To continue the discussion on Bruce's 1556 of 26/03 I want to focus on
a misunderstanding which I hope will deal with the objections raised
by Bruce and others.
Bruce writes:
"presumably capitalists make their decisions on the basis of what they expect
to happen in the next period, so they *would* institute this highly profitable
change; then, given the effects produced in period 2, they would presumably
look around for any further labor-using technical opportunities in order to
further raise their rate of profit by adding to the workforce and shrinking the
means of production (and the composition of capital)."
Many others have followed up by asking whether behavioural assumptions
such as the above lead to reasonable or unreasonable results.
But you cannot judge the temporal approach on this basis. As such
it makes no special behavioural assumptions. It is a paradigm, not a model.
It does not tell you what capitalists do; only the outcome of their actions,
whatever these actions are.
If you think the capitalists invest in methods of production that yield
the highest profit rate, then do so. If you wish to argue that they will invest
in the method of production that yields the highest ratio of physical input to
physical output, then do so. If you adopt Okishio's own assumption, that they
will adopt the technology that yields the lowest unit costs and hence the
highest profit at existing prices (in Bruces's case p=1) then feel free.
*All* these particular assumptions are compatible with the temporal approach.
They are all different 'models' within the temporal paradigm. I think it is
rather important to get this distinction clear.
The difference between the temporal and the simultaneous approach is how
it evaluates the outcome of the capitalists' decisions. In each case the
temporal calculation in this case yields one rate of profit and the
simultaneous calculation yields another. The dispute is not about which
behaviour is more reasonable but about these two estimates of the profit rate:
(1) which is Marx's profit rate?
(2) which is closest to the actual observed profit rate?
The temporal approach is a paradigm, not a model. You can plug in any
particular set of behavioural assumptions you like and you will get a
different model for each set of behavioural assumptions (as is actually
the case for the simultaneous model).
The point is that the correspondence between sets of behavioural assumptions,
and sets of outcomes, is different for the two paradigms; not that one paradigm
commits us to one set of behaviours, and the other paradigm commits us to
another set of behaviours.
Much confusion has arisen, I think, because like Bruce we have been
in an essentially defensive position for the last fifteen years. We have had
to accept the terrain of our opponents and we have therefore adopted assumptions,
like the rate of profit equalises in each period or the capitalists adopt
Okishio-behaviour, not because we necessarily believe this to be the
case, but because we are refuting theorems that adopt these assumptions. In order
to overturn someone else's theory, you have to adopt the assumptions of that
theory. But in order for us to develop our own theory, there is no need to
adopt these assumptions and there are many of them that I would reject.
The point about the temporal paradigm is it shows that *on Okishio's
own assumptions* the rate of profit falls when he says it rises. No other
refutation of Okishio has done this. All other refutations of Okishio
consist in one way or another in *violating* Okishio's assumptions, by
purporting to prove that capitalists would not in fact behave as he says.
On Okishio's assumptions, the capitalists in Bruce's example would not
adopt the new technology, because on Okishio's assumption they would assume
(perhaps wrongly) that output would be sold at its old price of $1.
They would conclude that their sales would yield $100 but that their costs
would be $45 for the seed corn and $15.50 for wages, so they would
estimate their new profits to be $39.50 and stick with the old technology.
The point is that if they *do* switch, the two paradigms yield a different
profit rate. Neither paradigm in and of itself predicts whether they actually
will switch.
I will continue this in a further post.
Alan