1. Skip to navigation
  2. Skip to content
  3. Skip to sidebar
Source link: http://archive.mises.org/8908/quants-forgot-about-the-human-factor/

Quants forgot about the human factor

November 6, 2008 by

Quantitative finance analysts, or “quants” in the financial jargon, played a prominent role in the subprime crisis. Using a blend of mathematics, statistics and computing, they were the Wall Street geniuses who assessed the risks on all those exotic mortgage-backed securities and credit-default swaps. Their models were telling them that the risks were widely spread and that it was safe for the institutions employing them to pour billions of dollars into such financial instruments. This was supposed to be the most up-to-date, scientifically based way to make such calculations. What went wrong? Well, they forgot to include one minor variable… the “human factor.”

An article published two days ago in the New York Times surveys the various defects that have been found in those models.

The models, according to finance experts and economists, did fail to keep pace with the explosive growth in complex securities, the resulting intricate web of risk and the dimensions of the danger. 

But the larger failure, they say, was human – in how the risk models were applied, understood and managed.

Yes, one of the fundamental problems of mainstream neoclassical economics was at play here again: the erroneous belief that one can reduce human action to a set of numbers and formulae, and that these can be calculated and extrapolated just like physicists do when they try to predict the position of some space object a few millions years in the future.

The Wall Street models, said Paul S. Willen, an economist at the Federal Reserve in Boston, included a lot of wishful thinking about house prices. But, he added, it is also true that asset price trends are difficult to predict. “The price of an asset, like a house or a stock, reflects not only your beliefs about the future, but you’re also betting on other people’s beliefs,” he observed. “It’s these hierarchies of beliefs – these behavioral factors – that are so hard to model.” 

Indeed, the behavioral uncertainty added to the escalating complexity of financial markets help explain the failure in risk management. The quantitative models typically have their origins in academia and often the physical sciences. In academia, the focus is on problems that can be solved, proved and published – not messy, intractable challenges. In science, the models derive from particle flows in a liquid or a gas, which conform to the neat, crisp laws of physics.

Not so in financial modeling. Emanuel Derman is a physicist who became a managing director at Goldman Sachs, a quant whose name is on a few financial models and author of “My Life as a Quant – Reflections on Physics and Finance” (Wiley, 2004). In a paper that will be published next year in a professional journal, Mr. Derman writes, “To confuse the model with the world is to embrace a future disaster driven by the belief that humans obey mathematical rules.”

Funny how we always tend to reinvent the wheel in economics – or at any rate that’s the impression one gets watching mainstream economists trying to make sense of the world from their confused perspective. I haven’t read Mr. Derman’s book and I don’t know who he is quoting, but that insight is not particularly new. Decades ago, in a book aptly titled Human Action, Mises denounced “those economists who want to substitute ‘quantitative economics’ for what they call ‘qualitative economics’ (…). The impracticality of measurement is not due to the lack of technical methods for the establishment of measure. It is due to the absence of constant relations.”

Human acts of choice, Mises explained, cannot be predicted with certainty because “different individuals value the same things in a different way, and valuations change with the same individuals with changing conditions.”

The quants were blinded by their too complex models. If they had only applied some common sense, they would have noticed that something wrong was going on in the wonderful world of these securities. Of course, the ultimate high-tech device to avoid being caught in this mess would have been a good book explaining the Austrian theory of the business cycle. But sadly, Wall Street is still behind the times when it comes to such sophisticated devices.

{ 13 comments }

Philip Mathis November 7, 2008 at 12:36 am

Which is the best book for explaining Austrian Business Cycle Theory? I’m looking especially for something that is easy to get into that I could recommend to people who aren’t necessarily interested in economics.

Robert C November 7, 2008 at 1:51 am

I suppose the best start (for a short book) would be “The Austrian Theory of the Trade Cycle (and other essays).” It’s only about 100 small-format pages, and you can read it here:

Webpage: http://mises.org/tradcycl.asp
PDF: http://mises.org/pdf/austtrad.pdf

If you think it’s well-suited to your educational needs, you can buy it at the Mises store at a quite reasonable price:

Store: http://mises.org/store/Austrian-Theory-of-the-Trade-Cycle-and-Other-Essays-The-P46.aspx

Also, someone earlier this month referred me to a very informative slideshow that those who like graphs could find useful. I can’t remember what the title was or where to find it, but I’m sure someone else knows what I’m talking about and will post a link shortly.

Beta Hater November 7, 2008 at 2:22 am

Philip,

The best book on ABCT is “Money, Bank Credit, and Economic Cycles” by Jesus Huerta de Soto. Roger Garrison’s “Time and Money” is also very good, but I recommend reading it after Huerta de Soto’s book.

The slides Robert mentioned are by Roger Garrison. They accompany his lectures, which can all be found under the media section on Mises.org

Here’s a link to the slides: http://www.auburn.edu/~garriro/lvmi.htm

Keith November 7, 2008 at 5:24 am

“The quants were blinded by their too complex models.”

And yet these are the same people still professing to know the ‘right’ way forward and are followed as blindly.

tesla November 7, 2008 at 7:08 am

I believe that most quants knew the limitations of their own models. They shut their traps because when times were good they were clearing huge salaries and bonuses.

Pat November 7, 2008 at 7:25 am

Tesla is correct and I should know, I have a degree in financial engineering. In any case, Nassim Taleb and some other people were pointing out that financial models are as solid as some people in the field seem to profess (Of course, he was being a little bit specific, e.g.: probability distribution cannot be found).
In any case, it is true that the most compelling reasons were compensation and hubris (which was dashed with the unfounded belief that they can tell good times when they see one). To be fair, this has always been a part of human behavior (e.g.: rain dance). This time, it is finance. And unfortunately, it has huge consequences.

Larry N. Martin November 7, 2008 at 8:20 am

“Financial engineering”? Really? There’s a degree for that?

Pat November 7, 2008 at 10:36 am

Yup. Several graduate schools offer master’s programs in financial engineering (Another names for financial engineering are quantitative finance and computational finance). Of course, this has to do with Wall Street firms needing people with a background in finance and maths.

pussum207 November 7, 2008 at 1:49 pm

What blows me away was that people actually thought that the various assets they constructed from bundled subprime mortgages (SIVs, CDOs,etc), which at the time were hailed as wonders of quantitative brilliance, were somehow “diversified” simply because of the large number of mortgages and their geographical mix. The bottom line was that most of these assets were constructed of mortgages whose returns were obviously highly positively correlated. In other words, the economic circumstances that would force one to go bad would force them all to go bad. I would suggest that anyone with a good theoretical grounding in finance (as opposed to very advanced arithmetic) would not make this mistake.

I am an economist and have worked with lots of engineers, mathematicians, statisticians, etc., and they usually have the same flaws. They have a profound distrust for theory which they view as “speculative”. When theory (aka common sense) and what they perceive to be the “facts” (i.e., whatever corrupted incomplete set of data they happen to be working with) diverge, they always believe the “facts”. They are always wrong.

newson November 7, 2008 at 6:10 pm

tesla’s right. many were aware of long term credit and the problems with the models (ltcm ’98 being a well-known textbook example). but it should be said that the probability distribution used in many pricing tools (eg black scholes option-pricing) works ok in most environments.

the fact that they work reasonably in benign conditions means more players use the devices. the more the liquidity, the better the models function…until you run into the “fat tails” of the real distribution curve, and things go pear-shaped.

but the black swans are comparatively few, and you could put your children through private schooling easily between meltdowns.

Walt D. November 7, 2008 at 10:02 pm

The securitization process passed the risk on to the entity buying the securitized product. The quants were there just to give the pretense that the securitizer and the credit rating agency knew what they were doing. The goal was to sell the securities, not to provide a superior opinion of loss rate or loss severity.
Research departments at retail brokerage houses fill the same role – to give the impression that they know whether the stock market, bond market, XYZ stock, oil the dollar, whatever,is going up or down or sideways when in fact they don’t.

fundamentalist November 9, 2008 at 11:28 am

What if the models were good but the input was bad? When the Feds monkey with interest rates, and the state intervenes in the market, they distort all kinds of prices. Anyone remember GIGO?

Walt D. November 9, 2008 at 2:35 pm

Fundamentalist:
If you want alternative statistics you might want to check out this site.
http://www.shadowstats.com/alternate_data
Obviously, you will need to read through and understand his methodology and decide whether you think that it is reasonable.
-Walt

Comments on this entry are closed.

Previous post:

Next post: