dispatches from the new world of work

The Black Swan Snaps!*

Any semi-regular reader of this Blog knows I can't say enough and enough that's laudatory about Nassim Nichols Taleb's two books, Fooled by Randomness and The Black Swan—they capture my world view with incredible clarity.** Black Swan came at exactly the right moment—it offers a more or less 100% explanation for our deeply disturbed financial markets. I was around Bill Sharpe, the Nobel-holding, more or less father of contemporary portfolio theory, engine of our current woes, when he did the work that won the prize—I was mostly ignorant of what was up, but appalled by the apparent arrogance of Sharpe and his devotees. It was clear—to them—that risk would be tamed, once and for all.

[*Swans do indeed snap—I have them on my farm, or at least their kindred Chinese geese.] [**Gawd, do people get upset when one dwells on the "significant" role of luck in life—stunning and troubling and amusing at once.]

I now eagerly refer you to NNT's 24 October (sorry for the delay) magical, terse piece in the Financial Times, "The Pseudo-science Hurting Markets." The pull quote goes like this: "Medicine used to kill more patients than it saved—just as financial economics endangers the system by creating risk."

The "pseudo-science," per NNT, is" 'Nobel-crowned' methods of modern portfolio theory." Aiming to erase risk, the quants' house of cards has increased said risk immeasurably and is blowing us all away—just ask Mssrs O'Neal/Merrill and Prince/Citi about their week—decidedly more harrowing than mine. NNT really rubs it in—and I literally rubbed my hands with glee, tough to do while reading a paper, but possible if the glee-meter was as high as mine—when he pointed out the little known fact that the economics Nobel ain't the real thing; instead it's the "Sveriges Riksbank Prize in Economic Science in Memory of Alfred Nobel."

Selfishly, I don't want the world to go to hell in a handbasket—but a part of me believes that the arrogant quantocracy deserve a serious public thrashing; some, like O'Neal, will just have to learn how to live off their tens of millions of severance pay.

Tom Peters posted this on 11/05/07.

Comments

Tom:

I'm a big fan of NTT, and appreciate you pointing out this article.

Is there anything like the 'Oh S**T' feeling one gets while reading his books, that the the statistical tools we use for financial modeling is completely analogous to wandering around with a hammer in our hand, looking for nails to hit? And then the one that really brings it home, the 'OH S**T!!' that occurs when he takes apart Modern Portfolio Theory?

A great read on a Monday in Q4 '07!

Posted by Dave at November 5, 2007 10:38 AM

Fellow NNT fan Steve couldnâ€™t agree more: The Black Swan exposes how we are suckers for patterns presented with logic - we are all too ready to infer cause and effect where we have no right to. Financial "experts" take special advantage of this phenomenon. The problem, argues Taleb, is that our world isn't ruled by moving averages anymore - it is ruled by wild exceptions - Black Swans.

Here are earlier tompeters.com posts on The Black Swan

http://tompeters.com/entries.php?note=009424.php

http://tompeters.com/entries.php?note=009722.php

Read it, everyone!

Posted by Steve Yastrow at November 5, 2007 10:59 AM

I have long been hesitant about using financial models. Four points underline my scepticism. First, the models use simplifying assumptions that move the conclusion away from the real wolrd. Second is the financial world's acceptance of what I feel is good first order approximation but at the same time a seriously flawed assumption: that future risk is defined by historical variation. I quite simply do not believe this is true. Third, the models often assume that the markets will act within historical norms. Fourth, I believe that many financial models are not independently derived or managed but instead are used by individuals and institutions to support their business, et cetera. All of this suggests that the modelling can not be dependable in the very times of stress that accent the need for accurate information.

The NY Times has an article "Go Ahead, Rationalize. Monkeys Do It Too" that helps explain this at http://www.nytimes.com/2007/11/06/science/06tier.html.

Posted by s g at November 6, 2007 9:40 PM

1. NNT Kicks Ass! (And provides glee to me too!)
2. The science, financial models, etc. are based on assumptions...too few check the assumptions...and even fewer have enough firing neurons to understand. So much easier to ride on laurels and labels.
3. I'm in healthcare. I know how much scientific crap is out there...and a big part of my work is to sift through the crap for the value.
3. The South Park Episode below may have some relevance to this string of comments, measures, laurels, egos, and validity. (Yes, South Park can and often does parody things that matter to us in work....I use this stuff in workshops!)

Have fun.

(FYI.."How to LIe with Statistics" is a good book for those of us who seek to take the "red pill" (or is it the "blue pill" in the Matrix trilogy)

http://www.southparkzone.com/episodes/1109/More-Crap.html

Posted by Manoj Pawar at November 7, 2007 1:24 AM