A Response to Sander

Hi Sander,

You've written a very long text; I don't intend to reply to all of it - my main interest is in value theory as it relates to software, not in services in general. I will make one comment on services: you think my distinction between creating value and being productive for capital is untenable, and reduces to a Smithian view that only creation of physical objects can be productive; I think your restatement of some of Marx's rather circuitous and experimental comments on the issue leaves the problem as unresolved as it was when we began. Neither of us can possibly prove the other 'wrong'. So I would like to stay with the issue of value creation and software (which has some overlap but is certainly not identical with the problem of services), but where I have more motivation to try to be 'right'.

I suspect our differences may relate to underlying political attitudes, and that these in turn are connected to your complaint that "I am trying to understand your [position] but don't find it easy". Of course, it may just be that I'm not explaining the formal side of my position well enough, so I'll start with that, but I'd like to spend more time on how this relates to political interpretations of both Marx and the world.

First, the formal side: you say that all I am giving are 'more authority-arguments'. Well, yes, you could say that. But I think you are underestimating the importance of such arguments. Economic theories may be ideological, but if all of them - including Marx's critique - agree on a point, then it is unlikely to be unmotivated. At that point rather than dismissing it as 'authority arguments' you need either to show the ideological reasons behind this agreement, and how Marx managed to miss those reasons, or to create an alternative theory which integrates fixed costs in a radically new way. Without one of those two you are stuck with the 'authorities', like it or not. Of course I'm not an economist so it may be I'm factually wrong about this agreement, but I doubt it. Just to be clear, what I am saying is that:

1. Fixed costs have no major role in any economic theory (of capitalism) beyond that of their role in the formation of oligopolies through presenting barriers to entry to an industry.

2. Fixed costs are of course incredibly important in practice to anyone actually running a company, and they are a major topic in accountancy, as opposed to economics.

3. There is a very straightforward reason why 1. should be so. Capital without growth is not capital. Fixed costs do not grow, by definition (of course in the real world there is no absolute division between fixed and other costs, since some costs grow in sudden jumps at different scales of operation and so are fixed from one point of view and variable from another). The differential of a fixed cost is zero. By definition, therefore, they cannot be an important part of marginalist economics. Marx's economics is also based on a definition of capital as something that grows; his formulas are always of the form x -> x + delta x (and the role of surplus value is always as delta x, never as a fixed quantity).

In both marginalist and Marxist economics, fixed costs are a premise (and so take the form of sunk costs) for the topics which really interest the economists.

4. Clearly R and D and similar costs fixed costs have in reality become increasingly important in the modern world. For the normal functioning of a capitalist economy these costs need to somehow become part of commodity circulation. In addition to the social and legal problems this creates (which I'll skip for now as I'm staying with the formal side of things at the moment) this means that capitalism needs a theory/ideology which shows that these costs are a normal part of commodity prices, and can be encompassed by the 'hidden hand' to give an ideal distribution of these goods to those who can pay for them, just as happens with physical goods. Marginalist economics as an ideology is based on the claim that free competition leads to an optimal set of prices. But it was proved in the 1950s that there can be no optimal price for a good which has fixed creation costs but no reproduction costs (it's a standard theorem in welfare economics). The theory therefore completely fails to prove what it is needed to, although this fact does not seem to be publicised by economists (1).

5. The really fetishistic nature of capitalism means that the theoretical problem is not simply an ideological distortion of an underlying reality which is quite different; the underlying reality itself has these problems, and the theory itself is a part of them. So, to take one example, the EU has imposed a directive on access to government managed data (maps, meteorological data, census data, food composition data etc) based on a study (the 'PIRA study' (2) ) which has shown that charging for this data is grossly inefficient, reducing the use of the data and so the tax revenue from the data. Governments are now trying to turn the directive into law, but finding that not charging for the data leaves no way to pay the departments which generate the data. The result is complete confusion and laws which have no coherence at all. There truly is no optimal price for such data. To take another example, UK tax regulations require companies to value their 'intangible assets'. But the regulations can provide no way to do this at all; the costs of creating them are irrelevant to price, but there is also no market value. You suggested that while copies of these goods may have no value, the original does; but in practice there is no market to establish this: if we both produce shoes, the market will tell us the real value (ie. socially necessary labour) to produce the shoes; it can do this, since shoes are comparable. But if we both write songs, the market will not tell us the real value of 'a song'. In fact for tax purposes you do not value the 'intangible asset' itself, but the rights to the asset - you must have a patent, or trade mark, or other right which can be sold, and it is the degree of monopoly given by the right which determines the value (eg. the royalties which can be expected from the right). Thus price is a function of law with minimal mediation, and the function of the market in determining necessary labour for this type of goods is nonexistent. It is not simply the abstract economic theory which struggles; it is also the practical accountant who needs to draw up a tax return (3).

6. Marxism in so far as it is an economic theory, or has an economic aspect, is trying to explain the same reality as is marginalism. The explanation is of course entirely different - and ultimately explains marginalism itself; but both are attempts at an explanation of the system we live in. Marx explained the price of goods, which apparently emerged as the sum of costs due to wages, parts, rent, and interest, as in fact being ultimately determined by value, which has its roots in labour alone. Price appears to be an entirely natural phenomenon; Marx showed that this appearance derives from a social reality. Marginalism denies this reality and says that price is all (and further that price is eternal). But now we have a situation where we have goods with no 'natural' price; the price has to be fixed by granting a legal monopoly, then allowing the owner to charge a fee for the use of the good (so that no sale actually takes place), with the enforcement of this fee backed up by additional laws and technology; the whole basis for the capitalist ideology of sale as 'exchange of equivalents' is abandoned. In this situation, where there is no apparent natural price to explain, why is there a need to bring in value? What is value to explain here? Of course, if Marxism can come up with a coherent way of associating value with software and similar goods, it may be able to rescue marginalism from its difficulties, and so provide a theoretical justification for the repressive methods needed to enforce pricing of these goods, but for the reasons given above and in earlier texts I do not believe it can do so.

So far with the formal argument. Now for the more political side of things (completely separable from the discussion over value in one way, but for me important in terms of motivation and relevance of the discussion). Many Marxists have a tendency to look at capitalism as a system where 'tout se tient', both at an instant and through time. At most, it is split into a few large phases: for example, 'ascendant capitalism', 'monopoly capitalism', and 'capitalism in decline'. This goes along with a vision of a single determining contradiction - between working class and capitalist - driving the system forward until it reaches a point of explosion. Before that point nothing essentially changes: capitalism expands intensively by commodifying more and more of life, and extensively by absorbing more and more of the world, but the inescapable machinery of surplus value creation, commodity exchange, and reproduction of labour power is essentially reproducing the same system.

From this point of view there is a strong motivation to insist that the underlying mechanics of capitalism must be unchanging; all work for capital must produce value, which can be incorporated in capital, so the cycle can continue. Only financial breakdowns - crises, slumps, bubbles etc - are allowed for; these are either the exit point from the system as a whole or a temporary interruption before normal service is resumed. The thought of a commodification process which fails, because the underlying ratio of design to reproductive work has changed in such a way that commmodification becomes resistible, is unacceptable. The thought of people deliberately and self- consciously producing non-commodities in high-tech, collaborative production and in opposition to commodity production, is unacceptable. From my point of view, this produces an automatically defeatist attitude.

So, you write about education 'I've got news for you - the transition in this regard is over. Education has become a commodity in which value is materialized' when in fact recent years have seen a broad offensive attempting to use the new technologies to increase commodification of education stumble and fail, resisted successfully enough to lead to a current, unstable, stalemate - and even the start of attempts to reverse the process. And when people take advantage of the inability of capital to fully impose commodification in software to create free software, you see this as an illusory attempt to create an impossible island of communism within a coherent capitalist system, doomed either to defeat or triviality, rather than asking how it can be defended, used and expanded to intensify the contradictions of the system.

The ironic thing is that both points of view can be defended with reference to Marx. Quoting from the Grundrisse (notebook 7) (a quote I believe you also referenced in this discussion though I can't find it now):

'As soon as labour time in the direct form has ceased to be the great well-spring of wealth, labour time ceases and must cease to be its measure, and hence exchange value [must cease to be the measure] of use value... With that, production based on exchange value breaks down...'

I believe that this central insight of Marx was one that was absolutely constant for the rest of his life. The HOW of the breakdown was not; it was something he searched for wherever he could see a glimpse of it. Some of the possibilities are very mediated (the fall in the rate of profit in particular, which is just a rather complex expression of exactly this point).

A breakdown of value in the field of software - clearly not the end of capitalism as a whole, but equally clearly a direct expression of exactly this contradiction - is perfectly compatible with this insight of Marx. The alternative reading says that value is always the measure of labour time in capitalism; if it has ceased to be the measure of value, than capitalism must have ended - as capitalism is a system so the end of value must be all or nothing. You write: "The fragility of value, as a form of suspended disbelief, a powerful but increasingly absurd social construct, a story, becomes more and more visible. It must all come crashing down of course because the real value of the hoard depends on the value it commands and thus on the real value that is created in production and realised in circulation". That is, value ends with a systemic financial crash terminating capitalism as a whole, which suddenly brings the lack of reality of value home to people in the mass (as it did in 1929, I assume, though the lack of visibility of any alternatives apart from Stalinism played a large part in allowing the system to recover).

I think this second reading is harder to defend than mine, which allows for both possibilities; there may be such a final crash, but in the meantime the increasing importance of science relative to direct labour causes the value relationship to begin to fail at the leading edge of technology even before the systemic crash.

Your defence, as I understand it, depends on a interpretation of value as a kind of substance injected into commodities by labour, rather than (as you interpret it when discussing other topics, as in the quote above) a social relationship. So you write about bankrupt companies which have spent on R&D to the extent that they cannot profit sufficiently from sales of the final product to survive "since the R&D cost must not be ['does not have to be'?] repeated, the reproduction cost and thus the social value fell and the innovative capitalist couldn't recuperate the value he created". So we have a new kind of value, 'social value', which has fallen, while the metaphysical kind stays constant. I would say that this metaphysical kind never existed. Furthermore, you then end up justifying monopoly: "the expansion of legal protection of monopolies over new products [...] makes it possible to recuperate the value created in the prototype". Once again, there is the concept of value as something fixed ("created in the prototype") once work has been done. Marx assumed that it was an essential feature of capitalism that commodities sold at their value - profit is not to be explained by theft, but by extraction of surplus value. Only where monopolies are created do commodities sell above their value. You have instead created a new type of monopoly; a benign kind which allows commodities to sell at their real value as they would fail to do otherwise. From my point of view this kind of legally backed monopoly is an enemy - one which is being fought by large numbers of people already, and which we may defeat. From yours it becomes just capitalism as normal, and we cannot fight successfully until the last battle.

You write "the capitalist world is a complex totality so interwoven it can only change as a totality, not by nibbling at its flanks and creating islands of communism". I agree about the "islands of communism"; there can be no such thing except as part of a process; for example, the worker managed factories in Argentina will not survive in the longer term unless they are part of a process which replaces the entire system. But that is not a reason why they should not be supported. And a rejection of 'nibbling at flanks' sounds dangerously similar to all the arguments that have been put forward in the past to say that Marxists should not support trade unions, or should not get involved in strikes. Where there is movement, why sit and wait on the grounds that a more complete movement will inevitably emerge later? And where we are being attacked, why refuse to defend ourselves on the grounds that this is a complex totality about which we can do nothing until the final crash?

OK, that's all for now. This arguing in semi-public is a strange thing; positions get exaggerated and pushed to extremes they don't really merit. I hope at least my position is a little clearer now.

Best wishes

Graham Seaman

March 13, 2006

1. Any textbook on welfare economics should include this proof; it is usually presented in a very mathematical form, but basically all it says is: if you can copy something for free but charge anything for it, some people who would otherwise use it, won't - which is sub-optimal. But if you don't charge for it you can't cover the cost of creating it, and it won't be created - which is also sub-optimal. There is therefore no optimal price for such goods. Yew-Kwang Ng, Welfare Economics, Macmillan Press 1979 has a typical explanation.

2. http://www.primet.org/documents/Weiss%20-%20Borders%20in%20Cyberspace.htm is a recent summary of the PIRA study.

3. http://www.tax.org.uk/showarticle.pl?n=379&id=390 is a typical set of suggestions from a rather baffled accountant, with a lot of illustrations of the problem.

Home IP Archive Texts Discussion IP's French site Links