[OPE-L:3489] Re: accumulation: disequilibrium dynamics

Steve Keen (s.keen@uws.edu.au)
Sun, 20 Oct 1996 19:04:34 -0700 (PDT)

[ show plain text ]

Replying to Jerry's points:

>Steve K wrote in [OPE-L:3482]:
>> In those cases, in effect, workers are issued with "chits" by one
>> capitalist which are honoured by all others.
>Well ... I don't think "chits" can serve as a substitute for money under
>capitalism. Marx discusses this topic in the chapter on money in the
>_Grundrisse_ re Darimon and Proudhon.

Come one Jerry, "chits" is in inverted commas above mate! A proper
expression would have been "negotiable bills of exchange drawn upon the
bank", which is nothing like Proudhon's labor chits. They are in money terms
for a start, not labor-time equivalent.

Re technical change being modelled first vs finance, this revives my little
comment on the nature of abstractions: simplifying, domain and heuristic.
Starting in the absence of technical change, but with finance, is a
heuristic. I could have started, for example, with technical change but
without finance, and then later attempted to introduce finance. But in my
opinion that is a less sensible path, for 2 reasons:

(a) introducing technical change introduces lots of other complications. For
a start, if I have technical change, then I also have to have an expanding
population, otherwise in the limit my model will predict declining
employment. So the initial model with tech change first is more complicated,
and potentially misleading because of:

(b) With technical change but without finance, you must have a "real",
commodity-money perspective on finance. This is how Marx began Capital
I--everything is a commodity. It can work as a starting point, but only if
you're cogniscant of the need to introduce non-commodities later on. That
doesn't necessarily happen: people can continue to try to make a
commodity-money perspective (developed without a finance sector) fit a
non-commodity-money situation (when a finance sector is introduced).

Despite my protestations below about the order in which things are
introduced and the difficulty of modelling technical change, it is quite
do-able. Realism doesn't take a back seat to mathematical considerations:
it's more a question of constructing your model in the rigt order, and in my
opinion that is finance first, tech change after.


>> The answer to (1) is also another reason why technical change comes
>> after finance in this model: a whole raft of things have to be taken
>> account of once you introduce technical change. But capitalism can exist
>> in its absence, and capitalism requires finance to operate, technical
>> change or no.
>The thrust for technical change under capitalism is an expression of the
>capital-form and the valorization process. On the simplest level, it is an
>expression of attempts to increase relative surplus value.
>Can capitalism exist without technical change? This is similar to asking
>whether capitalism can exists -- over the long-term -- without
>accumulation. While we can hypothesize a model of simple reproduction in
>which there is no technical change or accumulation, this does *not* mean
>that capitalism does not require technical change.
>Does capitalism require finance to operate? Yes ... but, ...
>(a) the requirement for finance by business firms is, in large part, a
>function of business firm's needs for money capital *for* technical
>change. In that sense, the analysis of technical change should happen
>prior to the analysis of the financial sector.
>(b) The funds that are needed for technical change (and expansion of v +
>c) can be *financed out of the surplus value created in those firms*,
>i.e. there is a source of *internally generated funds* for capital
>expansion under capitalism. Of course, at a more concrete level, there is
>a need for outside financing, but ... the question here concerns the
>*logical ordering* of different determinations of the accumulation
>> Introducing technical change is "simply" a matter of allowing the A
>> matrix and L vectors to alter from one period to the next, but there are
>> lots of issues about how one does that (is technical change a sigmoidal
>> function of profit, for example; how does change of the A matrix and L
>> vector relate to movements in commodity prices and the real wage?).
>> Such a model is always going to be a pale imitation of technical change
>> in the real world, in any event. You have to maintain the fiction of a
>> fixed number of aggregated sectors, when in the real world, technical
>> change results in new industries (aeroplanes, computers, robots, etc.)
>> which simply didn't exist in earlier times.
>I agree that there are lots of issues associated with modeling technical
>change, but that is not a sufficient reason to not undertake that
>investigation. You seem to be saying that it would be very difficult in
>practice to mathematically model technical change. I agree, but I again
>insist that the desire for mathematical simplicity must take a back seat
>to *theoretical* considerations. In other words, we can't begin, IMHO, by
>saying: let's build a determinate math model. We have to, instead, first,
>examine the subject at hand and _then_ *ask* _whether_ it can be expressed
>in a formal math model.
>In Solidarity,
Steve Keen
Senior Lecturer
Economics & Finance
University of Western Sydney
PO Box 555 Campbelltown NSW 2560
s.keen@uws.edu.au (046) 20-3254 Fax (046) 26-6683