Did Physicists and Mathematicians Cause the Financial Crisis?

[Pangloss]

Interesting story on 60 Minutes tonight about the financial crisis. One thing they said that caught my ear was that some of the weird securities that involved the breaking up of mortgage-backed securities into tiny pieces and then remarketing them in creative ways were designed by "Nobel-track" physicists and mathematicians working on the side for the Wall Street firms. The complex algorithms designed by these people were supposed to reduce risk, and the managers took their word for it and leveraged to the hilt.

 

Much of the crisis is blamed on mortgage failures, but in fact (again according to this story) 94% of all US mortgages are currently being paid on time. So you can kinda see where this is going -- scientists getting left holding the ball. However the story does go on to say that most of the blame resides with "insurance"-like entities called "swaps".

 

And the problem with that anti-scientist reasoning, of course, is that no security should be banked to the extent that these were -- with something like $60 trillion in securities being traded -- with whole companies riding on them -- unless that security is fully understood by its own managers. Even worse, these products were given high security ratings by Moody and S&P!

 

60 Minutes has the 12-minute story online here:

http://www.cbsnews.com/video/watch/?id=4502673n

 

What do you all think? I think they're making scientists the fall guys for their own screw-ups.

[D H]

From the early 90's on, I have known quite a few engineers, mathematicians, and physicists who have abandoned the real field of "rocket science" to go into IP law or to become "rocket scientists" -- that is, a bright boy in finance. A six figure salary (and a high six figure salary too boot) was a rather compelling reason, plus it was all legal.

 

They certainly helped contribute to the financial crisis. Was it their fault? No. Their management chose to ignore that the mathematics was over their heads when the quants were making their employers billions (trillions?) of dollars in profits. Management could have asked about the underlying modeling assumptions, could have had a different group of "Nobel-track" physicists assess the models against risk, and could have had yet another group of "Nobel-track" physicists independently verify and validate the models. Did they do any of these things? Not that I can see.

 

For those of you who don't like videos (e.g., me),

 

The transcript of the 60 Minutes report

 

A 1999 Physics World feature article on the exodus of physicists to finance

 

A 1999 Physics World editorial on the same

 

A Sept 16 Physics World feature article on the collapse of Lehman Bros.

 

Did Physics PhDs cause current financial crisis, a Sept 16 blog by a physicist

[CaptainPanic]

Engineers, mathematicians, and physicists who work for financial institutions are called economists.

[Sisyphus]

Whew. Finally somebody figured out how to blame some nerds.

[john5746]

I bet they believe in evolution.

[Phi for All]

Nobel-track?! Holy spin, Batman! What a piercingly nebulous explanatory sound byte!

 

How far down the line of possible science educations does this Nobel track go?

[D H]

Engineers, mathematicians, and physicists who work for financial institutions are called economists.

No. They are called economists only if they have a degree in economy as well as a degree in engineering, math, or science. By the time someone has gone through grad school and received a non-economic PhD they are not too likely to go back to school to study economics, particularly when financial institutions will hire top-notch PhDs with little or no economic training. The generic term for this specialty is "quantitative analyst", or "quant" for short.

 

Therein lies part of the problem. Quants don't necessarily know much about the economics of the systems they are modeling, and the management in those companies don't necessarily know much about the models developed by the quants.

[npts2020]

I wonder if any of those physicists and mathematicians were named Ponzi?

[iNow]

Interesting story on 60 Minutes tonight about the financial crisis. One thing they said that caught my ear was that some of the weird securities that involved the breaking up of mortgage-backed securities into tiny pieces and then remarketing them in creative ways were designed by "Nobel-track" physicists and mathematicians working on the side for the Wall Street firms. The complex algorithms designed by these people were supposed to reduce risk, and the managers took their word for it and leveraged to the hilt.

I saw the special last night, too. Here's my quick response.

 

They wrote the models, they didn't take action based on them. The fault lies with the regulatory and oversight folks who were using said models to make decisions of monstrous importance without actually understanding them (pretty important thing to understand if you're in charge of making decisions based on their outputs).

 

Their lack of understanding the models when they were in the sole position of assessing risk was the problem, IMO.

 

 

Someone should have realized there was a problem and pressed the brakes hard when AAA ratings were going belly up.

 

 

A line that caught my attention in the 60 Minutes special was (paraphrased):

 

"Even trainees and interns making less than $40K a year would know better than to risk the entire company on such dangerous investments, yet the people at the very top did it anyway."

 

 

 

That's bad leadership and short-sightedness, not intentional misdirection by physicists and mathematicians if you ask me.

[npts2020]

Well the physicists and mathematicians were hired to be "experts" in a particular matter, if you are a CEO do you ignore your own experts' advice? It is probably true that not enough questions were asked for clarification, but the fact of the matter is that many if not most made their money anyway and will never have to give any of it back. IMO many knew the whole thing was untenable from the start but didn't really care because it was too complicated for more than a very few to understand and they would be gone before things collapsed.

[D H]

Well the physicists and mathematicians were hired to be "experts" in a particular matter, if you are a CEO do you ignore your own experts' advice?

 

In most other industries that involve significant risk, a separate group (and sometimes multiple separate groups) independently double-check the work of the designers and developers. This independent assessment is a core concept in the aerospace, electrical power and transportation industries, just to name a few. Problems happen even with safeguards in place (e.g., NASA's Challenger and Columbia disasters). For one thing, the safety experts have to be treated as key partners rather than as interlopers.

 

The fault in the Challenger and Columbia disasters was deemed to lie primarily with NASA management - even though NASA had safety checks, independent assessments, a hoard of safety experts, ... The financial industry:

• Safety checks: The financial industry required the developers of the financial products to do their own safety checks. That should cover the safety check issue, right?
  
• Independent assessments: That would have required the financial institutions to hire another group of "Nobel-track" physicists whose sole job purpose would have been to get in the way of their existing group of "Nobel-track" physicists who were making money hand-over-fist.

In the end, I predict the financial crisis will be viewed as not only a failure of management, but as a criminal failure of management. I also predict that much of the money those executives paid themselves will be seized by authority of the RICO act.

[Pangloss]

That was how I saw it as well, but I thought you guys had some comments (and amusing quips) above. I suppose I could see a case being made against (hypothetically) a physicist or mathematician who knowingly deceived his employer with a bogus algorithm (fraud?). But there's just no way this can be construed as a let-down by science in general.

 

That having been said, I wonder if some sort of new guidelines might be called-for. Something that would sit as a protective procedural barrier between uneducated managers and brilliant-but-untested methods. Perhaps some sort of peer review process for high-tech financing schemes. (Maybe such a thing already exists?)

[Realitycheck]

There are just too many variables to accurately model anything with a 100% confidence rate. When you start factoring in drops of consumer confidence due to unpopular wars and their effect on oil prices, for example, anything can happen. It's all connected.