Interview with J.
Doyne Farmer
Chaos Theorist Discusses
the Markets and Nonlinear Theory
Interview by Will McClatchy, Editor
"It is clear from classic arguments that the market
can't be fully efficient. If it were, everybody would
go home. There has to be some kind of stasis where
the market is pretty efficient but not perfectly efficient."
- J. Doyne Farmer
Indexers use the efficient market hypothesis to help
explain generally why stock market averages are so
hard to beat with consistency. But efficient market
theory does a poor job of explaining bouts of speculation
or panic - irrational investors.
One intriguing theoretical approach that incorporates
irrationality well is the evolutionary biology model.
It requires no particular level of rationality, efficiency
or other assumptions that have plagued the classical
efficient market theory. A wealth of statistical techniques
support it, and it incorporates computerized analysis
easily.
"In the biological world species that do well are
rewarded by proliferating and those who don't die
off," says J. Doyne Farmer, McKinsey Research Professor
at the Santa Fe Institute, a complex systems think-tank.
His seminal paper 'Frontiers
of Finance: Evolution and Efficient Markets' sums
up this approach.
As a twenty-four year old physics graduate student
he was a member of the "Dynamical Systems Collective"
of UC Santa Cruz, a group of pioneer nonlinear dynamics
scholars featured in the bestseller "Chaos" by James
Gleick. He is also a successful institutional trader
of indexes for Wall St. through his firm Prediction
Company.
"A similarity in the investment world involves managers
who do well and increase their funds and managers
who don't and go out of business. It's quite clear
that there is some strong selection pressure in financial
markets. Good investors are selected and poor investors
are weeded out. That is very different than saying
things are optimal. In biology there is little sense
of perfection."
The commonsense notion of relative efficiency works
fine in evolutionary competition. Species need not
perform at optimal levels to be successful. Humans,
for instance, have chronically bad backs but they
still compete well as a species.
"Performance in markets is very much relative to
players in the market," he said. "The system is co-evolving.
It is dynamic. Efficiency doesn't happen instantaneously."
"It is clear from classic arguments that the market
can't be fully efficient. If it were, everybody would
go home. There has to be some kind of stasis where
the market is pretty efficient but not perfectly efficient.
What does efficiency mean? What are reasonable profits
for taking risk that might involve skill? Once you
say the market is not perfectly efficient, you need
a way to talk about what that means."
What does this mean for the average investor? Inefficiencies
of boom and bust cycles are likely to exist, but it
is hard to say whether a successful investor is methodically
exploiting them or just lucky.
"My gut feeling is that there are substantial inefficiencies,"
he said. "Irrationality makes a difference. It s not
just that people don't have the information but that
it's hard to make reasonable inferences. I myself
am often quite confused."
"Often managers who do well are very lucky. On average
skill is rewarded and those without skill die out.
Track records are indicative but it's important to
know how to read the fine print. Someone making a
lot of bets probably has a more meaningful track record
than someone adjusting their portfolio once a year."
And, yes, that is what he does on the side with Prediction
Company, a technical analysis firm he founded 9 years
ago. The company makes thousands of small trades each
year for money managers Warburg Dillon Read.
Professor Farmer gained notoriety in the 1970's
as the de facto spokesman for the so-called Chaos
Cabal of young physicists who were at the forefront
of Chaos Theory. Known for his efforts to crack the
mystery of the roulette wheel, Farmer worked for the
Theoretical Division of Los Alamos National Laboratory.
Currently, Professor Farmer works at the Santa
Fe Institute. In April of 1999, he jointly published
a research paper with MIT economist Andrew W. Lo,
the coauthor of 'A Nonrandom Walk Down Wall Street.'
The paper, entitled, 'Frontiers
of Finance: Evolution and Efficient Markets' is
a fascinating comparison of the market to an evolving
biological entity
Other recent papers by Professor Farmer:
'Physicists
Attempt to Scale the Ivory Towers of Finance'
'Market
Force, Ecology, and Evolution'
07/10/00