs and the ability to sell mortgages as (increasingly risky) securities the more times they went through the "meat grinder"

 

But it seems like your only bone to pick is with the Fed...

[abskebabs]

I'm curious if you see the converse happening... the housing bubble was exacerbated by the creation of CDOs and the ability to sell mortgages as (increasingly risky) securities the more times they went through the "meat grinder"

 

I don't rule it out as being entirely impossible, in the same way I don't doubt that that it may be possible for the valuations of all individual actors on this planet to change such that tommorow they decide that they should commit suicide.

 

I deem it highly unlikely however, since without central bank interest rate manipulation and moral hazard creation, since individuals would be highly disincentivised from undertaking such activities, and in fact would be selectively removed from the market via the price and profit-loss mechanism if this was unmanipulated and not perturbed from its proper operation politically, as has happened during this crisis.

[bascule]

I suppose more to the point: would the bursting of the housing bubble have lead to the worldwide financial crisis without CDOs?

[abskebabs]

I suppose more to the point: would the bursting of the housing bubble have lead to the worldwide financial crisis without CDOs?

 

The housing bubble was an artificial boom generated by credit expansion, so it is inevitable that it would be followed by a bust and a process involving general price correction, and capital and labour reallocation.

 

Whether the latent features accompanying the crisis would have been the same with or without CDOs is an interesting question perhaps. I think the fundamental features of the crisis would have been the same, though the way, and the places in which capital were misallocated may have changed in nature slightly.

[bascule]

The housing bubble was an artificial boom generated by credit expansion, so it is inevitable that it would be followed by a bust and a process involving general price correction, and capital and labour reallocation.

 

What about a complete collapse of the financial system? Does that seem somewhat historically aberrant to you?

 

Do you think maybe, just maybe, CDOs were involved in that?

[abskebabs]

What about a complete collapse of the financial system? Does that seem somewhat historically aberrant to you?

 

Do you think maybe, just maybe, CDOs were involved in that?

 

I don't think that such a collapse was warranted given the fact the major problems were with large, highly leveraged banks and investment banks that did make bad decisions dabbling in mortgage securities and credit default swaps, while albeit being misled by false interest rates and moral hazard.

 

But given the corrections that were already taking place, by no means was commercial lending overall(many banks, especially smaller banks were not making these mistakes, and in ordinary circumstances would have taken over, gaining market share) even dropping overall until the TARP was passed as shown in the chart I posted on your other thread.

 

Also you're charge is a little unwarranted in the sense that you believe the propaganda Paulson spewed out in order to pass his bill. Nobody would back a package to bail out the fat cats on Wall Street, so instead it was sold that small businesses would lack short term financing.

 

Have you ever thought to yourself what kinds of business would pay their employees would pay their employees with borrowed money, and do you really think this represents the cross section of small businesses you cry about the collapse of if no bailout is given?

[iNow]

Those banks were making those risky choices well before they were guaranteed bailouts from the government, so your assertion that moral hazard was a factor is an example of the post hoc propter hoc fallacy.... again.

[abskebabs]

Those banks were making those risky choices well before they were guaranteed bailouts from the government, so your assertion that moral hazard was a factor is an example of the post hoc propter hoc fallacy.... again.

 

Ever hear of the Greenspan put, or do you selectively read what I write so you can make false accusations?

[bascule]

I don't think that such a collapse was warranted given the fact the major problems were with large, highly leveraged banks and investment banks that did make bad decisions dabbling in mortgage securities and credit default swaps, while albeit being misled by false interest rates and moral hazard.

 

You don't think such a collapse was warranted? As if there were some choice in the matter?

 

But given the corrections that were already taking place, by no means was commercial lending overall(many banks, especially smaller banks were not making these mistakes, and in ordinary circumstances would have taken over, gaining market share) even dropping overall until the TARP was passed as shown in the chart I posted on your other thread.

 

You mean in the weeks between the fall of Lehman Brothers and the passage of TARP?

 

Do you think maybe, just maybe, lending went down after the financial crisis, not TARP.

 

Honestly, getting a direct response out of you is like pulling teeth.

 

Also you're charge is a little unwarranted in the sense that you believe the propaganda Paulson spewed out in order to pass his bill.

 

You're opinions are thoroughly riddled with Austrian confirmation bias, but you don't see me calling you out on it (whoops)

 

Nobody would back a package to bail out the fat cats on Wall Street

 

I would, if the alternative were the collapse of the financial sector/economy. That was the case here, unless you're saying otherwise.