|
PureBytes Links
Trading Reference Links
|
"ARE SPECULATIVE STOCK MARKETS USEFUL FOR ANYTHING?"
by Damian H. Zanette
Centro Atomico Bariloche
In his recent contribution D. Stauffer concludes that "if econophysics is
able to predict many crashes with good reliability... I think this is still
a fundamentally helpful new contribution and will help to prevent the stock
markets to crash." Indeed, it is apparent that most of the work in this
recent interdisciplinary research field refers, in a more or less explicit
manner, to the dynamics of stock markets [1]. The prospective of having at
hand a complex system at a human scale, whose interacting elements seem to
be driven by rather simple rules and whose output is presently accessible
with a resolution of the order of seconds, is most appealing for we
physicists. The effort focused on the physics-like modeling and the
statistical description of stock markets is therefore not surprising.
Now, it is clear that stock markets work in two well differentiated time
scales. In the long run, over the years, the average trend of share prices
reflects the evolution of real economy, varying according to the profits of
the firms represented in the market. This trend depends of course not only
on the individual performance of firms, but also on their complicated
interaction along the productive cycles. In any case, at such time scales,
the stock market is influenced by a definite input from the "real world" and
evolves accordingly. On the other hand, within periods of a few minutes to
some days or months, stock markets work on a purely speculative basis.
Information from outside is used by agents to build expectations concerning
future events but, in the process, elements such as suggestibility, gambling
behavior, fashion, group pressure, and opinion diffusion [2] play the
leading role. The output of this process is essentially unpredictable and,
indeed, share prices vary in a seemingly random manner, which justifies the
physicists' approach to stock evolution by means of probabilistic models.
Note that whereas every-day facts are essentially irrelevant to the
short-term dynamics of the stock market, the way back is not true: it
suffices to recall that the Great Depression of the Thirties begun with a
stock market crash in 1929 [3].
If econophysicists get strong evidence supporting the fact that speculative
stock markets work much like an autonomous complex system, able to
spontaneously develop episodes of large internal fluctuations similar to SOC
cascades or to "mass extinction" events in evolutionary models [4], their
contribution to economics should go far beyond predicting crashes or
inventing smoothing mechanisms for such events --which remains however a
relevant task, as D. Stauffer asserts. I think that the main message to the
financial world should be that, in contrast with long-term investment,
speculation in the stock market is a rather stupid and very dangerous game,
able to suddenly drive thousands of investors to bankruptcy, without any
reason connected with reality.
This conclusion raises immediately the question in the title of this note.
What is the speculative stock market useful for? Brokers, who may make money
even when stocks fall, would not have problems to answer. However, it is
surprisingly difficult to find a serious answer to such an elementary
question in the literature. Only the most enthusiastic supporters of
capitalism seem to agree that "because a booming capitalistic economy
requires large accumulation of investment capital, a market of capital was
developed as well... It was the capital obtained in these (stock) markets
that dug the mines and built the factories of the industrial revolution and
which continues to this day to make the capitalistic societies productive
colossuses" [5]. In the light of what econophysics teaches us about stock
markets, such assertion is hard to grasp. Would the whole industrialized
world have got its financial resources from an essentially random system,
able to develop catastrophic fluctuations within a few hours?
Meanwhile, it would be really stimulating if the focus of econophysics
shifts towards problems more related to real economy. Of course, such
problems are typically more difficult than the stock market dynamics since,
among other reasons, they cannot be formulated in terms of autonomous
systems, they involve input data from the real world, and they are not
expected to reach stationary regimes. Nevertheless, they are not less
interesting or less adapted to physics-like modeling. Here are a few:
Concentration of economical power
Role of economical activities in the ecology of the human population
Dynamics of money and its role as a purely conventional exchange tool
I am sure that a physicist's approach to these problems may significantly
contribute to their understanding under a new light.
|