Kaos (theory) on Wall Street

Is it really true that lapsed physicists who work in financial institutions applying physics to stock market modelling have caused the current world economic turmoil? This idea has been raised in a NYT opinion piece and has been discussed in other blogs.

The idea is that physicist influenced modelling became so esoteric that no one else could understand it or the now toxic packages of debt. Since no one understands where or exactly what the debt is, no bank wants to lend any more, ergo the crisis.

You see its just the packaging of debt that's the problem, not the shyster 100+% housing loans and the insane housing price bubble it produced. Nor the wholesale privatisation or the government approval of banking excesses. Its not really surprising that bankers are trying to spread the heat around. Not when the public sentiment towards bankers is something like,

"I'll tell you what we should do, spray them with wildebeest odour and make them run through the Serengeti, with a commentary by Attenborough"

But an investment bank cant be a very nice place to work at the best of times. One former physicist who works on Wall Street

"termed the market a wild beast that cannot be controlled, [...] There are a thousand physicists on Wall Street, she estimated, and many, she said, talk nostalgically about science. 'They sold their souls to the devil'."

Now thats a sad story. Though it is true, in some of my darkest moments, in the face of some crazy research grant cut, I entertained the thought of taking up the devil's offer. But my motives were noble you understand. I saw myself as a latter day Parvus who would expose the system for what it was. But since Parvus' scheme ended in disaster and that the bank interviewers would more than likely see through me to my bad attitude, I thought better of it.

Benoit Mandelbrot gave a lecture recently in which he railed against the prevailing economic models as (clearly) wrong. Attempts to model data as "random walks which assume that the price at any given moment depends on what it was the moment before" ignored the fact that stock prices were often discontinuous - jumping wildly from time to time. In fact, Mandelbrot went on the say, economists often ignore "outliers" - data from extreme events (like the one we are going through at the moment). But its the wild swings which constitute the reality

[Mandelbrot] contends that more realistic models of economics—including, naturally, models based on fractals—are driven by "wild randomness," wherein things don't average out and individual freak occurrences matter. This wildness, he said, "imitates real phenomena in a very strong way."

Mandelbrot leaves open the possibility that the stock markets could be modelled using fractals. As a guide I point out one feature of fractals - that they display the same level of complexity at all scales. To that end I took the liberty of posting here a plot of US GNP since 1950.us_gnp_1950

Now I would say, that plot shows a regularity that is simply missing from say, daily currency fluctuations (which I look up sometimes to watch my modest savings in UK pounds become ever more modest). The regularity is indicated by the grey vertical bands - the recessions. If you allow for the special case of the mid-70s "oil-shock" recession and the (relatively mild) mid-50s recessions, then there is a recession every, ooh lets say, 10 years:

since 1825, when the first general crisis broke out, the whole industrial and commercial world, production and exchange [...] are thrown out of joint about once every ten years. Commerce is at a standstill, the markets are glutted, products accumulate, as multitudinous as they are unsaleable, hard cash disappears, credit vanishes, factories are closed
Fredrich Engels 1877 Anti-Duhring Part 3, Chapter 2

If referring to Engels(and Marx) makes you uncomfortable, then you can take a respectable bourgeois economist like Ricardo who recognised these cycles even before Marx and Engels. Or German chancellor Angela Merkel - herself a former physicist - a couple of days ago, "The world stands at a watershed. We cannot afford crises like this every 10 years".

Indeed. And to make a value judgement, why should we listen to our "leaders" when they spent the last 10 years encouraging greed, speculation and debt, and now we are suposed to trust them that they've seen the light? Actually I dont think they have any real idea about how the economic system works.

So that brings me to my point. We should bemoan the fact, not that so many physicists attempt to understand the economy, but that so much effort is wasted on daily stock fluctuations in order to make the rich richer. The system shows regularity at larger scales and its these fluctuations that really need to be studied and prevented - even if the system has to be stood on its head as Marx and Engels argued.