Re: [ox-en] Re: Value of software
- From: Graham Seaman <graham seul.org>
- Date: Wed, 24 Jul 2002 18:40:58 -0400 (EDT)
Hi Stefan,
Looks like this one will run and run ;-)
On Wed, 24 Jul 2002, Stefan Meretz wrote:
Hi Graham, Stefan, and all,
some small comments;-)
On Monday 08 July 2002 00:43, Graham Seaman wrote:
Stefan Mn wrote:
Of course every commodity has some development costs, i.e. the
immaterial labor needed to develop it. The labor needed for that is
fixed / constant for a given product because it is all done when the
development process has come to an end. (Is this similar to the fixed
capital of Marx? I don't know.)
I am saying that it is NOT similar.
I am not so sure. First, I spontaneously agreed with you. But following
arguments make me hesitating:
Fixed capital is "objectified dead work" ("vergegenstaendlichte tote
Arbeit"), meaning: work, which was done and which had objectified in
machines *etc.*. What can *etc.* be? Can it be software, knowledge,
experience? And "objectification" ("Vergegenstaendlichung") does not
neccessarily mean materialization, it can have an immaterial form. This
implies, that you can sell these "knowledge products" -- like machines.
Than the same is true for these immaterial products: They transmit its
value in portions to the end-user products -- like machines.
Well, I disagree. But I am not sure if there's any way we can settle the
argument - it may be as much a question of definitions as anything. I
would claim that my definition is closer to the classical one, where
the basic factor on which everything else depends is the material
transmission of value as machines wear out. But against that you could
argue that Marx already allows for 'moral depreciation' (not sure if
the same words are used in German - meaning depreciation as a result
of machinery becoming out-of-date) so that he accepted cases where value
is transmitted with no material change in the machinery. Personally,
that makes sense to me when talking about an exceptional case, but if
it becomes the typical case it seems to me that the whole theory is
becoming so abstract (and value is tending to become something mystical)
that the theory no longer applies very well.
The reason I think it is not similar, is that fixed capital has value,
and part of this value is transmitted to the product (financially, as
depreciation; physically, as wear-and-tear). Produce more of the same
product, and the fixed capital wears out more quickly.
Disagree, only the portions of transmitted value become smaller.
This is not a question of definition but of fact - and surely it must
depend on the type of machine or processing system. But maybe we are
talking about two slightly different things:
1. (my case above) - producing more goods by running the same physical
machines faster
2. (your case?) - producing more goods by expanding the scale of
production, in the region of positive returns to scale.
Even in case two I'm not sure that the value from the fixed capital
becomes smaller, though the direct labour value usually will.
Same hint
I gave before: These categories are societal ones, they are only valid
"in the societal mean" ("im gesellschaftlichen Durchschnitt").
But the 'societal mean' for value is established through its
transformation into price, via the stock exchange and many other
institutions. What we are discussing is much more abstract than that.
Than,
following your arguments, you can conclude similarity.
No, following YOUR arguments ;-)
Design (whether
of software or anything else) does not wear out physically. Whether you
produce 500 or 5 million copies of a design is irrelevant. So, if the
design had value (like fixed capital) this value would never be
transmitted to the product itself, and value without realization on a
market does not exist.
Above you said the contrary ("...NOT similar")?
Yes, and here I said only 'IF it were similar' (ie. IF it had value).
It is not irrelevant how
many copies you can produce (and sell: of course, only realized "value"
is value). If you produce a higher amount of products the relative
portion of transmitted value becomes smaller and your profit increases.
Oh. So, you are saying that price is independent of value. Then you are
agreeing with me: the price of software (and other IP goods) is not
determined by its value. :-)
With 'normal' goods (say bread, for example) if you produce more, then
you can generally make savings in total labour input, the value falls,
and then competition pushes the price down to somewhere close to the
value. To make increased profits you have to be the first to use the
new system, so you can sell the bread above its value but below the
'social mean' value. And this is a temporary situation (or you have
a monopoly on bread).
But with software monopoly is built-in by copyright/patent laws,
there is no true competition (apart from free software), and so software
can be sold far above its value.
And in addition every commodity has a production cost being the labor
needed to actually produce it. If you're looking from the information
side of the product it would be the reproduction cost BTW.
The fundamental - not to say: historical - shift here is that the
reproduction cost drops relative to the development cost. And
software is the most exteme example.
Is this what you're saying, basically?
Yes. But there are two ways of looking at this fact. I believe that the
most important thing here (as you said) is that reproduction costs have
dropped hugely, while development costs have stayed static or fallen
less than reproduction costs (because they are more labour intensive).
The pharmaceutical companies/Hollywood/Microsoft would have us believe
that the basis of the change is that development costs have INCREASED
hugely. This is the whole basis for their justification of more
restrictive IP laws. But there are many pressures (tax in particular)
that make them inflate their development costs so that the figures that
are available cannot be trusted.
In general, I think they _are_ right. This in other terms show, that the
immaterial part of production increases -- and this is the part which can
be freed first!
Now we have a purely empirical question :-) But a very hard one to answer
- I don't know any way to get enough historical data to solve it. One
very crude way to get an approximation might be to find the ratio of the
wage bill of Edisons research department in 1900 to the wage bill of
the lightbulb factories, and then do the same for Intel now. Repeat for
a range of industries. But it would be very crude, even if you could find
the data. The cost of production of a chip factory is I believe around
2 billion dollars (part of which is again research, since new machines
have to be designed for each new feature size of the chips themselves),
and that would have been excluded from the calculation.
The question also has political/moral consequences: I think record
companies, drug companies, microsoft etc are thieves, even in their own
terms, charging for things with almost no value. You think their prices
reflect real value.
But whichever is correct, it is still true that the immaterial part is
being freed first, as you say :-) And the fact that free software may
also be free in price demonstrates how a large (and increasing) part of
the economy is based on theft - and needs to be, to survive in its
existing form.
best,
Graham
Ciao,
Stefan
--
Vereinte Dienstleistungsgewerkschaft ver.di
Internetredaktion
Potsdamer Platz 10, 10785 Berlin
_______________________
http://www.oekonux.org/