Re: zero average profit

From: Philip Dunn (pscumnud@DIRCON.CO.UK)
Date: Mon Jun 09 2003 - 08:21:35 EDT


Quoting Ian Wright <ian_paul_wright@HOTMAIL.COM>:

> For example, I like very much the paper "Statistical Mechanics of Money"
> by two physicists working in the field of econophysics:
> http://www2.physics.umd.edu/~yakovenk/papers/EPJB-17-723-2000.pdf
> Their model is extremely simple: N agents, M money, random encounters of
> pairs of agents who randomly transfer an amount of money between them.
> That's it. It turns out that the statistical equilibrium of the system
> generates
> an exponential income distribution (lots of agents with not much money,
> and a decreasing number of agents with a great deal). This distribution is
> in empirical agreement with a large section of the income distribution in
> the
> US, as measured from tax returns (the model breaks down for capitalist
> incomes, which are better fitted by a power law, not an exponential law,
> which is interesting in itself). Their model is exactly analoguous to a
> perfect
> gas, in which elastic collisions between molecules transfer energy between
> them. So this work is very much in the tradition of Farjoun and Machover,
> who also apply statistical mechanics to political economy. I wonder
> why the physicists' very simple model is able to capture something essential
> about the economy, and whether this kind of modelling approach can be
> extended.

Duncan Foley has written a paper applying thermodynamics to economics, rather
than statistical mechanics
Classical thermodynamics and economic general equilibrium theory (with Eric
Smith) http://cepe.newschool.edu/~foleyd/econthermo.pdf

It replaces Walrasiam initial endowments followed by adjustment to equilibrium
with the thermodynamic concept of reversible near-equilibrium changes.  The
analogue of temperature is different from that in "Statistical Mechanics of
Money".  Econophysics raises questions like 'is capital analogous to heat, or
whatever?'



Philip Dunn


This archive was generated by hypermail 2.1.5 : Tue Jun 10 2003 - 00:00:00 EDT