Re: [OPE-L] Stocks and flows in Marx's theory

[GERALD LEVY <gerald_a_levy@xxxxxxx>]

> I think the way forward is the computational agent-based modeling that
> Ian Wright advocates. It may be possible to extend his model
> (http://arxiv.org/abs/cond-mat/0401053) to deal with capital stocks.

 
Hi Dave Z:
 
I agree that's a promising area of research.
 
> I think of each unit of production, i.e. firm, as a node in a network
> that connects flows of labour, either direct or materialized as
> commodities. If the net flow of labour into a firm is positive it is
> building up capital stocks, if the net flow is negative its capital
> stock is shrinking through wear and tear. The trouble with this approach
> is that it is not easy to cast into a dynamic model that makes
> meaningful predictions.

 
I am not so worried about the predictive value of the model.
(Nor am I worried about how faithful a model is to Marx's
conceptions). 
 
The predictions of a model almost always follow from the
specifications of the model. What would be valuable, imo, is to
develop a reasonably realistic model of a capitalist economy
and then to see what outcomes there would be in _alternative_
scenarios.  This has long been done in economic forecasting,
but not as frequently with truly dynamic models. Of course
there are other issues (methodological, ontological, choice of
dynamic model, etc.).
 
In solidarity, Jerry